DNA sequence-specific ligands. XX. Synthesis, spectral properties, virological and biochemical studies of fluorescent dimeric trisbenzimidazoles DB3P(n)

A new series of AT-specific minor groove DNA ligands (DB3P(n); n = 1,2,3,4) was synthesized and their spectral, biological and virological properties were investigated. With a methylene spacers of different lengths blocks of three AT pairs located at different distances from each other could be recognized. The compounds synthesized suppressed the activity of HIV-1 integrase at submicromolar concentrations (0.25–0.50 µМ). Also, DB3P(n) were found to be effective inhibitors of simplex virus type I and DNA topoisomerase I. The synthesized DB3P(n) demonstrated good solubility in water, could penetrate through cell and nuclear membranes, and stain DNA in live cells.


Introduction
Currently one of the most promising approaches to suppress transcription processes is the synthesis of ligands with high binding constants with DNA and site selectivity. Lowmolecular-weight site-specific forming noncovalent complexes with DNA can be divided into two groups depending on their type of action. They are intercalators and agents binding to DNA either in the major or minor grooves. The DNA minor groove plays an important role in functioning of a large number of DNA-dependent enzymes and in the allosteric regulation of transcription factors. Therefore, it is a promising target for evaluation of biological activity of synthetic therapeutic agents and compounds specifically interacting with DNA sequences [1,2]. Minor groove ligands form hydrogen bonds in each of the DNA strands thus linking them. Such interactions result in more stable complexes, which was evidenced by their higher melting temperatures. In particular, antibiotics netropsin and distamycin A, fluorescent dyes Hoechst 33258, DAPI as well as berenyl and similar diamidines belong to minor grove specific ligands [3]. Benzimidazole derivatives are structurally close to DNA purine bases and were found in a number of biologically important natural compounds, such as vitamin B 12 [4]. Benzimidazole derivatives demonstrated a wide spectrum of biological activity [5][6][7][8][9][10][11].
Previously we described the design and synthesis of several sets of dimeric bisbenzimidazoles [12][13][14][15][16][17] bearing a bisbenzimidazole motif of the fluorescent AT-specific Hoechst 33258 dye. These dimeric bisbenzimidazoles were synthesized with the goal of developing AT specific ligands noncovalently binding to the dsDNA minor groove and occupying nearly one coil. These dimeric molecules could bind to two sites, each consisting of two AT pairs separated by two or more nucleotide pairs. Dimeric bisbenzimidazoles demonstrated a wide spectrum of biological activity and were shown to be effective agents in biochemical, virological, and genetic studies [12][13][14][15][16][17].
In this work we described the synthesis and physicochemical studies of monomeric trisbenzimidazole MB 3 and its dimerization to dimeric DB 3 P(n), where n is the number of methylene groups in the linker joining two trisbenzimidazole blocks (Fig. 1).
The structure difference of DB 3 P(n) from dimeric DBP(n) [13] was the presence of two additional benzimidazole residues, which were introduced in the molecule with the goal of increasing the sequence specificity and affinity to dsDNA (Fig. 1). In DB 3 Р(n) molecules trisbenzimidazole fragments were joined in a tail-to-tail manner with oligomethylene linkers of different lengths (n = 1, 2, 3, 4), due to which their structure was isogeometrical to that of DNA in the minor groove. Due to such a structure DB 3 Р(n) could function as bidentate ligands capable of recognizing nucleotide sequences of blocks containing three AT pairs located at different distances from each other. A 1,4piperazine residue in the bridge linking two benzimidazole blocks enhanced solubility of the molecule by transforming it into a tetracation at neutral pH. The scheme of the DB 3 Р(n)-DNA interaction is shown in Fig. 2.

