Growth-Dependent DNA Breakage and Cell Death in a Gyrase Mutant of Salmonella

Corresponding author: Nara Figueroa-Bossi, Centre de Génétique Moléculaire, CNRS, 91198 Gif-sur-Yvette Cedex, France., figueroa{at}cgm.cnrs-gif.fr (E-mail)

Communicating editor: G. R. SMITH

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A class of gyrase mutants of Salmonella enterica mimics the properties of bacteria exposed to quinolones. These mutants suffer spontaneous DNA breakage during normal growth and depend on recombinational repair for viability. Unlike quinolone-treated bacteria, however, they do not show accumulation of cleavable gyrase-DNA complexes. In recA or recB mutant backgrounds, the temperature-sensitive (ts) allele gyrA208 causes rapid cell death at 43°. Here, we isolated "suppressor-of-death" mutations, that is, secondary changes that allow a gyrA208 recB double mutant to survive a prolonged exposure to 43° and subsequently to form colonies at 28°. In most isolates, the secondary change was itself a ts mutation. Three ts alleles were mapped in genes coding for amino acyl tRNA synthetases (alaS, glnS, and lysS). Allele alaS216 completely abolished DNA breakage in a gyrA208 recA double mutant. Likewise, treating this mutant with chloramphenicol prevented death and DNA damage at 43°. Additional suppressors of gyrA208 lethality include rpoB mutations and, surprisingly, icd mutations inactivating isocitrate dehydrogenase. We postulate that the primary effect of the gyrase alteration is to hamper replication fork movement. Inhibiting DNA replication under conditions of continuing macromolecular synthesis ("unbalanced growth") activates a mechanism that causes DNA breakage and cell death, reminiscent of "thymineless" lethality.

TYPE II DNA topoisomerases are enzymes that can change the linking number of DNA by a mechanism involving transient double-strand cleavage and passage of a separate segment of the DNA molecule through the cut site (reviewed in BERGER 1998 ). Representative of this class, bacterial DNA gyrase is a tetrameric protein made up of two pairs of subunits, the products of gyrA and gyrB genes (reviewed in MENZEL and GELLERT 1994 ). Upon cleaving the DNA, the GyrA subunits become covalently attached to the 5' termini of cleaved strands. Passage of a DNA segment through the protein-held gate triggers the sealing of the break, regenerating the enzyme for a new cycle (KAMPRANIS et al. 1999 ). Unlike other type II DNA topoisomerases, gyrase has the ability to negatively supercoil DNA by a processive ATP-driven mechanism (WILLIAMS et al. 2001 ). In vivo, this activity is largely responsible for maintaining bacterial DNA within a physiological range of negative superhelical density (ZECHIEDRICH et al. 2000 ). In addition, gyrase intervenes in the removal of positive supercoils generated by gene transcription (LIU and WANG 1987; EL HANAFI and BOSSI 2000 ) and DNA replication (LEVINE et al. 1998 ; KHODURSKY et al. 2000 ). Not surprisingly, due to this multiplicity of functions, mutants in either gyrase subunit are typically highly pleiotropic, showing alterations in DNA supercoiling, transcriptional patterns, and DNA replication (MENZEL and GELLERT 1994 ; LEVINE et al. 1998 ).

Bacterial type II topoisomerases are the targets of a family of antibacterial agents of clinical relevance: the quinolone antibiotics and their derivatives (reviewed in DRLICA and ZHAO 1997 ; HEDDLE et al. 2000 ). Quinolones bind to the enzyme/DNA complex blocking the topoisomerization reaction at the level of the covalent intermediate. Treatment of the ternary complex with a protein denaturant releases cleaved DNA with topoisomerase subunits bound at each end. Hence, the complex is often referred to as the "cleavable" complex (LIU 1989 ). Although it appears that the cytotoxicity of quinolone drugs results from the induction of double-strand breaks in DNA, the actual mechanism that generates the breaks in vivo remains elusive. The cleavable complex is not thought to collapse spontaneously (i.e., in the absence of denaturing treatment) and is fully reversible. Thus, a prevalent view is that some active mechanism must be responsible for denaturing the complex in vivo.

Several lines of evidence point to the involvement of DNA replication in this event. An in vitro study showed that the collision between a moving replication fork and a stalled topoisomerase can convert the complex to a nonreversible form (HIASA et al. 1996 ). However, the collision did not release a double-strand break, leading to the proposal that the topoisomerase poisoning involves two steps: the conversion of the ternary complex to a nonreversible form and its denaturation by an unknown mechanism trying to repair the lesion (HIASA et al. 1996 ). A related aspect concerns the ability of transcriptional and translational inhibitors to relieve the cytotoxicity of topoisomerase poisons (DEITZ et al. 1966 ; CHEN et al. 1996 ). Chloramphenicol, an inhibitor of protein synthesis, was shown to block DNA breakage and cell death in quinolone-treated bacteria, leading to the suggestion that a short-lived repair factor participates in the removal of quinolone-gyrase complexes from DNA (CHEN et al. 1996 ). A further indication of the link between gene expression and cytotoxicity is provided by the finding that mutations affecting RNA polymerase or components of the translational apparatus in Salmonella confer increased resistance to nalidixic acid, a first-generation quinolone (BLANC-POTARD and BOSSI 1994 ; BLANC-POTARD et al. 1995 ).

