Control of Residues in Live Animals and Animal Products. Results 2005, plan 2006. Originating from the Faroe Islands (FO), Pursuant to Council Directive 96/23/EC. PRESENTATION OF THE RESIDUE CONTROL 2005 RESIDUE CONTROL PLAN 2006 Country: Faroe Date: 30th March 2006 Commission Reference Number (Stamp): Period Covered:
1471-2199-8-44.fmBMC Molecular Biology
Hin-mediated DNA knotting and recombining promote replicon
dysfunction and mutation
Richard W Deibler†1,2,3, Jennifer K Mann†2,4, De Witt L Sumners4 and
Address: 1Interdepartmental Program in Cell and Molecular Biology, Baylor College of Medicine, Houston, Texas 77030-3411 USA, 2Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, Texas 77030-3411 USA, 3Department of Systems Biology, Harvard Medical School, Boston, Massachusetts 02115 USA and 4Department of Mathematics, Florida State University, Tallahassee, Florida 32306-4510 USA Email: Richard W Deibler* - [email protected]; Jennifer K Mann - [email protected]; De Witt L Sumners - [email protected]; Lynn Zechiedrich - [email protected] * Corresponding author Published: 25 May 2007 Received: 23 January 2007Accepted: 25 May 2007 BMC Molecular Biology 2007, 8:44
2007 Deibler et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: The genetic code imposes a dilemma for cells. The DNA must be long enough to
encode for the complexity of an organism, yet thin and flexible enough to fit within the cell. The
combination of these properties greatly favors DNA collisions, which can knot and drive
recombination of the DNA. Despite the well-accepted propensity of cellular DNA to collide and
react with itself, it has not been established what the physiological consequences are.
Results: Here we analyze the effects of recombined and knotted plasmids in E. coli using the Hin
site-specific recombination system. We show that Hin-mediated DNA knotting and recombination
(i) promote replicon loss by blocking DNA replication; (ii) block gene transcription; and (iii) cause
genetic rearrangements at a rate three to four orders of magnitude higher than the rate for an
unknotted, unrecombined plasmid.
Conclusion: These results show that DNA reactivity leading to recombined and knotted DNA is
potentially toxic and may help drive genetic evolution.
crystalline state of 80 – 100 mg/ml ]. Understanding Much of DNA metabolism is understood in the context of how DNA functions requires understanding its conforma- the linear sequence of nucleotides that compose the tion under such compact conditions.
nucleic acid. For example, gene promoters, replication ori-gins, partitioning sequences and genes themselves are DNA conformation is affected not only by crowding but defined by their particular DNA sequences. However, the also by its physical structure. Intuitively, anything long, physical, mechanical and topological properties of DNA thin and flexible can become self-entangled. Interestingly, also exert significant influence over DNA metabo].
for 200 kb DNA molecules at thermal equilibrium, the Inside cells, the long (1.6 mm for Escherichia coli) and flex- most energetically favorable conformation is the trefoil ible (persistence length ≈ 50 nm) DNA must be com- knot, 31 is 20-fold smaller than the chromo- pacted into a very small volume, achieving a liquid some of E. coli. Thus, it is not surprising that when cells are (page number not for citation purposes) BMC Molecular Biology 2007, 8:44
lysed, a small portion ( 1%) of plasmid DNA, which is intracellular DNA can alter its activity. This system ties only on the order of 4 kb, is found knotted [. The knots topologically identical to those observed in vivo propensity for DNA to knot is predicted to be even greater . Although studying the effects of knots in chro- for the longer and more folded eukaryotic chromosomes mosomal DNA would be optimal, it is not technically fea- ver, if we apply the figure of 1% DNA knotting sible because there is no direct way to measure to human chromosomes, then nearly every other diploid chromosomal knotting. Therefore, we have examined human cell would have a knot.
what happens when DNA strands collide to recombineand knot a 5.4 kb plasmid containing a gene required for Although DNA knotting is clearly energetically favorable cell survival. Plasmids appear to be a reasonable model for DNA, several observations suggest that the intracellu- for chromosomal metabolism. For example, supercoiling lar environment should further exacerbate knotting.
changes in reporter plasmids ] mirror changes in the Experiments with the bacteriophage P4 demonstrated that supercoiling of the chromosome . The recombined the confinement of DNA in a small volume stimulates the plasmid products generated by Hin are easily analyzed knotting of DNA . Furthermore, DNA inside the cell is because of their small size. A recombination event occur- negatively supercoiled. Negative supercoiling promotes a ring in the chromosome would be much more difficult to number of genetic processes, including gene expression detect. Although Hin recombines and knots at the hix and DNA replication, in part because it promotes opening sites, the resulting knots can move during DNA metabo- of the DN]. DNA supercoiling also com- lism. On the chromosome, this knot sliding could be as pacts the DNA and brings distant strands into close prox- far as the size of a topological dom], which imity [uence, supercoiling promotes would be more difficult to detect experimentally.
strand collision and DNA tangling. Indeed, computer sim-ulations have revealed that supercoiling should drive Here we show that Hin-mediated site-specific recombina- DNA knotting because writhe in a knot is less stressful on tion and knotting led to dysfunction of the replicon and the DNA than writhe in an unknotted, supercoiled mole- blocked expression of a gene on the plasmid. This process is highly mutagenic, and our results suggest that unlessrecombination and knotting are carefully controlled, Collisions of DNA helices with one another are poten- intracellular DNA can be unstable. We suggest that such tially problematic because DNA is a self-reactive mole- instability of the genetic material could help drive evolu- cule. The repair of double strand breaks, single strand gaps tionary variation.
and stalled replication forks involve recombination,which requires physical contact with a homologous DNA molecule. Similarly, transposition, site-specific recombi- nation and modulation of transcription (by enhancers The experimental approach we use here to study the cellu- and other cis-regulatory elements) often involve DNA- lar effect of recombining and knotting DNA is outlined in DNA interactions. However, it has not been well estab- Figure ave shown previously that Hin recom- lished whether DNA strand collisions and the potential bines and knots plasmid DNA in E. coli that topoisomer- resulting entanglements affect DNA metabolism in the ase IV unti]. The Hin site-specific recombination system models two in vivo processes: it tangles the DNA tocreate knots identical to those formed inside the cell and One indication that DNA knotting is deleterious to cells is shuffles the DNA sequence to model DNA recombination the universal prevalence of type-2 topoisomerases. These e hin recombinase gene is provided by the plas- are essential enzymes that cleave both strands of a DNA mid pKH66 (hereafter referred to as pHIN) and is double helix, pass another duplex through this transient expressed from the tac promoter following induction by gate and reseal the break. Type-2 topoisomerases are the isopropyl-β-D-thiogalactopyranoside (IPTG). pHIN also enzymes responsible for unknotting DNA, and, in E. coli, encodes for spectinomycin resistance. E. coli cells harbor- the responsibility falls solely on topoisomerase IV ing pHIN also contained either pBR322 (pBR), which The loss of topoisomerase IV activity has additional affects lacks recombination sites and serves as a negative control, in cells that include hyper-negative supercoiling and the or one of two pBR22-derived plasmids pTGSE4 (pREC) or inability to segregate newly replicated DN pRJ862 (pKNOT) that carry sites recognized by the Hin Therefore, the effects of knots needed to be evaluated sep- recombinase. All three plasmids contain the bla gene, arately from supercoils and catenanes.