Chemical synthesis
The synthesis of dimeric trisbenzimidazoles is shown in Fig. 3.
Monomeric trisbenzimidazole МВ 3 was synthesized from 5-(4-methylpiperazin-1-yl)benzene 1,2-diamine (I). It was coupled with imidoester (II) obtained by the Pinner reaction from the corresponding nitrile. Catalytic hydrogenation of the resulting compound (III) yielded diamino derivative (IV), interaction of which with imidoester (II) yielded nitroamino derivative (V). The hydrogenation of the latter in the presence of palladium catalyst resulted in diamine (VI). Bisbenzimidazole (VII) prepared by the reaction of (VI) with N-tert-butyloxycarbonyl glycine using the mixed anhydride approach was refluxed in acetic acid to give N-acetyl derivative (VIII), whose deacetylation in boiling concentrated hydrochloric acid resulted in monomeric trisbenzimidazole MB 3 .
Dimeric trisbenzimidazoles DB 3 P(n) were obtained in yields of 89-94% by the reaction of MB 3 with 1,4-piperazine dialkylcarboxylic acids (2 : 1 molar ratio) in the presence of HBTU. All the synthesized compounds were homogeneous by the TLC data and well soluble in water. All of them manifested good fluorescent properties. Their structures were confirmed by 1 H NMR, 13 C NMR and MALDI-TOF spectra.

Physicochemical studies
Interaction of DB 3 P(n) with DNA was studied by a number of spectral methods. Absorption spectra of DB 3 Р(1-4) were registered both in the absence and presence of DNA and were similar. The spectra of DB 3 P(2) and DB 3 P(4) in the presence of DNA at different concentrations are shown in Fig. 4 (A, B, curves 1-6) as an example. An increase in the DNA concentration (relative to a base pair) resulted increase in optical density and bathochromic shift confirming formation of DB 3 Р(n)-DNA complexes.
Fluorescence spectra of DB 3 Р(1-4) were registered both in the absence and presence of DNA. DB 3 P(n) fluoresced at the excitation at 350 nm with the emission maximum of 475 nm. The DB 3 P(2) and DB 3 P(4) spectra in the presence of DNA at different concentrations are shown in Fig. 5 (A and B, curves 2-7). Fluorescent spectra of DB 3 P(n) also evidenced the formation of their complexes with DNA, since a manifold increase in the fluorescence intensity in the presence of DNA was observed (Fig. 5).
We determined the DB 3 P(n) localization in the DNA groove using the method of circular dichroism (CD) with DNA cholesteric liquid crystal dispersions (preparation of such Fig. 2 Scheme of the DB 3 P(n) complex with two DNA sites composed of three consecutive AT pairs. Dotted lines are hydrogen bonds. DNA is shown in bold Fig. 1 The structure of monomeric trisbenzimidazole MB 3 , dimeric bisbenzimidazole DBP(n) and dimeric trisbenzimidazole DB 3 P(n) dispersions was described in [13] (Fig. 6)). An appearance of the positive values of CD in the region of the absorption of DB 3 P(n) chromophores indicate that the angle between the plane of the chromophores of the ligand and the helical axis of the DNA molecule is less than 54 0 . This could only be possible if the ligand chromophores were located in the DNA groove [18,19]. Since according to the X-ray diffraction data, Hoechst 33258 binds to DNA in the minor groove [20,21] and DB 3 P(n) are Hoechst 33258-derived compounds, we assumed that they are also localized in the DNA minor groove, i.e., are minor groove binding ligands.
Comparison of the fluorescence spectra of dimeric trisbenzimidazoles DB 3 P(2) and DB 3 P(4) with polydeoxyribonucleotides poly(dG-dC) * poly(dG-dC) and poly(dA-dT) * poly(dA-dT) demonstrated their unambiguous AT-specificity. There is practically no increase in fluorescence intensity with increasing concentration of poly(dG-dC) * poly(dG-dC) (Fig. S1A), in contrast to the sequential addition of poly(dA-dT) * poly(dA-dT) (Fig. S1), accompanied by intense fluorescence build-up. This confirms the presence of a clearly expressed AT-specific interaction of the studied compounds with poly(dA-dT) * poly(dA-dT). The binding constants of DB 3 P(2) and DB 3 P(4) were 2 orders of magnitude higher when bound to poly(dA-dT) * poly(dA-dT) compared to poly(dG-dC) * poly(dG-dC) ( Table 1).