In a previous study, we described a peculiar class of DNA gyrase mutants of Salmonella that mimics the response of wild-type bacteria to quinolone treatment (GARI et al. 1996 ). These mutants chronically derepress the SOS system and are highly dependent on recombinational repair for viability. The phenotype is conditional in some of the mutants. Bacteria carrying the gyrA208 allele can grow at 37° provided that they are recombination proficient, but die at 43°. In a recA^- background, death is accompanied by extensive RecBCD-dependent degradation of chromosomal DNA, suggesting the occurrence of double-strand breaks. We found no evidence for a gratuitous accumulation of cleavable gyrase-DNA complexes in the gyrA208 mutant (GARI et al. 1996 ). To gather insight on the mechanism responsible for lethality, in this study we isolated and characterized mutations that prevent death of the gyrA208 mutant at restrictive temperature. We found such suppressor mutations in genes encoding aminoacyl tRNA synthetases and in the gene for isocitrate dehydrogenase (icd). A previously isolated RNA polymerase (rpoB) mutation was also found to effectively suppress gyrA208 lethality. These mutations have one property in common: they all slow or completely stop growth under suppressive conditions. The lack of DNA degradation in the gyrA208 recA mutant under these conditions suggests that growth is also involved in the mechanism leading to DNA breakage.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Bacteria and phage:
All strains used in this study are derivatives of Salmonella enterica serovar Typhimurium strain LT2. These strains are listed in Table 1 except for those carrying suppressor mutations that remain uncharacterized and the strains of the Mud-P22 mapping kit (BENSON and GOLDMAN 1992 ). Transductional crosses were carried out using the high-frequency generalized transducing mutant of phage P22 (HT 105/1 int-201; SCHMIEGER 1972 ) as described by MALOY et al. 1996 .

Plasmids:
Plasmid pTAC875N carries the Escherichia coli alaS gene under the control of the Tac promoter (VINELLA et al. 1992 ). Plasmid pXLysSK2 is a pBluescript derivative carrying a 3.5-kb insert that spans the E. coli lysS gene (LEVEQUE et al. 1990 ). Plasmid pYY105 is a pBR322 derivative carrying the E. coli glnS gene (PLUMBRIDGE 1987 ). The above three plasmids were kindly donated to us by P. Bouloc, P. Plateau, and J. Plumbridge, respectively.

Media and growth conditions:
Bacteria were cultured at 28° or 37° in liquid media or in media solidified with 1.5% (w/v) Difco agar. Difco nutrient broth (NB; 0.8% w/v) supplemented with NaCl (0.5% w/v) or Luria-Bertani (LB) broth (MALOY et al. 1996 ) were used as complex media. Vogel-Bonner "E" medium (MALOY et al. 1996 ) supplemented with 0.2% (w/v) glucose was used as minimal medium. In some experiments, E-glucose medium was supplemented with Difco casaminoacids (0.5% w/v). Antibiotics were used at the following final concentrations: tetracycline hydrochloride, 25 µg/ml; chloramphenicol, 20 µg/ml; kanamycin monosulfate, 50 µg/ml; sodium ampicillin, 50 µg/ml. Loss of tetracycline (Tc) resistance was scored by plating cells in Tc-sensitive selection medium (MALOY et al. 1996 ). Liquid cultures were grown in gyratory shakers (New Brunswick, Edison, NJ, or Infors HT, Bottmingen, Switzerland) and growth monitored measuring the optical density at 450 nm (OD₄₅₀) with a Milton-Roy (Rochester, NY) Spectronic 301 spectrophotometer.

Cytological methods:
For visualization of nucleoid morphology, bacteria were stained with 4',6-diamino-2-phenyl-indole (DAPI) and examined through a fluorescence microscope (Reichert-Jung) as previously described (BLANC-POTARD and BOSSI 1994 ).

Isolation and characterization of suppressors of gyrA208 lethality:
Cultures of strain MA2383 (gyrA208 [recB] din-1001::MudJ) were grown at 28° in LB to stationary phase. A sample of 0.1 ml from each culture (10⁸ bacteria) was spread on LB plates, incubated for 24 hr at 43°, and then shifted to 28° and incubated for an additional 48 hr. Survivors occurred at a frequency of 50–100 cfu/plate. Independent isolates were picked and initially screened for the ability to grow at 37° and 43°. Rare revertants of the gyrA208 mutation were readily identified from their ability to form large white colonies at 43° and were discarded. The remaining isolates could be separated into two classes: those still incapable of growing at 37° (class A, representing the majority of isolates) and those that grew at this temperature (class B). Next, the suppressor mutations were transferred to a wild-type genetic background. This was achieved in three steps: (i) rendering the strains RecB⁺ (by transduction, selecting Arg⁺); (ii) isolating Tn10- dTc insertions near each of the suppressor loci: a phage P22 lysate made on a pool of random Tn10dTc transposition mutants was used to transduce the suppressor-harboring strains to Tc resistance scoring for the loss of suppression (in the presence of din1001::MudJ, unsuppressed gyrA208 causes bacteria to form wrinkled blue colonies on X-gal indicator plates at 37°); and (iii) backcrossing the insertions so obtained into the suppressor background and transductionally transferring the suppressor mutations into a wild-type strain selecting Tc resistance. The tentative map location of the suppressor gene or of the nearby Tn10dTc element was determined using the Mud-P22 lysogen collection of BENSON and GOLDMAN 1992 . When possible, the map position was further refined by measuring transductional linkages to markers in the region. We were able to carry out all of the above steps with eight independent isolates: six from class A (s-41, s-43, s-54, s-56, s-57, s-59) and two from class B (s-45, s-46).

