Introduction of interrupted secondary structure in supercoiled DNA as a function of superhelix density: consideration of hairpin structures in superhelical DNA.

PM2 DNA was prepared with different superhelical densities (sigma) in order to examine the relationship betweenn supercoiling and the occurrence of a region(s) of unpaired bases in this DNA. A previous study showed that CH3HgOH reacts with native superhelical PM2 DNA more rapidly than the nicked form II. This evaluation of binding, monitored through the change of sedimentation velocity, was repeated on PM2 DNA I with different superhelical densities. Early binding is detected by an increase in sedimentation velocity and occurs with molecules with sigma' values betwee -0.025 and -0.037. The conversion of form I to form II with the single-strand-specific endonuclease from Neurospora crassa also occurs above a sigma value of -0.025. This data strongly supports the view that supercoiling produces interrupted secondary structure. The question whether the interrupted regions remain single stranded in character or form small intrastrand hairpin regions is considered by examining which model best fits the CH3HgOH- induced sedimentation velocity changes and the standard sedimentation velocity versus the superhelical density curve for the in vitro made DNAs. The hairpin model offers the most satisfactory explanations for all the results of this and previous studies.     

Sedimentation and intrinsic viscosity behavior of PM2 bacteriophage DNA in alkaline solution

The sedimentation coefficient and intrinsic viscosity of nicked and closed circular PM2 bacteriophage DNA have been measured as a function of pH in the alkaline region. A gradual increase in the sidimentation coefficient, and a corresponding decrease in the intrinsic viscosity, are observed for the superhelical (closed) circle in the pH region from 10.5 to about 10.9. This has been tentatively interpreted in terms of the known dependence of sedimentation coefficient upon the number of superhelical turns. At slightly higher pH values, the curve passes through the minimum (sedimentation coefficient) and maximum (intrinsic viscosity) expected when the superhelical turns present at neutral pH are unwound by partial alkaline denaturation. Sedimentation studies of the relaxed (nicked) circular species have revealed the existence of DNA forms in the pH region from 11.27 to 11.37 which sediment considerably faster than the closed circle in the same pH region. These have been identified as partially denatured nicked circles, in which varying fractions of the duplex structure have undergone alkaline denaturation, but strand separation has not yet occurred. Varying fractions of a slower species, either undenatured or completely denatured nicked circles, are also observed in some of these experiments. A corresponding result is observed in the intrinsic viscosity vs. pH curve. When nicked circular PM2 DNA is exposed to various alkaline pH's, rapidly neutralized, and sedimented at neutral pH, the expected sharp transition from native to denatured (strand-separated) molecules is seen. However, a very narrow pH range is noted in which native and denatured forms coexist in a single experiment. The above experiments carried out upon the closed form also reveal a narrow pH range in which the bulk of the transition from native closed circles to the collapsed cyclic coil takes place, in acccord with an earlier study on a different DNA. This transition is shown never to be completely effected, however, as there is a fraction (7–8%)of the closed circles which renature to the native form, regardless of the alkaline pH employed. This same phenomenon was not observed in the case of artificially closed λb₂b₅c DNA circles. Possible explanations for some of the above results are discussed.     

Circular dichroism of superhelical DNA

The circular dichroism (CD) spectra of a number of superhelical DNA's have been measured. The introduction of negative superhelical turns causes an increase in magnitude of the positive band around 280 mμ, while the trough around 250mμ is little affected. For two samples of λb2b5c DNA (20 Mdalton) containing different number of negative superhelical turns, the magnitude of the positive band relative to that of the nicked control increases with increasing number of superhelical turns. In 2M NaCl, the small (1.45 Mdalton) superhelical DNA from E. coli 15 shows an unusually large difference in CD compared with that of the same DNA with a few single-chain scissions per molecule. This large difference is not observed in a medium containing p. 0.11M NaCl. These results indicate that the double helix in a superhelical DNA is perturbed somewhat due to the bending and torsional forces in such a molecule. The magnitude of such structural alteration seems to depend on the number of superhelical turns per unit length, the size of the DNA molecule, as well as the ionic medium.     

Bleomycin-specific fragmentation of double-stranded DNA.