Biological results
The study of antiviral activity of dimeric trisbenzimidazoles DB 3 P(n) and bisbenzimidazoles DBP(n) towards HIV-1 integrase The compounds were evaluated as HIV-1 integrase inhibitors. This enzyme supports the viral DNA integration into a human genome and is an attractive target for HIV-1 therapy [22]. Monomeric trisbenzimidazole МВ 3 was found to inhibit the 3'-processing reaction catalyzed by recombinant integrase with IC 50 1,2 ± 0,3 µМ. Dimeric trisbenzimidazoles DB 3 P(n) were 2-to 5-fold more active than MB 3 (Таble 1) and could block integrase at submicromolar concentrations. The most active compound was DB 3 P(3), IC 50 0,24 ± 0,07 µМ ( Table 2). It is noteworthy that the activity of dimeric trisbenzimidazoles DB 3 P(n) was 10-to 20-fold higher than that of structurally similar dimeric bisbenzimidazoles DBP(n) [13] (Table 2). Trisbenzimidazole residue ensured more stable interaction with DNA than the bisbenzimidazole fragment. Evaluation of antiviral activity of DB 3 P(n) against herpes simplex virus type 1 and cytomegalovirus Antiviral activity of dimeric trisbenzimidazoles DB 3 P(n) was tested on a cell model infected with herpes simplex virus type 1 (HSV-1) and cytomegalovirus (CMV) [23]. Antiviral activity of the compounds was assessed by their capacity to inhibit HSV-infection at different stages of the virus life cycle. To this end, the compounds were introduced into cells both before and after infection, which enabled modelling of different modes of action, microbicidal (pretreatment assay) and therapeutic (post-treatment assay).

Evaluation of cytotoxicity
For the evaluation of cytotoxic properties of the tested compounds the MTT assay was used. Most of them were shown to be low toxic towards both cell lines, Vero (African green monkey kidney) and human fibroblast cells (HF) cells. The toxicities of two compounds, DB 3 P(3) and DB 3 P(4), were slightly higher in Vero cells ( Table 3).
Studies of antiviral properties of the DB 3 P(n) towards HSV-1 and CMV Antiviral activity of DB 3 P(n) towards HSV-1 and CMV was tested in three infection variants: infection prior to introduction of the compounds under study (microbicidal assay), post-treatment (therapy assay), and infection with the virus preliminarily incubated with the compounds under study (virulicidal assay). The tested compounds inhibited the HSV-1 infection at lower concentrations if compared with CMV (Table 3). Compounds DB 3 P(1) and MB 3 demonstrated outstanding virulicidal properties: their selectivity indices SI (СС 50 /IС 50 ) were 7021 and 6369 respectively. Also, these compounds manifested microbicide properties with SI 1112 and 381. Compounds DB 3 P(3) and DB 3 P(4) demonstrated antiviral therapeutic effects (SI 133 and 1333 respectively). The latter also displayed virus-neutralizing activity (SI 800 in the virulicidal assay). According to current views, the agents with SI higher than 100 are promising for further in vivo studies on laboratory animals. Compounds DB 3 P(1-4) and МВ 3 were shown to be low active inhibitors of the CMV-induced infection.   The results evidenced that the compounds under study can interact with various targets. When interacting with viruses the compounds could prevent the virion penetration into cells (virulicide activity). The interaction with molecules on the cell surface supported masking or modification of glycoproteins required for adsorption thus preventing virus penetration (microbicide activity). When interacting with intracellular molecules they could inhibit the infection development (therapeutic activity).

DB 3 P(n) and MB 3 cytotoxicity
Cytotoxic studies were performed using the MTT assay on tumor human glioblastoma U251 cell line and normal human kidney epithelium fibroblasts (NKE-hTERT). The comparison of DB 3 P(n) and DBP(n) series demonstrated that cytotoxicity of DB 3 P(n) was higher towards rapidly amplifying NKE-hTERT cells and similar with DBP(n) towards U251 (Fig. S2).

