|
|
J Bacteriol. 2007 August; 189(16): 5839–5849. Published online 2007 March 30. doi: 10.1128/JB.00083-07. | PMCID: PMC1952050 |
Copyright © 2007, American Society for Microbiology Growth Rate Toxicity Phenotypes and Homeostatic Supercoil Control Differentiate Escherichia coli from Salmonella enterica Serovar Typhimurium ![[down-pointing small open triangle]](/corehtml/pmc/pmcents/x25BF.gif) † Keith Champion and N. Patrick Higgins* Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Birmingham, Alabama 35294-0024 Received January 16, 2007; Accepted March 19, 2007. Escherichia coli and Salmonella enterica serovar Typhimurium share high degrees of DNA and amino acid identity for 65% of the homologous genes shared by the two genomes. Yet, there are different phenotypes for null mutants in several genes that contribute to DNA condensation and nucleoid formation. The mutant R436-S form of the GyrB protein has a temperature-sensitive phenotype in Salmonella, showing disruption of supercoiling near the terminus and replicon failure at 42°C. But this mutation in E. coli is lethal at the permissive temperature. A unifying hypothesis for why the same mutation in highly conserved homologous genes of different species leads to different physiologies focuses on homeotic supercoil control. During rapid growth in mid-log phase, E. coli generates 15% more negative supercoils in pBR322 DNA than Salmonella. Differences in compaction and torsional strain on chromosomal DNA explain a complex set of single-gene phenotypes and provide insight into how supercoiling may modulate epigenetic effects on chromosome structure and function and on prophage behavior in vivo. Dichotomous growth is an adaptation that allows certain bacteria to divide rapidly under nutrient-rich conditions. During dichotomous growth, wild-type (WT) Escherichia coli can divide every 20 min, even though it takes >50 min to replicate the genome. Cells can divide rapidly only as long as initiation of bidirectional DNA synthesis with the DnaA “initiator” complex at the oriC “replicator” ( 42) is well coordinated with the formation of nucleoids and the distribution of replicated copies of the chromosome to each daughter cell at the time of partition. Recent studies with live cells show that a highly organized traffic pattern deposits each replisome arm at opposing edges of growing nucleoids ( 2, 90). At high growth rates, the average cell has a complex chromosome structure, with nucleoids containing two or four partially replicated arcs spanning the oriC initiator region. To distribute nucleoids, the cell division machinery needs to find the cell midpoint, decatenate all cross-links between chromosomal replicas at the dif site ( 19, 49), and move the nucleoids into each new daughter cell. The initiation frequency at oriC is critical for cell division. Hyperinitiation is toxic or lethal ( 77), with one problem being fork overrun. If a fork from a recent initiation cycle overtakes a previously initiated fork, large chromosome fragments with double-strand ends can be generated ( 4). We previously showed that the Salmonella gyrB652 mutation causes “growth rate toxicity” that includes topological chaos near the dif site ( 68). Gyrase hypomorphs in Salmonella lose supercoiling near the terminus of replication, even though plasmids in the same cell maintain near-normal superhelical densities ( 68). Recent experiments suggest that a similar fate may occur in gyrase mutants of E. coli ( 29, 45). The GyrB protein is highly conserved in E. coli and Salmonella. Of its 804 amino acids, 97% are identical in the two species, with most substitutions being in the C-terminal region of the protein (18 of 28 substitutions are beyond H567). To see if the Salmonella GyrB652 mutation would cause terminal chaos in E. coli, we designed an oligonucleotide to introduce a single base pair substitution (Fig. ). The R436-S substitution was lethal in E. coli. Moreover, ectopic expression of the Salmonella WT GyrB or GyrB652 protein from a plasmid was toxic for E. coli whereas a multicopy plasmid containing the E. coli GyrB protein was tolerated in both E. coli and Salmonella. Using reporter plasmids, we found that the average supercoil density (σ) for mid-log cultures of WT Salmonella (−0.060) is 13% lower than that for E. coli (−0.069). Plasmids having 56 bp of alternating GC sequence proved that E. coli also has a significantly higher level of DNA torsional strain. These observations explain several mysterious phenotypic differences between the species, such as the growth rate toxicity of mukA- and seqA-null mutants and the behavior of Mu lysogens. Media. All cells were grown in LB broth ( 36). Antibiotics were added to the media at concentrations of 50 μg/ml for kanamycin (Kan) and ampicillin (Amp), 20 μg/ml for chloramphenicol, 15 μg/ml for tetracycline, 10 μg/ml for gentamicin, and 5 μg/ml for nalidixic acid (Nal). Plasmids and strains. All bacterial strains and plasmids in this work are described in Tables 1 and 2. Plasmid pRW478 is derived from pBR322 by the introduction of a (CG) 28 repeat into the EcoRI site ( 43). Chromosomal modification. Chromosome modifications were made with the λ RED recombination system ( 95). Synthetic oligonucleotides were from IDT (Coralville, IA), and the sequences are given in Table 3. PCR products used for recombineering were made using a 1:1 mixture of Taq polymerase (Sigma) and Taq extender additive (Stratagene). Plasmid maintenance assay. The abilities of cells to maintain plasmids were measured by comparing serial dilutions of the culture after plating them on selective and nonselective media. LB medium containing 50 μg/ml Amp (and IPTG [isopropyl-β-d-thiogalactopyranoside] when indicated) was inoculated with fresh overnight culture by 1:100 dilution. For stationary-phase measurements, cultures were allowed to grow 8 h or more at 30°C. A portion of each culture was serially diluted 1:10 in 96-well microtiter plates. Finally, 3-μl aliquots of each dilution were spotted onto LB plates both with and without 50 μg/ml Amp and plates were incubated overnight in a 30°C incubator. Supercoil density measurements. Plasmid supercoil density was determined by the band counting method using agarose gels containing chloroquine ( 75). DNA was prepared from log phase cells grown at 30°C. At indicated times, cells were collected and concentrated by centrifugation. Plasmid DNA was extracted with the Promega Wizard Plus midi prep DNA purification system and eluted into sterile water. DNA was further purified with phenol, phenol-chloroform, and chloroform extraction and concentrated with isobutanol. Ether was used to remove trace isobutanol. Purified plasmid DNA was separated in 25-cm 1.0% agarose gels containing chloroquine. For one-dimensional (1-D) gels, 1 to 2 μg of DNA was loaded into a gel containing 1.5 μM chloroquine and electrophoresis was carried out at room temperature at 2 V/cm for 46 h in 0.5× Tris-borate-EDTA (TBE), also containing 1.5 μM chloroquine. For 2-D gels, a topoisomer set from the first dimension was cut out, soaked in 0.5× TBE with 20 μM chloroquine for 4 hours, turned 90°, and recast in a new 25-cm 1.0% agarose gel containing 20 μM chloroquine. Electrophoresis in the second dimension was done at room temperature at 2 V/cm for 46 h in 0.5× TBE containing 20 μM chloroquine. After electrophoresis, the gels were soaked in deionized water for 30 min each for four changes and stained 60 min with ethidium bromide. DNA was visualized with UV light and captured on a Bio-Rad Zeta probe membrane using vacuum transfer. Southern blot