Brief exposure of covalently closed circular duplex PM2 DNA to low concentrations of the clinical bleomycin mixture (Blenoxane) resulted in specific fragmentation of the genome that does not depend on the presence of superhelical turns. The double-strand breaks are in fact produced at several discrete sites on the PM2 genome but frequently occurring near the HpaII restriction endonuclease cleavage site. Initial rates of formation of nicked circular and linear duplex PM2 DNAs are reduced to different extents as the ionic strength of the reaction is increased. Increasing ionic strength is most effective in reducing the initial rate and overall yield of apparent double-strand scissions compared with single-strand scissions in the bleomycin-treated PM2 DNA.     

Extracellular nucleases of Alteromonas espejiana BAL 31.IV. The single strand-specific deoxyriboendonuclease activity as a probe for regions of altered secondary structure in negatively and positively supercoiled closed circular DNA.

The dependence of the initial rate of introduction of the first single-chain scission (initial nicking rate) into covalently closed circular phage PM2 DNA by the single strand-specific nuclease from Alteromonas espejiana BAL 31 upon the superhelix density (sigma) of the DNA has been examined. The initial nicking rate decreases with decreasing numbers of negative superhelical turns (decreasing values of -sigma), which behavior is characteristic of other single strand-specific nucleases as reported earlier. In contrast to earlier work, the initial nicking rates of closed circular DNAs by the action of the Alteromonas nuclease have been shown to be readily measurable at values of -sigma as low as 0.02. However, even at the elevated concentrations of enzyme and extended digestion periods required to cause nicking at an appreciable rate at near-zero values of sigma, closed circular DNA containing very few superhelical turns (form IO DNA) is not cleaved at a detectable rate. When this DNA is rendered positively supercoiled by ethidium bromide (EtdBr), it is not affected by the nuclease until very high positive values of sigma are attained, at which low rates of cleavage can be detected at elevated enzyme concentrations. The effects of EtdBr on the enzyme activity have been tested and are entirely insufficient to allow the interpretation of zero nicking rates as the result of inhibition of the nuclease activity by the dye. Positively supercoiled DNA is concluded not to contain regions having significant single-stranded character until values of sigma are reached which are very much higher than the values of -sigma for which negatively supercoiled DNAs behave as if they contain unpaired or weakly paired bases.     

The helix-coil transition of closed and nicked DNAs in aqueous neutral trichloroacetate solutions.

The melting transition for closed, underwound DNAs and for nicked or linear DNAs was monitored by velocity sedimentation and by absorbance spectroscopy in aqueous NaCCl3CO2 (NaTCA) and RbTCA. The addition of neutral trichloroacetate lowers the midpoint of the helix-coil transition by 26% C/M for RbTCA and by 32% C/M for NaTCA, depressing the denaturation region to near room temperature at neutral pH. The melting of nicked DNA is cooperative, occurring over a temperature range of about 5.6 degrees C. The melting profile for closed DNA is broad and noncooperative with a transition breadth greater than 45 degrees. Closed DNAs undergo a structural alteration, as revealed by velocity sedimentation, resulting in a reduction in the number of superhelical turns at temperatures and salt concentrations substantially below the melting temperatures and salt concentrations substantially below the melting temperature of the nicked DNA. The reduction in the extent of supercoiling continues upon isothermal addition of salt up to the salt concentration at which all superhelical turns are removed. The salt concentration at the principal minimum in the sedimentation velocity profile (3.16 M NaTCA for PM-2 DNA) is approximately the same as that at the midpoint of the helix-coil transition for the nicked DNA.     

The degree of unwinding of the DNA helix by ethidium. I. Titration of twisted PM2 DNA molecules in alkaline cesium chloride density gradients