The DB3P(n) and MB 3 penetration into cells
The compounds under study DB 3 P(n) and MB 3 can penetrate through plasma and nuclear membranes for a day and followed by chromatin staining. In the experiments a human glioblastoma U251 cell line was used. Visualization was performed using a phase contrast microscope in the ultraviolet wavelength range (Fig. S3). As a control, a classical fluorescent dye Hoechst 33258 was used.
The DB 3 P(1-4) and Hoechst 33258 stained the nuclei U251 cell as well. The intensive increase of fluorescence demonstrates very good signal-to-noise ratio, what eliminates any autofluorescence. The DB 3 P(4) fluorescence is lower due to less effective accumulation.
Inhibition of the catalytic activity of eukaryotic topo I Using a DNA-relaxation assay we compared the effect of the synthesized compounds on the catalytic activity of topo I, which is one of the most widely used targets for anticancer agents. The compounds under study were added to scDNA at various concentrations. The mixture was incubated for 20 min at 37 0 C for the complex formation and topo I and the reaction buffer were added.
Inhibitory properties of dimers DB 3 P(n) towards topo I were more potent than those of monomers MB 3 . In a single experiment inhibition of topo I was observed for DB 3 P(n) at a concentration of 10 µМ, whereas for monomer MB 3 it was not detected even at a concentration of 20 µМ (Fig. S4).
Inhibitory activity towards topo I of dimeric trisbenzimidazoles DB 3 P(n) was higher than that of dimeric bisbenzimidazoles DBP(n) [13]. As is seen in Fig. S5, scDNA was accumulated in the presence of 20 µM DB 3 P(2 ) and DB 3 P(4), whereas for DBP(2) and DBP(4) at the same concentrations scDNA was not observed.
Compounds DB 3 P(n) smoothly penetrate into cell nuclear. They are more cytotoxic towards conditionally normal cells if compared with tumor cells but less cytotoxic than a commonly used drug cisplatin.
The inhibition efficiency of topo I function is dosedependent for DB 3 P(n), which indicates that activity suppressed by DNA-ligand. The obtained results indicate that the greater the number of methylene groups in the structure of the DB 3 P(n) spacer, the more effective the inhibition occurs. The DB 3 P(4) is at least two times more efficient than DB 3 P(2), and sufficiently productive than dimeric bisbenzimidazoles DBP(n) series and Hoechst 33258 (Fig.  S6). By this point of view the greater length of molecule and higher affinity to DNA enhance the inhibition efficiency of topo I function.

Conclusions
In this work we described the synthesis of fluorescent monomeric trisbenzimidazole (MB 3 ) as well as its dimerization to give dimeric DB 3 P(n), where n is the number of methylene groups in the linker joining two trisbenzimidazole blocks. The residue of 1,4-piperazine has been introduced into the structure of the oligomethylene linker of the DB 3 P(n) molecules in order to increase their solubility in aqueous solutions and affinity of the DNA complexes. Due to such a structure DB 3 Р(n) can function as bidentate ligands capable of recognizing nucleotide sequences of blocks of three AT pairs located at different distances from each other. Interaction of DB 3 P(n) with DNA was investigated by a complex of spectral methods using UV-vis absorption spectroscopy, fluorescence spectroscopy and the method of circular dichroism (CD) with DNA cholesteric liquid crystal dispersions. All methods indicated the localization of the DB 3 P(n) in the DNA groove. Comparison of the fluorescence spectra of dimeric trisbenzimidazoles DB 3 P(2) and DB 3 P(4) with polydeoxyribonucleotides poly(dG-dC) * poly(dG-dC) and poly(dA-dT) * poly(dA-dT) demonstrated their unequivocally AT-specificity. Virological and biochemical studies of dimeric trisbenzimidazoles DB 3 P(n) were performed. The DB 3 P(n) were shown to effectively inhibit the HIV-1 integrase catalytic activity at submicromolar concentrations (0.25-0.50 µМ). It was found that DB 3 P(n) could inhibit herpes simplex virus type I and weakly suppress the cytomegalovirus activity. Also, DB 3 P(n) inhibited DNA topoisomerase I at low micromolar concentrations (~10 µМ). All of the dimeric trisbenzimidazoles DB 3 P(n) manifested fluorescent properties, were well soluble in water, and could penetrate into the cell nucleus with nuclear staining.

Materials and methods
In this work, HBTU and NMM were from Alfa Aesar (England); DIPEA was from Fluka (Germany); DMF, AcOEt, iPrOH, and NH 4 OH were from Reachim (Russia). Poly(dA-dT) * poly(dA-dT) and Poly(dG-dC) * poly(dG-dC) from Sigma (USA). Organic solutions were dried with Na 2 SO 4 . The solvents were evaporated in vacuum of a water-jet pump at 30-50°C. The compounds were dried in vacuum with P 2 O 5 and NaOH. Melting points are uncorrected and were measured on a Boethius apparatus (Germany). Hydrogenation was carried out in the presence of 10% Pd/C (Merck, Germany) at atmospheric pressure and room temperature until hydrogen evolution was stopped. The compound purity was determined by TLC on Kieselgel 60 F 254 plates (Merck, Germany) in the following systems: A, MeOH-conc.NH 4 OH (25:1); B, MeOH-TFA-H 2 O (5:1:2). The compounds were visualized using UV irradiation (254 nm) and/or by fluorescence at 365 nm. 1 H and 13 C NMR spectra were registered using an Avance III 300 MHz spectrometer (Bruker, Germany) and Bruker Avance III 800 MHz spectrometer (Bruker, Germany), equipped with the triple resonance TCI cryogenic probe (Bruker Biospin Gmbh, Germany) in DMSO-d 6 at 30°C. Chemical shifts (δ) are given in ppm relative to the resonances of solvents ( 1 H, δ 2.50 for DMSO-d 6 ; 13 C, δ 39.52 for DMSO-d 6 ). Hydrogen atoms of benzimidazole cycles are defined as (Ar), N-H group of benzimidazole, as (bim), and piperazine, as (pip). Mass spectra were registered on a time-of-flight AB SCIEX 4800 plus mass-spectrometer (AB SCIEX, United States) in the positive ion mode; 2,5-dihydroxybenzoic acid was used as a matrix.

