Skip to main content
NCBI home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2002 Jul 1;30(13):2980–2986. doi: 10.1093/nar/gkf409

Structural perturbations in DNA caused by bis-intercalation of ditercalinium visualised by atomic force microscopy

Torunn Berge a, Nigel S Jenkins, Richard B Hopkirk 1, Michael J Waring, J Michael Edwardson, Robert M Henderson
PMCID: PMC117064  PMID: 12087184

Abstract

Atomic force microscopy (AFM) has been used to examine perturbations in the tertiary structure of DNA induced by the binding of ditercalinium, a DNA bis-intercalator with strong anti-tumour properties. We report AFM images of plasmid DNA of both circular and linearised forms showing a difference in the formation of supercoils and plectonemic coils caused at least in part by alterations in the superhelical stress upon bis-intercalation. A further investigation of the effects of drug binding performed with 292 bp mixed-sequence DNA fragments, and using increment in contour length as a reliable measure of intercalation, revealed saturation occurring at a point where sufficient drug was present to interact with every other available binding site. Moment analysis based on the distribution of angles between segments along single DNA molecules showed that at this level of bis-intercalation, the apparent persistence length of the molecules was 91.7 ± 5.7 nm, approximately twice as long as that of naked DNA. We conclude that images of single molecules generated using AFM provide a valuable supplement to solution-based techniques for evaluation of physical properties of biological macromolecules.

INTRODUCTION

Ligands binding to DNA so as to alter its structure and function occupy a central place in the current therapy of infectious diseases (protozoal, bacterial and viral) as well as cancer, and offer hope of treating a wider spectrum of diseases having a genetic origin via the expedient of gene targeting (1,2). Ditercalinium (NSC 335153) was developed as a novel addition to the arsenal of existing chemotherapeutic agents. It is a synthetic compound created by the linking of two 7H-pyrido [4,3-c] carbazole ring chromophores via a rigid bis-ethylpiperidine chain (3,4) (Fig. 1). The molecule is the result of a systematic search for improved anti-cancer drugs based on planar compounds coupled by rigid linkers. In ditercalinium, the linker chain serves to inhibit the intramolecular stacking of the two chromophores as well as increasing DNA association lifetime by obstructing drug displacement from the duplex.

Figure 1.

Figure 1

Open in a new tab

Ditercalinium is composed of two 7H-pyrido [4,3-c] carbazoles linked by a rigid bis-ethylpiperidine chain.

Ditercalinium bis-intercalates into DNA with high affinity (108 M–1) and the rigid linker chain becomes positioned in the major groove (3,5,6). No clear sequence specificity for the drug has yet been established; however, investigations involving proton NMR (5) and DNase I footprinting (7) have suggested a binding preference for GC-rich regions over AT-tracts. The three-dimensional structure of ditercalinium complexed to [d(CGCG)]2 has been resolved by X-ray crystallography and shows that the helix is both kinked by 15° and unwound by 36° upon bis-intercalation (8,9). This structural alteration is accompanied by a widening of the minor and major grooves.

The geometry of the drug–DNA complex is strongly linked to anti-tumour activity, and considerable effort has gone into resolving its mechanism of toxicity. Activity is conferred by the rigid nature of the linker, giving an asymmetrical drug–DNA complex and leading to the occurrence of structural polymorphism that is related to activation of the cellular DNA repair systems (6). In contrast to most other DNA-binding anti-tumour drugs, ditercalinium is not a conspicuous inhibitor of cellular processes such as DNA replication or transcription; rather, recognition of the bulky drug–DNA complexes causes the malfunction of DNA repair systems. This atypical feature of ditercalinium has been attributed to its ability to induce enzyme-mediated lesions in the non-covalently complexed DNA, an effect that has been traced to the UvrABC exinuclease excision repair system in Escherichia coli (10–12). The repair system is unable to cope with the reversible nature of the ditercalinium–DNA complex, which induces a futile and abortive repair process, eventually leading to cell death.

Initial studies using ditercalinium demonstrated a strong anti-tumour activity against L1210 murine leukaemia cells (3,13). Unfortunately, further clinical trials have been interrupted as a consequence of its irreversible hepatotoxic side effects, which may be linked to interactions between the drug and mitochondrial diphosphatidylglycerol (14,15). Attempts to eliminate these toxic effects have so far proved unsuccessful (16).