A series of covalently closed bacteriophage PM2 DNA samples with varying degrees of superhelicity were prepared in vitro. The amount of bound ethidium per DNA nueleotide needed for the removal of all superhelical turns, v_c⁰, was determined for each sample by a number of methods. In order to evaluate the unwinding angle for the binding of one ethidium molecule to a DNA double helix, the pH dependence of the buoyant densities in CsCI of these samples was examined. A new calibration relating the change in buoyant density of a DNA to the fraction of bases titrated has been obtained, by measuring the buoyant densities of a number of catenanes (interlocked rings) containing both single-stranded and double-stranded λ DNA rings, at a pH such that the single-stranded DNA is fully titrated while the double-stranded DNA is not titrated. This calibration was used to obtain the pH dependence of the fraction of DNA bases titrated for the phage PM2 DNAs with differing extents of supercoiling. A simple theoretical analysis shows that in a restricted pH range close to pH_m, the melting pH of the DNA in the absence of the topological constraint associated with covalently closed double-stranded DNAs, the difference in the fraction of bases titrated at a certain pH between two covalently closed DNAs with different degrees of superhelicity is directly proportional to the difference in the v_c⁰ values of the DNAs. The unwinding angle per bound ethidium molecule can be obtained from the proportionality constant. In this way, it is not necessary to know precisely the actual pH value for either DNA, pH_e, at which the DNA is titrated to the extent that it contains no superhelical turns. The conclusion of the theoretical analysis and the experimental results is that the binding of an ethidium molecule to a double-stranded DNA unwinds the DNA helix by an angle φ_e = 26 °. The uncertainty in this value is estimated to be less than 10%. The new value for φ_e is approximately a factor of two larger than the value 12 °, which has been in use in the past decade. In the earlier alkaline titration results for polyoma DNA (Vinograd et al., 1968), which had been interpreted as supporting the 12 ° value, the calculation of φ_e was critically dependent on knowing pH_e. It is believed that pH_e was underestimated in the earlier work, resulting in a low φ_e value. Since the previous value φ_e = 12 ° has been widely used in the determination of the number of superhelical turns for many DNAs, and in measurements on the angular alterations of the DNA helix by the binding of a variety of small and large molecules and by solvent and temperature changes, the new value φ_e = 26 ° requires proportional adjustments of many previous results.     

Left-handed Z-DNA regions are present in negatively supercoiled bacteriophage PM2 DNA

Bacteriophage PM2 DNA is a 10-kb covalently closed circular (ccc) molecule with a reported superhelical density of sigma = -0.12. Here we describe the binding of anti-Z-DNA antibodies to PM2 form I DNA under high and low salt conditions. The binding to PM2 DNA has been demonstrated by competitive radioimmunoassay (RIA), retardation of the DNA:antibody complexes in agarose gels and visualization by electron microscopy. The antibody binding is dependent on the degree of negative supercoiling. Thus, PM2 form II and form III did not bind the antibody. The low salt RIA results indicated the presence of 200-400 bp of left-handed DNA per PM2 molecule. This could reduce the effective superhelical density to sigma = -0.04 to -0.08, a range comparable with those found for other ccc DNAs in vivo. Electron microscopy revealed that a maximum of 22 antibody molecules bind to PM2. Single-site restriction with HpaII of the fixed DNA:antibody complex showed a cluster of four to five antibody molecules bound near one end of the linear DNA molecule. The evidence presented indicates that PM2 DNA contains regions of left-handed conformation under physiological conditions (low salt concentration) as well as at high salt concentrations. In addition, electrophoretic analyses of PM2 topoisomers indicate the presence of left-handed regions at superhelical densities less than that of isolated PM2 DNA.     

An endonucleolytic activity of nuclease TT1 specific for superhelical DNA.

Highly purified nuclease TT1 from T. thermophilus HB8 acts on a linear single- and double-stranded DNA as an exonuclease and produces 5'-mononucleotides either from the 5'- or 3'-terminus. It was found that the enzyme also possesses an endonuclease activity specific for superhelical (form I) and single-stranded circular DNA. Form I of various kinds of DNA (phi X174, PM2, Co1E1 and RF 1010 etc.) is nicked to yield first relaxed circles (form II) and then nicked at the opposite site to yield unit length linear DNA (form III), which is subsequently hydrolyzed from the 5'- or 3'-terminus. A single cleavage of the form I of phi X174 DNA seemed to occur at a limited number of unique sites. Both endonuclease and the known exonuclease activities co-migrate on polyacrylmide gels, show the same pH and temperature optima, are stimulated by Mg2+ and are inactivated by EDTA similarly.     