Spectral measurements
UV-vis absorption spectra were recorded on a Shimadzu UV-3101 PC spectrophotometer (Japan) in the range from 250 to 650 nm in quartz cells with an optical pathlength of 1 cm, unless otherwise noted. Stationary fluorescence measurements were performed on a FluoTime 300 fluorimeter, a Xe lamp was used as an excitation source (350 nm). The binding constant was determined by the hyperbolic function from the equation for approximating the dependence of the fluorescent signal intensity on the polydeoxyribonucleotide concentration [24]. The binding constants of DB 3 P(n) with polydeoxyribonucleotides were determined using fluorescence spectroscopy. The binding constant K b was calculated from the curves in accordance with the formula (1) [24]: where Θ = (F -F 0 )/(F ∞ -F 0 ) is the proportion of the dye associated with the polydeoxyribonucleotide; F 0 , F ∞ and F are the fluorescence intensities at [DNA] = 0 and at complete and intermediate binding of DB 3 P(n) to polydeoxyribonucleotide, respectively. CD spectra were recorded on a portable SKD-2 dichrometer (Troitsk, Russia). DNA from salmon sperm (Derinat, Russia) was used.

Monomeric trisbenzimidazole MB 3
A solution of compound (VII) in AcOH (30 mL) was refluxed for 10 h, evaporated, and the resulting oil was diluted with conc. NH 4 OH (30 mL). The mixture was kept for 4 days at 4°С and the grey powder without further purification was dissolved in conc. HCl. The mixture was refluxed for 5 h and poured into iPrOH (100 mL General procedure for preparation of compounds DB 3 P(1-4) To a solution of 1,4-piperazine dialkylcarbonic acids dihydrochloride (0.1 mmol) [16] in abs. DMF (2 mL) HBTU (0.25 mmol) and DIPEA (0.50 mmol) were added. The mixture was stirred at room temperature for 30 min and MB 3 (0.10 g, 0.2 mmol) was added. The mixture was stirred for 1 h and kept overnight at room temperature. The solvent was evaporated and the residue was treated with abs. ethanol to give a suspension. A solution of 35% HCl in dioxane (0.5 mL) was added and the precipitate, a yellow bulk powder, was filtered and dried in vacuum with NaOH/P 2 O 5 . The target DB 3 P(n) was homogeneous according to TLC data (system B).

Preparation of Recombinant Integrase
The recombinant HIV-1 integrase was expressed in Escherichia coli strain Rosetta and purified without addition of detergent as described in [25].