which encodes β-lactamase and provides resistance toampicillin. We used the bla gene as a reporter to assess the Here we use the previously characterized Hin site-specific effects of recombining and knotting the DNA.
recombination and DNA knotting system [, tounderstand how the physical constraints placed upon (page number not for citation purposes)
BMC Molecular Biology 2007, 8:44
Figure 1 cal effects of Hin-mediated recombination/knotting
Physiological effects of Hin-mediated recombination/knotting. (A) Assay for the effect of knotting on the function of a
gene. The ovals represent E. coli cells. The Hin expression vector, pHIN, and plasmid substrates pBR, pREC and pKNOT con-
taining the bla gene (encoding β-lactamase) are depicted. Wild-type recombination sites are depicted as black arrows. The
mutant hix site is shown as a grey arrow. (B) Effect of DNA knotting on ampicillin sensitivity of E. coli strain W3110 containing
pHIN and either pBR, pREC or pKNOT. Single colonies were streaked from left to right across LB-agar that contained an amp-
icillin gradient and constant IPTG (1 mM) and spectinomycin (50 μg/ml) for Hin overexpression and maintenance. The experi-
ment was repeated five times in either strain C600 or W3110, and was carried out either from high to low or from low to high
ampicillin concentration with identical results. (C) Ampicillin sensitivity (MIC ) was quantified using the plate dilution method.
(page number not for citation purposes) BMC Molecular Biology 2007, 8:44
Hin binds two 26-bp recognition sites and makes double- Hin-mediated recombination and knotting of a plasmid
stranded breaks at the center of these sites leaving a two- alter function of a reporter gene
bp overhang within each break ext, Hin rotates We first assessed the effect of Hin-mediated DNA recom- the DNA strands in a right-hand direction as dictated by bination and knotting on resistance to ampicillin con- the required DNA negative superco. If both sites ferred by the bla gene on pBR, pREC and pKNOT. LB-agar are wild-type hix (or gix) sites (black arrows Figure contained a gradient of ampicillin], a constant con- in pREC, the two-bp overhangs are complimentary and centration of spectinomycin to maintain the Hin expres- religation may occur after a 180° rotation. However, if the sion vector and IPTG to induce expression of Hin. Wild- Hin substrate has one wild-type and one mutant site (grey type E. coli K12 strains, C600 or W3110, containing pHIN arrow Figure NOT, the overhangs are not and either pBR, pREC or pKNOT were streaked across the complimentary and a 360° rotation (or some multiple of LB-agar. Whereas the strains containing pBR and pREC 360°) is necessary for religation to occur. Thus, Hin were able to grow on the highest ampicillin concentra- recombines pREC and knots pKNOT. Although we ini- tions, growth of the strain carrying pKNOT was limited tially anticipated that pREC would serve to differentiate We determined whether this effect was spe- between effects caused by recombination and those cific to the gene encoded by the plasmid being targeted, caused by DNA knotting, we (data not shown) and others pKNOT, or caused a general increased susceptibility to observed that, in vivo, Hin will occasionally mediate antimicrobial agents. Knotting (and recombination) had processive recombination events to knot plasmids con- no effect on resistance encoded on a separate plasmid or taining wild-type recombination sites at a steady-state on the chromosome: strains harboring the three plasmids level of 2 – 3% (Tab) [. pKNOT is extensively knot- were all growth inhibited at identical concentrations of ted by Hin because the mismatch between the sites spectinomycin (resistance encoded by pHIN) or nor- requires processive recombination [,] for religa- floxacin (targets encoded by the chromosome) (data not tion. Hin expression increases the steady-state knotting of shown). These results indicate that the sensitivity of E. coli pKNOT approximately 5- to 10-fold over endogenous lev- to ampicillin is affected negatively when a knotted plas- els in the presence of topoisomerase IV function and 35- mid encodes its resistance. Knots impair the function of to 45-fold when topoisomerase IV function is blocked the replicon on which they form rather than cause a gen- ]. Twist knots with three (31), five (52) and seven eral effect on the cell.
(72) negative nodes as well as composite knots with six(31#31), eight (31#52) and nine (31#31#31) nodes are We determined minimal inhibitory concentration readily generated when Hin knots pKNOT [. Higher (MIC50) values (the ampicillin concentration that inhibits noded knots are seen in decreasing abundance. Only twist 50% of bacterial growth) to quantify the Hin-mediated knots with three (31) and four (41) nodes are seen when sensitivity to ampicillin. Strains harboring pKNOT were Hin knots a plasmid with two wild-type recognition sites killed at a lower ampicillin concentration (1.4 mg/ml) Previous studies examined the effect of Hin and than pBR (4.7 mg/ml) or pREC (2.3 mg/m other site-specific recombinases on gene expression The intermediate sensitivity of the pREC-containing strain ]. However, a key distinction between those studies to ampicillin compared to the pBR- and pKNOT-contain- and the experiments performed here is that in those exper- ing strains could be caused either by the intermediate level iments the recombinase binding sites were placed of DNA knotting that occurs in pREC or by Hin-mediated between the promoter and the gene whereas here the recombination of pREC.
reporter gene is distant to the site of recombination. Thus,here we examine the global effects on the DNA molecule Hin recombination and knotting alter β-lactamase levels
rather than the local effects on promoter function.
To dissect the molecular mechanism by which Hin affectsthe function of the bla gene on pREC and pKNOT, we per-formed immunoblots to assay β-lactamase levels. Strainswere grown in liquid medium containing IPTG and spec-tinomycin until mid-logarithmic phase (OD600 = 0.3).