Because of the unique behaviour of ditercalinium, specific interest is still being directed towards both its mechanism of cytotoxicity and its DNA binding mode. In this study, the perturbations in the structure of DNA caused by bis-intercalation with ditercalinium were investigated using the direct imaging technique of atomic force microscopy (AFM). Previous studies showing AFM images of DNA–ligand complexes have demonstrated the potential of the technique for assessment of structural effects induced by small-molecule binding. For example, the mono-intercalator ethidium bromide has been used in investigations of tertiary structure changes in both relaxed and supercoiled plasmid DNA (17–20), while lambda-phage DNA has been used for intercalation studies with the dye YOYO-1 (21), resulting in images that have been employed for calculations of binding affinity as well as site exclusion number. Single molecule resolution also allows detailed examinations of specific DNA fragments, resulting in estimates of parameters such as the bend angle induced in the helix, as demonstrated by the specific binding of benzo[a]pyrene diol epoxide (BPDE) to DNA (22). The AFM force measurement technique has also been exploited in investigations of ligand binding to both single- and double-stranded DNA (23).

Thus, in addition to providing unique visual evidence of the conformational effects of DNA bis-intercalation, this approach also allows a statistical treatment of the drug binding process with quantitative determination of parameters such as contour lengths and bend angle distributions of DNA molecules as a function of drug concentration. Here, we report the effects of ditercalinium binding using long closed-circular and linear DNA molecules as well as short DNA fragments of a few hundred base pairs. Our results provide a detailed insight into the structural pathway involved in the bis-intercalation process and demonstrate further the versatility of AFM as an imaging technique for determining physical properties of biological macromolecules.

MATERIALS AND METHODS

Ditercalinium

Ditercalinium (NSC 335153) was a generous gift synthesised by Dr Christianne Garbay-Jaureguiberry in the laboratory of Professor B. P. Roques at the Département de Pharmacochimie Moléculaire et Structurale, INSERM U266–CNRS UMR 8600, Faculté de Pharmacie, 75270 Paris, France. We are extremely grateful to these workers for supplying the sample used here.

DNA constructs

292 bp DNA fragments were generated by PCR amplification of a region of the plasmid pBR322 (Sigma, St Louis, MO) using the primers 5′-CGTTGTGAGGGTAAACAACTGGCGG-3′ (sense) and 5′-GGTTAGCAGAATGAATCACCGATACGC-3′ (antisense) (PNAC Facility, Department of Biochemistry, University of Cambridge, UK). The amplified fragments had a GC content of 52%.

The plasmid pBR322 was used both in its circular form and as a linear fragment produced by digestion with PvuII (New England BioLabs, Beverly, MA). All DNA molecules were purified using a QIAquick purification procedure kit (Qiagen Ltd, Crawley, Surrey, UK) and eluted in 10 mM Tris–HCl, pH 8.5, prior to ditercalinium binding.

Sample preparation

Ditercalinium was dissolved in MilliQ water (Millipore System, Bedford, MA) and bound to DNA by mixing using the 292 bp DNA fragments at molar drug:DNA ratios ranging from 5:1 to 75:1. After incubation for 20 min at room temperature, the drug–DNA complexes were diluted in 5 mM MgCl2 to a DNA concentration of ∼1 nM, suitable for AFM imaging of individual molecules. A 50 µl droplet of the sample was rapidly deposited on freshly cleaved ruby mica (Goodfellow, Huntingdon, UK). After 10 min incubation, the sample was rinsed gently with 10× 1 ml of MilliQ water and dried under N2 gas. Preparation of samples as described ensured equilibration of the DNA in two dimensions on the surface after deposition (24), allowing a statistical treatment of the imaged DNA molecules as described below.

Linear and circular plasmid pBR322 molecules were complexed with ditercalinium in a similar manner and samples prepared for AFM imaging as described above.

A reference was created using the mono-intercalator ethidium bromide (Sigma, St Louis, MO), which was incubated with the 292 bp DNA fragments at an ethidium bromide:base pair stoichiometry of 1:1. The DNA–ligand complexes were diluted in 5 mM MgCl2 for AFM sample preparation as described above.

AFM imaging

Imaging was performed using a Multimode atomic force microscope (Veeco/Digital Instruments, Santa Barbara, CA). Samples were imaged in air using tapping mode operating at a root mean square amplitude of 0.7 V (∼9 nm) and a drive frequency of ∼300 kHz. Commercially available silicon cantilevers were used (NCH Pointprobes; Nanosensors, Wetzlar-Blankenfeld, Germany).

Image analysis

Contour and persistence lengths were determined using an image analysis program written in MatLab, version 6.00 (The MathWorks Inc., Natick, MA). Prior to analysis using this program, all images were flattened using the Nanoscope software (Veeco/Digital Instruments) to remove tilt and slope. The program extracted digital paths for the DNA molecules based on colour contrast and then fitted smooth polynomial curves to the paths for subsequent measurements of the statistical parameters. The angle θ, between segments separated by a distance l, was measured for the whole population of DNA paths as previously described (24,25). Apparent persistence length, Lpe, was determined from the dependence of the mean square angle on segment length as expressed by <θ2> = l/Lpe for each set of drug–DNA complexes.