• Class A: All mutations from this class were found to be temperature-sensitive alleles. In the case of alleles s-54 and s-59 (27.5–30.5 map unit interval) and s-56 (8.0–12.5 map unit interval) lack of linkage to any marker tested precluded more refined mapping. Suppressor s-41 was highly linked to supK (prfB) at 66 map units. Complementation analysis with plasmid pXLysSK2 (LEVEQUE et al. 1990 ) confirmed that the s-41 mutation affects the lysS gene coding for lysine-tRNA synthetase. Suppressors s-43 and s-57 were transductionally linked to several markers in the 16–17 and 62–63 map unit intervals. Gene orders and linkages are summarized in Fig 1. Examination of the corresponding regions in the E. coli map suggested that mutations s-43 and s-57 affected glnS, the gene for glutamine tRNA synthetase, and the alaS gene for alanine tRNA synthetase, respectively. This was confirmed upon finding the ts phenotype associated with s-43 and s-57 to be complemented by plasmids pYY105 (PLUMBRIDGE 1987 ) and pTAC875N (VINELLA et al. 1992 ), respectively.

  
  

  View larger version (17K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Position of alaS and glnS mutations on the Salmonella genetic map. (A) Transductional linkages of alaS216 to nearby markers. Linkages were determined in six two-point crosses involving the strains TT521, MA1154, MA2856, MA2862, MA3071, MA3062, MA4093, and MA4217 (see Table 1). Gene order was determined in four three-point crosses involving strains TT521, MA2856, MA2862, MA3071, MA4093, MA4217, and MA4310 (see Table 1). At least 100 transductants were analyzed in all cases. (B) Transductional linkages of glnS1521 to nearby markers. Linkages were determined in four two-point crosses involving the strains TR2679, DB7155, TT10258, MA1154, MA2724, and MA2745 (see Table 1). Gene order was determined in three three-point crosses involving strains TR2679, DB7155, TT10258, and MA2745 (see Table 1). At least 100 transductants were analyzed in all cases.

• Class B: Strains carrying mutations s-45 or s-46 (in an otherwise wild-type background) grew (slowly) at all temperatures. The specific growth requirements of these strains (the requirement for proline or glutamate and the failure to grow on acetate as sole carbon source) suggested that they result from icd gene alterations. Genetic linkage to a purB marker, as well as the specific reversion pattern of two mutants (see RESULTS), confirmed this identification.

DNA labeling:
Chromosomal DNA was uniformly labeled by growing cells in the presence of [³H]thymine as previously described (GARI et al. 1996 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Mutation gyrA208 causes cell filamentation and death:
Our previous work showed that some phenotypic traits associated with a particular class of gyrase mutations in Salmonella do not result from DNA supercoiling defects. This was most evident with the highly lethal gyrA208 allele, which affects negative supercoiling levels only marginally (GARI et al. 1996 ). When incubated at the restrictive temperature (43°), gyrA208 mutant cells continue to elongate without dividing while their nucleoids become increasingly decondensed and diffuse (Fig 2). The picture is unlike that observed with common gyrase mutants, which show defective chromosome partitioning but remain proficient at septum formation and, as a result, segregate small anucleate cells (HUSSAIN et al. 1987 ; KATO et al. 1989 ; BLANC-POTARD and BOSSI 1994 ). The near absence of anucleate cells in the gyrA208 mutant suggests that an active mechanism blocks cell division in this strain. Derepression of the SOS regulon, a gyrA208 phenotype (GARI et al. 1996 ), may be mainly responsible for the inhibition. However, an additional mechanism must also be involved since some heterogeneous filamentation is still observed in the presence of recB mutations, which prevent SOS induction (data not shown). Mutations in recA or recB genes exacerbate the lethality of gyrA208, causing bacteria to no longer grow at 37°. This is not due to inability to induce SOS, since lexA mutations that gratuitously induce the SOS regulon (CLERCH et al. 1996 ) fail to suppress the 37° lethality (Table 2). In contrast, growth of a gyrA208 recB double mutant at 37° is restored by mutations activating alternative recombination pathways—i.e., by sbcB and sbcE mutations (GARI et al. 1996 ; FIGUEROA-BOSSI et al. 1997 )—suggesting that recombinational repair is needed for the gyrA208 mutant to grow at a semipermissive temperature (Table 2).

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Nucleoid morphology of the gyrA208 mutant incubated at permissive and restrictive temperatures. Overnight cultures grown in NB at 28° were diluted 1:100 in NB and incubated for 2 hr at 28° (A) or at 43° (B). Cells were stained with DAPI and examined under a fluorescence microscope (objective x100).