Locating Interrupted Hydrogen Bonding in the Secondary Structure of PM2 Circular DNA by Comparative Denaturation Mapping

Previous studies with HCHO have revealed a reaction with superhelical DNA that strongly suggests that this DNA consists of small regions of interrupted secondary structure. To map these sites in PM2 DNA, the following set of experiments was performed using electron microscopy. (i) A denaturation map of nicked form II was obtained using Inman's alkaline-HCHO conditions. (ii) The superhelical form I was reacted with HCHO at 30 C until equilibrium was achieved at the interrupted sites (3.6% reactivity). The excess HCHO was removed rapidly and X-ray treatment was employed to nick these prereacted molecules. These form II molecules containing HCHO (form II HCHO) were also subjected to denaturation mapping. It would be expected that the HCHO-unpaired regions would serve as induction sites for the propagation of melting. Hence, depending on the location of the induction sites; we would anticipate either the creation of new regions of melting or a normal denaturation map shifted to lower pH values. Comparison of the development of progressive denaturation of form II and form II HCHO reveals that the latter is the case. The denaturation maps of form II are highly organized patterns of adenine-thymine (AT)-rich regions, with a total of five regions at extreme pH conditions. There are six highly organized regions for form II HCHO, i.e., smaller adjacent loops, at low denaturation conditions where no denaturation is seen for form II. These coalesce into the pattern for form II containing four of five A-T-rich regions observed for form II. Hence we conclude that the regions of altered hydrogen bonding in superhelical PM2 DNA are four to six in number and they map in the A-T-rich regions of the DNA.     

The number of superhelical turns in native virion SV40 DNA and minicol DNA determined by the band counting method.

By a method of overlapping the results obtained after agarose gel electrophoresis under two different sets of conditions, it has become possible to determine the number of superhelical turns in a given DNA by counting the bands present after partially relaxing the DNA (Keller and Wendel, 1974) with highly purified nicking-closing (N-C) enzyme from LA9 mouse cell nuclei. Because native supercoiled DNA is heterogeneous with respect to superhelix density, an average number of superhelical turns was determined. Virion SV40 DNA contains 26 +/- 0.5 superhelical turns, and native Minicol DNA contains 19 +/- 0.5 superhelical turns. The above are values at 0.2 M NaCl and at 37 degrees C, the condition under which the enzymatic relaxations were performed. The superhelix densities determined by the band counting method have been compared with superhelix densities determined by buoyant equilibrium in PDl-CsCl gradients. The Gray, Upholt, and Vinograd (1971) calculation procedure has been used for evaluating the superhelix densities by the latter method with the new statement, however, that relaxed DNA has zero superhelical turns. Comparison of the superhelix densities obtained by both methods permits a calculation of an unwinding angle for ethidium. The mean value from experiments with SV40 DNA is 23 +/- 3 degree. The average number of superhelical turns in SV40, 26, combined with the value, 21, obtained by both Griffith (1975) and Germond et al. (1975) for the average number of nucleosomes per SV40 genome, yields an average of 1.25 superhelical turns per 1/21 of the SV40 genome. If the regions of internucleosomal DNA are fully relaxed, 1.25 correesponds to the average number of superhelical turns with a nucleosome. When analyzed under identical conditions, the limit product generated by ligating a nicked circular substrate in the presence of 0.001 M Mg2+ at 37 degrees C (ligation conditions) is slightly more positively supercoiled than the limit product obtained when the N-C reaction is performed in 0.2 M NaCl at 37 degrees C. The difference in superhelix density as measured in gels between the two sets of limit products for both Minicol and SV40 DNAs is 0.0059 +/- 0.0005. This result indicates that the DNA duplex is overwound in the ligation solvent relative to its state in 0.2 M NaCl.     

The time-resolved kinetics of superhelical DNA cleavage by BamHI restriction endonuclease

The kinetics of the cleavage of superhelical plasmid DNA (pBR322) by the restriction endonuclease, BamHI, have been analyzed in terms a compartmental model consistent with the chemistry first proposed by Rubin and Modrich (Rubin, R. A., and Modrich, P. (1978) Nucleic Acids Res. 5, 2991-2997) for analysis of the kinetics of the restriction endonuclease, EcoRI. The model was defined in terms of two compartments representing DNA substrate (bound and free), two compartments representing nicked intermediate (bound and free), one compartment representing linear product, and one compartment for free enzyme. A simultaneous analysis of concentration changes over time of the three DNA forms (superhelical, nicked, and linear) at six different enzyme concentrations was undertaken employing this compartmental model using SAAM (Simulation Analysis And Modeling) software. Results showed that rate constants characterizing the association of enzyme with superhelical DNA (6.0 x 10(5) M-1 s-1) and nicked DNA (2.8 x 10(5) M-1 s-1) were similar in magnitude and rate constants characterizing cleavage of the first (1.2 x 10(-2) s-1) and second phosphodiester bonds (3.1 x 10(-2) s-1) were also similar. The analysis yields a kinetically determined equilibrium constant of 12.9 nM for the dissociation of nicked intermediate from the enzyme. The rate constant describing the release of the nicked intermediate from the enzyme has a value of 3.7 x 10(-3) s-1. By comparing the value of this release rate constant to the value of the constant describing the second cleavage event, it can be determined that only 10% of the nicked intermediate bound to the enzyme is released as free nicked DNA and that 90% of the nicked intermediate is processed to the linear form without being released. Hence, most of the DNA is cleaved as the result of a single enzyme-DNA recognition event. No steady state assumptions were made in the analysis. The approach was to directly solve the differential equations which described the kinetic processes using an interactive method. This study demonstrates the usefulness of this approach for the analysis of kinetics of protein-DNA interactions for the restriction endonucleases.     