RESULTS

Ditercalinium binding to plasmid DNA

Samples for AFM imaging were prepared by adsorption onto freshly cleaved mica in the presence of 5 mM Mg2+. This process allowed the DNA molecules to reach equilibrium in two dimensions after deposition as described by Rivetti et al. (24). Thus, the DNA should be able to rearrange on the mica surface after adsorption with loss of one degree of freedom being confined to the rotation around the helical axis. Initially, a qualitative investigation of the effects of bis-intercalation was performed using circular and linearised plasmid pBR322 in the absence and presence of ditercalinium. Representative AFM images of ligand-free DNA showed populations of the circular molecules appearing as loosely interwound supercoiled structures (Fig. 2A). The corresponding linearised DNA molecules appeared as relaxed and well-separated strands (Fig. 2C). Crossing strands were rarely observed on linear fragments. Both the negative charge of the DNA sugar-phosphate backbone and the excluded volume effect experienced by long chain molecules such as these (>1 µm) (24) are contributing factors that mediate this behaviour of naked DNA. Excluded volume interactions cause an extended conformation, leading to perturbation of chain dimensions such as the end-to-end distance, a parameter that is commonly used to describe a population of polymer molecules.

Figure 2.

Figure 2

Open in a new tab

Bis-intercalation into linear and circular plasmid DNA. Naked circular (A) and linear (C) forms of the plasmid pBR322 (4.36 kb) were imaged by AFM, using tapping mode in air. The circular plasmids show a degree of supercoiling, while a relaxed, extended conformation with few overlapping strands is observed for the linear molecules. After bis-intercalation using ditercalinium, here shown at a molar drug:DNA ratio of 25:1, both circular (B) and linear (D) DNA molecules were found to exhibit increased looping and close DNA–DNA contacts. (E) Characteristic loop structures found on linearised plasmids. The thick-stranded features are made by the tight coiling of two individual DNA strands. Height information for all images is colour coded [dark (low) → light (high)]. Scale bars: 200 µm (A–D), 100 µm (E).

Incubation with ditercalinium at various molar drug:DNA ratios was found to affect the supercoiled helical state and twisted appearances for both the circular and linear DNA conformations, giving typical images as shown for a ratio of 25:1 (Fig. 2B and D). Ditercalinium is known to cause local unwinding of the DNA helix by 36° upon binding (8,9), producing changes in superhelical stress that are correlated with transitions reflecting an increase in the negative supercoiling of the circular plasmid DNA molecules, and eventually causing the generation of tightly interwound structures with a marked increase in the number of inter-strand contacts. The effect of bis-intercalation was sometimes even more evident on the linearised plasmid molecules (Fig. 2D), where, in addition to an increase in the formation of simple loops, several plectonemic coils were also observed. A panel of typical loop structures is shown (Fig. 2E), all observed on linear DNA molecules. The plectonemic coils appear as thick-stranded features composed of two interwound DNA strands.

Bis-intercalation of ditercalinium into short DNA molecules

Even at a low level of bis-intercalation as shown in Figure 2, the formation of loops and close strand contacts on plasmid DNA made measurements of contour lengths and bend angles difficult. Thus, DNA molecules of 292 bp lengths were chosen for further investigations of the bis-intercalation process based on polymer chain statistics. Using shorter DNA molecules, the complex morphologies were avoided, producing images suitable for computer-aided analysis of single molecules. A program written in MatLab was specifically designed for this study; this program analysed the images on the basis of colour contrast, creating digital paths for the imaged DNA molecules. Any DNA fragments that were touching other fragments were excluded from the analyses. The digital paths were the basis for estimates of contour length and angles between segments along the molecules. The angle distributions were subjected to moment analysis as previously described (24,25).

A titration of ditercalinium with DNA was carried out using a range of molar drug:DNA ratios from 5:1 to 75:1. Contour length distributions for the drug–DNA complexes as well as for naked DNA are illustrated in Figure 3 (data not shown for 75:1). The average length of ligand-free DNA molecules was estimated as 96.6 ± 6.6 nm. A comparison with the theoretical length of 99.3 nm for 292 bp DNA molecules, calculated assuming B-form DNA and a rise of 0.34 nm/bp, shows good agreement for the naked DNA, confirming that the molecules deposited as described above remained predominantly in the B-form on the mica surface.

Figure 3.

Figure 3

Open in a new tab

Ditercalinium binding to 292 bp DNA fragments. The histograms show the length distribution for naked DNA molecules (A), as well as for fragments with ditercalinium bis-intercalated at molar drug:DNA ratios of (B) 5:1, (C) 10:1, (D) 25:1, (E) 35:1 and (F) 50:1. The broken line indicates the theoretical length of the 292 bp fragments assuming B-form DNA, with a rise of 0.34 nm/bp.