Strains carrying gyrA208 in combination with either recA or recB mutations rapidly die when incubated at 43° (Fig 3). The rate of killing is more dramatic if bacteria are incubated on solid medium (conditions under which even a recombination-proficient gyrA208 mutant suffers a marked loss of viability following overnight incubation; data not shown). This suggests that one might gather insight into the cytocidal mechanism from the analysis of mutations that suppressed death at 43°.

View larger version (19K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Effect of recA and recB mutations on gyrA208 lethality. Bacterial cells were grown at 28° in E-glucose medium supplemented with casaminoacids (0.5% w/v) and thymine (5 µg/ml). At an OD450 of 0.2 (time zero), cultures were transferred to 43°. Samples were periodically withdrawn and plated on LB medium supplemented with thymine (5 µg/ml) following appropriate dilution. Colony-forming units were plotted normalized to the value at time zero []. Solid circles, strain MA4017 (gyrA208); open circles, strain MA4199 (gyrA208 recA1); solid squares, strain MA4094 (gyrA208 recB546::Tn10dCm); open squares, strain MA4115 (gyrA208 recA1 recB546::Tn10dCm).

Suppressors of gyrA208-mediated cell death:
Aminoacyl tRNA synthetase mutations: We searched for suppressor mutations in strain MA2383, which harbors gyrA208 and a nonrevertable recB deletion. The selection protocol involved incubating bacteria spread on LB plates (lawns of 10⁸ cells) for 24 hr at 43° and then shifting the plates back to 28° to allow surviving cells to form colonies. Except for rare revertants of the gyrA208 mutation (readily identified and discarded; see MATERIALS AND METHODS), all of the isolates that were obtained behaved as expected for "suppressor-of-death" mutations: they did not grow at 43° but experienced little or no loss of viability during prolonged incubation at this temperature (data not shown). It soon became apparent that the majority of these "death suppressors" resulted from mutations that by themselves—i.e., in a gyr⁺ genetic background—conferred a temperature-sensitive growth phenotype. Genetic mapping and complementation tests identified three such suppressors (s-43, s-57, s-41) as mutant alleles of alanine, glutamine, and lysine tRNA synthetases, respectively. The ability of these mutations to suppress the lethal phenotype correlates with their property to cause cytostasis at the restrictive temperature (Fig 4). Because the primary consequence of aminoacyl tRNA synthetase defects is the inhibition of protein synthesis, we tested whether chloramphenicol, a protein synthesis inhibitor, would also suppress killing of a gyrA208 recB mutant at 43°. Treatment with this drug (20 µg/ml) effectively prevented death of the mutant at the restrictive temperature (data not shown). Inhibition of protein synthesis, by drug or mutation, also relieved the loss of viability suffered by a recombination-proficient gyrA208 mutant when subjected to a prolonged exposure to 43° on solid medium (data not shown).

View larger version (13K): [in this window] [in a new window] [Download PPT slide]   Figure 4. Effect of alaS216 ts on viability of a gyrA208 recA1 mutant. Bacterial cells were grown at 28° in E-glucose medium supplemented with casaminoacids (0.5% w/v) and thymine (5 µg/ml). At an OD450 of 0.2 (time zero), cultures were transferred to 43°. Samples were periodically withdrawn to determine OD450 and plated on LB medium supplemented with thymine (5 µg/ml) following appropriate dilution. (A) OD450 as a function of time normalized to the value at time zero []. (B) Colony-forming units as a function of time normalized to the value at time zero []. Open circles, strain MA4199 (gyrA208 recA1); open triangles, strain MA4120 (recA1 alaS216); solid circles, strain MA4122 (gyrA208 recA1 alaS216).

Icd mutations: A class of mutations resulting from the suppression-of-death selection above did not confer a ts phenotype. Strains carrying this class of suppressors appeared to require glutamate or proline to grow in minimal medium, were incapable of using acetate as the sole carbon source, and grew slowly in all media, including LB. These characteristics, as well as genetic mapping, allowed us to identify the auxotrophic mutations as alleles of the icd gene encoding isocitrate dehydrogenase. Repeating the suppressor-of-death search by transposon mutagenesis, we were able to obtain Tn10dTc insertion mutants of the icd gene. Significantly, icd mutants were originally obtained in E. coli during a search for bacteria resistant to nalidixic acid (HELLING and KUKORA 1971 ). LAKSHMI and HELLING 1976 correlated the nalidixic acid (Nal)-resistant phenotype with the accumulation of a growth-inhibitory Krebs cycle intermediate (possibly citrate). These authors showed that secondary mutations in the gltA gene (coding for citrate synthase) prevented the accumulation of the inhibitory metabolite restoring near normal growth rate and a normal Nal sensitivity (LAKSHMI and HELLING 1976 ). Prompted by this report, we isolated fast-growing derivatives of strain MA4985 (gyrA208 icd::Tn10dTc) at 37°. Unlike the parental strain, the fast-growing variants (presumably resulting from gltA changes) were unable to survive at 43°, indicating the loss of gyrA208 suppression. These results strengthen the analogy between gyrA208 and quinolone toxicity and suggest that cell growth—as opposed to the synthesis of any protein in particular—is the key factor in gyrase-dependent lethality.