Table 1: Hin-mediated knotting.
in vivo + NORb 35% , 45%  ano Hin inductionbnorfloxacin (to block unknotting by topoisomerase IV) (page number not for citation purposes)
BMC Molecular Biology 2007, 8:44
Equal amounts of whole cell extracts were submitted to shown). Therefore, the reduction in protein levels is spe- SDS-PAGE and either stained with Coomassie blue or cific to genes encoded on the plasmid rather than a gen- subjected to Western blotting with anti-β-lactamase antis- eral inhibition of gene expression. Subsequently, AcrA er. The Coomassie blue stained gels indicated levels were used to standardize loading.
that equal amounts of protein had been loaded (data notshown). The pHIN-containing C600 strain with pKNOT If it is knotting that caused the increased susceptibility to produced four- and three-fold less β-lactamase than the ampicillin, then, because topoisomerase IV resolves knots pBR- or pREC-containing strain, respectively, in either rich in E. coli, inhibiting topoisomerase IV should increase the (Lal M9 medium (Figure amount of knots and cause an additional reduction of β- reduction in β-lactamase levels correlates well with the lactamase production from pKNOT. To test whether knot- reduction in ampicillin resistance (compare Figure ting increased ampicillin susceptibility, we utilized a tem- wi There was no effect of pKNOT on levels of perature-sensitive topoisomerase IV mutant, parCts.
the chromosomally encoded topoisomerase IV subunits, Although cell growth and viability are reduced at the non- ParC or ParE, or on levels of AcrA (Fand data not permissive temperatures for parCts, cell division contin- fect on β-lactamase protein levels Hin-mediated effect on β-lactamase protein levels. Cultures of C600 (left) and parCts (right) were grown in rich (LB) or
minimal (M9) medium. Results here are from a typical experiment performed at 42°C. Immunoblotting was performed on total
cellular lysates. Blots were probed with anti-AcrA, anti-β-lactamase, anti-ParC or anti-ParE antibodies. Shown below the blot
from cells grown in LB is the mean of four independent experiments (except for parCts pREC, which was performed three
times) and standard deviations. The values below the M9 experiment show the quantification of that blot, but the M9 experi-
ment was repeated with the same results.
(page number not for citation purposes) BMC Molecular Biology 2007, 8:44
ues to occur and produces enough viable offspring that we pBR or pREC relative to pHIN (Figure e was even were able to obtain sufficient growth (OD600 = 0.3) to per- less pKNOT DNA isolated from the parCts strain grown at form immunoblots. When either of the non-permissive 42°CA, right side). When knotted, the copy temperatures, 37°C or 42°C, for the parCts allele was number of pKNOT was reduced from 20 – 40 copies/cell used, the results were the same. pKNOT in the parCts ,] to lower than pHIN levels in parCts cells (Figure strain produced 8.5-fold less β-lactamase than pREC and β-lactamase levels were determined from immunob- 17-fold less than pBR when cells were grown in LB or M9 lotting and densitometry of total cell lysates following medium at the non-permissive temperatur, SDS-PAGE. Although the DNA knots caused a reproduci- right panels). Therefore, inhibiting the enzyme that unties ble reduction of pKNOT copy number, the magnitude of the knots exacerbates the reduction in β-lactamase pro- this effect (two- to six-fold) was never as large as the effect duction. As with the MIC50 data above, it is unclear on β-lactamase levels (eight- to twenty-fold). In addition, whether the β-lactamase differences between pBR and pREC copy number was unchanged although there was a pREC are caused by Hin binding, recombining or knotting less than two-fold decrease in β-lactamase production pREC. There is less β-lactamase produced in the parCts from pREC in parCts cells at 42°C. Therefore, decreased β- harboring pKNOT than C600 containing pKNOT. How- lactamase levels seen for pREC must not be at the level of ever, the plasmids carried in the two strains have different replicon copy number. The difference between the effect superhelical densities. In the parCts strain, DNA is more on β-lactamase levels and the effect on plasmid copy negatively supercoiled at the non-permissive temperature number suggests that DNA knots mediate their effects e increased negative supercoiling should, if any- through a combination of promoting replicon loss and thing, slightly stimulate β-lactamase production. How- blocking gene transcription.
ever, the knots counter this increase in β-lactamaseproduction. Thus, the inhibitory effects of DNA knotting It was possible that some cells had lost all their plasmid may be greater than measured because some effects are DNA to become plasmid free and other cells retained nor- potentially being masked by the increase in negative mal plasmid levels or that all cells generally had reduced plasmid levels. To distinguish between these possibilities,we grew cells harboring pHIN and either pBR, pREC or Molecular analysis of Hin-mediated effects
pKNOT in the presence of IPTG and spectinomycin, under It has been observed in vitro that DNA knots can diminish conditions identical to those used to evaluate plasmid lev- transcript]. Thus, the effect on β-lactamase produc- els and β-lactamase production, except ampicillin was not tion and ampicillin resistance we observed could be included. Cell culture dilutions were spread on LB-agar explained by an inhibition of bla transcription. However, and incubated overnight at 30°C. The following day, the it could also result from knots interfering with DNA repli- colonies were replica plated onto LB-agar ± ampicillin, cation, which would reduce the number of copies of the grown overnight at 30°C and counted. The frequency of bla gene and, consequently, the amount of β-lactamase plasmid-free cells was the same among all the three strains generated. An effect of DNA knotting on replication in and was similar to what others have observed for loss of vitro or in vivo has not been documented previously. Addi- pBR in E. coli grown in a rich medium (LB) over the time tionally, knots could reduce bla expression by causing period comparable to the one used here ( 3 h) [. Thus, pKNOT to break by weakening the tensile strength of in these experiments, the Hin-mediated effect does not DNA. Precedence of knots weakening and breaking poly- lead to complete loss of pKNOT. However, it is possible mers has been predicted by molecular dynamics simula- that given enough time complete plasmid loss might tions of polyethyleneown using optical tweezers on actin filaments trated withsoft macroscopic strings [ ]. It DNA catenanes are produced as intermediates of replica- is possible that Hin mediates its effects through a combi- tion and they accumulate in temperature-sensitive topoi- nation of blocking transcription, interfering with replica- somerase IV mutants at the non-permissive temperature tion and breaking pKNOT.
,]. When DNA replication is disrupted, replicationcatenanes do not accumulate and data not shown).
To determine whether plasmid stability and copy number We examined the levels of catenanes in parCts carrying are affected by Hin activity we quantified the levels of pHIN and either pBR, pREC or pKNOT. Plasmid DNA was pBR, pREC or pKNOT DNA. Cultures were grown to mid- isolated from parCts strains grown for 40 min. at 42°C as logarithmic phase and divided in half. Plasmid DNA lev- before, nicked to remove supercoiling and analyzed by els were measured from one half and β-lactamase levels high-resolution agarose gel electrophoresis ]. Cate- from the other half. DNA levels were determined by den- nated pBR and pREC products were clearly visible, but sitometric analysis of agarose gels. Following Hin induc- DNA catenanes were greatly reduced for pKNOT (Figure tion, cells contained roughly two-fold less pKNOT than e catenanes were seen under conditions where (page number not for citation purposes)
BMC Molecular Biology 2007, 8:44
Hin-mediated effect on plasmid replication Hin-mediated effect on plasmid replication. (A) Plasmid DNA was isolated from strains C600 or parCts containing the
plasmids indicated. The DNA was linearized with HindIII, which cuts all the plasmids once, including pHIN. Shown is a repre-
sentative ethidium bromide-stained agarose gel from an experiment performed at 42°C. Given below the gel image are the
mean plasmid level values of three independent experiments with standard deviations. (B) Plasmid DNA was isolated 50 min.
after IPTG induction of Hin, nicked, displayed by high-resolution gel electrophoresis and visualized by Southern blotting. Shown
is an autoradiogram. All lanes contain plasmid DNA from the pHIN-harboring parCts cells with pBR, pREC or pKNOT. The
positions of nicked dimer ( ), linear dimer ( ) catenanes (e.g., ) are indicated.
