Opinion TRENDS in Microbiology Vol.12 No.2 February 2004

Bacterial programmed cell death systems as targets for antibiotics

Hanna Engelberg-Kulka, Boaz Sat, Myriam Reches, Shahar Amitai and Ronen Hazan

Department of Molecular Biology, The Hebrew University-Hadassah Medical School, Jerusalem 91120, Israel

Growing experimental evidence has revealed the exist- encodes a labile antitoxin that interferes with the lethal ence of programmed cell death (PCD) systems in bac- action of the toxin. Such genetic toxin–antitoxin (TA) teria. Among these is the mazEF system, which is a systems for bacterial PCD have been studied in several regulable suicide module located on the chromosome laboratories, including those of Yarmolinsky [6], Couturier of E. coli and of some other bacteria, including patho- [7] and Gerdes [8]. The TA systems have been found gens. Several well-known antibiotics have recently mainly in E. coli on low-copy number plasmids (and some been found to cause cell death in E. coli by indirectly other extrachromosomal elements) where they are respon- activating this built-in suicide module. These antibiotics sible for the post-segregational killing effect [8]: the death belong to two groups: (i) inhibitors of transcription of cells from which these extra-chromosomal elements and/or translation; and (ii) inhibitors of folic acid metab- have been lost. Such a genetic unit has been termed an olism resulting in thymine starvation. These data, ‘addiction module’: it causes the cells to be ‘addicted’ to the together with the recent elucidation of the crystal struc- short-lived antitoxin product because its de novo synthesis ture of mazEF-directed components, hold promise for a is essential for cell survival [6]. Therefore, these modules rational chemical design of a new class of antibiotics have been implicated as having a role in maintaining the that directly activate chromosomal suicide modules by stability of the extra-chromosomal elements in the host on interacting with their components. Because multi-drug which they are borne [6–9]. resistance among bacterial pathogens is becoming Over the past decade, pairs of genes homologous to some more widespread, the results obtained might be useful of these extrachromosomally borne addiction modules as a basis for producing alternative drugs. have been found on the chromosome of E. coli and of various other bacteria [10–17]. The first addiction module From the late 1940s, the use of antibiotics to treat carried on a bacterial chromosome that was described as infectious diseases in humans and animals has drastically regulable and responsible for PCD was the E. coli mazEF disturbed microbial ecology [1,2]. In response to the threat module [13]. The mazEF system was described as a suicide of extinction, microbes have found genetic and biochemical module on the basis of its properties [18,19]. This module routes to develop resistance to every anti-microbial agent consists of two adjacent genes, mazE and mazF, located used to date. Resistance to these agents might develop downstream from the relA gene in the rel operon [10]. The through mutations of target sites or the acquisition of mazEF suicide module [13,20] has all the basic properties novel biochemical functions [1,2]. Therefore, the develop- that are required to be an addiction module: (i) MazF is ment of bacterial resistance to antibiotics can be seen as toxic and MazE is antitoxic; (ii) MazF is long-lived, the price that we are paying for medical progress. Perhaps whereas MazE is a labile protein degraded in vivo by the the only way that we might be permitted to continue to rely ATP-dependent ClpPA serine protease; (iii) MazE and on antibiotic therapy is through persistent efforts to MazF interact; (iv) MazE and MazF are co-expressed; and discover new antibiotics, and to identify additional (v) mazEF is negatively auto-regulated at the level of biochemical pathways as novel targets for their action. transcription by the combined action of both MazE and Here we describe the process of programmed cell death MazF proteins. Moreover, the mazEF chromosomal suicide (PCD) in bacteria as a newly discovered target for the module has unique characteristics that were not pre- action of antibiotics. viously described for the extrachromosomal addiction modules. Initially, it was found that mazEF transcription 0 0 Programmed cell death in bacteria is inhibited by guanosine 3 ,5 -bispyrophosphate (ppGpp) PCD refers to any form of cell death mediated by an [13], which is synthesized by the RelA protein under intracellular death program, no matter what triggers it conditions of extreme amino acid starvation [21]. On the and whether or not it displays all the characteristics of basis of these combined properties and the requirement for apoptosis [3–5]. In bacteria, one of the most studied PCD the continuous expression of MazE to prevent cell death, a systems is mediated through a unique genetic unit that model was proposed for PCD under conditions of nutrient consists of a pair of genes that specify two components. The starvation (Figure 1) [13]. This model was supported by second gene encodes a stable toxin, and the first gene results that showed that mazEF-mediated cell death is triggered by the artificial overproduction of ppGpp [13,22]. Corresponding author: Hanna Engelberg-Kulka ([email protected]). The other unique characteristics will be described below. www.sciencedirect.com 0966-842X/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2003.12.008 Opinion TRENDS in Microbiology Vol.12 No.2 February 2004 67