RNA polymerase mutations: Earlier, we described a class of RNA polymerase (rpoB) mutations conferring low-level resistance to nalidixic acid in Salmonella (BLANCPOTARD et al. 1995 ). One of these mutations, Rif^S allele rpoB1044, results in temperature-sensitive growth. In this study, we tested the effect of this allele on gyrA208 suppression and found that the double mutant (strain MA3010) had >10⁴-fold increased viability relative to a strain carrying gyrA208 alone (MA2301) after a 24-hr exposure to 43° (data not shown).

Induction of the stringent response can contribute to gyrA208 suppression:
In enteric bacteria, the synthesis of cell envelope constituents, such as peptidoglycan and membrane phospholipids, is negatively controlled by guanosine 3',5'-bis(diphosphate) (ppGpp), the mediator of the stringent response (reviewed in CASHEL et al. 1996 ). Synthesis of ppGpp by the relA gene product occurs in response to nutritional deprivation. Conceivably, this regulatory mechanism helps coordinate cell wall biosynthesis with the rate of protein synthesis. Mutations in the relA gene prevent ppGpp synthesis, resulting in the "relaxed" phenotype, characterized, among other traits, by defective cell division and filamentation (CASHEL et al. 1996 ). Mutations in the spoT gene, coding for an activity involved in ppGpp turnover, cause gratuitous accumulation of the nucleotide and result in slow growth (RUDD et al. 1985 ).

We studied the effects of a spoT mutation on gyrA208 lethality. The spoT1 allele greatly increased survival of the gyrA208 mutant at 43° (strain MA3170). As expected, this effect was completely eliminated by the subsequent introduction of a relA mutation (in strain MA3382; data not shown). Unexpectedly, the relA1 mutation also inhibited gyrA208 suppression in the glnS1521 and lysS103 aminoacyl tRNA synthetase mutants, whereas suppression by alaS216 remained unaffected (Table 3). Closer examination revealed that the differential sensitivity to the relA mutation correlates with the degree of "leakiness" of the synthetase alleles. While alaS216 brings bacterial growth to a sudden and complete arrest at 43°, glnS1521 and lysS1031 both show a leaky phenotype, allowing bacterial cells to continue dividing at a slow pace (data not shown). In conclusion, these results show that the induction of the stringent response (elicited by poor tRNA aminoacylation) critically contributes to gyrA208 suppression in the glnS and lysS mutants (Table 3; note that the three synthetase mutations suppress gyrA208 lethality with approximately the same efficiency in a relA⁺ background). No such contribution is needed when protein synthesis is completely inhibited.

Inhibition of protein synthesis prevents "reckless" DNA degradation in gyrA208 recA1 mutant:
Formation of DNA breaks in bacteria defective in RecA activity leads to extensive DNA degradation. This results from the "reckless" activity of the ExoV nuclease (RecBCD) that uses double-strand DNA ends as entry sites (WILLETTS and CLARK 1969 ; SMITH 1988 ). Therefore, chromosome degradation in a recA mutant background provides an indicator for the occurrence of double-strand breaks. We showed previously that a gyrA208 recA double mutant suffers reckless DNA degradation when incubated at 43°; the process is recB mediated since it is completely abolished in a gyrA208 recA recB triple mutant (Fig 4 in GARI et al. 1996 ).

In this study, we sought to determine whether mutations or treatments suppressing gyrA208 recA lethality concomitantly blocked degradation. Results in Fig 5 show that DNA degradation no longer occurs following the introduction of the alaS216 allele (strain MA4122) or upon treating cells with chloramphenicol during the 43° incubation. Since the strains used for these experiments are devoid of detectable recombinational activity, it seems unlikely that these results could be explained by postulating that the inhibition of protein synthesis simply allows more time for DNA lesions to be repaired. Rather, the most likely conclusion is that the gyrase-mediated lesions are no longer formed when protein synthesis is inhibited or, more generally, when bacterial growth is arrested.

View larger version (16K): [in this window] [in a new window] [Download PPT slide]   Figure 5. Inhibiting protein synthesis prevents "reckless" DNA degradation in a gyrA208 recA1 mutant. Chromosomal DNA was uniformly labeled with tritium by growing cultures overnight at 28° in E-glucose medium supplemented with casaminoacids (0.5% w/v), thymine (5 µg/ml), and [3H]thymine (50 µCi/ml). Cells were then harvested and resuspended in the medium lacking [3H]thymine. The cell suspension was diluted 100-fold and growth resumed in the presence of an excess of cold thymine (50 µg/ml). Following 2 hr of incubation at 28°, cultures were shifted to 43° and one part of the culture was treated with chloramphenicol (20 µg/ml) 10 min after the shift. Aliquots were withdrawn at 30-min intervals and trichloroacetic acid-precipitable radioactivity was measured. Open squares and circles, strain MA4199 (gyrA208 recA1); solid circles, strain MA4122 (gyrA208 recA1 alaS216).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The gyrA208 allele is representative of a class of gyrase alterations that have little effect on DNA supercoiling but cause lethality by inducing chromosomal breakage. Under conditions of inadequate recombinational repair, these mutations lead to irreversible DNA damage and cell death. In this study we sought to gather insight into the mechanism of killing through the isolation of second-site suppressors preventing death at the restrictive temperature (43°). Three such suppressors were found to result from aminoacyl tRNA synthetase changes that either block or strongly slow growth at 43°. In the synthetase mutants that still grow at 43°, suppression of gyrA208 lethality depends on the activity of the relA gene product, suggesting that induction of the stringent response—consequent to poor tRNA amino-acylation—contributes to suppression. In turn, this suggests that the mechanism of killing is in relation to some aspect(s) of cell development rather than to protein synthesis per se. Consistent with this interpretation, suppression of death can also result from accumulation of a growth-inhibitory Krebs cycle intermediate consequent to the inactivation of iso-citrate dehydrogenase.