(page number not for citation purposes) BMC Molecular Biology 2007, 8:44
either bla was fully functioning (pREC) or impaired ance (Table Ampicillin at 4.8 mg/ml (3.4-fold higher (pKNOT), it does not seem likely that catenanes block than the MIC50 for pKNOT) was high enough to block all replication and transcription (or cause mutagenesis). This growth and select for hyper-resistant mutants in the C600 experiment also provided an indirect method to examine strain containing pHIN and pKNOT. At this concentration the effect of knotted DNA on DNA replication. Because of ampicillin, using the MSS maximum likelihood DNA replication is the only source of the catenanes, the method OT strain yields 3.4 × 10-6 muta- observation that the level of DNA catenanes is reduced tions/cell/generation. At an ampicillin concentration of provides additional support that DNA replication is 16.1 mg/ml (3.4-fold higher than the MIC50), C600 con- impaired in pKNOT.
taining pHIN and pBR yields 4.8 × 10-10 mutations/cell/generation. At an ampicillin concentration of 7.9 mg/ml Hin-mediated recombination/knotting is mutagenic
(3.4-fold higher than the MIC50), C600 containing pHIN We propose two models to explain how Hin blocks the and pREC had a mutation rate of 4.7 × 10-7 mutations/ function of the bla gene (Fig. These possibilities are cell/generation. Thus, Hin-mediated recombination and not mutually exclusive. The first is the "roadblock" model: knotting increased the mutation rate three to four orders Hin, bound to or cleaving DNA, or the knots themselves of magnitude compared to the spontaneous mutation rate form an impasse to RNA and/or DNA polymerases. The of cells with pBR and Hin expression (Table second is the "breakage" model. Although it is not clearhow knots localize in DNA, it has been suggested from To determine the molecular basis for the hyper-resistance numerical studies that knots may spontaneously pull into to ampicillin, plasmid DNA was isolated from mutant col- a tight conforA within the cell is con- onies and analyzed (FigurThere were two notable stantly being subjected to a number of pulling forces gen- and unanticipated features of these rearrangements. First, erated by transcription, replication and segregation. A the isolated plasmid DNA was much larger than the force (15 pN) comparable to that generated by a single parental pKNOT. This result was surprising because any replication or transcription comp], has been number of deletions or substitution mutations could dis- shown to tighten a DNA trefoil to a diameter less than 25 rupt Hin recombination and these types of changes would nm]. Either the roadblock or the breakage model either result in a smaller plasmid or no change in plasmid predicts that DNA knots would be mutagenic through rep- size. However, these latter types of alterations were not lication fork arrest or through a DNA double-strand break.
apparent. Second, we found that not only was pKNOT In addition, either could induce the SOS response, which altered, but pHIN was also changed in the hyper-resistant could lead to a genome-wide increase in mutation fre- mutants. Gross genetic rearrangements of the plasmid were visible by restriction endonuclease digestion of eachsample (data not shown). These results suggest that While measuring the ampicillin resistance of pBR-, pREC- recombination between pHIN and pKNOT is responsible and pKNOT-containing cells, we observed that, following for the rearrangements and the phenotype of hyper-resist- overnight growth, C600 or parCts strains harboring ance to ampicillin. Although Hin does not directly recom- pKNOT formed large, robust colonies in the zone of drug- bine or knot pHIN, it is likely that recombination between mediated clearing (Figure trast, no colonies pHIN and pKNOT results in a fused plasmid that is either were observed in the cleared zones around filter discs con- refractory to additional knotting/recombination or taining any concentration of ampicillin in lawns of C600 expresses β-lactamase at a sufficient level to confer hyper- or parCts strains containing pBR or pREC (Figur resistance to ampicillin. Without causing ampicillin data not shown). We found that the effect was specific to hyper-resistance, pKNOT-pKNOT fusions would not be β-lactam (ampicillin or cefotaxime) resistance, as no col- selected. In an attempt to analyze the role of homologous onies were found in the zones of drug clearance for pBR-, recombination in this plasmid rearrangement, we tried, pREC- or pKNOT-containing C600 or parCts when nor- but were unable to transform mutant strains lacking recA floxacin was used (data not shown). These results suggest or recD with pHIN.
that increased mutation rate is occurring specifically forpKNOT and not the genome as a whole.
The plasmid changes and ampicillin hyper-resistance wereheritable. Plasmid DNA was isolated from the colonies One hundred percent of the colonies that grew in the that arose in the zones of clearance and transformed into drug-cleared zone had a higher resistance to ampicillin C600 cells harboring pHIN. The plasmids conferred a than the original pKNOT-containing C600 cells and 69% higher level of ampicillin resistance than pKNOT as deter- had higher resistance than pBR-containing cells (Figure mined by Kirby-Bauer assay (data not shown). We found ing the MIC50 values of the original pBR, pREC and that in four of five transformants tested, the mutant plas- pKNOT strains (Figurermed fluctuation mid-transformed cells retained hyper-resistance to ampi- assays to determine the mutation rate to ampicillin resist- cillin and there were no visible colonies in the new zones (page number not for citation purposes) BMC Molecular Biology 2007, 8:44
Intracellular DNA is supercoiled, compacted and highlyconcentrated. Consequently, DNA will collide frequently with itself, and the result of these collisions increases the potential for DNA recombining and knotting. We have analyzed what can happen when the collisions lead to recombining and knotting. The results are that both repli-cation and transcription are blocked and genetic rear-rangements are increased.
Mechanism of the Hin-mediated effect
ls for the Hin-mediated effect DNA knots, and not recombination, are likely the pre- Potential models for the Hin-mediated effect. Plasmid
dominant cause of Hin-dependent replication and tran- pKNOT is recombined by Hin to knot the DNA (a single line scription blocks and mutagenesis because the effect for represents the central axis of the double helix). In the road- pKNOT is more severe than for pREC. The effects were not block model, the knot (or possibly Hin bound to or cleaving caused by inherent differences in the three plasmids.
DNA) is impassable and stalls polymerase. Alternatively, in Ampicillin MIC50 values of C600 strains harboring pHIN the breakage model, knots may break DNA as a result of and either pBR, pREC or pKNOT grown in the absence of forces on the plasmid.