Transcription

P2 mazE mazF Translation

ppGpp Toxin Antitoxin

Antibiotics

ClpPA protease Cell death

TRENDS in Microbiology

Figure 1. The ‘suicide module’ mazEF and how it can be activated. ppGpp (guanosine 30,50-bispyrophosphate) and antibiotics that inhibit transcription and/or translation triggers mazEF-dependent death. In each case, mazEF-expression is inhibited. Because MazE is a labile protein, its cellular concentration is reduced below the level required to antagonize the toxic MazF.

Antibiotics that inhibit transcription and/or translation MazE falls so low that there is not enough to prevent the trigger mazEF-mediated cell death stable MazF from killing the cells (Figure 1). The described model for the activation of the mazEF suicide module was subsequently broadened to include Antibiotics that cause thymine starvation trigger the conditions under which RNA and protein synthesis is mazEF system and lead to cell death inhibited [18,23]. First, the effects on the system of several Recently, an additional group of antibiotics that induce antibiotics, including rifampicin, chloramphenicol and mazEF- mediated cell death was found [26]. These spectinomycin, were tested. Rifampicin inhibits RNA antibiotics, trimethoprim and the sulfonamide drugs synthesis (and subsequently protein synthesis) by binding cause thymine starvation. The effect of thymine starvation to the b-subunit of RNA polymerase [2]. Chloramphenicol on bacterial PCD deserves special attention because of its and spectinomycin inhibit protein synthesis by binding to wide implications and also its historical significance. As the A site of the 50S ribosomal subunit [2,24,25]. In wild- early as 1954, Cohen and Barner [27] discovered that a type cells, it was found that even briefly inhibiting thymine auxotrophic mutant (thyA)ofE. coli undergoes transcription and/or translation by these antibiotics was cell death in response to thymine starvation. This sufficient to induce mazEF-mediated death. By contrast, phenomenon, called thymineless death (TLD), is also deleting mazEF from the bacterial chromosome prevented found in many other organisms ranging from prokaryotes cell death in the presence of these antibiotics [18,19].It to eukaryotes [28]. TLD is a unique effect as starvation of should be noted that mazEF-dependent bacteriocidity of bacteria to other growth factors is bacteriostatic but not chloramphenicol and spectinomycin was observed when bactericidal. Although TLD has been investigated inten- cells were growing on a minimal but not a rich medium sively for almost 50 years, its molecular mechanism [18]. Therefore, it appears that the bacteriocidity of this remains unclear [28]. However, thymidine metabolism antibiotic depends on the environmental conditions. The and the coupled folic acid metabolism are well known proposal that the inhibition of translation triggers (Figure 2). Thymine starvation in E. coli can be induced in mazEF-mediated death was further supported by the several ways that inhibit the synthesis of thymidylate, results of experiments in which the phd-doc addiction either directly or indirectly, by interfering with folate module, borne on the P1 plasmid prophage, was used. It metabolism [28]. The synthetic antibacterial agents, was found that Doc protein, the toxic product of this trimethoprim and the sulfonamides, inhibit specific steps module, is a general inhibitor of translation and a trigger in the biosynthesis of tetrahydrofolate (Figure 2). of mazEF-mediated death [23]. Therefore, inhibiting As was recently described in E. coli, thymine starvation protein synthesis either by antibiotics or the Doc protein triggers mazEF-mediated cell death [26] (Figure 3). In triggers E. coli mazEF-mediated death; when the trans- these experiments, thymine starvation was induced in an lation of mazEF mRNA is inhibited, the level of the labile E. coli thyA auxotroph by omitting thymine from the www.sciencedirect.com 68 Opinion TRENDS in Microbiology Vol.12 No.2 February 2004