After bis-intercalation, contour lengths for the DNA complexes were clearly greater than for the ligand-free fragments. According to simple intercalation theory (26), contour length is a reliable measure for intercalation, giving a rise of 0.34 nm per intercalated ring compound. Binding of ditercalinium to DNA should result in the stacking of two carbazole rings between the base pairs, thus giving a total length increment of 0.68 nm. Figure 4 shows the change in mean contour lengths for the DNA molecules as a function of ditercalinium concentration. It is evident that the interaction approaches a point of saturation at a ditercalinium:DNA ratio of ∼50:1. At saturation conditions, the contour length was increased by ∼28%, corresponding to an interaction of drug molecules with every other available binding site. As a reference, binding of the mono-intercalator ethidium bromide was carried out using a concentration of one ethidium bromide molecule per base pair (Fig. 5). Due to a significantly lower affinity for DNA [1.5 × 105 M–1 (27) versus 108 M–1 (3)], construction of a titration curve as shown in Figure 4 was not possible as the final DNA concentration had to be kept at a level where individual DNA molecules were still visible, making samples suitable for computer-aided analysis. At this concentration, the average length of the DNA strands was 103.4 ± 6.2 nm (n = 161), i.e. an increment of ∼7%, equivalent to the intercalation of 20 molecules of ethidium bromide per DNA strand. However, it is likely that saturation of the DNA molecules is not reached at this concentration, as indicated by the right-handed tail of the histogram (Fig. 5).

Figure 4.

Figure 4

Open in a new tab

Change in DNA length in response to the bis-intercalation of ditercalinium. The curve (broken line) fitted to the experimental data reveals that saturation of the DNA molecules is reached at a molar drug:DNA ratio of ∼50:1. The error bars indicate standard deviations.

Figure 5.

Figure 5

Open in a new tab

Ethidium bromide binding to 292 bp DNA fragments. The histogram shows the length distribution for DNA molecules mono-intercalated with ethidium bromide at a molar drug:DNA ratio of 300:1. The broken line represents the theoretical length of 292 bp B-form DNA.

The characteristic lengthening of the bis-intercalated molecules was also immediately apparent in the AFM images showing drug–DNA complexes (Fig. 6). In addition to the increase in DNA length, ditercalinium bis-intercalation also caused a marked increase in the rigidity of the DNA. The effect was quantified using plots showing the bend angle distribution, <θ2>, between two segments along the DNA fragments as a function of their separation, l. Apparent persistence lengths were determined from the inverse slope of the regression lines for these plots. The results are shown in Figure 7, where, for clarity, only the linear region of the plots is shown. At longer segment separations, deviations from linearity are caused by the limited number of observations. At saturation conditions, the apparent persistence length was found to increase by 109% from 43.9 ± 1.2 nm (naked DNA) to 91.7 ± 5.7 nm (50:1). Apparent persistence length for the mono-intercalated DNA molecules was 53.1 ± 3.5 nm (data not shown) with an average of 20 molecules bound per DNA strand. This value is comparable to the persistence length of the bis-intercalated molecules at a molar ratio of 25:1. Ethidium bromide has previously been shown to unwind the helix by 26°, and insertion of the dye has a definite stabilising effect, as demonstrated both by improved base stacking energies and higher melting temperatures (28,29).

Figure 6.

Figure 6

Open in a new tab

AFM images showing the effect of ditercalinium binding to 292 bp DNA fragments. Tapping-mode AFM was performed in air, and images are of naked DNA molecules (A) and drug–DNA complexes generated using molar ratios of (B) 5:1, (C) 10:1, (D) 25:1, (E) 35:1 and (F) 50:1. Bis-intercalation of ditercalinium into DNA causes an evident lengthening of the DNA molecules as well as increasing rigidity.

Figure 7.

Figure 7

Open in a new tab

Apparent persistence lengths of the DNA molecules were evaluated using plots of mean square angles versus segment lengths and the inverse slope of these plots. For naked DNA, a value of Lpe = 43.9 ± 1.2 nm was determined (filled circles). For molecules incubated with ditercalinium, the persistence lengths were Lpe= 53.8 ± 1.8 nm [5:1 (molar ditercalinium:DNA ratio), open squares], Lpe = 51.6 ± 1.6 nm (10:1, filled triangles), Lpe = 55.0 ± 2.6 nm (25:1, open circles), Lpe = 74.1 ± 4.2 nm (35:1, filled squares), Lpe = 91.7 ± 5.7 nm (50:1, open triangles) and Lpe = 135.1 ± 12.4 nm (75:1, filled diamonds).

A summary of the results derived from the titration using 292 bp DNA fragments and ditercalinium is reported in Table 1.

Table 1. Summary of the effects of ditercalinium bis-intercalation on the properties of DNA.