Priming of superhelical SV40 DNA by Escherichia coli RNA polymerase for in vitro DNA synthesis.

When closed circular SV40 DNA containing 58 negative superhelical turns is used as a template for RNA synthesis with Escherichia coli RNA polymerase, a fraction of the RNA product remains complexed with the DNA. The RNA in the complex is resistant to ribonuclease in high salt, and the Tm indicates that it is hydrogen bonded to the DNA. The mole ratio of RNA to DNA nucleotides in the complex ranges from 0.01 to 0.08; the RNA ranges in length from 80 to 600 nucleotides. The formation of the complex is dependent on the circular DNA being topologically underwound since no complex is formed when closed circular DNA containing zero superhelical turns is used as the template. The DNA-RNA complex can serve as a primer-template combination for in vitro DNA synthesis by E. coli DNA polymerase I. After synthesis with (alpha-32P)-labeled deoxyribonucleoside triphosphates followed by alkaline hydrolysis, the isolation of 32P-labeled ribonucleotides is evidence for a covalent linkage between the RNA and the DNA synthesized. During the in vitro DNA synthesis, the template is nicked at a low rate, and the nicked molecules support extensive DNA synthesis. This observation indicates that only limited synthesis can occur on unnicked molecules possibly owing to the topological constraints against unwinding of the helix. Possible models for in vivo priming of double-stranded DNA by E. coli RNA polymerase are discussed.     

The effect of divalent cations on the mode of action of DNase I. The initial reaction products produced from covalently closed circular DNA

The effect of different divalent metal ions on the hydrolysis of DNA by DNase I was studied with an assay which distinguishes between cleavage of one or both strands of the DNA substrate during initial encounters between enzyme and DNA. Using covalently closed superhelical SV40(I) DNA as substrate, initial reaction products consisting of relaxed circles or unit-length linears are resolved by electrophoresis of radioactively labeled DNA in agarose gels. Only in the presence of a transition metal ion, such as Mn2+ or Co2+, and only under certain reaction conditions, is DNase I able to cut both DNA strands at or near the same point, generating unit-length linears. This ability to cut both DNA strands is inhibited by such factors as temperature decrease, the addition of a monovalent ion or another divalent cation which is not a transition metal ion, or a reduction in the number of superhelical turns in the DNA substrate. All of these factors lead to a winding of the duplex helix and antagonize the unwinding of the duplex promoted by transition metal ion binding. Transition metal ions may thus convert the DNA substrate locally to a form in which DNase I can introduce breaks into both strands. In the presence of Mg2+, DNase I introduces single strand nicks into SV40(I), generating exclusively the covalently open, relaxed circular SV40(II) as the initial product of the reaction. In the presence of Mn2+, DNase I generates as initial products a mixture of SV40(II) and unit-length SV40 linear DNA molecules, formed by two nicks in opposite strands at or near the same point in the duplex. These circular SV40(II) molecules consist of two types. A minority class is indistinguishable from the nicked SV40(II) produced by DNase I in the presence of Mg2+. The majority class consists of molecules containing a gap in one of the two strands, the mean length of the gap being 11 nucleotides. The SV40(L) molecules produced in the presence of Mn2+ appear to have single strand extensions at one or both ends.     

On the degree of unwinding of the DNA helix by ethidium. II. Studies by electron microscopy.