include sulfonamide drugs and trimethoprim, which by p-amino benzoic acid (PABA) inhibiting folic acid metabolism induce thymine starvation Sulfonamides leading to cell death. Dihydropteroic acid (DHPA) On the basis of the results showing that TLD depends on mazEF in E. coli, a model for the induction of mazEF-mediated cell death by thymine starvation was Folic acid suggested [26]. Previous research has shown that thymine starvation provokes DNA damage that involves a unique breaking or twisting of the chromosome into a configur- Dihydrofolic acid (DHFA) Trimethoprim ation that defies all the repair and protective mechanisms of the cell [29]. Such damage to the DNA might be Tetraydrofolic acid (THFA) responsible for a preferential reduction in the activity of mazEF promoter P2 (Figure 3b), probably because of its unique structure that consists of two alternating palin- 5 10 N -N -Methylene-tetraydrofolic droms [20]. Therefore, thymine starvation might directly affect the activity of the P2 promoter. However, the dUMP dTMP dTDP dTTP DNA interference of transcription from P2 as a result of thymine Thymidylate synthase starvation might also be caused indirectly by the induction (thyA) of ppGpp synthesis, which was previously shown to inhibit TRENDS in Microbiology transcription from the mazEF P2 promoter [13].In addition, the involvement of some specific proteins that Figure 2. Sulfonamide drugs and trimethoprim cause thymine starvation by inter- fering with folic acid metabolism. Shown are the main steps in folic acid (blue) and can sense the DNA damage cannot be excluded. Taken as a thymine (green) metabolism in E. coli. The steps at which sulfonamides or tri- whole, when continuous expression of the labile antitoxin methoprim interrupt metabolism by inhibiting a specific reaction [28] are indicated MazE is prevented as a consequence of thymine starvation in red. cell death occurs. Although thymine starvation affects the growth medium [Figure 3a(iii)], and in an E. coli wild-type activity of the mazEF P2 promoter, it does not affect the strain by adding either the drug trimethoprim activity of the E. coli trpR promoter [26]. This does not [Figure 3a(i)] or a sulfonamide [Figure 3a(ii)]. As shown exclude the possibility that thymine starvation might (Figure 3), TLD depends on E. coli mazEF. Moreover, affect other promoters, including those of other genes that thymine starvation by trimethoprim drastically reduces might be involved in the mazEF-mediated death. transcription from the mazEF promoter P2 (Figure 3b), which is responsible for transcription and regulation of Programmed cell death in other bacteria E.coli mazEF [13,20]. Therefore, the list of antibiotics The discovery that thymine starvation in E.coli triggers a known to trigger mazEF-mediated PCD has grown to built-in death system provides a new insight into an old

(a) (b) (i) TM (ii) SMZ (iii) thyA (iv) Control

100 100 (%)

80 2 80

60 60

promoter P promoter of survivors) of 40 40

mazEF CFU (% CFU

20 20 Activity of of Activity

0 0 WT ∆EF ∆EF/ WT∆EF WT∆EF WT ∆EF ÐTM +TM pEF E. coli strains TRENDS in Microbiology