analysis was done with probes made using Invitrogen's random-primed DNA labeling system with a pBR322 template. The membranes were exposed on a Molecular Dynamics phosphor screen and quantitated with Bio-Rad Gel Doc Software. The GyrB652 protein is lethal in E. coli. The Salmonella gyrB652 mutation ( 27) results from a single C 1306-to-A transversion mutation that substitutes serine for arginine at residue 436 (R436-S) ( 68). The 3-D model of the gyrase tetramer places this residue near a GyrB-GyrA interface ( 7, 47, 72). The GyrB652 form of gyrase has a lower in vitro catalytic efficiency ( kcat) than the WT enzyme. The temperature-sensitive (TS) growth defect is not caused by enzyme inactivation at 42°C ( 68) but is due to the inability of gyrase to keep up with high metabolic rates and multiple rounds of chromosomal DNA replication. The signature of this defect is topological chaos near the dif site, which is the terminus of replication ( 68). To learn whether this suboptimal form of gyrase would cause terminal chaos in E. coli, we tried to introduce at a permissive temperature (30°C) the GyrB R436-S substitution into strains of E. coli derived from W3110, NH3423, and NH3612. Bacteriophage λ “recombineering” technology ( 95) exploits the regulated expression of λ recombination proteins Beta, Gam, and Exo to modify bacterial chromosomes efficiently using double-stranded or short single-stranded DNA substrates with only 30 bp of targeting homology ( 18). To introduce the gyrB652 mutation, a synthetic 70-mer oligonucleotide (oligonucleotide 1) was made with 35 nucleotides of 5′-flanking homology followed by a single mismatch (the gyrB652 C-to-A transversion mutation) and 34 nucleotides of 3′-flanking homology (Fig. ) (see Table 3 for oligonucleotide 1 sequence). Because the primary gyrB sequences of this segment are identical in E. coli and Salmonella, we used λ recombineering to modify two E. coli strains (NH4323 and NH3612) and a WT Salmonella strain (NH3702). Expression of λ RED recombination proteins was induced for 15 min, oligonucleotide 1 was introduced into cells by electroporation, and 1-ml cultures were incubated overnight at 30°C in LB before plating them onto selective media. The R436-S substitution was easy to introduce into the Salmonella strain NH3702 by using the λ recombineering technology. The phenotype includes TS growth at 42°C and resistance to 5 μg/ml of Nal ( 27). After 100-μl aliquots of the 1-ml NH3702 electroporation culture was spread onto Nal plates, an average of 125 Nal r colonies was found on multiple plates. When 10 independent colonies from these plates were restreaked on LB agar at 30°C and 42°C, all 10 Nal r strains failed to grow at 42°C. Sequence analysis of DNA PCR amplified from four of the Salmonella NH3702-derived Nal-resistant TS colonies confirmed that each had the gyrB652 C-to-A transversion mutation. The efficiency of this chromosome modification reaction is similar to what others and we have observed for introducing single base pair changes into many different sites in both E. coli and Salmonella ( 13, 18). However, introducing the R436-S substitution into E. coli was not easy. The same volume of the NH3423 or NH3612 strain electroporated with oligonucleotide 1 yielded an average of six Nalr colonies. None of these colonies had a TS phenotype upon restreaking at 42°C. Low-level Nal resistance can be caused by mutations in several genes, including gyrA, rpoB, and tolC. The number of Nalr colonies arising from the E. coli electroporation experiment was the same as that for the control experiment. When 100 μl of unmodified stationary-phase cultures of E. coli or Salmonella was spread on Nal plates, they gave an average of seven Nalr colonies. Sequence analysis showed that none of the control strains of E. coli or Salmonella Nalr colonies and none of the Nalr E. coli NH3612 strains derived from plating electroporated cells carried the GyrB R436-S substitution. The difficulty of establishing the gyrB652 mutation in E. coli could be caused by intolerance of this variant protein. However, it could also be caused by a species-specific difference in gyrB652 phenotype or by differences in λ recombineering efficiency between E. coli and Salmonella. To increase target homology, a second strategy (Fig. ) combined oligonucleotide 1 with a downstream 20-mer oligonucleotide (oligonucleotide 2) to generate a double-stranded, 339-bp PCR product. The gyrB652 C-to-A transversion had 35 bp of upstream gyrB homology and 303 bp of downstream homology. Strains NH3702 and NH3612 were thermo-induced and electroporated with the PCR product of oligonucleotides 1 and 2. E. coli NH3612 produced fewer than 10 Nalr colonies (the same as for the control plates), and none were TS. For the same experiment with Salmonella, 100-μl aliquots of the overnight electroporation culture yielded an average of more than 100 Nalr TS colonies on each. To eliminate the possibility that the GyrB652 substitution in E. coli does not exhibit low-level Nal resistance, we changed our genetic selection scheme. A new strain (NH3703) was made using λ recombineering to insert a Kan r module in the intergenic space between gyrB and yidB (Fig. ). Using linkage to the Kan r module, we could introduce mutant alleles by selection on Kan plates followed by screening of Kan r recombinants for a linked TS phenotype and/or the presence of the C-to-A transversion in the gyrB gene. DNA from NH3703 was isolated and used as a template with two new PCR primers (oligonucleotides 5 and 6) to make a double-strand DNA product for electroporation. Oligonucleotide 6 had the gyrB652 C-to-A transversion at position 54 of 74 nucleotides. The 2,808-bp PCR product includes 53 bp of upstream homology before the C-to-A transversion and 133 bp of homology beyond the Kan r inset toward yidB (Fig. and Table 3). E. coli NH3702 was electroporated with this PCR product, cells were incubated overnight, and 100-μl aliquots were spread onto Kan plates. An average of 20 Kan r recombinants per plate spread with 100 μl of this electroporation culture was recovered. Each Kan r colony was tested for Nal r and TS phenotypes. None were found. The gyrB regions of 20 Kan r strains were PCR amplified and sequenced. No C-to-A mutation was found among the Kan r recombinants. The failure to isolate an R436-S mutation linked to the Kan r gene in the last experiment was surprising and unique in our recombineering experience. The result could mean that the C-to-A mutation was lethal and that all Kan r recombinants from the last experiment resulted from a mismatch repair that corrected a heteroduplex recombination intermediate to the WT sequence during recombination. Alternatively, the PCR product might have been degraded from the upstream end in vivo to eliminate the gyrB652 segment, which would yield Kan r recombinants that did not include the C-to-A transversion. To test these possibilities, the experiment was carried out with a mutS::Tn 10 strain (NH3704) that cannot perform mismatch repair ( 53). To determine how many Kan r recombinants would be expected from this experiment, a WT gyrB sequence was made with oligonucleotides 5 and 7 (Table 3 and Fig. ). Both the mutant and the WT PCR products have >1,100 bp of gyrB homology upstream of the Kan module and 139 bp beyond the Kan module. Aliquots (100 μl) of the culture electroporated with the WT GyrB PCR (oligonucleotides 5 and 7) averaged 150 Kan r colonies per plate. Not a single Kan r colony was found on any plate using 100-μl aliquots of the gyrB652 PCR product (oligonucleotides 5 and 6). This