The changes or conditions that suppress gyrA208 lethality are the same as those that protect bacteria against killing by quinolones. This was already evident in previous studies in which mutations affecting RNA or protein synthesis were isolated by selecting for low-level resistance to Nal and found to suppress gyrA208 (BLANC-POTARD and BOSSI 1994 ; BLANC-POTARD et al. 1995 ). In this work, mutations selected as gyrA208 suppressors concomitantly conferred resistance to low levels of Nal (4 µg/ml). Finally, chloramphenicol, a protein synthesis inhibitor, was similarly effective in preventing chromosomal DNA breakage and cell death in quinolone-treated bacteria and in bacteria harboring the gyrA208 allele. These analogies strongly suggest that mutation gyrA208 and quinolone antibiotics cause cell death through a common mechanism.

At first glance, the gyrA208 mutation could act in a drug-like manner causing the gyrase reaction to stall at the stage of the cleavable complex. However, we found no evidence for a spontaneous accumulation of cleavable complexes in the gyrA208 mutant under any condition tested (GARI et al. 1996 ). Cleavable complexes were readily detected in bacteria exposed to as little as 2 µg/ml of Nal (a sublethal concentration); however, their levels did not increase significantly at higher drug doses and were unaffected by treatments that completely prevented bacterial killing (GARI et al. 1996 ; our unpublished data). Thus, cleavable complex formation does not correlate with cell death and it is not lethal in resting bacteria.

In both the gyrA208 mutant and Nal-treated bacteria, DNA breakage occurs in dividing cells, suggesting that the mechanism responsible for damage is connected with cell growth. This dependence is strongly reminiscent of the phenomenon known as "thymineless" death, the rapid killing of thymine auxotrophic microorganisms in response to thymine starvation (reviewed by AHMAD et al. 1998 ). Thymineless death is accompanied by extensive DNA breakage. Early studies showed that only cells active in replication are sensitive to the lethal effect of thymine deprivation (HANAWALT et al. 1961 ; MAALOE and HANAWALT 1961 ). This suggests that the arrest of DNA replication under conditions in which bacterial cells continue to elongate (unbalanced growth; BARNER and COHEN 1957 ) results in DNA damage. The DNA/mass ratio and the average cell length are kept constant during growth of gram-negative bacteria (DONACHIE and BEGG 1989 ). Possibly, checkpoints in the cell cycle verify the synchrony of cell wall growth and chromosome replication. Loss of synchrony might cause the replication fork to collapse and break, resulting in SOS induction and inhibition of septation (MUKHERJEE et al. 1998 ). The implication of this study is that there is no need to invoke the formation of an enzyme-DNA adduct to envision how a defective gyrase could cause DNA damage. It is sufficient to postulate that some gyrase defects hamper the progression of replication forks—for example, by inefficiently removing positive supercoils ahead of the fork, causing replication to stall or slow. In growing cells, this event would have the same effect as thymine starvation and lead to DNA damage and death. Therefore, according to a unifying view, thymineless death, gyrA208-induced killing, and killing by first-generation quinolones could all be the manifestation of the same basic mechanism.

	FOOTNOTES

¹ Present address: Departament de Ciències Mèdiques Bàsiques, Universitat de Lleida, 25006 Lleida, Spain.

	ACKNOWLEDGMENTS

We are grateful to Montserrat Llagostera and John Roth for providing strains and to Philippe Bouloc, Pierre Plateau, and Jacqueline Plumbridge for the generous gift of plasmids. We thank Amando Flores for first identifying the icd mutants and contributing to their characterization. E.G. was the recipient of a fellowship from the Human Capital and Mobility Program of the European Community Commission. This work was supported by the Centre National de la Recherche Scientifique.

Manuscript received May 4, 2001; Accepted for publication September 7, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AHMAD, S. I., S. H. KIRK, and A. EISENSTARK, 1998  Thymine metabolism and thymineless death in prokaryotes and eukaryotes. Annu. Rev. Microbiol. 52:591-625[Medline].

BARNER, H. D. and S. S. COHEN, 1957  The isolation and properties of amino acid requiring mutants of a thymineless bacterium. J. Bacteriol. 74:350-355[Free Full Text].