Figure 3. Thymine starvation by trimethoprim and a sulfonamide in E. coli induces mazEF-mediated cell death. (a) The effect of thymine starvation on viability was measured by colony forming units (CFUs) in E. coli K12 wild-type (WT), DmazEF(DEF), or in DmazEF harboring a plasmid carrying mazEF (DEF/pEF). Thymine starvation was induced in the following ways: (i) trimethoprim (TM) (0.1 mg/106cells/ml) for 2.5 hours; (ii) sulfamethoxazole (SMZ) (10 mg/106cells/ml) for 6.0 hours; (iii) a thyA mutant of the WT and the DmazEF derivative grown for 2.5 hours in a medium lacking thymine; (iv) a control of unstarved cells. (b) The effect on the activity of the mazEF promoter P2 of thymine starvation by the addition of trimethoprim. E. coli K12 WT cells were grown with trimethoprim (5 mg/108cells/ml) or without, and the activity of the mazEF pro- moter P2 was analyzed [25]. www.sciencedirect.com Opinion TRENDS in Microbiology Vol.12 No.2 February 2004 69 enigma that might have implications on PCD and toxins and antitoxins is appropriate for studying the anitibiotics in other bacteria, including pathogens. The mechanisms of their action. However, ectopic overexpres- mazEF system is not unique to E. coli; it has been found on sion drastically affects bacterial pathways and networks so the chromosome of some other bacteria [16,19,30]. Accord- that the conditions in the cell no longer reflect the actual ing to these data, bacteria carrying mazEF-like systems on physiological conditions under which TA systems mediate their chromosomes include the pathogenic E. coli 0157:H7, cell death. Our experiments (described above), in which we Neisseria meningitidis, Mycobacterium tuberculosis and found that the lethality of some well-known antibiotics is Leptospira interrogans. According to our current database mediated by the mazEF system, were done when the research, the list of bacteria that carry mazEF-like module was located on the bacterial chromosome as a systems is broadened to also include the pathogens single copy and in its natural context [18,26]. We suggest Staphylococcus aureus, Staphylococcus epidermidis, that experiments similar to ours are more appropriate Streptococcus mutans, Enterococcus faecium, Clostridium than ectopic overexpression for studying bacterial path- perfringens, Listeria monocytogenes, Shigella flexneri and ways and physiological networks. Among the antibiotics Bacillu santhracis. The effect of thymine starvation on cell that we used were rifampicin [18], a sulphonamide drug death has been described in a wide range of Gram-negative and trimethoprim [26], which are known to induce cell and Gram-positive bacteria [28]. Some of these, such as death [2]. Our results clearly show that this phenomenon Salmonella typhimurium, lack mazEF-like systems, is mazEF-dependent. suggesting that they might have analogous suicide Based on the existing results we hypothesize that modules that participate in PCD. These systems could inhibition of protein synthesis by MazF and/or other also be targets for new antibiotics. toxins, such as ChpBK and RelE, might be one of the initial steps in the PCD pathway of E. coli. Nevertheless, we The theory of programmed cell death versus the theory believe that MazF (and probably RelE and ChpBK) does of bacteriostasis not execute cell death, but rather mediates cell death. Recently, it has been suggested that rather than inducing Therefore, we have always called this phenomenon ‘cell PCD, chromosomal TA systems induce a state of reversible death mediated by the E. coli mazEF module’ [18,23,26].It bacteriostasis [31–34]. This view is based on two exper- is possible that the pathway to bacterial cell death might imental approaches that include mazEF and its homolo- be mediated through some unique mechanism in which gous module chpBIK [12], and another chromosomal MazF (or ChpBK or RelE) inhibits translation. The module, relBE, which is not homologous to mazEF [17]. specificity with which each toxin might cleave the The results of the first approach showed that ectopic mRNA could be the clue for understanding the particular overexpression of the toxins MazF or RelE inhibits pathway to cell death. The presence of such toxins might translation and cell growth, which can be resumed if the cause a selective synthesis of proteins encoded by mRNAs cognate antitoxin is expressed at a later time [31].In that are resistant to inactivation by MazF (or ChpBK, the second approach, the mode of action of RelE on the RelE or other yet uncharacterized toxins). We suspect that translation system was studied both in vitro [32] and these proteins might be part of an elaborate pathway to in vivo [33]. It was found that RelE has a ribonucleolytic cell death, making the chromosomal TA systems and PCD activity that cleaves mRNA codons between the second intriguing fields of research. and third nucleotide in the ribosomal A site. This cleavage is highly codon-specific with a preference for the UAG stop What could be the function of a programmed cell death codon. More recently, it was reported that the ectopic system in bacteria? overexpression of toxins MazF and ChpBK inhibits The presence of a PCD system on the bacterial chromo- translation by a similar ribosome-dependent ribonucleo- some raises an intriguing question; what could be the lytic activity, although each of the systems MazF, ChpBK function of such a system? For an individual bacterium, and RelE has its own codon specificity [34]. However, it PCD is counterproductive but becomes effective by the was recently reported that, in contrast to RelE, MazF is simultaneous action of a group of cells. A chromosomal TA a sequence-specific endoribonuclease that inhibits trans- system might be a device that is activated by stress lation in a manner that is independent of ribosomes; MazF conditions, so that a sub-population of cells within a preferentially cleaves mRNA between A and C at the ACA bacterial culture die to permit the survival of the sequence [35]. The reason for the discrepancy between the remaining cells. Therefore, a bacterial culture might act two reports is unclear [35]. as a multi-cellular organism that, when challenged by These results do not exclude our contention that the stressful conditions, sacrifices part of itself to keep the mazEF module, at least, mediates PCD. We base our whole alive [36]. We suggest three possible functions for conclusion on the following arguments. The experiments PCD systems in bacteria. First, response to nutritional showing that the ectopic overexpression of MazE reverses stress; the death of a sub-population would provide food for MazF lethality were carried out for only five hours the surviving cells. Such a role was previously suggested following MazF expression [31]. Using the same system, for mazEF [9], and was recently shown to participate we found that overexpressing MazE after a prolonged during sporulation of Bacillus subtilis as a result of the overexpression of MazF did not lead to the reversal of cell activation of a novel PCD system that differs from the TA death. Therefore, it appears that there is a point of no systems [36,37]. Second, a defense mechanism that return (Amitai and Engelberg-Kulka, unpublished). We prevents spreading of phage infection. This possible role agree that the methodology of ectopic overexpression of of PCD is supported by the results of recent experiments www.sciencedirect.com 70 Opinion TRENDS in Microbiology Vol.12 No.2 February 2004 showing that mazEF prevents spreading of phage P1 5 Hengartner, M.O. (2000) The biochemistry of apoptosis. Nature 407, (Hazan and Engelebrg-Kulka, unpublished). Third, PCD 770–776 6 Yarmolinsky, M.B. (1995) Programmed cell death in bacterial could act as a guardian of the bacterial chromosome; when populations. Science 267, 836–837 other systems fail, PCD systems might lead to the 7 Couturier, M. et al. (1998) Bacterial death by DNA gyrase poisoning. maintenance of genomic stability by causing the elimin- Trends Microbiol. 