experiment confirms that the Kan r colonies generated with the gyrB652-containing PCR product in the Mut + background (NH3702) arose after a MutS-dependent correction of a gyrB652 heteroduplex to the WT gyrB sequence. The striking bias between WT and gyrB652 PCR products is strong evidence that the GyrB R436-S substitution is lethal for E. coli under our experimental conditions. WT Salmonella gyrB is toxic in E. coli. To study the gyrB652 allele in E. coli, we introduced a plasmid with a copy of the Salmonella gyrB652 gene. pRC03 is a pGEM-derived plasmid containing a copy of the Salmonella gyrB652 gene cloned adjacent to a bacteriophage T7 promoter. This plasmid was used to express and purify the GyrB652 protein from E. coli BL21::DE3, a strain that harbors a Lac-inducible T7 RNA polymerase ( 68). However, in strains lacking the T7 RNA polymerase, the T7 promoter is slightly “leaky.” pRC03 was introduced into E. coli NH3706 and maintained by continuous selection on Amp. But the strain grew more erratically than strains with the pGEM vector alone or pGEM vectors with other cloned inserts (not shown). To test for a gyrB652-associated growth penalty, overnight cultures of NH3706 (with pRC03) were diluted 1:100 into triplicate LB-plus-Amp cultures and grown into stationary phase at 30°C. Aliquots (3 μl) of serially diluted samples were spotted onto both LB and LB-Amp plates (Fig. ). There was a 4-order-of-magnitude difference between the numbers of CFU detected on LB and LB-Amp plates; pRC03 was lost from 99.99% of the stationary cells in culture. Control experiments with pGEM vector or pGEM vectors with other cloned genes showed that >99% of all cells retained these plasmids under the same experimental protocol. During growth of strains carrying plasmids with the Amp r gene, Studier noted that Amp becomes degraded by the secreted penicillinase in early log phase, and during the latter stages of growth there is no selection for maintaining Amp-resistant plasmid ( 83). Thus, Salmonella GyrB652 protein is toxic and selectively eliminated from fast-growing cells, even when E. coli contains a WT chromosomal copy of gyrB. This explains our failure to isolate an E. coli strain bearing the R436-S substitution in the chromosome. Since the mutant form of gyrB is not well tolerated by E. coli, we carried out the same experiment with strain NH3707, an E. coli strain that contains a pGEM plasmid with a cloned copy of the WT Salmonella gyrB gene. Surprisingly, after inoculation of cultures carrying pRC05 into fresh LB-Amp medium and growth to stationary phase, the plasmid was eliminated from 99.9% of the cells. Thus, even WT Salmonella GyrB is deleterious in E. coli. To determine if the E. coli gyrB gene is toxic when cloned on a multicopy plasmid, pAG111 was introduced into both E. coli and Salmonella. In pAG111, the WT E. coli gyrB gene is cloned in the pTTQ-18 vector under Lac promoter control. Cells inoculated at a 1:100 dilution into fresh LB-Amp medium containing the indicated amounts of IPTG were grown to stationary phase at 30°C. Each culture was serially diluted, and 3-μl aliquots were spotted onto an LB-Amp plate (Fig. ). In E. coli, there was no significant growth penalty when IPTG was absent. Over the range from 10 μM to 1 mM IPTG, a 1,000-fold plasmid destabilization was noted. However, IPTG induction of the lacZ gene in E. coli causes stochastic all-or-none high-level expression ( 66, 67, 76). GyrB protein expressed from the Lac promoter accumulates up to 40% of the total soluble cell protein in expressing cells (about 500-fold in excess of the normal level) ( 31). The E. coli GyrB protein was not toxic in Salmonella, which lacks the lacY permease, except when incubated with 1 mM IPTG. Salmonella may not give the all-or-none expression pattern observed in E. coli strains with the lacY gene, which actively concentrates the inducer in expressing cells. Nonetheless, E. coli is sensitive to large increases in its own GyrB protein and is extremely sensitive to the presence of low levels of the Salmonella GyrB protein (see below). Species-specific supercoil set points. The phenotypic differences described above led us to ask whether a difference in mean supercoil density (σ) exists between these organisms. To compare values of σ, the linking difference of pBR322 was analyzed in strains with a WT complement of the four topoisomerases. E. coli JTT1 (NH0572) has been used in several important topological analyses ( 15, 52, 82). Experiments were carried out with Salmonella LT2 (NH0742) and SL4213 (NH0405), a strain with WT topoisomerases that carries a restriction mutation that makes it easy to move plasmids from E. coli directly to Salmonella. Actively growing cultures at mid-log phase (50 to 70 Klett units) were harvested, and plasmid DNA was purified. Topoisomers were separated in 2-D 1.0% agarose gels, and DNA was blotted to nylon membranes for quantitation. On 2-D gels, one can measure two quantities. First is the linking difference, which is obtained by assigning a number to each topoisomer relative to a relaxed DNA. The second parameter that can be scored is a supercoil-induced change in DNA structure, like the cooperative transition from right-handed to left-handed conformation in plasmids that have sequences that adopt the Z conformation ( 70). Remarkably, the σ of pBR322 in Salmonella was significantly lower than that in E. coli (Fig. ). With DNA isolated from cells in mid-log phase, a three- to four-topoisomer difference existed between E. coli and Salmonella. By use of the band counting method in multiple experiments ( 75), the mean supercoil density (σ) values for plasmid DNA were found to be −0.069 for E. coli and 0.060 for Salmonella, a 13% difference. This difference means that Salmonella grows normally at a supercoil level that would be toxic for E. coli ( 16, 89). The gyrB652 mutation of Salmonella lowers the linking number of pBR322 by approximately 1.5 topoisomers relative to the WT ( 68, 80), representing an additional 5% loss of supercoil density. This level is nonviable for E. coli. If the GyrB protein contributes to the control of mean supercoil density, this accounts for the dramatic effects of the GyrB652 protein in E. coli. Because gyrase is a heterotypic tetramer, GyrB652 expressed as only 10% of the total GyrB protein could alter nearly 20% of the gyrase molecules with a protomeric structure of GyrA 2-GyrB-GyrB652. Bacteria partition supercoils into unconstrained and constrained conformations (Δ LK = SU + SC) (see reference 43). Constrained supercoiling ( SC) is created by the unwound region of DNA bound to 3,000 molecules of transcribing RNA polymerase ( 26) and by the writhe associated with binding of abundant DNA binding proteins like HU, H-NS, and FIS ( 35, 46). Unconstrained supercoiling ( SU) causes torsion that stabilizes inter- and intramolecular triplex structures, left-handed Z-DNA, and cruciforms ( 35). In E. coli, SU and SC are distributed roughly 50:50 ( 5, 43, 71). To compare SU in E. coli and Salmonella, the Z-DNA-forming plasmid pRW478 was introduced in both organisms. pRW478 has a 56-bp repeating (CG) insert at the EcoRI site of pBR322; when SU is −0.025, 50% of the plasmid adopts a left-handed Z-DNA form in vitro ( 96). pRW478 was isolated from E. coli NH3716 and Salmonella NH3713 and analyzed on 2-D gels (Fig. ). Relative to that of Salmonella, the E. coli pRW478 gel pattern showed a striking difference. Most of the plasmid was hypersupercoiled with bands migrating into the upper-right sector of the gel. This pattern indicates Z-DNA formation in vivo, because