Kenn Gerdes

Prokaryotic toxin–antitoxin stress response loci

Although toxin–antitoxin gene cassettes were first found in plasmids, recent database mining has shown that these loci are abundant in free-living prokaryotes, including many pathogenic bacteria. For example, Mycobacterium tuberculosis has 38 chromosomal toxin–antitoxin loci, including 3 relBE and 9 mazEF loci. RelE and MazF are toxins that cleave mRNA in response to nutritional stress. RelE cleaves mRNAs that are positioned at the ribosomal A-site, between the second and third nucleotides of the A-site codon. It has been proposed that toxin–antitoxin loci function in bacterial programmed cell death, but evidence now indicates that these loci provide a control mechanism that helps free-living prokaryotes cope with nutritional stress.

The Bacterial Toxin RelE Displays Codon-Specific Cleavage of mRNAs in the Ribosomal A Site

The Escherichia coli relBE operon encodes a toxin-antitoxin pair, RelE-RelB. RelB can reverse inhibition of protein synthesis by RelE in vivo. We have found that although RelE does not degrade free RNA, it cleaves mRNA in the ribosomal A site with high codon specificity. Among stop codons UAG is cleaved with fast, UAA intermediate and UGA slow rate, while UCG and CAG are cleaved most rapidly among sense codons. We suggest that inhibition of protein synthesis by RelE is reversed with the help of tmRNA, and that RelE plays a regulatory role in bacteria during adaptation to poor growth conditions.

Bacterial persistence by RNA endonucleases

Bacteria form persisters, individual cells that are highly tolerant to different types of antibiotics. Persister cells are genetically identical to nontolerant kin but have entered a dormant state in which they are recalcitrant to the killing activity of the antibiotics. The molecular mechanisms underlying bacterial persistence are unknown. Here, we show that the ubiquitous Lon (Long Form Filament) protease and mRNA endonucleases (mRNases) encoded by toxin-antitoxin (TA) loci are required for persistence in Escherichia coli. Successive deletion of the 10 mRNase-encoding TA loci of E. coli progressively reduced the level of persisters, showing that persistence is a phenotype common to TA loci. In all cases tested, the antitoxins, which control the activities of the mRNases, are Lon substrates. Consistently, cells lacking lon generated a highly reduced level of persisters. Moreover, Lon overproduction dramatically increased the levels of persisters in wild-type cells but not in cells lacking the 10 mRNases. These results support a simple model according to which mRNases encoded by TA loci are activated in a small fraction of growing cells by Lon-mediated degradation of the antitoxins. Activation of the mRNases, in turn, inhibits global cellular translation, and thereby induces dormancy and persistence. Many pathogenic bacteria known to enter dormant states have a plethora of TA genes. Therefore, in the future, the discoveries described here may lead to a mechanistic understanding of the persistence phenomenon in pathogenic bacteria.

RelE, a global inhibitor of translation, is activated during nutritional stress

The stringent response is defined as the physiological changes elicited by amino acid starvation. Many of these changes depend on the regulatory nucleotide ppGpp (guanosine tetraphosphate) synthesized by RelA (ppGpp synthetase I), the relA-encoded protein. The second rel locus of Escherichia coli is called relBE and encodes RelE cytotoxin and RelB antitoxin. RelB counteracts the toxic effect of RelE. In addition, RelB is an autorepressor of relBE transcription. Here we reveal a ppGpp-independent mechanism that reduces the level of translation during amino acid starvation. Artificial overexpression of RelE severely inhibited translation. During amino acid starvation, the presence of relBE caused a significant reduction in the poststarvation level of translation. Concomitantly, relBE transcription was rapidly and strongly induced. Induction of transcription occurred independently of relA and spoT (encoding ppGpp synthetase II), but instead depended on Lon protease. Consistently, Lon was required for degradation of RelB. Replacement of the relBE promoter with a LacI-regulated promoter indicated that strong and ongoing transcription of relBE is required to maintain a proper RelB:RelE ratio during starvation. Thus relBE may be regarded as a previously uncharacterized type of stress-response element that reduces the global level of translation during nutritional stress.

(p)ppGpp Controls Bacterial Persistence by Stochastic Induction of Toxin-Antitoxin Activity

Persistence refers to the phenomenon in which isogenic populations of antibiotic-sensitive bacteria produce rare cells that transiently become multidrug tolerant. Whether slow growth in a rare subset of cells underlies the persistence phenotype has not be examined in wild-type bacteria. Here, we show that an exponentially growing population of wild-type Escherichia coli cells produces rare cells that stochastically switch into slow growth, that the slow-growing cells are multidrug tolerant, and that they are able to resuscitate. The persistence phenotype depends hierarchically on the signaling nucleotide (p)ppGpp, Lon protease, inorganic polyphosphate, and toxin-antitoxins. We show that the level of (p)ppGpp varies stochastically in a population of exponentially growing cells and that the high (p)ppGpp level in rare cells induces slow growth and persistence. (p)ppGpp triggers slow growth by activating toxin-antitoxin loci through a regulatory cascade depending on inorganic polyphosphate and Lon protease.