6, 269–275 ation of cells carrying genomic defects and mutations from 8 Jensen, R.B. and Gerdes, K. (1995) Programmed cell death in bacteria: the culture. These possible roles for PCD would provide an proteic plasmid stabilization systems. Mol. Microbiol. 17, 205–210 9 Engelberg-Kulka, H. and Glaser, G. (1999) Addiction modules and evolutionary advantage for the presence of various PCD programmed cell death and antideath in bacterial cultures. Annu. Rev. systems on the bacterial chromosome. Microbiol. 53, 43–70 10 Metzger, S. et al. (1988) The nucleotide sequence and characterization Conclusions and future directions of the relA gene of Escherichia coli. J. Biol. Chem. 263, 15699–15704 We have summarized experimental results showing that 11 Masuda, Y. et al. (1993) chpA and chpB, Escherichia coli chromosomal some well-known antibiotics cause cell death in E. coli by homologs of the pem locus responsible for stable maintenance of plasmid R100. J. Bacteriol. 175, 6850–6856 indirectly activating mazEF, the built-in suicide module. 12 Masuda, Y. and Ohtsubo, E. (1994) Mapping and disruption of the So far, there are two known groups of such antibiotics: chpB locus in Escherichia coli. J. Bacteriol. 176, 5861–5863 (i) inhibitors of transcription and/or translation [18], and 13 Aizenman, E. et al. (1996) An Escherichia coli chromosomal addiction 0 0 (ii) agents that cause thymine starvation and cell death by module regulated by guanosine [corrected] 3 ,5 -bispyrophosphate: a model for programmed bacterial cell death. Proc. Natl. Acad. Sci. inhibiting folic acid metabolism [26]. We suggest that these U. S. A. 93, 6059–6063 two groups of antibiotics can be used as models for the 14 Gerdes, K. et al. (1997) Antisense RNA-regulated programmed cell rational chemical design of a new generation of antibiotics. death. Annu. Rev. Genet. 31, 1–31 These new drugs will be designed to directly activate 15 Gotfredsen, M. and Gerdes, K. (1998) The Escherichia coli relBE genes bacterial suicide modules, therefore causing cell death. belong to a new toxin-antitoxin gene family. Mol. Microbiol. 29, 1065–1076 Designing these new drugs will be facilitated by the 16 Mittenhuber, G. (1999) Occurrence of mazEF-like antitoxin/toxin knowledge of the three dimensional structures of the systems in bacteria. J. Mol. Microbiol. Biotechnol. 1, 295–302 toxins and antitoxins directed by chromosomal suicidal 17 Christensen, S.K. et al. (2001) RelE, a global inhibitor of translation, is modules, such as mazEF. Loris and colleagues [38,39] have activated during nutritional stress. Proc. Natl. Acad. Sci. U. S. A. 98, already determined the crystal structure of the anti-toxin 14328–14333 18 Sat, B. et al. (2001) Programmed cell death in Escherichia coli: some MazE, and Burley and colleagues [40] have determined antibiotics can trigger mazEF lethality. J. Bacteriol. 183, 2041–2045 the crystal structure of the TA complex MazE–MazF. 19 Engelberg-Kulka, H. et al. (2001) Bacterial programmed cell death and MazE has a structured N-terminal domain and an antibiotics. ASM News 67, 617–624 unstructured C-terminal domain. Interactions between 20 Marianovsky, I. et al. (2001) The regulation of the Escherichia coli MazE and MazF are primarily mediated by the C-termi- mazEF promoter involves an unusual alternating palindrome. J. Biol. Chem. 276, 5975–5984 nus domain of MazE. MazE and MazF form a linear 21 Cashel, M.D. et al. (1996) The stringent response. In Escherichia coli heterohexamer composed of alternating toxin and anti- and Salmonella: Cellular and Molecular Biology (Neidhardt, F.C. et al., toxin homodimers (MazF2–MazE2–MazF2).The knowledge eds), pp. 1458–1496, American Society for Microbiology of these crystal structure is a crucial turning point. It might 22 Engelberg-Kulka, H. et al. (1998) rexB of bacteriophage lambda is an contribute to a deeper understanding of the mechanism of anti-cell death gene. Proc. Natl. Acad. Sci. U. S. A. 95, 15481–15486 23 Hazan, R. et al. (2001) Postsegregational killing mediated by the P1 action of the MazF toxin (and probably some other phage addiction module phd-doc requires the Escherichia coli chromosomally directed toxins). It might clear the way programmed cell death system mazEF. J. Bacteriol. 183, 2046–2050 for the structurally based chemical design of antibiotics 24 Spahn, C.M. and Prescott, C.D. (1996) Throwing a spanner in the that will directly interact with specific components of works: antibiotics and the translation apparatus. J. Mol. Med. 74, 423–439 chromosomal suicide modules. Because multi-drug resist- 25 Schlunzen, F. et al. (2001) Structural basis for the interaction of ance among bacterial pathogens is becoming more wide- antibiotics with the peptidyl transferase centre in eubacteria. Nature spread, the results obtained could be useful as a basis for 413, 814–821 alternative drugs, and might help to solve what has 26 Sat, B. et al. (2003) The Escherichia coli mazEF suicide module become a major public health problem. mediates thymineless death. J. Bacteriol. 185, 1803–1807 27 Cohen, S.S. and Barner, H.D. (1954) Studies on the unbalanced growth in Escherichia coli. Proc. Natl. Acad. Sci. U. S. A. 40, 885–893 Acknowledgements 28 Ahmad, S.I. et al. (1998) Thymine metabolism and thymineless death We thank F.R. Warshaw-Dadon for critical reading of the manuscript. The in prokaryotes and eukaryotes. Annu. Rev. Microbiol. 52, 591–625 research described here was supported by grant No. 215/99–2 from the 29 Nakayama, K. et al. (1994) Thymine starvation-induced structural Israel Science Foundation administrated by the Israel Academy of Science changes in Escherichia coli DNA. Detection by pulsed field gel and Humanities. electrophoresis and evidence for involvement of homologous recombi- nation. J. Mol. Biol. 243, 611–620 References 30 Picardeau, M. et al. (2001) Killing effect and antitoxic activity of the 1 Schlessinger, D. and Eisenstein, B. (1999) Biological bases for Leptospira interrogans toxin-antitoxin system in Escherichia coli. antibacterial action. In Mecanisms of Microbial Diseases (Schaechter, J. Bacteriol. 183, 6494–6497 M., ed.), pp. 52–60, Williams and Wilkins 31 Pedersen, K. et al. (2002) Rapid induction and reversal of bacteriostatic 2 Davies, J. and Webb, V. (1998) In Emerging Infections (Krause, R.M., conditions by controlled expression of toxins and antitoxins. Mol. ed.), pp. 239–273, Academic Press Microbiol. 45, 501–510 3 Jacobson, M.D. et al. (1997) Programmed cell death in animal 32 Pedersen, K. et al. (2003) The bacterial toxin RelE displays codon- development. Cell 88, 347–354 specific cleavage of mRNAs in the ribosomal A site. Cell 112, 131–142 4 Raff, M. (1998) Cell suicide for beginners. Nature 396, 119–122 33 Christensen, S.K. and Gerdes, K. (2003) RelE toxins from Bacteria and www.sciencedirect.com Opinion TRENDS in Microbiology Vol.12 No.2 February 2004 71