B-form DNA
  
 Contour length (nm)
 Persistence length (nm)
  
292 bp
  
 99.3a
 ∼50b
  
Ditercalinium: DNA (molar ratio) Sample size     Bound fraction (%)
Pure DNA 146  96.6 ± 6.6  43.9 ± 1.2 –
5:1 145 101.7 ± 7.3  53.8 ± 1.8 150
10:1 148 103.1 ± 7.8  51.6 ± 1.6  96
25:1 124 114.1 ± 8.7  55.0 ± 2.6 103
35:1 142 119.8 ± 9.3  74.1 ± 4.2  97
50:1 147 124.2 ± 10.2  91.7 ± 5.7  81
75:1 128 121.9 ± 12.9 135.1 ± 12.4  54
Open in a new tab

Contour and persistence lengths were estimated for populations of 292 bp DNA molecules after incubation with the drug at increasing molar ratios. The bound fraction was calculated using the theoretical increment in contour length following stacking of two carbazole rings within the double helix (0.68 nm for each drug molecule).

aTheoretical rise per base pair: 0.34 nm.

bPreviously reported (33–35).

DISCUSSION

The bis-intercalator ditercalinium is endowed with strong anti-tumour properties and functions by a mechanism that is completely different from that of mono-intercalating compounds. Binding of the drug causes a structural deformation of the DNA helix that is recognised by repair systems, and it is the reversible nature of the deformation that causes malfunctioning of this DNA repair, eventually leading to cell death (11).

In this study, the structural transitions of DNA following bis-intercalation of ditercalinium were examined using AFM. In addition, a statistical evaluation of the molecules was carried out using contour lengths and bend angles extracted from the AFM images. Preparation of samples for AFM imaging involves a shift from a three-dimensional conformation in solution to a two-dimensional conformation on a surface. In order to extract quantitative data from the AFM images, the method of deposition is crucial and must allow the adsorbed molecules to equilibrate on the surface after the loss of one degree of freedom, but also provide continued freedom of movement in the remaining two dimensions. As a result of this requirement, surface treatments giving mica strong adsorption properties, such as silanisation (30) or coating with poly-l-lysine (31), were not considered suitable for this type of investigation. In contrast, the presence of 5 mM Mg2+ in the deposition buffer supports two-dimensional equilibration conditions on the mica surface. In this way, it was possible to ensure that the sample preparation had a negligible influence on the statistical parameters describing each population of molecules.

Ditercalinium bis-intercalation into both circular and linear forms of the plasmid pBR322 was found to increase coiling and promote the establishment of close DNA–DNA contacts. Ditercalinium has been shown to cause kinking as well as unwinding of the double helix (8,9), a structural change that will increase the helical stress in the DNA molecule. In circular plasmid DNA, this should cause increased supercoiling, and this has been illustrated in Figure 2. Ditercalinium also carries a net positive charge, which will reduce the negative charge on the DNA phosphate backbone at the site of bis-intercalation and consequently favour the formation of close intra-strand contacts and tight coiling. This should lead to an increased number of simple loops, as observed on the linear DNA molecules. Plectonemic coiling of these molecules is most likely explained by a combination of the increased helical stress following the interaction with ditercalinium and the loss of one degree of freedom associated with the adsorption of the DNA on to a flat substrate. Loss of rotation around the helical axis upon contact with the surface could cause strands to coil around each other in order to adopt a more energetically favourable conformation.

Even at a very low concentration, for example at a molar ratio of 25:1 (corresponding to only one drug molecule per 175 bp), ditercalinium interacted with DNA to produce complicated structures, with crossing DNA strands and increased coiling. Thus, for the titration study of ditercalinium–DNA complexes, short DNA fragments were chosen in order to generate conformations that could be evaluated easily using an image analysis program. Contour lengths were measured using polynomial curves fitted to tracings of the DNA paths. The mean contour length for naked DNA corresponded well with the theoretical value for the B-form conformation. The experimental value was within ∼3% of the theoretical value, a result that is comparable to other AFM studies on nucleic acids (24,32).

Changes in contour lengths represented a readily quantifiable consequence of bis-intercalation. By determining the mean lengths of each population of drug–DNA complexes it was easy to extract a parameter such as the apparent bound fraction, i.e. the percentage of the total drug present that was actually bound to the polymer (Table 1). At very low drug concentrations (e.g. 5:1), the bound fraction was calculated to exceed 100%. This anomalous value may, in part, reflect an underestimate of the actual drug concentration caused by the tendency of ditercalinium to form aggregates in solution.

Non-linear fitting of the length-versus-concentration data (Fig. 4) shows that saturation of the DNA molecules occurs at a molar ratio of ∼50:1, corresponding to the occupation of every other available binding site. This result confirms the high affinity of ditercalinium for DNA (3,12). The DNA fragments prepared for this study were amplified by PCR from a region of the plasmid pBR322 chosen to generate molecules lacking any obvious sequence anomalies and having a GC content of 52%. Sequence-specific binding has not been established for ditercalinium; however, previous investigations have suggested a preference for GC-rich regions over AT-tracts (5,7).