once the 56-bp GC segment flips to the left-handed conformation, gyrase can introduce about a dozen additional negative supercoils ( 43). One can also see the transition between topoisomers that had a left-handed conformation in the first dimension but were relaxed to the right-handed conformation by the higher chloroquine concentration in the second dimension (seen as a break in the continuous line in the upper-right part of Fig. , right). Most of the E. coli plasmid population (>70% of the total) is hypersupercoiled relative to the 2-D profile for pBR322 (Fig. and ). For Salmonella, some plasmids have adopted the left-handed conformation, but the fraction is substantially less than that for E. coli (<40% of the total). Therefore, the σ value and the mean torsional strain are both higher in E. coli than in Salmonella. Species-specific phenotypes for SeqA and MukB. The differences in σ and SU in E. coli and Salmonella predict different regulatory response thresholds in each organism. For example, a supercoil-sensitive promoter tuned to the E. coli SU level would be inappropriately expressed in Salmonella unless a retuning mechanism alters the system. E. coli responded in a very negative way to the gyrB652 mutation, but how do other chromosome dynamics proteins behave? Null mutation in two genes, seqA and mukB, showed substantially different phenotypes when these two organisms were compared. The seqA gene encodes an A-methylation-sensitive DNA binding protein that recognizes the sequence GATC ( 51, 78, 88). SeqA binds preferentially to hemi-methylated GATC sites and sequesters the oriC region immediately after replication to block DnaA-dependent reinitiation of DNA synthesis for several minutes ( 6). E. coli seqA deletions were sensitive to growth in rich medium and failed to plate efficiently on rich LB plates at 30°C. A Salmonella seqA deletion plated efficiently on LB at 30°C (Fig. ) and at temperatures up to 40°C (not shown). Cell cytometry also showed that Salmonella seqA mutants grown in LB at room temperature did not have the filamentation phenotype that is characteristic of E. coli seqA mutants (data not shown). The second case is that of the bacterial “condensin” ( 37), a protein composed of MukB, MukE, and MukF ( 63, 64). E. coli deletions of mukB, mukE, or mukF exhibit the same phenotype, which is associated with growth rate toxicity, frequent loss of F plasmids, and segregation of DNA-less cells ( 62, 91). E. coli mukB deletions were nearly 100% viable in minimal medium, and they plated efficiently on LB medium as long as the temperature was 30°C or lower ( 74) (Fig. .) Salmonella mukB deletion mutants plated efficiently on minimal medium at 30°C but did not form colonies on LB plates (Fig. ). Growth in the 10° dilution spot (Fig. , bottom panel) is due to a low growth rate caused by high cell density and competition for nutrients, because spreading these cells across a whole plate leads to only a few colonies that appear to have suppressor mutations (data not shown). We interpret these results to indicate that WT Salmonella is more dependent on MukBEF for condensing and segregating chromosomes than E. coli ( 74). All organisms store DNA in a supercoiled conformation to condense the material into a functionally useful form. Eukaryotic organisms organize supercoils on the surface of a highly conserved histone octamer and separate the chromosome from most of the protein synthetic activity by the nuclear membrane. Bacteria have no nuclear membrane to separate chromosomal DNA from protein translation. Rather, they exploit the ATP-dependent enzyme DNA gyrase to supercoil DNA, which compacts the molecule by using torsional stress ( 10, 59, 69, 89). A bacterial cell must compact the DNA over 1,000-fold while it is being simultaneously transcribed and replicated ( 38, 85). Compaction in both prokaryotes and eukaryotes is assisted by structural maintenance of chromosome proteins, which are ATP-binding, pincer-like proteins that are conserved from bacteria to humans ( 9). The central importance of supercoiling in nucleoid formation requires that bacteria maintain DNA torsion levels within a narrow range (±15% of the normal value) ( 16, 89). Supercoil density is modulated by the activities of three topoisomerases, gyrase ( 56), Topo I ( 56), and Topo IV ( 97). These three enzymes are under strong selection to produce sufficient torsional strain to condense DNA, but not too much strain, which can trigger formation of alternative DNA conformations, such as Z-DNA, R loops, cruciforms, or intramolecular triplexes. These alternative DNA conformations stop replication forks, stall transcription complexes, and cause introduction of single- and double-strand chromosome breaks by several mechanisms ( 34). Species-specific gyrB mutations. The A-to-C transversion that causes the R436-S substitution in GyrB is nonviable in an otherwise WT strain of E. coli. The strongest evidence of a severe gyrB652 defect in E. coli comes from the electroporation of two PCR products into a mutS recombination-competent strain of E. coli (NH3704) (Fig. ). The single C-to-A substitution in a >2-kb PCR product (Fig. ) caused at least a 1,000-fold reduction in the transduction efficiency of a Kan r gene tightly linked to the gyrB locus. Moreover, the Salmonella gyrB652 allele had a deleterious impact on E. coli as a merodiploid. Modest expression of the WT Salmonella GyrB was toxic in WT E. coli, and the GyrB652 mutation was >10 times more toxic. A pGEM plasmid bearing Salmonella gyrB cloned behind a T7 promoter was selectively eliminated from 99.99% of viable cells that did not contain T7 RNA polymerase (Fig. ). Cells that spontaneously lost these plasmids rapidly overgrew cells with the plasmid. Although we do not know precisely how leaky the pRC03 GyrB652 expression was, it did not add enough protein to change the Western blot band intensity of WT E. coli GyrB (data not shown). Two- to threefold overexpression of E. coli gyrase subunits does not hinder E. coli's growth rate ( 44, 79), so Salmonella GyrB protein is exceptionally toxic in E. coli. Species-specific supercoil set points. WT strains of E. coli growing exponentially in LB medium maintained significantly higher supercoiled densities than Salmonella (σ values of −0.069 versus −0.059, respectively [Fig. ]). The analysis with plasmid pRW478 showed that E. coli also generated more torsional strain than Salmonella. In vitro, at SU levels of −0.025, half of the pRW478 molecules adopt a left-handed conformation. Most of the plasmid molecules isolated from E. coli were hypersupercoiled, indicating that the mean in vivo SU was >−0.025. The fraction of hypersupercoiled plasmid in Salmonella was less than half, indicating that the mean SU was <−0.025 (Fig. ). These facts suggest a supercoil explanation for phenotypic differences of mukB- and seqA-null mutations in otherwise WT strains of E. coli and Salmonella ( 38, 74). Our results suggest that GyrB is part of the mechanism controlling σ. There are 28 species-specific substitutions (3.4%) among 804 amino acid residues of GyrB and 78 changes (8.9%) in 878 amino acids for GyrA. A significant fraction of the 15% difference in σ measured under identical growth conditions is likely caused by one or more of these substitutions. Internal metabolic flux changes the cellular phosphorylation state and ATP/ADP ratio, which modulates gyrase supercoiling activity ( 41, 44, 79, 92). But changes of metabolic flux are modest relative to the 15% difference shown in Fig. and . Previous work indicated that E. coli had higher