When a negatively twisted covalently closed DNA is annealed with single-stranded fragments of the same DNA, under proper conditions a loop (or loops) may form by the disruption of a segment (or segments) of base pairs between the complementary strands of the covalently closed DNA, and the formation of base pairs between the strands of the covalently closed DNA and the single-stranded fragments. Since such a process involves essentially no net gain or loss of the number of base pairs, it is driven by the free energy favoring the reduction of the number of superhelical turns. If the fragments are sufficiently long or are present at a sufficiently hig concentration during annealing, the most stable product between a covalently closed DNA and the DNA fragments (under conditions favoring the formation of double-stranded DNA) is a looped molecule devoid of superhelical turns. The size of the looped region or regions, which can be measured by electron microscopy, provides a way to determine the degree of superhelicity of the covalently closed DNA in the absence of the fragments. When this is compared with the degree of superhelicity of the covalently closed DNA determined by titration with the intercalative dye ethidium, the unwinding angle of the DNA double helix due to the intercalation of an ethidium can be calculated. Such measurements were done on two samples of phage PM2 DNA with different extents of supercoiling. The results are in agreement with the value 26 degree obtained recently by alkaline titration of covalently closed PM2 DNA samples in CsC1 density gradients (Wange, J.C., (1974) J. Mol. Biol. 89, 783-801).     

Determination of the number of superhelical turns by the hyperchromicity of partially denatured covalently-closed DNA molecules

A novel method of determining the number of superhelical turns of a covalently-closed plasmid DNA is described. It relies on the determination of the hyperchromicity, and hence the proportion of unstacked basepairs, of a partially heat-denatured sample which co-migrates during electrophoresis with nicked circular duplex DNA. The values obtained for plasmid pBR beta G DNA at 4 degrees C (-29.8 and -33.5 in the two buffers used) agree closely with the values obtained in parallel by topoisomer band-counting. Our method is less precise than band-counting but is readily applicable to determining the superhelicity of very large DNA molecules. Our results confirm earlier findings that magnesium-containing buffers cause an increase in the duplex winding angle, and hence an increase in the number of negative superhelical turns.     

Kinetic parameters of superhelical DNA cleavage by endonuclease S1

Major kinetic parameters of endonuclease S1 were determined on superhelical bacteriophage PM2 DNA and on relaxed nicked circular PM2 DNA. At 37 degrees and 0,25 M NaCl, the Michaelis constants were respectively 1.7 . 10(-8) M and 1 . 10(-9) M, and catalytic constants were respectively 0.36 sec-1 and 1.2 . 10(-2) sec-1. The inhibition of the enzyme reaction by its product was detected.     

AMP-dependent DNA relaxation catalyzed by DNA ligase occurs by a nicking-closing mechanism.

In the presence of AMP and Mg2+, a covalently closed duplex DNA containing negative superhelical turns was treated with DNA ligase isolated from bacteriophage T4-infected E. coli. This resulted in the gradual and not sudden loss of superhelical turns as for example in the case of type I DNA topoisomerase. All DNA products remain covalently closed. Since T4 enzyme-mediated DNA relaxation is inhibited by both pyrophosphate and by ATP this suggests that DNA relaxing and DNA joining activities probably coincide. EDTA addition in the presence of a large excess of enzyme, induces the formation of nicked DNA products while protein denaturing treatments are not very effective. Our observations might suggest an involvement of the relaxing activity of DNA ligase during the ligation process.     

Agarose gel electrophoretic analysis of damage to supercoiled DNA by adriamycin in the presence of beta-NADH dehydrogenase

In a cell-free system, the anticancer anthracycline antibiotic adriamycin was able to convert purified covalently closed circular, superhelical, form I bacteriophage PM2 DNA to relaxed circular form II DNA in the presence of either sodium borohydride (NaBH4), NADPH cytochrome P-450 reductase or beta-NADH dehydrogenase isolated from myocardial cells. There was no detectable increase in the amount of form III linear duplex DNA formed during the reaction even at high drug concentrations. Less drug was required for the conversion of form I to form II DNA in the presence of the enzymic reducing agents than in the presence of NaBH4. Form II DNA, prepared by irradiation using a Cs-137 source, was not degraded to form III linear duplex DNA. However, form I0 DNA, covalently closed circular DNA without superhelical turns, freshly prepared using topoisomerase I, was converted to form II DNA similar to the conversion of superhelical form I to form II DNA. Again, no increase in the amount of form III linear duplex DNA could be detected.