Inhibition of 3'-End Processing
Inhibition of 3'-end processing was performed as described in [25]. Briefly, a 32P-labeled duplex U5B/U5A (3 nM) was Viruses Reference HSV-1 F strains and CMV AD169 were used. Their activities were evaluated on Vero (HSV-1) and HF (CMV) cell cultures using the modified plaque reduction assay. Cytotoxicity of the tested compounds towards Vero cells sensitive to HSV-1 and HF cells sensitive to CMV were assessed by the MTT method. The percentage of viable vs the total number of cells in the population 72 h after addition of the tested compounds was calculated. The CC 50 values were calculated using the Microsoft Excel program. Antiviral activity of the tested compounds was studied using a standard plaque reduction assay.
Microbicidal assay (express prophylaxis) A medium supplemented with 2% serum (supporting medium) and the tested compounds at various concentrations was loaded on the cell monolayer and the mixture was incubated for 1 h at 37°С. The viruses were introduced the mixture was incubated for 1 h at 37°С. The cell monolayer was washed and the supporting medium was added.
Post-treatment assay (therapy) The viruses were introduced on the monolayer and the mixture was incubated for 1 h at 37°С. The cells were washed twice and the supporting medium containing the tested compounds at various concentrations was added.
Virucidal (virus-neutralization) assay The viruses were incubated with the tested compounds at various concentrations for 1 h at 37°С. The incubation mixture was loaded on the cell monolayer and the mixture was incubated for 1 h at 37°С. The cell monolayer was washed twice and the supporting medium was added.
As a control, the infected cells untreated with the tested compounds were used. Antiviral activity was assessed by calculating the number of infected cells in the experiment vs control in 48 h (HSV-1) and 96 h (CMV) after infection. For the calculation of IC 50 , a concentration inhibiting virus plaque formation by 50% relative to the control, the dependence of the inhibition degree of viral activity vs the compound concentration was plotted. The IC 50 values were calculated using a Microsoft Excel packet. The selectivity index (SI) of compounds was calculated as the ratio of CC 50 to IC 50 .

Inhibition of the catalytic activity of eukaryotic topo I
The human cell lines U251 were purchased from American Type Culture Collection; ATCC, Manassas, VA, USA. A normal human kidney epithelia lcell line NKE-hTERT cell line was kindly provided by K. Gurova (Roswell Park Comprehensive Cancer Center, USA). DMSO and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT-reagent) were purchased from Sigma Chemical Co., topoisomerase from Invitrogen (ThermoFisher, USA). The stock solutions of DB 3 P(n) and MB 3 were made in DMSO.

Cell culture
All cell lines were cultured in DMEM (Dulbecco's Modified Eagle Medium) supplemented with 10% fetal calf serum (BioWhittaker, Belgium), 2mM L-glutamine, 100U/ml penicillin, and 100 U/ml streptomycin at 37°С, 5% СО 2 in humidified atmosphere. Cells in logarithmic phase of growth were used in all experiments. The final concentration of DMSO was 0,5% in the culture medium. DMSO at this concentration didn't affect the viability of the cells.

Cytotoxicity and penetration into cells
Various human cells were seeded in 96-well plates containing 190 µl of medium at 3000-4000 cells/well and allowed to attach overnight. After 24 h plates were incubated with varying concentration of the studied compounds during 72 h. The cell viability was determined by the MTTassay. The light absorbance of converted dye in living cells was measured at 570 nm wavelength using a microplate spectrophotometer (Benchmark Plus, BioRad).
Penetration of the tested compounds into U251 human glioblastoma cells was evaluated using a fluorescent microscope (Zeiss, Germany) in the UV area at 40X magnification. Dimeric trisbenzimidazoles DB 3 P(n) were added to the cell culture at a final concentration of 20 µM. After one day the cells were washed with PBS and fixed with paraformaldehyde (4%).
Inhibition of catalytic activity of eukaryotic topoisomerase I in the relaxation reaction of supercoiled DNA Modulation of topo I activity in vitro was studied using the Topoisomerase I Drug Screening kit (TopoGen). 3,5 units of purified topoisomerase from calf thymus (Fermentas, Lithuania) and the studied compounds were incubated with 0.2 μg of supercoiled plasmid DNApHOT1 (TopoGen) in reaction buffer (10 mM Tris . HCl, pH 7.9: 1 mM EDTA, 0.15 M NaCl, 0.1% BSA, 0.1 mM spermidine, 5% glycerol). The mixture was incubated for 30 min at 37°C, the reaction was stopped by addition of SDS to the final concentration 1%, and the reaction mixture was treated by proteinase K at the final concentration 50 μg/ml for 30-60 min at 37°C. The reaction products were separated electrophoretically in 1% agarose gel with TAE buffer (2 M Tris_base, 0.05 M EDTA, 1.56 M acetic acid) at the maximal voltage 3-4 V/cm. The gel was then stained with 0.5 μg/ml ethidium bromide. The presence of DNA in the gel was visualized by UV fluorescence with wavelengths from 240 to 360 nm. In the absence of inhibitor, topo I relaxed scDNA with formation of a series of relaxed topoisomers. Topo I inhibition was revealed by the ability of the studied compounds to retard scDNA relaxation, that is via the decreased composition of migrating topoisomers and restoration of scDNA.