supercoiling levels than three other gram-negative organisms ( 43), so E. coli may have unusually high supercoiling among gram-negative organisms. Species-specific phenotypes of chromosome dynamics. Supercoiling is critical during dichotomous growth for efficient formation of new nucleoids. Significant differences exist between E. coli and Salmonella in the phenotypes associated with several topologically sensitive genes other than gyrase, including topA, seqA (Fig. ), mukB (Fig. ), and hns ( 21, 61). Although there may be alternative explanations in each specific case, the difference in WT supercoiling level between E. coli and Salmonella provides a unifying explanation for each of these observations. First, topA amber mutations are lethal in WT strains of E. coli but are relatively healthy in Salmonella enterica serovar Typhimurium or Shigella flexneri ( 3, 15, 52). High torsional strain caused by elimination of topA from E. coli leads to formation of transcription-driven intermolecular triplexes or R loops ( 17, 40). R loops impede transcription and stall replication forks ( 35). Because Salmonella maintains a 15% lower level of σ, its supercoil level is already lower than that of E. coli strains carrying gyrB compensatory mutations that make introduction of topA mutations permissible ( 15, 43). Thus, R-loop formation would not cause the same type of problem in Salmonella that it does in E. coli. Second, seqA-null mutants in E. coli exhibit growth rate toxicity in rich LB medium while Salmonella does not (Fig. ). In E. coli, SeqA helps quell the lethal consequence of overinitiation by sequestering the hemi-methylated oriC region for many minutes after initiation, thus dampening DnaA-dependent replisome reassembly ( 8). Rapid reinitiation of replication at oriC leads to filamentation and cell death due replication fork failure ( 28, 77). One critical step in oriC initiation is the unwinding of A/T-rich sequences adjacent to DnaA binding sites at the origin ( 1, 24, 25). The 15% lower supercoil density that Salmonella attains during dichotomous replication may decrease oriC initiation to acceptable levels in rich media without the contribution of SeqA sequestration. Third, an exception that proves the rule is the mukB mutation, which has a much more striking phenotype in Salmonella than in E. coli (Fig. ). An E. coli mukB mutant plates with equal efficiency on minimal and LB medium at 30°C. Our Salmonella mukB mutant is viable on minimal medium but forms colonies on LB medium 6 orders of magnitude less efficiently than E. coli. In E. coli, supercoiling influences the phenotypes associated with mukB mutants. Sawitzke and Austin demonstrated that growth in the presence of low levels of the GyrB inhibitor novobiocin increased the reliance of E. coli on MukB, whereas increasing the median supercoil density by introducing a topA mutation made it possible for E. coli mukB mutants to plate on LB up to 42°C ( 38, 74). Salmonella fits this pattern. The lower level of σ correlates with growth of E. coli on novobiocin and a consequential heavier reliance on MukB for condensation and segregation in rich medium (Fig. ). The fourth example is the hns gene, which encodes a highly conserved gene for the abundant DNA binding protein H-NS. Like MukB, deletion of H-NS results in a more severe phenotype in Salmonella than it does in E. coli, where the growth rates and general physiologies of WT and hns-null mutations are nearly indistinguishable ( 21, 58). H-NS participates in chromosome condensation, regulation of gene expression, and the targeting of transposons ( 46). In E. coli, H-NS influences expression of about >200 genes directly or indirectly ( 39), while experiments with Salmonella indicate that expression levels of over 400 genes are altered in an hns-null mutant ( 61). Most of the genes in both organisms are derepressed when H-NS is eliminated. In addition to regulating genes, H-NS is one of the proteins that may organize loops in the bacterial chromosome ( 14). H-NS-induced looping in vitro has been demonstrated by atomic force microscopy ( 12) as well as by single-molecule manipulation using optical tweezers ( 11). When expressed, H-NS can cause a dramatic condensation of the nucleoid, which leads to inhibition of global transcription and an artificial stationary phase ( 55). The lower supercoiling level of Salmonella may contribute to the increased dependence of this organism on H-NS for maintaining nucleoid structure. Many of the genes that are bound by H-NS in Salmonella appear to be A/T rich and transferred horizontally from distantly related organisms ( 61). Why different supercoil densities in related enterics? Whereas the superhelical difference in Salmonella and E. coli provides a reasonable explanation for phenotypic differences in the mukB, seqA, gyrB, topA, and hns mutants, what selective force could lead to an altered state of mean superhelicity? We do not know the answer to this question. One strong selective pressure may come from prophages in the Salmonella genome (Fig. ). E. coli K-12 harbors no prophage that can make an infectious virus in a lytic cycle, whereas S. enterica serovar Typhimurium contains four lysis-competent prophages; they are Gifsy-1, Gifsy-2, Fels-1, and Fels-2 ( 54). How long Salmonella enterica serovar Typhimurium has harbored four prophages is uncertain, but Gifsy-1 and Gifsy-2 contain genes that contribute to pathogenicity in infected mice ( 22). This could explain the selective pressure to preserve these elements. Phage development changes in complex ways with the different growth conditions of the host. For lytic phages like T4, adsorption rate and burst size decrease at least an order of magnitude under poor growth conditions while eclipse time increases ( 30). For lysogenic phages Mu and λ, the lytic/lysogeny decision changes with physiology. Neither λ nor Mu produces significant single-cell bursts in stationary phase, and gyrase modulates the lytic/lysogeny outcome in both viruses. λ excision is reduced from some locations in the bacterial chromosome in himB ( gyrB) mutants ( 57, 58). In E. coli, Mu burst size and plaque size are reduced in gyrB mutants ( 94) and the lysis/lysogeny fate is directly linked to negative supercoiling ( 33, 48). To test the influence of host environment on the lytic/lysogenic behavior of Mu, plaque-forming MupAp1 monolysogens were isolated in E. coli (NH1126) and Salmonella (NH742), which also carries the musA allele that changes the surface polysaccharide structure to permit Mu adsorption ( 20). Mu cts62 pAp1 carries the TS cts62 repressor allele ( 87). At 30°C, the Mu cts62 repressor binds to the operator and blocks p E promoter activity, but a shift to 42°C induces conformation change in the C-terminal domain of the repressor, causing immediate release of DNA and initiation of the lytic transcription pathway ( 86). Duplicate overnight cultures were diluted 100-fold into fresh LB broth and incubated at 30°C to an optical density at 600 nm (OD 600) of 0.6, which corresponds to approximately 5 × 10 8 CFU/ml. Cultures were shifted to a 42°C shaking water bath, and the OD 600 as well as the number of CFU that plated on LB agar at 30°C were sampled at 10-min intervals for the next 80 min. The OD 600s of E. coli cultures dropped 30 min after the temperature shift, while Salmonella cultures started lysis 10 min later (Fig. ). Thus, the eclipse phase of phage development was longer in Salmonella than in E. coli. More dramatically, the numbers of cells surviving the shift to 42°C were different. In E. coli, the CFU number fell