IPTG were identical (data not shown). In addition, themagnitude of the pKNOT-mediated effects was increased of clearance. Because of this, it appears that either no fur- by compromising the activity of the enzyme, topoisomer- ther DNA rearrangements are occurring or, if they are, ase IV, responsible for unknotting DNA. However, in these additional rearrangements do not confer ampicillin addition to unknotting, topoisomerase IV carries out two hyper-resistance. The transformant (1/5) that behaved other cellular tasks: decatenation (reviewed in ) and similarly to pKNOT-containing strains indeed harbored DNA supercoil re. Removal of the decate- pKNOT. Thus, the fused mutant plasmid appears to have nation activity of topoisomerase IV did not account for resolved back into pHIN and pKNOT. To compensate for the increased pKNOT-mediated effects because far more reduced production of β-lactamase, the mutant plasmids catenated replication intermediates were seen in parCts could contain either a mutated bla gene that produces an cells containing either pBR or pREC than in those that enzyme more efficient at metabolizing ampicillin, or the contain pKNOT (Figuree DNA supercoiling shift mutations could allow for increased production of the resulting from the inhibition of topoisomerase IV is not enzyme. Using immunoblot analysis as described above, enough to stimulate either the transcription of the super- we found that all the cells carrying the rearranged plas- coiling-dependent leu-500 prom or the λ inte- mids that were examined had increased β-lactamase pro- grase recombination system [in vivo, suggesting that duction relative to pKNOT (Fig the increase in negative supercoiling resulting from inhib-iting topoisomerase IV activity is unlikely to affect Hin To determine whether the larger molecular weight plas- recombination. It is possible that mechanistic differences mid that had replaced pKNOT and pHIN contained DNA in recombination on a substrate with two wild-type sites originally present in both pHIN and pKNOT, we trans- (pREC) compared to a substrate with one wild-type and formed C600 with total plasmid DNA from the ampicillin one mutant site (pKNOT) could account for the Hin- hyper-resistant isolates. Plasmid DNA from four of the mediated effects. For example, in a purified system, DNA ampicillin resistant colonies was used in independent cleavage by Hin is stimulated by a single mutant recombi- transformations. For each transformation, half of the nation ally, in vivo, DNA cleavage of transformed cells was spread on LB-agar containing amp- pKNOT by pHIN has been detected].
icillin (100 μg/ml, sufficient to select for the parentalpKNOT), and half was spread on LB-agar containing spec- Plasmids replicate completely in less than six seconds and tinomycin (50 μg/ml, to select for pHIN). We found that do so asynchronously. Moreover, they transcribe con- 64/64 spectinomycin resistant transformants were also stantly. Thus, a slight increase of a lethal DNA form could resistant to ampicillin and 28/32 ampicillin resistant have large consequences. Although topoisomerase IV rap- transformants were also resistant to spectinomycin. These idly unties knots, perhaps knot-induced problems, such as results are consistent with a fusion between pHIN and stalled replication forks, or stalled or blocked transcrip- pKNOT being responsible for the ampicillin hyper-resist- tion, persist longer than the knots themselves. Indeed ance phenotype observed in the majority of mutants.
because topoisomerase IV can resolve DNA knots as theyare formed, then, as the copy number of the plasmid goesdown, there should come a point at which topoisomerase (page number not for citation purposes)
BMC Molecular Biology 2007, 8:44
MT7 MT19 MT1 MT4 MT5 MT9 MT10 MT1 MT12 MT21 MT8 MT13 MT2 MT3 MT6 MT27 MT15 MT16 MT17 MT18 MT20 MT22 MT23 MT24 MT25 MT26 nicked pREC/pKNOT sc pREC/pKNOTsc pBR Plasmid pBR pREC pKNOT MT9 MT10 MT11 MT12 MT13 MT14 0.11 0.59 0.65 0.34 0.48 0.43 0.41 Relative Hin-mediated mutag Hin-mediated mutagenesis. (A) Ampicillin resistant colonies growing in the zone of clearance around a filter containing 4 mg ampicillin. (B) Quantitation of ampicillin resistance of individual colonies. (C) Ethidium bromide-stained gel of plasmid DNA iso-lated from mutant colonies growing within the zone of clearance and separated by agarose gel electrophoresis. Lane a is a supercoiled molecular weight standard. Lanes b, c and d contain plasmid DNA from the parental strains harboring pHIN and either pBR, pREC or pKNOT, respectively. Lanes e-j contain plasmid DNA isolated from mutant pKNOT colonies. (D) Total cell lysates of mutants grown in 1 mM IPTG were separated by SDS-PAGE and submitted to immunoblotting. Immunoblots were probed with anti-AcrA antibodies (for a loading control) or anti-β-lactamase antibodies. Shown below the blot are signal intensities in arbitrary units. AcrA and anti-β-lactamase levels for C600 strains containing pHIN and either pBR, pREC or pKNOT are shown for comparison.
(page number not for citation purposes) BMC Molecular Biology 2007, 8:44
Table 2: Hin-mediated mutation rates
Mutation rate per cell division
Mutation rate normalized to pBR
IV can resolve all the knots produced by the Hin system.
helped drive selection and genetic change. In addition, The result would not be a complete loss of plasmid, but cellular stress causes a number of recombinases and trans- instead a steady-state level lower than that found with posases to be acps this activation unknotted DNA, which is what we observed (Figur creates a transient "hypermutable state" that allows cells It is difficult to envision a process analogous to topoi- to develop a mechanism to overcome the stress. Such an somerase IV unknotting that would reverse the effects of event would be similar to that suggested to occur during Hin-mediated site-specific recombination. Thus, if recom- adaptive mutagenesis when E. coli are starved for lactose bination were leading to the loss of plasmids, it would -istent with this idea, cells harboring trans- seem that the unchecked altered plasmid would be lost posons such as Tn10, which can recombine and knot completely from a population of cells in the absence of DNA, will out-compete cells lacking Tn10 that are other- selection, which was not observed.
wise isogenic, which suggests that the transposon confersa greater evolutionary fitnes The DNA knot- or recombination-created blockage couldimpinge upon either the initiation or elongation of gene transcription or DNA replication. Gene promoters and Our results suggest that recombined and knotted forms of replication origins are small relative to plasmids. Unless DNA are problematic for the cell. Thus, it is the DNA con- DNA knots preferentially form in or are localized to pro- formation, rather than the primary sequence, that causes moters or origins, or are hotspots for recombination, then malfunctions. Effects of transient changes in conforma- it is expected that the polymerase roadblocks would occur tion may then persist through induced mutations in the at arbitrary positions on the DNA. Thus, such blockages primary sequence. Unexpectedly, the DNA molecule would likely be outside of where transcription or DNA undergoing site-specific recombination/knotting can "attack" a bystander DNA, and thus both DNA moleculesmay be altered.
It has been demonstrated that when topoisomerase IVactivity is reduced by mutation, priA, which encodes the PriA protein that plays an important role in restarting Strains and Plasmids
blocked replication forks, becomes an essential gene E. coli strains C600, ParC1215 (parCts) and W3110 were ]. It is possible that the stalling of replication forks described previomid pKH66 (pHIN) at knots is the cause of this need for PriA and would contains the S. typhimurium hin gene under control of the explain why the presence of gyrase, which can remove tac promoter and expresses Hin upon addition of isopro- positive supercoils, but not knots, is insufficient to keep pyl-1-thio-β-galactoside (IPTE4 (pREC) replication moving in these cells.