BENSON, N. R. and B. S. GOLDMAN, 1992  Rapid mapping in Salmonella typhimurium with Mud-P22 prophages. J. Bacteriol. 174:1673-1681[Abstract/Free Full Text].

BERGER, J. M., 1998  Type II DNA topoisomerases. Curr. Opin. Struct. Biol. 8:26-32[Medline].

BLANC-POTARD, A.-B. and L. BOSSI, 1994  Phenotypic suppression of DNA gyrase deficiencies by a deletion lowering the gene dosage of a major tRNA in Salmonella typhimurium.. J. Bacteriol. 176:2216-2226[Abstract/Free Full Text].

BLANC-POTARD, A.-B., E. GARÍ, F. SPIRITO, N. FIGUEROA-BOSSI, and L. BOSSI, 1995  RNA polymerase (rpoB) mutants selected for resistance to gyrase inhibitors in S. typhimurium. Mol. Gen. Genet. 247:680-692[Medline].

CASHEL, M., D. R. GENTRY, V. J. HERNANDEZ and D. VINELLA, 1996 The stringent response, pp. 1458–1496 in Escherichia coli and Salmonella: Cellular and Molecular Biology, Ed. 2, edited by F. C. NEIDHARDT, R. CURTISS, III, J. L. INGRAHAM, E. C. C. LIN, K. B. LOW, B. MAGASANIK, W. S. REZNIKOFF, M. RILEY, M. SCHAECHTER and H. E. UMBARGER. American Society for Microbiology, Washington, DC.

CHEN, C. R., M. MALIK, M. SNYDER, and K. DRLICA, 1996  DNA gyrase and topoisomerase IV on the bacterial chromosome: quinolone-induced DNA cleavage. J. Mol. Biol. 258:627-637[Medline].

CLERCH, B., X. GARRIGA, E. TORRENTS, C. M. ROSALES, and M. LLAGOSTERA, 1996  Construction and characterization of two lexA mutants of Salmonella typhimurium with different UV sensitivities and UV mutabilities. J. Bacteriol. 178:2890-2896[Abstract/Free Full Text].

DEITZ, W. H., T. M. COOK, and W. A. GOSS, 1966  Mechanism of action of nalidixic acid on Escherichia coli. III. Conditions required for lethality. J. Bacteriol. 91:768-773[Abstract/Free Full Text].

DONACHIE, W. D. and K. J. BEGG, 1989  Cell length, nucleoid separation, and cell division of rod-shaped and spherical cells of Escherichia coli. J. Bacteriol. 171:4633-4639[Abstract/Free Full Text].

DRLICA, K. and X. ZHAO, 1997  DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol. Mol. Biol. Rev. 61:377-392[Abstract].

EL HANAFI, D. and L. BOSSI, 2000  Activation and silencing of leu-500 promoter by transcription-induced DNA supercoiling in the Salmonella chromosome. Mol. Microbiol. 37:583-594[Medline].

FIGUEROA-BOSSI, N., E. COISSAC, P. NETTER, and L. BOSSI, 1997  Unsuspected prophage-like elements in Salmonella typhimurium. Mol. Microbiol. 25:161-173[Medline].

GARÍ, E., N. FIGUEROA-BOSSI, A.-B. BLANC-POTARD, F. SPIRITO, and M. B. SCHMID et al., 1996  A class of gyrase mutants of Salmonella typhimurium show quinolone-like lethality and require Rec functions for viability. Mol. Microbiol. 21:111-122[Medline].

HANAWALT, P. C., O. MAALØE, D. J. CUMMINGS, and M. SCHAECHTER, 1961  The normal DNA replication cycle. II. J. Mol. Biol. 3:156-165[Medline].

HEDDLE, J. G., F. M. BARNARD, L. M. WENTZELL, and A. MAXWELL, 2000  The interaction of drugs with DNA gyrase: a model for the molecular basis of quinolone action. Nucleosides Nucleotides Nucleic Acids 19:1249-1264[Medline].

HELLING, R. B. and J. S. KUKORA, 1971  Nalidixic acid-resistant mutants of Escherichia coli deficient in isocitrate dehydrogenase. J. Bacteriol. 105:1224-1226[Abstract/Free Full Text].

HIASA, H., D. O. YOUSEF, and K. J. MARIANS, 1996  DNA strand cleavage is required for replication fork arrest by a frozen topoisomerase-quinolone-DNA ternary complex. J. Biol. Chem. 271:26424-26429[Abstract/Free Full Text].

HUSSAIN, K., E. J. ELLIOTT, and G. P. SALMOND, 1987  The parD^- mutant of Escherichia coli also carries a gyrA_am mutation. The complete sequence of gyrA. Mol. Microbiol. 1:259-273[Medline].

KAMPRANIS, S. C., A. D. BATES, and A. MAXWELL, 1999  A model for the mechanism of strand passage by DNA gyrase. Proc. Natl. Acad. Sci. USA 96:8414-8419[Abstract/Free Full Text].

KATO, J., Y. NISHIMURA, and H. SUZUKI, 1989  Escherichia coli parA is an allele of the gyrB gene. Mol. Gen. Genet. 217:178-181[Medline].