immediately after the shift and rapidly declined. By 50 min, the CFU number had dropped over 8 logs and no survivors were detectable. In Salmonella, the CFU number also dropped immediately after the temperature shift, but it leveled off at 20 min, and the viable cell count remained at about 10 7 CFU/ml for the next hour. Whereas both species are sensitive to the Mu lytic cycle, Salmonella survives thermo-induction 7 orders of magnitude better than E. coli. The E. coli-Salmonella comparison. The supercoiling behavior of Salmonella differs from that of E. coli in at least two ways, either or both of which might contribute to the dominant-negative phenotype of expressing the Salmonella GyrB protein in E. coli. The first difference is that E. coli generates 13% more negative supercoils than Salmonella during rapid growth in rich medium (Fig. and ). Recent data from single-molecule experiments present new insight into the gyrase reaction mechanism. Nollmann et al. found that E. coli gyrase relaxes negative supercoils efficiently in the absence of ATP or other cofactors when DNA is held under strain from 0.5 to 1 pN ( 65). A model in which relaxation of negative supercoils occurs by a branched topoisomerase pathway in which supercoil relaxation is triggered by the energetic strain of negative supercoils has been proposed ( 32), i.e., the supercoil set point could be controlled by the level of DNA strain that causes gyrase to stop supercoiling and start relaxing negative supercoils. If this is correct, then a subset of the 28-amino-acid substitution that differentiates the E. coli and Salmonella GyrB proteins controls supercoil levels. However, a second species difference is shown in Fig. S10 in the supplemental material. As E. coli enters stationary phase, the σ of pBR322 falls from −0.069 at the peak growth rate to below −0.055 as cells enter early stationary phase. This has been seen before in E. coli ( 43, 73). But in Salmonella, there is no loss of σ when cells enter or exit stationary phase. What controls the changes in supercoil levels in E. coli is not known, but the supercoiling recovery does not require new protein synthesis ( 73). One possibility is the GyrI gyrase inhibitory protein, which is under SOS regulatory control ( 60). Mechanisms notwithstanding, there is at least one system that modulates gyrase activity in stationary-phase E. coli that does not work in Salmonella. The separation of E. coli and Salmonella from a common ancestor occurred about 200 million years before the present ( 50), which is about the time of separation of the lineages for mouse and human. Both organisms retain approximately 65% of their homologous genes, with 35% of the genome moving via horizontal transfer, but the genetic map is largely conserved. Maintaining four prophages that can all kill the cell may mean that Salmonella's fitness is improved by damping the lysis/lysogeny trigger. The four Salmonella prophages have expression and gene amplification effects that can be monitored using microarray technology under stress conditions ( 23). On the other hand, selection pressure in E. coli may have increased the supercoiling level to maximize its growth rate under dichotomous conditions ( E. coli outgrows Salmonella in mixed culture). One added benefit might be that higher supercoiling levels lead to rapid curing once E. coli is lysogenized by a prophage. Recent work shows that Mu transposition hotspots in Salmonella are different from those in E. coli (D. Manna, S. Porwollik, M. McClelland, and N. P. Higgins, submitted for publication), and lower negative supercoiling could account for different locations of unusual DNA structure that act as transposition hotspots ( 93). If the supercoiling set point is established by the gyrA and gyrB genes, it would be very interesting to see how Salmonella fitness and dichotomous growth rates would change after swapping the normal Salmonella gyrase alleles with E. coli gyrA and gyrB sequences. We are grateful to Stuart Austin for carrying out flow cytometry on WT and seqA mutants of Salmonella and for providing seqA and mukB mutant strains of E. coli, to Don Court for providing the recombineering plasmid pSim5 prior to publication, and to members of the Higgins laboratory for critical comments on the manuscript. This work was supported by grant GM33143 to N.P.H. 1. Baker, T. A., K. Sekimizu, B. E. Funnell, and A. Kornberg. 1986. Extensive unwinding of the plasmid template during staged enzymatic initiation of DNA replication from the origin of the Escherichia coli chromosome. Cell 45:53-64. [PubMed] 2. Bates, D., and N. Kleckner. 2005. Chromosome and replisome dynamics in E. coli: loss of sister cohesion triggers global chromosome movement and mediates chromosome segregation. Cell 121:899-911. [PubMed] 3. Bhriain, N. N., and C. J. Dorman. 1993. Isolation and characterization of a topA mutant of Shigella flexneri. Mol. Microbiol. 7:351-358. [PubMed] 4. Bidnenko, V., S. D. Ehrlich, and B. Michel. 2002. Replication fork collapse at replication terminator sequences. EMBO J. 21:3898-3907. [PubMed] 5. Bliska, J. B., and N. R. Cozzarelli. 1987. Use of site-specific recombination as a probe of DNA structure and metabolism in vivo. J. Mol. Biol. 194:205-218. [PubMed] 6. Boye, E., T. Stokke, N. Kleckner, and K. Skarstad. 1996. Coordinating DNA replication initiation with cell growth: differential roles for DnaA and SeqA proteins. Proc. Natl. Acad. Sci. USA 93:12206-12211. [PubMed] 7. Cabral, J. H. M., A. P. Jackson, C. V. Smith, N. Shikotra, A. Maxwell, and R. C. Siddington. 1997. Crystal structure of the breakage-reunion domain of DNA gyrase. Nature 388:903. [PubMed] 8. Camara, J., and E. Crooke. 2005. Initiation of chromosomal replication, p. 177-191. In N. P. Higgins (ed.), The bacterial chromosome. ASM Press, Washington, DC. 9. Cobbe, N., and M. M. Heck. 2000. Review: SMCs in the world of chromosome biology—from prokaryotes to higher eukaryotes. J. Struct. Biol. 129:123-143. [PubMed] 10. Cozzarelli, N. R., and J. C. Wang. 1990. DNA topology and its biological effects. Cold Spring Harbor Press, Cold Spring Harbor, NY. 11. Dame, R. T., M. C. Noom, and G. J. Wuite. 2006. Bacterial chromatin organization by H-NS protein unravelled using dual DNA manipulation. Nature 444:387-390. [PubMed] 12. Dame, R. T., C. Wyman, and N. Goosen. 2000. H-NS mediated compaction of DNA visualised by atomic force microscopy. Nucleic Acids Res. 28:3504-3510. [PubMed] 13. Datta, S., N. Costantino, and D. L. Court. 2006. A set of recombineering plasmids for gram-negative bacteria. Gene 379:109-115. [PubMed] 14. Deng, S., R. A. Stein, and N. P. Higgins. 2005. Organization of supercoil domains and their reorganization by transcription. Mol. Microbiol. 57:1511-1521. [PubMed] 15. DiNardo, S., K. A. Voelkel, R. Sternglanz, A. E. Reynolds, and A. Wright. 1982. Escherichia coli DNA topoisomerase I mutants have compensatory mutations in DNA gyrase genes. Cell 31:43-51. [PubMed] 16. Drlica, K. 1992. Control of bacterial supercoiling. Mol. Microbiol. 6:425-433. [PubMed] 17. Drolet, M., S. Broccoli, F. Rallu, C. Hraiky, C. Fortin, E. Masse, and I. Baaklini. 2003. The problem of hypernegative supercoiling and R-loop formation in transcription. Front. Biosci. 8:D210-D221. [PubMed] 18. Ellis, H. M., D. Yu, T. DiTizio, and D. L. Court. 2001. High efficiency mutagenesis, repair, and engineering of chromosomal DNA using single-stranded oligonucleotides. Proc. Natl. Acad. Sci. USA 98:6742-6746. [PubMed] 19. Espeli, O., and K. J. Marians. 2004. Untangling intracellular DNA topology. Mol. Microbiol. 