] is a pBR322-derived plasmid containing the Ginrecombination (gix) sites and enhancer from bacteri- Implications for cellular physiology and evolution
ophage Mu. Gin, Hin and their respective recombination Given (i) the abundance of recombinases, transposases sites are interchange62 (pKNOT) contains and topoisomerases found in both prokaryotic and hix recombination sites and the enhancer binding site for eukaryotic organisms, (ii) the lack of sequence specificity the Hin recombinase from S. typhimurium ]. One hix by these enzymes, (iii) the confined space for the chromo- site contains a single base pair change, which forces a sec- somes and (iv) the propensity of DNA to react and entan- ond round of recombination to tie knots by preventing gle with itself, DNA rearrangements that lead to cellular religation after only one roun. To create the strains transformation or death, or that contribute to the muta- used throughout this work, we used a CaCl2 method to tions that shape evolution seem likely to occur. In other transform wild-type E. coli with plasmid DNA (typically words, an intrinsic lack of DNA stability might have (page number not for citation purposes) BMC Molecular Biology 2007, 8:44
Norman), anti-ParC or anti-ParE (kind gifts of the late Gradient pl and Kirby-Bauer [ disc diffusion N.R. Cozzarelli, University of California, Berkeley), incu- assays were used to measure antibiotic resistance. Satu- bated in SuperSignal West chemiluminescent reagent rated overnight cultures containing the strains were (Pierce, Rockford, IL) and visualized with a charge cou- diluted 30- to 100-fold in LB containing 1 mM IPTG and pled display camera.
50 μg/ml spectinomycin. The freshly diluted cultures weregrown at 37°C until they reached OD600 = 0.3. For the Plasmid loss assay
Kirby-Bauer disc diffusion assays, cells were spread on LB- Cells were grown as for Western blot analysis. Plasmid agar containing 1 mM IPTG and 50 μg/ml spectinomycin.
DNA was isolated by the alkaline lysis method , line- The plates were allowed to dry for 20 min. and discs con- arized with HindIII (which cuts pKNOT and pHIN once) taining 10 μl of different ampicillin concentrations (0 – and separated by electrophoresis on 1% agarose (TAE) 500 mg/ml) were placed onto the agar. The plates were gels. Plasmid levels were quantified by densitometric then incubated overnight at 37°C. The diameter of the scanning (NucleoVision software, NucleoTech Corp., San cleared zone around each disc was measured. For the gra- Mateo, CA) of images of ethidium bromide-stained gels.
dient plate assay, cells were spread on square plates con- Assuming pHIN levels do not change among the strains, taining a gradient from 0 to 17.5 mg/ml ampicillin, and plasmid bands were first normalized within each lane to then incubated overnight at 37°C. The plate dilution the pHIN vector. These standardized band values are method was used to determine the ampicillin MIC50 val- shown relative to the value for pBR within each strain ues. The three E. coli strains harboring pHIN and either background. To determine whether entire plasmid popu- pBR, pREC or pKNOT were grown overnight in LB lations were lost from cells, various dilutions of cells medium containing 100 μg/ml ampicillin, 50 μg/ml spec- grown in LB medium were spread onto LB-agar and rep- tinomycin and no IPTG. These cultures were diluted 500- lica plated on agar ± 100 μg/ml ampicillin. Colonies were fold in LB medium containing 50 μg/ml spectinomycin counted following overnight incubation at 30°C (C600 and 1 mM IPTG, but no ampicillin. The freshly diluted and parCts) or 37°C (C600).
cultures were grown with shaking to mid-logarithmicphase (OD600 = 0.3 – 0.4) at 37°C. Appropriate dilutions DNA catenane analysis
(to final cell counts of approximately 100 and 1000 per DNA catenanes were analyzed as done pr].
plate) were spread onto LB-agar alone and LB-agar con- parCts cells containing pHIN and either pBR, pREC or taining ampicillin concentrations from 1.3 to 4.8 mg/ml, pKNOT were grown at 30°C to mid-logarithmic phase 50 μg/ml spectinomycin, but no IPTG. Colonies were (OD600 = 0.3 – 0.4). IPTG was added to a final concentra- counted following overnight incubation at 37°C. For each tion of 1 mM to induce Hin expression. After 10 min., of the three strains, regression analysis was performed to cells were shifted to 42°C to inactivate the mutant topoi- determine the best-fit curve through the data points somerase IV. Forty minutes later, plasmid DNA was iso- (2670: n = 10, 2671: n = 9 and 2672: n = 8) in the plot of , nicked with DNase I to remove supercoiling survival as a function of ampicillin concentration. From ] and displayed by high-resolution gel electrophoresis this best-fit curve, the ampicillin MIC50 values were ]. The DNA was then transferred to a Zeta Probe nylon membrane (Bio-Rad Laboratories, Hercules, CA) andprobed with [α-32P]-dCTP (GE Healthcare, Little Chal- Antibodies and immunoblotting
font, UK) labeled pBR322 (made by random priming, Isogenic C600 and ParC1215 (parCts) strains were grown Amersham Megaprime™ DNA labeling systems, GE overnight in LB medium without IPTG. Cells were diluted Healthcare, Little Chalfont, UK), which hybridizes all 1/100 into LB medium and grown with shaking ± 1 mM three plasmids.
IPTG and 50 μg/ml spectinomycin to mid-logarithmicphase (OD600 = 0.3 – 0.4) at 37°C or 42°C. Duplicate sets Isolation of ampicillin resistant colonies and fluctuation
of whole cell extracts were made by resuspending equal amounts of pelleted cells in loading buffer (125 mM Tris- Ampicillin resistant colonies that grew inside the zone of HCl, pH 6.8; 1.4 M β-mercaptoethanol; 20% glycerol; 2% clearance (FigurA) were streaked onto LB-agar plates SDS; 0.1% Bromophenol blue), boiling for 3 min. and containing 1 mM IPTG, 50 μg/ml spectinomycin and 1 subjecting to 10% SDS-PAGE. One set was stained with mg/ml ampicillin and incubated overnight at 30°C. These Coomassie blue to ensure equal protein amounts were conditions were used to prevent the accumulation of loaded. The other set was blotted to a nitrocellulose Pro- revertants to ampicillin sensitivity. To determine the tran membrane. The blots were probed with (1:10,000 mutation rate, E. coli harboring pHIN and either pBR, dilution for all) antisera to β-lactamase (a kind gift of T.