The well-characterised chromophore ethidium bromide is often used as a general model for biological activity of intercalating agents. Limitations set by the assay presented here, however, made it impossible to construct a titration curve corresponding to that of ditercalinium using the mono-intercalator. A homogenous spread of individual DNA molecules is required for producing AFM images suitable for analysis, thus defining an upper limit for the DNA concentration in the sample. This had to be balanced against the compound’s lower affinity for DNA, resulting in a reference produced using a ratio of one ethidium bromide molecule per base pair. At this concentration, an average of 20 molecules was intercalated into the DNA, giving the DNA molecules an average length of 103.4 ± 6.2 nm. It is likely that saturation of the DNA molecule occurs at a higher concentration, as indicated by the right-handed tail of the histogram (Fig. 5) with several DNA strands exceeding the average value.

Persistence lengths for the DNA molecules were estimated using an analysis of bend angle distributions, <θ2>, giving a value of 43.9 ± 1.2 nm for naked DNA. This value is somewhat smaller than previous estimates obtained by independent methods (Lpe ∼50 nm) (33–35). It is likely that the size of the chosen fragment is a contributing factor to the underestimation of the rigidity of the DNA helix. Although no high-flexibility sequence motifs were identified in the amplified fragments, a longer molecule would have ensured a wider range of mixed sequence motifs. The use of longer fragments could also have yielded a larger number of angle measurements at greater segment separations, extending the linear region of the angle distribution graph, although at the risk of introducing confusion due to strand crossover(s).

The anti-tumour activity of ditercalinium is closely related to its linker composition and is improved, for example, with increasing conformational restraints on the linker (6,16). The stiff linking chain causes a deformation of the helix, leading to asymmetry in the DNA complexes and structural polymorphism that is recognised by DNA repair systems. Rigidity of the drug–DNA complexes was found to increase with increasing concentration of drug, in good agreement with the lengthening of the molecules as described in Figure 4. However, at concentrations above those required to fill all the bis-intercalation sites, rigidity continued to rise, even though the length of the fragments remained constant. It is possible that the ditercalinium molecules, which carry a net positive charge, can continue to interact with the negatively charged DNA phosphate backbone at concentrations exceeding nominal saturation conditions. ‘Outside binding’, mediated principally by electrostatic forces, has long been recognised to occur after saturation of primary DNA binding sites by cationic ligands (36). Because of the intrinsic rigidity of the ditercalinium molecule, such secondary interaction could contribute to the observed increase in persistence length.

It is interesting to note that ethidium bromide binding at an average of 20 molecules per DNA fragment resulted in complexes having an apparent persistence length comparable to that of DNA molecules bis-intercalated with ditercalinium at a similar level. It is known that ethidium bromide binding causes a transition to a structure that is stabilised as compared to non-complexed DNA molecules. The contribution to this stability is predominantly dispersion energy (28). Our results illustrate some of the complex effects on the properties of DNA caused by intercalation. Of course, DNA is a target for a great variety of ligands, many of which have therapeutic or toxicological significance. Further investigations of concentration dependence and sequence specificity of the observed effects using other compounds will help to correlate these observations to the structural changes induced by small molecule binding.

Acknowledgments

ACKNOWLEDGEMENTS

This work was funded by the Biotechnology and Biological Sciences Research Council (Grant BI 11179 to R.M.H. and J.M.E.) and by a grant from the Cancer Research Campaign (to M.J.W.). T.B. acknowledges the support of a Research Fellowship at Emmanuel College, Cambridge. N.S.J. was supported by grants from the Cuthbert Fund of Fitzwilliam College and the Cambridge University H.E.Durham Fund.