52:925-931. [PubMed] 20. Faelen, M., M. Mergeay, J. Gerits, A. Toussaint, and N. Lefebvre. 1981. Genetic mapping of a mutation conferring sensitivity to bacteriophage Mu in Salmonella typhimurium LT2. J. Bacteriol. 146:914-919. [PubMed] 21. Falconi, M., V. McGovern, C. Gualerzi, D. Hillyard, and N. P. Higgins. 1991. Mutations altering chromosomal protein H-NS induce mini-Mu transposition. New Biol. 3:615-625. [PubMed] 22. Figueroa-Bossi, N., and L. Bossi. 1999. Inducible prophages contribute to Salmonella virulence in mice. Mol. Microbiol. 33:167-176. [PubMed] 23. Frye, J. G., S. Porwollik, F. Blackmer, P. Cheng, and M. McClelland. 2005. Host gene expression changes and DNA amplification during temperate phage induction. J. Bacteriol. 187:1485-1492. [PubMed] 24. Funnell, B. E., T. A. Baker, and A. Kornberg. 1986. Complete enzymatic replication of plasmids containing the origin of the Escherichia coli chromosome. J. Biol. Chem. 261:5616-5624. [PubMed] 25. Funnell, B. E., T. A. Baker, and A. Kornberg. 1987. In vitro assembly of a pre-priming complex at the origin of the Escherichia coli chromosome. J. Biol. Chem. 262:10327-10334. [PubMed] 26. Gamper, H. B., and J. E. Hearst. 1982. A topological model for transcription based on unwinding angle analysis of E. coli RNA polymerase binary, initiation and ternary complexes. Cell 29:81-90. [PubMed] 27. Gari, E., N. Figueroa-Bossi, A.-B. Blanc-Potard, F. Spirito, M. B. Schmid, and L. Bossi. 1996. A class of gyrase mutants of Salmonella typhimurium show quinolone-like lethality and require Rec functions for viability. Mol. Microbiol. 21:111-122. [PubMed] 28. Grigorian, A. V., R. B. Lustig, E. C. Guzman, J. M. Mahaffy, and J. W. Zyskind. 2003. Escherichia coli cells with increased levels of DnaA and deficient in recombinationalrepair have decreased viability. J. Bacteriol. 185:630-644. [PubMed] 29. Grompone, G., S. D. Ehrlich, and B. Michel. 2003. Replication restart in gyrB Escherichia coli mutants. Mol. Microbiol. 48:845-854. [PubMed] 30. Hadas, H., M. Einav, I. Fishov, and A. Zaritsky. 1997. Bacteriophage T4 development depends on the physiology of its host Escherichia coli. Microbiology 143:179-185. [PubMed] 31. Hallett, P., A. J. Grimshaw, D. B. Wigley, and A. Maxwell. 1990. Cloning of the DNA gyrase genes under tac promoter control: overproduction of the gyrase A and B proteins. Gene 93:139-142. [PubMed] 32. Higgins, N. P. 2007. Under DNA stress, gyrase makes the sign of the cross. Nat. Struct. Mol. Biol. 14:256-258. [PubMed] 33. Higgins, N. P., D. A. Collier, M. W. Kilpatrick, and H. M. Krause. 1989. Supercoiling and integration host factor change the DNA conformation and alter the flow of convergent transcription in phage Mu. J. Biol. Chem. 264:3035-3042. [PubMed] 34. Higgins, N. P., S. Deng, Z. Pang, R. Stein, K. Champion, and D. Manna. 2005. Domain behavior and supercoil dynamics in bacterial chromosomes, p. 133-153. In N. P. Higgins (ed.), The bacterial chromosome. ASM Press, Washington, DC. 35. Higgins, N. P., and A. Vologodskii. 2004. Topological behavior of plasmid DNA, p. 181-201. In B. E. Funnell and G. J. Phillips (ed.), Plasmid biology. ASM Press, Washington, DC. 36. Higgins, N. P., X. Yang, Q. Fu, and J. R. Roth. 1996. Surveying a supercoil domain by using the gd resolution system in Salmonella typhimurium. J. Bacteriol. 178:2825-2835. [PubMed] 37. Hirano, T. 2002. The ABCs of SMC proteins: two-armed ATPases for chromosome condensation, cohesion, and repair. Genes Dev. 16:399-414. [PubMed] 38. Holmes, V. F., and N. R. Cozzarelli. 2000. Closing the ring: links between SMC proteins and chromosome partitioning, condensation, and supercoiling. Proc. Natl. Acad. Sci. USA 97:1322-1324. [PubMed] 39. Hommais, F., E. Krin, C. Laurent-Winter, O. Soutourina, A. Malpertuy, J. P. Le Caer, A. Danchin, and P. Bertin. 2001. Large-scale monitoring of pleiotropic regulation of gene expression by the prokaryotic nucleoid-associated protein, H-NS. Mol. Microbiol. 40:20-36. [PubMed] 40. Hraiky, C., M.-A. Raymond, and M. Drolet. 2000. RNase H overproduction corrects a defect at the level of transcription elongation during rRNA synthesis in the absence of DNA topoisomerase I in Escherichia coli. J. Biol. Chem. 275:11257-11263. [PubMed] 41. Hsieh, L.-S., J. Rouviere-Yaniv, and K. Drlica. 1991. Bacterial DNA supercoiling and ATP/ADP ratio: changes associated with salt shock. J. Bacteriol. 173:3914-3917. [PubMed] 42. Jacob, F., S. Brenner, and F. Cuzin. 1963. On the regulation of DNA replication in bacteria. Cold Spring Harbor Symp. Quant. Biol. 28:329-348. 43. Jaworski, A., N. P. Higgins, R. D. Wells, and W. Zacharias. 1991. Topoisomerase mutants and physiological conditions control supercoiling and Z-DNA formation in vivo. J. Biol. Chem. 266:2576-2581. [PubMed] 44. Jensen, P. R., C. C. van der Weijden, L. B. Jensen, H. V. Westerhoff, and J. L. Snoep. 1999. Extensive regulation compromises the extent to which DNA gyrase controls DNA supercoiling and growth rate of Escherichia coli. Eur. J. Biochem. 226:865-877. 45. Jeong, K. S., J. Ahn, and A. B. Khodursky. 2004. Spatial patterns of transcriptional activity in the chromosome of Escherichia coli. Genome Biol. 5:R86. [PubMed] 46. Johnson, R. C., L. M. Johnson, J. W. Schmidt, and J. F. Gardner. 2005. The major nucleoid proteins in the structure and function of the E. coli chromosome, p. 65-132. In N. P. Higgins (ed.), The bacterial chromosome. ASM Press, Washington, DC. 47. 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. [PubMed] 48. Krause, H. M., and N. P. Higgins. 1986. Positive and negative regulation of the Mu operator by Mu repressor and Escherichia coli integration host factor. J. Biol. Chem. 261:3744-3752. [PubMed] 49. Kuempel, P., J. Henson, L. Dircks, M. Tecklenburg, and D. Lim. 1991. dif, a recA-independent recombination site in the terminus region of the chromosome of Escherichia coli. New Biol. 3:799-811. [PubMed] 50. Lawrence, J. 2005. The dynamic bacterial genome, p. 19-37. In N. P. Higgins (ed.), The bacterial chromosome. ASM Press, Washington, DC. 51. Lu, M., J. L. Campbell, E. Boye, and N. Kleckner. 1994. SeqA: a negative modulator of replication initiation in E. coli. Cell 77:413-426. [PubMed] 52. Margolin, P., L. Zumstein, R. Sternglanz, and J. C. Wang. 1985. The Escherichia coli supX locus is topA, the structural gene for DNA topoisomerase I. Proc. Natl. Acad. Sci. USA 82:5437-5441. [PubMed] 53. Marti, T. M., C. Kunz, and O. Fleck. 2002. DNA mismatch repair and mutation avoidance pathways. J. Cell. Phys. 191:28-41. 54. McClelland, M., K. E. Sanderson, J. Spieth, S. W. Clifton, P. Latreille, L. Courtney, S. Porwollik, J. Ali, M. Dante, F. Du, S. Hou, D. Layman, S. Leonard, C. Nguyen, K. Scott, A. Holmes, N. Grewal, E. Mulvaney, E. Ryan, H. Sun, L. Florea, W. Miller, T. Stoneking, M. Nhan, R. Waterston, and R. K. Wilson. 2001. The complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413:852-856. [PubMed] 55. McGovern, V., N. P. Higgins, S. Chiz, and A. Jaworski. 1994. H-NS over-expression induces an artificial stationary phase by silencing global transcription. Biochimie 76:1030-1040. [PubMed] 56. Menzel, R., and M. Gellert. 1983. Regulation of the genes for E. coli DNA gyrase: homeostatic control of DNA supercoiling. Cell 34:105-113. [PubMed] 57. Miller, H. I., J. Abraham, M. Benedik, A. Campbell, D. Court, H. Echols, R. Fischer, J. M. Galindo, G. Guarneros, T. Hernandez, D. Mascarenhas, C. Montanez, D. Schindler, U. Schmeissner, and L. Sosa. 1981. Regulation of the integration-excision reaction by bacteriophage lambda. Cold Spring Harbor Symp. Quant. Biol. 45 Pt. 1:439-445. 