pREC or pKNOT were grown overnight in LB medium Palzkill, Baylor College of Medicine, Houston), anti-AcrA containing 100 μg/ml ampicillin, 50 μg/ml spectinomy- (a kind gift of H. I. Zgurskaya, University of Oklahoma, cin and no IPTG. The overnight cultures were diluted (page number not for citation purposes) BMC Molecular Biology 2007, 8:44
6,000-fold into LB medium ( 105 cells/ml) containing no ampicillin, 50 μg/ml spectinomycin and 1mM IPTG and Cozzarelli NR, Cost GJ, Nollmann M, Viard T, Stray JE: divided into ten 1.2-ml aliquots. These aliquots were Nat Rev Mol Cell Biol 2006, 7(8):580-588.
grown with shaking to mid-logarithmic phase (OD600 = Bohrmann B, Haider M, Kellenberger E: 0.3 – 0.4) at 37°C to obtain parallel, independent cul- tures. The number of ampicillin resistant mutants that originated in each culture was determined by spreading 2 × 70 μl (pBR- and pREC-containing strains) or 2 × 200 μl (pKNOT-containing strain) of undiluted culture onto LB- Annu Rev Biophys Biomol Struct 2004, agar containing various ampicillin concentrations, 50 μg/ ml spectinomycin, but no IPTG. 16.1 mg/ml ampicillin Yan J, Magnasco MO, Marko JF: Nature was used for the strain harboring pBR; 7.9 mg/ml ampicil- lin was used for the strain harboring pREC; and 4.8 mg/ Shishido K, Komiyama N, Ikawa S: Increased production of a
knotted form of plasmid pBR322 DNA in Escherichia col
ml ampicillin was used for the strain harboring pKNOT.
J Mol Biol 1987, 195(1):215-218.
Each of these ampicillin concentrations is 3.4-fold higher Martin-Parras L, Lucas I, Martinez-Robles ML, Hernandez P, Krimer than the corresponding strain's ampicillin MIC DB, Hyrien O, Schvartzman JB: total number of cells was determined by spreading dilu- Nucleic Acids Res 1998, tions of each culture on LB-agar. Colonies were counted Deibler RW: The biological implications of DNA knots and the in vivo activ- after incubation overnight at 37°C. The probable number ity of topoisomerase IV Houston: Baylor College of Medicine; PhD the- of mutations per culture (m) was calculated from the dis- tribution of hyper-resistant mutants in the independent Shishido K, Ishii S, Komiyama N: The presence of the region on
pBR322 that encodes resistance to tetracycline is responsi-
cultures using the MSS maximum likelihood method.
ble for high levels of plasmid DNA knotting in Escherichia co
Then the mutation rate (μ) was calculated as μ = m/2N Nucleic Acids Res 1989, t is the total number of cells per cultur].
Ishii S, Murakami T, Shishido K: Gyrase inhibitors increase the
content of knotted DNA species of plasmid pBR322 in
Escherichia coli J Bacteriol 1991, 173(17):5551-5553.
Sikorav JL, Jannink G: RWD carried out antibiotic resistance measurements, immunoblotting, plasmid loss assays, DNA catenane Biophys J 1994, 66(3 Pt 1):827-837.
analysis and isolation of ampicillin resistant colonies, Arsuaga J, Vazquez M, Trigueros S, Sumners D W, Roca J: conceived of and participated in the design of the study Proc Natl Acad Sci USA and drafted the manuscript. JKM carried out antibiotic Baker TA, Sekimizu K, Funnell BE, Kornberg A: Extensive unwind-
resistance measurements, plasmid loss assays, isolation of ing of the plasmid template during staged enzymatic initia-
ampicillin resistant colonies and fluctuation analysis, per- tion of DNA replication from the origin of the Escherichia col
formed the statistical analysis, participated in the design Cell 1986, 45(1):53-64.
Hatfield GW, Benham CJ: DNA topology-mediated control of
of the study and helped to draft the manuscript. DWS par- global gene expression in Escherichi Annu Rev Genet 2002,
ticipated in the design and coordination of the study, ana- lyzed experimental results and helped to draft the Liu Y, Bondarenko V, Ninfa A, Studitsky VM: Proc Natl Acad Sci manuscript. LZ conceived of and participated in the USA 2001, 98:14883-14888.
design and coordination of the study, analyzed experi- Steck TR, Franco RJ, Wang JY, Drlica K: Topoisomerase muta-
tions affect the relative abundance of many Escherichia coli
mental results and helped to draft the manuscript. All Mol Microbiol 1993, 10(3):473-481.
authors read and approved the final manuscript.
Vologodskii AV, Cozzarelli NR: J Mol Biol 1993, 232(4):1130-1140.
Vologodskii AV, Levene SD, Klenin KV, Frank-Kamenetskii MD, Coz- We thank Dr. Stacy Merickel and Dr. Reid Johnson for Hin reagents pKH66 J Mol Biol 1992, 227(4):1224-1243.
and pRJ862 and for sharing unpublished results. We are grateful to Dr. Tim- Podtelezhnikov AA, Cozzarelli NR, Vologodskii AV: othy Palzkill and Dr. Helena I. Zgurskaya for providing antibodies. We are Proc Natl Acad Sci USA 1999, grateful to Dr. Mary-Jane Lombardo for comments on the manuscript. Funding for this work was provided by the National Science Foundation Vologodskii AV, Marko JF: MCB-0090880, the National Institutes of Health RO1 AI054830, a Bur- Biophys J 1997, 73(1):123-132.
Deibler RW, Rahmati S, Zechiedrich EL: Topoisomerase IV,
roughs Wellcome Fund New Investigator Award in the Toxicological Sci- alone, unknots DNA in E. coliGenes Dev 2001, 15:748-761.
ences and the Curtis Hankamer Research Award to LZ. RWD and JKM Adams DE, Shekhtman EM, Zechiedrich EL, Schmid MB, Cozzarelli were supported by pre-doctoral fellowships from the Program in Mathe- matics and Molecular Biology at Florida State University with funding from the National Science Foundation and the Burroughs Wellcome Fund Inter- Cell 1992, 71(2):277-288.
Zechiedrich EL, Cozzarelli NR: Roles of topoisomerase IV and
DNA gyrase in DNA unlinking during replication in
Escherichia coli Genes Dev 1995, 9(22):2859-2869.
(page number not for citation purposes) BMC Molecular Biology 2007, 8:44
Zechiedrich EL, Khodursky AB, Bachellier S, Schneider R, Chen D, Lil- Davenport RJ, Wuite GJ, Landick R, Bustamante C: Single-molecule
ley DM, Cozzarelli NR: Roles of topoisomerases in maintaining
study of transcriptional pausing and arrest by E. co
steady-state DNA supercoiling in Escherichia coli J Biol Chem
Bao XR, Lee HJ, Quake SR: Merickel SK, Johnson RC: Topological analysis of Hin-catalysed
Phys Rev Lett 2003, 91(26 Pt 1):265506.
DNA recombination in vivo and in vitro Mol Microbiol 2004,
Beaber JW, Hochhut B, Waldor MK: Heichman KA, Moskowitz IPG, Johnson RC: Nature 2004, 427(6969):72-74.