KHODURSKY, A. B., B. J. PETER, M. B. SCHMID, J. DERISI, and D. BOTSTEIN et al., 2000  Analysis of topoisomerase function in bacterial replication fork movement: use of DNA microarrays. Proc. Natl. Acad. Sci. USA 97:9419-9424[Abstract/Free Full Text].

KOHNO, T. and J. R. ROTH, 1974  Proflavin mutagenesis of bacteria. J. Mol. Biol. 89:17-32[Medline].

LAKSHMI, T. M. and R. B. HELLING, 1976  Selection for citrate synthase deficiency in icd mutants of Escherichia coli. J. Bacteriol. 127:76-83[Abstract/Free Full Text].

LEVEQUE, F., P. PLATEAU, P. DESSEN, and S. BLANQUET, 1990  Homology of lysS and lysU, the two Escherichia coli genes encoding distinct lysyl-tRNA synthetase species. Nucleic Acids Res. 18:305-312[Abstract/Free Full Text].

LEVINE, C., H. HIASA, and K. J. MARIANS, 1998  DNA gyrase and topoisomerase IV: biochemical activities, physiological roles during chromosome replication, and drug sensitivities. Biochim. Biophys. Acta 1400:29-43[Medline].

LIU, L. F., 1989  DNA topoisomerase poisons as antitumor drugs. Annu. Rev. Biochem. 58:351-375[Medline].

LIU, L and J. C. WANG, 1987  Supercoiling of the DNA template during transcription. Proc. Natl. Acad. Sci. USA 84:7024-7027[Abstract/Free Full Text].

MAALØE, O. and P. C. HANAWALT, 1961  Thymine deficiency and the normal DNA replication cycle II. J. Mol. Biol. 3:144-155[Medline].

MALOY, S. R., W. J. STEWART and R. K. TAYLOR, 1996 Genetic Analysis of Pathogenic Bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MENZEL, R. and M. GELLERT, 1994  The biochemistry and biology of DNA gyrase. Adv. Pharmacol. 29:39-69.

MIESEL, L. and J. R. ROTH, 1994  Salmonella recD mutations increase recombination in a short sequence transduction assay. J. Bacteriol. 176:4092-4103[Abstract/Free Full Text].

MUKHERJEE, A., C. CAO, and J. LUTKENHAUS, 1998  Inhibition of FtsZ polymerization by SulA, an inhibitor of septation in Escherichia coli.. Proc. Natl. Acad. Sci. USA 95:2885-2890[Abstract/Free Full Text].

PLUMBRIDGE, J., 1987  Organisation of the Escherichia coli chromosome between genes glnS and glnU, V. Mol. Gen. Genet. 209:618-620[Medline].

RUDD, K. E., B. R. BOCHNER, M. CASHEL, and J. R. ROTH, 1985  Mutations in the spoT gene of Salmonella typhimurium: effects on his operon expression. J. Bacteriol. 163:534-542[Abstract/Free Full Text].

SANDERSON, K. E., A. HESSEL, and K. E. RUDD, 1995  Genetic map of Salmonella typhimurium, edition VIII. Microbiol. Rev. 59:241-303[Abstract/Free Full Text].

SCHMIEGER, H., 1972  Phage P22 mutants with increased or decreased transduction activities. Mol. Gen. Genet. 119:75-88[Medline].

SMITH, G. R., 1988  Homologous recombination in procaryotes. Microbiol. Rev. 52:1-28[Free Full Text].

VINELLA, D., R. D'ARI, A. JAFFÉ, and P. BOULOC, 1992  Penicillin binding protein 2 is dispensable in Escherichia coli when ppGpp synthesis is induced. EMBO J. 11:1493-1501. (erratum: EMBO J. 11: 5137).[Medline].

WILLETTS, N. S. and A. J. CLARK, 1969  Characteristics of some multiply recombination-deficient strains of Escherichia coli. J. Bacteriol. 100:231-239[Abstract/Free Full Text].

WILLIAMS, N. L., A. J. HOWELLS, and A. MAXWELL, 2001  Locking the ATP-operated clamp of DNA gyrase: probing the mechanism of strand passage. J. Mol. Biol. 306:969-984[Medline].

ZECHIEDRICH, E. L., A. B. KHODURSKY, S. BACHELLIER, R. SCHNEIDER, and D. CHEN et al., 2000  Roles of topoisomerases in maintaining steady-state DNA supercoiling in Escherichia coli. J. Biol. Chem. 275:8103-8113[Abstract/Free Full Text].

This article has been cited by other articles:

				Z. Pang, R. Chen, D. Manna, and N. P. Higgins A Gyrase Mutant with Low Activity Disrupts Supercoiling at the Replication Terminus J. Bacteriol., November 15, 2005; 187(22): 7773 - 7783. [Abstract] [Full Text] [PDF]

				E. A. Kouzminova, E. Rotman, L. Macomber, J. Zhang, and A. Kuzminov RecA-dependent mutants in Escherichia coli reveal strategies to avoid chromosomal fragmentation PNAS, November 16, 2004; 101(46): 16262 - 16267. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Garí, E. Articles by Figueroa-Bossi, N. Search for Related Content PubMed PubMed Citation Articles by Garí, E. Articles by Figueroa-Bossi, N.