REFERENCES

  • 1.Dervan P.B. and Burli,R.W. (1999) Sequence-specific DNA recognition by polyamides. Curr. Opin. Chem. Biol., 3, 688–693. [DOI] [PubMed] [Google Scholar]
  • 2.Ansari A.Z., Mapp,A.K., Nguyen,D.H., Dervan,P.B. and Ptashne,M. (2001) Towards a minimal motif for artificial transcriptional activators. Chem. Biol., 8, 583–592. [DOI] [PubMed] [Google Scholar]
  • 3.Pelaprat D., Delbarre,A., Le Guen,I., Roques,B.P. and Le Pecq,J.B. (1980) DNA intercalating compounds as potential antitumor agents. 2. Preparation and properties of 7H-pyridocarbazole dimers. J. Med. Chem., 23, 1336–1343. [DOI] [PubMed] [Google Scholar]
  • 4.Markovits J., Pommier,Y., Mattern,M.R., Esnault,C., Roques,B.P., LePecq,J.B. and Kohn,K.W. (1986) Effects of the bifunctional antitumor intercalators ditercalinium on DNA in mouse leukaemia-L1210 cells and DNA topoisomerase II. Cancer Res., 46, 5821–5826. [PubMed] [Google Scholar]
  • 5.Delbarre A., Delepierre,M., Langlois d’Estaintot,B., Igolen,J. and Roques,B.P. (1987) Bisintercalation of ditercalinium into a d[CpGpCpG]2 minihelix: structure and dynamics aspects—a 400-MHz 1H-NMR study. Biopolymers, 26, 1001–1033. [DOI] [PubMed] [Google Scholar]
  • 6.Peek M.E., Lipscomb,L.A., Haseltine,J., Gao,Q., Roques,B.P., Garbay-Jaureguiberry,C. and Williams,L.D. (1995) Asymmetry and dynamics in bis-intercalated DNA. Bioorg. Med. Chem., 3, 693–699. [DOI] [PubMed] [Google Scholar]
  • 7.Mendoza R., Markovits,J., Jaffrezou,J.P., Muzard,G. and Le Pecq,J.B. (1990) DNase I susceptibility of bent DNA and its alteration by ditercalinium and distamycin. Biochemistry, 29, 5035–5043. [DOI] [PubMed] [Google Scholar]
  • 8.Gao Q., Williams,L.D., Egli,M., Rabinovich,D., Chen,S.L., Quigley,G.J. and Rich,A. (1991) Drug-induced DNA repair: X-ray structure of a DNA–ditercalinium complex. Proc. Natl Acad. Sci. USA, 88, 2422–2426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Williams L.D. and Gao,Q. (1992) DNA–ditercalinium interactions: implications for recognition of damaged DNA. Biochemistry, 31, 4315–4324. [DOI] [PubMed] [Google Scholar]
  • 10.Lambert B., Roques,B.P. and Le Pecq,J.B. (1988) Induction of an abortive and futile DNA repair process in E.coli by the antitumor DNA bifunctional intercalator, ditercalinium: role in polA in death induction. Nucleic Acids Res., 16, 1063–1078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Lambert B., Jones,B.K., Roques,B.P., Le Pecq,J.B. and Yeung,A.T. (1989) The noncovalent complex between DNA and the bifunctional intercalator ditercalinium is a substrate for the UvrABC endonuclease of Escherichia coli. Proc. Natl Acad. Sci. USA, 86, 6557–6561. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Lambert B., Segal-Bendirdjian,E., Esnault,C., Le Pecq,J.B., Roques,B.P., Jones,B. and Yeung,A.T. (1990) Recognition by the DNA repair system of DNA structural alterations induced by reversible drug–DNA interactions. Anticancer Drug, 5, 43–53. [PubMed] [Google Scholar]
  • 13.Esnault C., Roques,B.P., Jacquemin-Sablon,A. and Le Pecq,J.B. (1984) Effects of new antitumor bifunctional intercalators derived from 7H-pyridocarbazole on sensitive and resistant L 1210 cells. Cancer Res., 44, 4355–4360. [PubMed] [Google Scholar]
  • 14.Fellous R., Coulaud,D., el Abed,I., Roques,B.P., Le Pecq,J.B., Delain,E. and Gouyette,A. (1988) Cytoplasmic accumulation of ditercalinium in rat hepatocytes and induction of mitochondrial damage. Cancer Res., 48, 6542–6549. [PubMed] [Google Scholar]
  • 15.Dupont J., Schwaller,M.A. and Dodin,G. (1990) Ditercalinium, a nucleic acid binder, inhibits the respiratory chain of isolated mammalian mitochondria. Cancer Res., 50, 7966–7972. [PubMed] [Google Scholar]
  • 16.Garbay-Jaureguiberry C., Barsi,M.C., Jacquemin-Sablon,A., Le Pecq,J.B. and Roques,B.P. (1992) Synthesis and biological activity of new dimers in the 7H-pyrido[4,3-c] carbazole antitumor series. J. Med. Chem., 35, 72–81. [DOI] [PubMed] [Google Scholar]
  • 17.Coury J.E., McFail-Isom,L., Williams,L.D. and Bottomley,L.A. (1996) A novel assay for drug–DNA binding mode, affinity, and exclusion number: scanning force microscopy. Proc. Natl Acad. Sci. USA, 93, 12283–12286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Pope L.H., Davies,M.C., Laughton,C.A., Roberts,C.J., Tendler,S.J.B. and Williams,P.M. (2000) Atomic force microscopy studies of intercalation-induced changes in plasmid DNA tertiary structure. J. Microsc., 199, 68–78. [DOI] [PubMed] [Google Scholar]