Archea cleaves mRNAs on translating ribosomes, which are rescued by 37 Gonzallez-Pastor, E. et al. (2003) Cannibalism of sporulating bacteria. tmRNA. Mol. Microbiol. 48, 1389–1400 Science 301, 510–513 34 Christensen, S.K. et al. (2003) Toxin-antitoxin loci as stress-response 38 Lah, J. et al. (2003) Recognition of the intrinsically flexible addiction elements: ChpAK/MazF and ChpBK cleave trasnlated mRNAs and are antidote MazE by a dromedary single domain antibody fragment. counteracted by tmRNA. J. Mol. Biol. 332, 809–819 Structure, thermodynamics of binding, stability, and influence on 35 Zhang, Y. et al. (2003) MazF cleaves cellular mRNA specifically at interactions with DNA. J. Biol. Chem. 278, 14101–14111 ACA to block protein synthesis in Escherichia coli. Mol. Cell 12, 39 Loris, R. et al. (2003) Crystal structure of the intrinsically flexible 913–923 addiction antidote MazE. J. Biol. Chem. 278, 28252–28257 36 Engelberg-Kulka, H. and Hazan, R. (2003) Perspective. Cannibals defy 40 Kamada, K. et al. (2003) Crystal structure of the MazE/MazF complex: starvation and avoid sporulation. Science 301, 467–468 molecular bases of antidote-toxin recognition. Mol. Cell 11, 875–884