Friedberg EC, Walker GC, Siede W: DNA repair and mutagenesis edn.
Washington: ASM Press; 1995. Genes Dev 1991, 5(9):1622-1634.
Rosche WA, Foster PL: Rochman M, Aviv M, Glaser G, Muskhelishv Methods 2000, 20(1):4-17.
Levine C, Hiasa H, Marians KJ: EMBO Rep 2002, 3:355-360.
Hardy CD, Cozzarelli NR: A genetic selection for supercoiling
Biochim Biophys Acta mutants of Escherichia coli
Mol Microbiol 2005, 57(6):1636-1652.
Kato J-I, Nishimura Y, Imamura R, Niki H, Hiraga S, Susuki H: New
Postow L, Hardy CD, Arsuaga J, Cozzarelli NR: Topological
topoisomerase essential for chromosome segregation in
domain structure of the Escherichia coli Genes
Escherichia coli Cell 1990, 63(2):393-404.
Dev 2004, 18(14):1766-1779.
Zechiedrich EL, Khodursky AB, Cozzarelli NR: Topoisomerase IV,
Glasgow AC, Bruist MF, Simon MI not gyrase, decatenates products of site-specific recombina-
J Biol Chem 1989, 264(17):10072-10082.
tion in Escherichia col Genes Dev 1997, 11(19):2580-2592.
Johnson RC, Bruist MF Grompone G, Bidnenko V, Ehrlich SD, Michel B: PriA is essential
for viability of the Escherichia coli topoisomerase IV
Embo J 1989, 8(5):1581-1590.
parEJ Bacteriol 2004, 186(4):1197-1199.
Lee SY, Lee HJ, Lee H, Kim S, Cho EH, Lim HM: Heller RC, Marians KJ: Nature 2006, J Bacteriol 1998, 180(22):5954-5960.
Tam CK, Hackett J, Morris C: Rate of inversion of the Salmonella
Posfai G, Plunkett G 3rd, Feher T, Frisch D, Keil GM, Umenhoffer K, Kolisnychenko V, Stahl B, Sharma SS, de Arruda M, Burland V, Har- Infect Immun 2005, 73(9):5568-5577.
cum SW, Blattner FR: Emergent properties of reduced-genome
Eisenstadt E, Carlton BC, Brown BJ: Gene Mutation. In Methods for
Escherichia coli Science 2006, 312(5776):1044-1046.
General and Molecular Bacteriology Edited by: Gerhardt P, Murray RGE, Haniford DB: Transpososome dynamics and regulation in Tn
Wood WA, Krieg NR. Washington, D.C.: American Society for 10 Crit Rev Biochem Mol Biol 2006, 41(6):407-424.
Microbiology; 1994:304. Hastings PJ, Slack A, Petrosino JF, Rosenberg SM: Portugal J, Rodriguez-Campos Acids Res 1996, 24(24):4890-4894.
PLoS Biol 2004, 2(12):e399.
Saitta AM, Soper PD, Wasserman E, Klein ML: Hersh MN, Ponder RG, Hastings PJ, Rosenberg SM: Adaptive muta-
Nature 1999, tion and amplification in Escherichia coli
Res Microbiol 2004, Arai Y, Yasuda R, Akashi K, Harada Y, Miyata H, Kinosita KJ, Itoh H Nature 1999, Mutat Res 2005, 569(1-2):3-11.
Pieranski P, Kasas S, Dietler G, Dubochet J, Stasiak A: Localization
Chao L, McBroom SM: Evolution of transposable elements: an
of breakage points in knotted strings. New Journal of Physics
IS10 insertion increases fitness in Escherichia coli Mol Biol Evol
Uehara H, Kimura H, Aoyama A, Yamanobe T, Komoto T: Effects of
Chao L, Vargas C, Spear BB, Cox EC: knot characteristics on tensile breaking of a polymeric
Nature 1983, 303(5918):633-635.
monofilament. New Journal of Physics 2007, 9:65.
Kato J-I, Nishimura Y, Yamada M, Suzuki H, Hirota Y: Gene organ-
Lee CL, Ow DS, Oh SK: ization in the region containing a new gene involved in chro-
mosome partition in Escherichia co J Bacteriol 1988,
J Microbiol Methods 2006, 65(2):258-267.
Lee C, Kim J, Shin SG, Hwang S: Absolute and relative QPCR
Hughes KT, Gaines PC, Karlinsey JE, Vinayak R, Simon MI: quantification of plasmid copy number in Escherichia coJ
Sequence-specific interaction of the Salmonella
Embo J 1992, Noack D, Roth M, Geuther R, Muller G, Undisz K, Hoffmeier C, Gaspar S: Maintenance and genetic stability of vector plas-
Crisona NJ, Kanaar R, Gonzalez TN, Zechiedrich EL, Klippel A, Coz- mids pBR322 and pBR325 in Escherichia coli
Mol Gen Genet 1981, 184(1):121-124.
Sundin O, Varshavsky A: J Mol Biol Cell 1981, 25(3):659-669.
Plasterk RHA, Brinkman A, van de Putte P: DNA inversions in the
Katritch V, Olson WK, Vologodskii AV, Dubochet J, Stasiak A: Tight-
chromosome of Escherichia coli
ness of random knotting. Phys Rev E 2000, 61(5 Pt B):5545-5549.
Metzler R, Hanke A, Dommersnes PG, Kantor Y, Kardar M: Proc Natl Acad Sci USA 1983, 80(17):5355-5358.
Phys Rev Lett 2002, 88(18):188101.
Bauer AW, Kirby WMM, Sherris JC, Turck M: Wuite GJ, Smith SB, Young M, Keller D, Bustamante C: Am J Clin Pathol 1966, 45:493-496.
Nature 2000, 404(6773):103-106.
Sambrook J, Fritsch EF, Maniatis T: Molecular cloning: a laboratory man- Maier B, Bensimon D, Croquette V: ual 2nd edition. Cold Spring Harbor: Cold Spring Harbor Laboratory Proc Natl Acad Sci USA 2000, 97(22):12002-12007.
Wang MD, Schnitzer MJ, Yin H, Landick R, Gelles J, Block SM: J Mol Biol 1973, 74(4):739-742.
Science 1998, 282(5390):902-907.
(page number not for citation purposes)
CLINICAL TRIALS AND OBSERVATIONS Eradication of minimal residual disease in hairy cell leukemia Farhad Ravandi, Jeffrey L. Jorgensen, Susan M. O'Brien, Srdan Verstovsek, Charles A. Koller, Stefan Faderl, Francis J. Giles,Alessandra Ferrajoli, William G. Wierda, Shirley Odinga, Xuelin Huang, Deborah A. Thomas, Emil J. Freireich, Dan Jones,Michael J. Keating, and Hagop M. Kantarjian