Rapid induction and reversal of a bacteriostatic condition by controlled expression of toxins and antitoxins

RelE and ChpAK (MazF) toxins of Escherichia coli have previously been described as proteins that mediate efficient cell killing. We show here that induction of relE or chpAK transcription does not confer cell killing but, instead, induces a static condition in which the cells are still viable but unable to proliferate. Later induction of transcription of the antitoxin genes relB or chpAI fully reversed the static condition induced by RelE and ChpAK respectively. We also provide a mechanistic explanation for these findings. Thus, induction of relE transcription severely inhibited translation, whereas induction of chpAK transcription inhibited both translation and replication. Hence, most likely, lack of colony formation is due to inhibition of translation in the case of relE and inhibition of translation and/or replication in the case of chpAK. Consistent with this proposal, later induction of transcription of the cognate antitoxin genes simultaneously reversed cell stasis and the inhibitory effects of RelE and ChpAK on macromolecular syntheses. These results preclude that RelE and ChpAK mediate cell killing during the conditions used here. In vivo and in vitro analyses of a mutant RelE protein supported that inhibition of colony formation was due to inhibition of translation.

Molecular Mechanisms Underlying Bacterial Persisters

All bacteria form persisters, cells that are multidrug tolerant and therefore able to survive antibiotic treatment. Due to the low frequencies of persisters in growing bacterial cultures and the complex underlying molecular mechanisms, the phenomenon has been challenging to study. However, recent technological advances in microfluidics and reporter genes have improved this scenario. Here, we summarize recent progress in the field, revealing the ubiquitous bacterial stress alarmone ppGpp as an emerging central regulator of multidrug tolerance and persistence, both in stochastically and environmentally induced persistence. In several different organisms, toxin-antitoxin modules function as effectors of ppGpp-induced persistence.

Programmed cell death in bacteria: proteic plasmid stabilization systems

Bacterial plasmids are stabilized by a number of different mechanisms. Here we describe the molecular aspects of a group of plasmid‐encoded gene systems called the proteic killer gene systems. These systems mediate plasmid maintenance by selectively killing plasmid‐free cells (post‐segregational killing or plasmid addiction). The group includes ccd of F, parD/pem of R1/R100, parDE of RP4/RK2, and phd/doc of P1. All of these systems encode a stable toxin and an unstable antidote. The antidotes prevent the lethal action of their cognate toxins by forming tight complexes with them. The antidotes are degraded by cellular proteases. Thus, the different decay rates of the toxins and antidotes seem to be the molecular basis of toxin activation in plasmid‐free cells. The operons encoding the toxins and antidotes are autoregulated at the level of transcription either by a complex formed by the toxins and the cognate antidotes or by the antidote alone. The cellular targets of the killer proteins have been determined to be DNA gyrase in the case of ccd of F and DnaB in the case of parD of R1. Surprisingly, the Escherichia coli chromosome encodes at least two of these peculiar gene systems.

The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex

MreB proteins of Escherichia coli, Bacillus subtilis and Caulobacter crescentus form actin‐like cables lying beneath the cell surface. The cables are required to guide longitudinal cell wall synthesis and their absence leads to merodiploid spherical and inflated cells prone to cell lysis. In B. subtilis and C. crescentus, the mreB gene is essential. However, in E. coli, mreB was inferred not to be essential. Using a tight, conditional gene depletion system, we systematically investigated whether the E. coli mreBCD‐encoded components were essential. We found that cells depleted of mreBCD became spherical, enlarged and finally lysed. Depletion of each mre gene separately conferred similar gross changes in cell morphology and viability. Thus, the three proteins encoded by mreBCD are all essential and function in the same morphogenetic pathway. Interestingly, the presence of a multicopy plasmid carrying the ftsQAZ genes suppressed the lethality of deletions in the mre operon. Using GFP and cell fractionation methods, we showed that the MreC and MreD proteins were associated with the cell membrane. Using a bacterial two‐hybrid system, we found that MreC interacted with both MreB and MreD. In contrast, MreB and MreD did not interact in this assay. Thus, we conclude that the E. coli MreBCD form an essential membrane‐bound complex. Curiously, MreB did not form cables in cell depleted for MreC, MreD or RodA, indicating a mutual interdependency between MreB filament morphology and cell shape. Based on these and other observations we propose a model in which the membrane‐associated MreBCD complex directs longitudinal cell wall synthesis in a process essential to maintain cell morphology.

RelE toxins from Bacteria and Archaea cleave mRNAs on translating ribosomes, which are rescued by tmRNA

RelE of Escherichia coli is a global inhibitor of translation that is activated by nutritional stress. Activation of RelE depends on Lon‐mediated degradation of RelB, the antagonist that neutralizes RelE. In vitro, RelE cleaves synthetic mRNAs positioned at the ribosomal A‐site. We show here that in vivo overexpression of RelE confers cleavage of mRNA and tmRNA in their coding regions. RelE‐mediated cleavage depended on translation of the RNAs and occurred at both sense and stop codons. RelE cleavage of mRNA and tmRNA was also induced by amino acid starvation. An ssrA deletion strain was hypersensitive to RelE, whereas overproduction of tmRNA counteracted RelE toxicity. After neutralization of RelE by RelB, rapid recovery of translation required tmRNA, indicating that tmRNA alleviated RelE toxicity by rescuing ribosomes stalled on damaged mRNAs. RelE proteins from Gram‐positive Bacteria and Archaea cleaved tmRNA with a pattern similar to that of E. coli RelE, suggesting that the function and target of RelE may be conserved across the prokaryotic domains.