Endeavour

the quarterly magazine for the history and philosophy of science

You can access Endeavour online either through your BioMedNet Reviews subscription or via ScienceDirect, where you’ll find a collection of beautifully illustrated articles on the history of science, book reviews and editorial comment.

featuring

‘Dr. Steinach coming to make old young!’: sex glands, vasectomy and the quest for rejuvenation in the roaring twenties by C. Sengoopta Cheese mites and other delicacies: the introduction of test objects into microscopy by J. Schickore An herbal El Dorado: the quest for botanical wealth in the Spanish Empire by P. De Vos Zones of inhibition: interactions between art and science by K. Davies Global science: the eruption of Krakatau by M. Do¨ erries Two pills, two paths: a tale of gender bias by M. Potts Joseph Banks: Pacific pictures by P. Fara and coming soon Mr Blandowski misses out: discovery and loss of fish species in 19th century Australia by P. Humphries The traffic and display of body parts in the early 19th century by S. Alberti and S. Chaplin Exhibiting monstrosity: Chang and Eng, the ’original’ Siamese twins by S. Mitchell The ancient mariner and the transit of Venus at Hudson Bay by R. Griffin-Short ‘I got rhythm’: Gershwin and birth control in the 1930s by P. Viterbo The market for Julia Pastrana by J. Browne and S. Messenger Race mixing and science in the United States by P. Farber Continental drift under the Third Reich by E. Buffetaut The first president of the Royal Society by P. Fara and much, much more . . . Locate Endeavour in the BioMedNet Reviews collection (http://reviews.bmn.com) or on ScienceDirect (http://www.sciencedirect.com) www.sciencedirect.com