58. Miller, H. I., A. Kikuchi, H. A. Nash, R. A. Weisberg, and D. I. Friedman. 1979. Site-specific recombination of bacteriophage l: the role of host gene products. Cold Spring Harbor Symp. Quant. Biol. 43:1121-1126. [PubMed] 59. Miller, W. G., and R. W. Simons. 1993. Chromosomal supercoiling in Escherichia coli. Mol. Microbiol. 10:675-684. [PubMed] 60. Nakanishi, A., S. Imajoh-Ohmi, and F. Hanaoka. 2002. Characterization of the interaction between DNA gyrase inhibitor and DNA gyrase of Escherichia coli. J. Biol. Chem. 277:8949-8954. [PubMed] 61. Navarre, W. W., S. Porwollik, Y. Wang, M. McClelland, H. Rosen, S. J. Libby, and F. C. Fang. 2006. Selective silencing of foreign DNA with low GC content by the H-NS protein in Salmonella. Science 313:236-238. [PubMed] 62. Niki, H., and S. Hiraga. 1997. Subcellular distribution of actively partitioning F plasmid during the cell division cycle in E. coli. Cell 90:951-957. [PubMed] 63. Niki, H., R. Imamura, M. Kitaoka, K. Ymanaka, T. Ogura, and S. Hiraga. 1992. E. coli MukB protein involved in chromosome partition forms a homodimer with a rod-and-hinge strucure having DNA binding and ATP/GTp binding activities. EMBO J. 11:5101-5109. [PubMed] 64. Niki, H., A. Jaffe, R. Imamura, T. Ogura, and S. Hiraga. 1991. The new gene mukB codes for a 177 kd protein with coiled-coil domains involved in chromosome partitioning of E. coli. EMBO J. 10:183-193. [PubMed] 65. Nollmann, M., M. D. Stone, Z. Bryant, N. J. Crisona, S. C. Hong, S. Mitelheiser, A. Maxwell, C. Bustamante, and N. R. Cozzarelli. 2007. Multiple modes of Escherichia coli DNA gyrase activity revealed by force and torque. Nat. Struct. Mol. Biol. 14:264-271. [PubMed] 66. Novick, A., and M. Weiner. 1957. Enzyme induction as an all-or-none phenomenon. Proc. Natl. Acad. Sci. USA 43:553-566. [PubMed] 67. Ozbudak, E. M., M. Thattai, H. N. Lim, B. I. Shraiman, and A. van Oudenaarden. 2004. Multistability in the lactose utilization network of Escherichia coli. Nature 427:737-740. [PubMed] 68. Pang, Z., R. Chen, D. Manna, and N. P. Higgins. 2005. A gyrase mutant with low activity disrupts supercoiling at the replication terminus. J. Bacteriol. 187:7773-7783. [PubMed] 69. Pavitt, G. D., and C. F. Higgins. 1993. Chromosomal domains of supercoiling in Salmonella typhimurium. Mol. Microbiol. 10:685-696. [PubMed] 70. Peck, L. J., and J. C. Wang. 1983. Energetics of B-to-Z transition in DNA. Proc. Natl. Acad. Sci. USA 80:6206-6210. [PubMed] 71. Pettijohn, D. E., and O. Pfenninger. 1980. Supercoils in prokaryotic DNA restrained in vivo. Proc. Natl. Acad. Sci. USA 77:1331-1335. [PubMed] 72. Reece, R., and A. Maxwell. 1991. DNA gyrase: structure and function. Crit. Rev. Biochem. Mol. Biol. 26:335-375. [PubMed] 73. Reyes-Dominguez, Y., G. Contreras-Ferrat, J. Ramirez-Santos, J. Membrillo-Hernandez, and M. C. Gomez-Eichelmann. 2003. Plasmid DNA supercoiling and gyrase activity in Escherichia coli wild-type and rpoS stationary-phase cells. J. Bacteriol. 185:1097-1100. [PubMed] 74. Sawitzke, J. A., and S. Austin. 2000. Suppression of chromosome segregation defects of Escherichia coli muk mutants by mutations in topoisomerase I. Proc. Natl. Acad. Sci. USA 97:1671-1676. [PubMed] 75. Shure, M., and J. Vinograd. 1976. The number and superhelical turns in native virion SV40 DNA and minicol DNA determined by the band counting method. Cell 8:215-226. [PubMed] 76. Siegele, D. A., and J. C. Hu. 1997. Gene expression from plasmids containing the araBAD promoter at subsaturating inducer concentrations represents mixed populations. Proc. Natl. Acad. Sci. USA 94:8168-8172. [PubMed] 77. Simmons, L. A., A. M. Breier, N. R. Cozzarelli, and J. M. Kaguni. 2004. Hyperinitiation of DNA replication in Escherichia coli leads to replication fork collapse and inviability. Mol. Microbiol. 51:349-358. [PubMed] 78. Slater, S., S. Wold, M. Lu, E. Boye, K. Skarstad, and N. Kleckner. 1995. E. coli SeqA protein binds oriC in two different methyl-modulated reactions appropriate to its roles in DNA replication initiation and origin sequestration. Cell 82:927-936. [PubMed] 79. Snoep, J. L., C. C. van der Weijden, H. W. Andersen, H. V. Westerhoff, and P. R. Jensen. 2002. DNA supercoiling in Escherichia coli is under tight and subtle homeostatic control, involving gene-expression and metabolic regulation of both topoisomerase I and DNA gyrase. Eur. J. Biochem. 269:1662-1669. [PubMed] 80. Staczek, P., and N. P. Higgins. 1998. DNA gyrase and Topoisomerase IV modulate chromosome domain size in vivo. Mol. Microbiol. 29:1435-1448. [PubMed] 81. Stark, M. J. R. 1987. Multicopy expression vectors carrying the lac repressor gene for regulated high-level expression of genes in Escherichia coli. Gene 51:255-267. [PubMed] 82. Sternglanz, R., S. DiNardo, K. A. Voelkel, Y. Nishimura, Y. Hirota, K. Becherer, L. Zumstein, and J. C. Wang. 1981. Mutations in the gene coding for Escherichia coli DNA topoisomerase I affect transcription and transposition. Proc. Natl. Acad. Sci. USA 78:2747-2751. [PubMed] 83. Studier, F. W. 1991. Use of bacteriophage T7 lysozyme to improve an inducible T7 expression system. J. Mol. Biol. 219:37-44. [PubMed] 84. Sutcliffe, J. G. 1979. Complete nucleotide sequence of the Escherichia coli plasmid pBR322. Cold Spring Harbor Symp. Quant. Biol. 43:77-90. [PubMed] 85. Trun, N. J., and J. F. Marko. 1998. Architecture of a bacterial chromosome. ASM News 64:276-283. 86. Vogel, J. L., V. Geuskens, L. Desmet, N. P. Higgins, and A. Toussaint. 1996. C-terminal deletions can stabilize temperature sensitive Mu repressor mutants and block Clp-dependent proteolysis. Genetics 142:661-672. [PubMed] 87. Vogel, J. L., J. L. Zhu, M. M. Howe, A. Toussaint, and N. P. Higgins. 1991. Temperature sensitive mutations in the bacteriophage Mu c repressor locate a 63-amino-acid DNA-binding domain. J. Bacteriol. 173:6568-6577. [PubMed] 88. von Freiesleben, U., K. V. Rasmussen, and M. Schaechter. 1994. SeqA limits DnaA activity in replication from oriC in Escherichia coli. Mol. Microbiol. 14:763-772. [PubMed] 89. Wang, J. C. 1985. DNA topoisomerases. Annu. Rev. Biochem. 54:665-697. [PubMed] 90. Wang, X., X. Liu, C. Possoz, and D. J. Sherratt. 2006. The two Escherichia coli chromosome arms locate to separate cell halves. Genes Dev. 20:1727-1731. [PubMed] 91. Weitao, T., S. Dasgupta, and K. Nordstrom. 2000. Role of the mukB gene in chromosome and plasmid partition in Escherichia coli. Mol. Microbiol. 38:392-400. [PubMed] 92. Westerhoff, H. V., M. H. ODea, A. Maxwell, and M. Gellert. 1988. DNA supercoiling by DNA gyrase. A static head analysis. Cell Biophys. 12:157-181. [PubMed] 93. Yanagihara, K., and K. Mizuuchi. 2002. Mismatch-targeted transposition of Mu: a new strategy to map genetic polymorphism. Proc. Natl. Acad. Sci. USA 99:11317-11321. [PubMed] 94. Yoshida, R. K., J. K. Miller, H. I. Miller, K. I. Friedman, and M. M. Howe. 1982. Isolation and mapping of Mu nu mutants which grow in him mutants of E. coli. Virology 120:269-272. [PubMed] 95. Yu, D., H. E. Ellis, E.-C. Lee, N. A. Jenkins, N. G. Copeland, and D. L. Court. 2000. An efficient recombination system for chromosome engineering in Escherichia coli. Proc. Natl. Acad. Sci. USA 97:5978-5983. [PubMed] 96. Zacharias, W., A. Jaworski, J. E. Larson, and R. D. Wells. 1988. The B- to Z-DNA equilibrium in vivo is perturbed by biological processes. Proc. Natl. Acad. Sci. USA 85:7069-7073. [PubMed] 97. Zechiedrich, E. L., A. B. Khodursky, S. Bachellier, R. Schneider, D. Chen, D. M. Lilley, and N. R. Cozzarelli. 2000. Roles of topoisomerases in maintaining steady-state DNA supercoiling in Escherichia coli. J. Biol. Chem. 275:8103-8113. [PubMed]
|