  • 19.Utsuno K., Tsuboi,M., Katsumata,S. and Iwamoto,T. (2001) Viewing of complex molecules of ethidium bromide and plasmid DNA in solution by atomic force microscopy. Chem. Pharm. Bull. (Tokyo), 49, 413–417. [DOI] [PubMed] [Google Scholar]
  • 20.Lillehei P.T. and Bottomley,L.A. (2001) Scanning force microscopy of nucleic acid complexes. Methods Enzymol., 340, 234–251. [DOI] [PubMed] [Google Scholar]
  • 21.Kim J.M., Ohtani,T., Sugiyama,S., Hirose,T. and Muramatsu,H. (2001) Simultaneous topographic and fluorescence imaging of single DNA molecules for DNA analysis with a scanning near-field optical/atomic force microscope. Anal. Chem., 73, 5984–5991. [DOI] [PubMed] [Google Scholar]
  • 22.Pietrasanta L.I., Smith,B.L. and MacLeod,M.C. (2000) A novel approach for analysing the structure of DNA modified by benzo[a]pyrene diol epoxide at single-molecule resolution. Chem. Res. Toxicol., 13, 351–355. [DOI] [PubMed] [Google Scholar]
  • 23.Wang J. and Bard,A.J. (2001) Monitoring DNA immobilization and hybridization on surfaces by atomic force microscopy measurements. Anal. Chem., 73, 2207–2212. [DOI] [PubMed] [Google Scholar]
  • 24.Rivetti C., Guthold,M. and Bustamante,C. (1996) Scanning force microscopy of DNA deposited onto mica: equilibration versus kinetic trapping studied by statistical polymer chain analysis. J. Mol. Biol., 264, 919–932. [DOI] [PubMed] [Google Scholar]
  • 25.Frontali C., Dore,E., Ferrauto,A. and Gratton,E. (1979) An absolute method for the determination of the persistence length of native DNA from electron micrographs. Biopolymers, 18, 1353–1373. [DOI] [PubMed] [Google Scholar]
  • 26.Lerman L.S. (1964) Acridine mutagens and DNA structure. J. Cell Comp. Physiol., 64, 1–18. [PubMed] [Google Scholar]
  • 27.Gaugain B., Barbet,J., Capelle,N., Roques,B.P. and Le Pecq,J.B. (1978) DNA bifunctional intercalators. 2. Fluorescence properties and DNA binding interaction of an ethidium homodimer and an acridine ethidium heterodimer. Biochemistry, 17, 5078–5088. [DOI] [PubMed] [Google Scholar]
  • 28.Reha D., Kabelac,M., Ryjacek,F., Sponer,J., Sponer,J.E., Elstner,M., Suhai,S. and Hobza,P. (2002) Intercalators. 1. Nature of stacking interactions between intercalators (ethidium, daunomycin, ellipticine, and 4′,6-diaminide-2-phenylindole) and DNA base pairs. Ab initio quantum chemical, density functional theory, and empirical potential study. J. Am. Chem. Soc., 124, 3366–3376. [DOI] [PubMed] [Google Scholar]
  • 29.Maeda Y., Nunomura,K. and Ohtsubo,E. (1990) Differential scanning calorimetric study of the effect of intercalators and other kinds of DNA-binding drugs on the stepwise melting of plasmid DNA. J. Mol. Biol., 215, 321–329. [DOI] [PubMed] [Google Scholar]
  • 30.Lyubchenko Y.L. and Shlyakthenko,L.S. (1997) Visualization of supercoiled DNA with atomic force microscopy in situ. Proc. Natl Acad. Sci. USA, 94, 496–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Berge T., Ellis,D.J., Dryden,D.T.F., Edwardson,J.M. and Henderson,R.M. (2000) Translocation-independent dimerization of the EcoKI endonuclease visualized by atomic force microscopy. Biophys. J., 79, 479–484. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Spisz T.S., Fang,Y., Reeves,R.H., Seymour,C.K., Bankman,I.N. and Hoh,J.H. (1998) Automated sizing of DNA fragments in atomic force microscope images. Med. Biol. Eng. Comput., 36, 667–672. [DOI] [PubMed] [Google Scholar]
  • 33.Hagerman P.J. (1988) Flexibility of DNA. Annu. Rev. Biophys. Biophys. Chem., 17, 265–286. [DOI] [PubMed] [Google Scholar]
  • 34.Porschke D. (1991) Persistence length and bending dynamics of DNA from electrooptical measurements at high salt concentrations. Biophys. Chem., 40, 169–179. [DOI] [PubMed] [Google Scholar]
  • 35.Bustamante C., Marko,J.F., Siggia,E.D. and Smith,S. (1994) Entropic elasticity of λ-phage DNA. Science, 265, 1599–1600. [DOI] [PubMed] [Google Scholar]
  • 36.Atwell G.J., Stewart,G.M., Leupin,W. and Denny,W.A. (1985) A diacridine derivative that binds by bis-intercalation at 2 contiguous sites on DNA. J. Am. Chem. Soc., 107, 4335–4337. [Google Scholar]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES