Spectroscopic and electrochemical study of interactions between DNA and different salts of 1,4-dihydropyridine AV-153

Chemicals

Different AV-153 salts (3,5-bis-ethoxycarbonyl-2,6-dimethyl-1,4- dihydropyridine-4-carboxylate) were synthesized in the Laboratory of Membrane Active Compounds at the Latvian Institute of Organic Synthesis. The structure of the AV-153 salts are shown in Fig. 1. The synthesis of the 2,6-dimethyl-3.5-bis(ethoxycarbonyl)-1,4-dihydroisonicotinic acid was performed as reported (Duburs & Uldrikis, 1969). Salts of monovalent metals (Na, Li, K, Rb) were prepared following the same procedure. At the final stage of the synthesis, the saturated water solution of the appropriate hydroxide was added to the hot 2,6-dimethyl-3,5-bis(ethoxycarbonyl)-1,4-dihydroisonicotinic acid solution in ethanol. Salts of divalent metals (Ca and Mg) were prepared by transmetallation reaction to the sodium salt solution with the appropriate metal chloride in a water solution at room temperature. The obtained salts were dried in vacuo on phosphorus pentoxide and were fully characterized by ¹H NMR, LC/MS and elemental analysis. The purity of the salts determined by HPLC exceeded 95%. Tris base, ethidium bromide (EtBr), calf thymus DNA (ct-DNA), human serum albumin (HSA), Na₂EDTA, and inorganic salts were purchased from Sigma-Aldrich (Taufkirchen, Germany).

Cyclic voltammetry

For the voltammetric experiments we used an EcoChemie Autolab PGSTAT 302T potentiostat/galvanostat (Utrecht, The Netherlands) and the electrochemical software package Nova 2.0. We used a three-electrode system: a two mm-sized Pt disk working electrode, an Ag/AgCl reference electrode (3 M KCl) and a Pt wire counter electrode purchased from Metrohm Co (Herisau, Switzerland). Voltammograms of the 5 mM AV-153 salt solution in 0.1 M Tris–HCl (pH = 7.4) were first registered. After that, 10 µM of DNA was added and the measurements were repeated. This was repeated twice or more. Scan rate was 100 mV/s. Electrodes were washed with double-distilled water before each measurement. Experiments were performed at 25 °C. Samples were deoxygenized with oxygen-free nitrogen gas.

The binding constant was determined using the equation: ${\displaystyle log\left(\frac{1}{DNA}\right)=log\left(K\right)+log\left[I_{\mathrm{free}}∕\left(I_{\mathrm{free}}-I_{\mathrm{bond}}\right)\right]}$

where K is the apparent binding constant; I_free is the peak current of free compound; and I_bond is the peak current of compound in the presence of DNA (Feng, Li & Jiang, 1997).

The number of binding sites was determined using the equation: ${\displaystyle \frac{I-I_{DNA}}{I_{DNA}}=\frac{K\left[DNA\right]}{2s}}$

where I is the peak current of a compound in the absence of DNA; A, IDNA is the peak current of a compound in the presence of DNA; A, K is the binding constant of compound-DNA complex; [DNA] is the DNA concentration, mol/L; s is the size of the binding site (bp) (Aslanoglu, 2006). The number of electrons (n) was calculated using the equation: ${\displaystyle E_p-\frac{E_p}{2}=47.7\mathrm{mV}∕\mathrm{\alpha }n}$

where Ep is the peak compound potential; mV; Ep/2 is the half-wave compound potential; mV, α –the assuming value = 0.539; n—number of electrons (Wang et al., 2011).

Fluorescence spectroscopic measurements

Spectrofluorimetry was performed using a Fluoromax-3 (Horiba JOBIN YVON, China).

To study the interaction of 1,4-DHP with the DNA-EtBr complex by means of the spectrofluorimetry, the calf thymus DNA (74.8 µM) and ethidium bromide (1.26 µM) were diluted in 5 mM Tris HCl; 50 mM NaCl at pH 7.4. Spectra were recorded in a 1-cm cuvette (2 ml) at room temperature. The AV-153 salts were added by 8 µl aliquots of the 2.5 mM solution. The solution was mixed thoroughly and kept for 5 min to equilibrate before the fluorescence measurement. Interactions of the tested compounds with DNA caused displacement of EtBr from the complex with DNA. The fluorescence was recorded at 109,600 nm using an indirect excitation wavelength of EtBr at 260 nm (Geall & Blagbrough, 2000). The linear Stern-Volmer equation, in which 1,4-DHPs were the quenchers, was applied for calculation of the quenching constants: ${\displaystyle \frac{I_0}{I}=1+K_{sv}\left[Q\right]}$

Where I₀ and I represent the fluorescence intensities in the absence and presence of quenchers, respectively; K_sv is a linear Stern-Volmer constant; and Q is the quencher concentration. K_sv values were evaluated from the slope of the plot (Geethanjali et al., 2015). Linearity of the plots was tested by calculation of the correlation coefficient (Gong & Zhu, 2013; Wahba, El-Enany & Belal, 2015).

Circular dichroism spectroscopy

CD spectra were recorded on a Chirascan CS/3D spectrometer (Applied Photophysics, Surrey, UK), DNA and compound binding measurements were done in 10 mM HEPES buffer, and pH 7.4 in a quartz cell of 10 mm path-length at room temperature. The CD spectra of ct-DNA and G4 structure were recorded in 220–320 nm range, and that of the HSA in a 200–260 nm range. The parameters for all spectra were as follows: scan rate (200 nm min⁻¹), averaging time (0.125 s), bandwidth (1 nm) and one recorded spectrum is the average of four scans. We carried out titrations in the DNA region by adding increasing amounts of AV-153 salts (10 µM at each step) to a 50 µM DNA solution. Titrations in the induced CD compound region were performed by adding DNA (62.5 µM at each step) to a 500 µM AV-153-Na solution.

UV/VIS spectroscopic measurements were applied for a study of the AV-153 salts interaction with bases (Sadeghi et al., 2016) and oligonucleotides corresponding to human telomere repeat sequence. The spectra were recorded with a Perkin Elmer Lambda 25 UV/VIS spectrophotometer. Spectra of a 25 µM solution of the tested compound in 50 mM NaCl and 5 mM Tris HCl at pH 7.4.were taken in a one cm quartz cell (2 ml) in the absence of bases and after adding 10 µM of bases.

In the case of human telomere repeats the titration was performed in 10 mM sodium phosphate buffer, 0.3 mM EDTA, 100 mM NaCl, pH 7.2, 50 nM of oligonucleotide were added each time.

The binding constants were calculated using the formula: ${\displaystyle \frac{1}{A_0-A}=\frac{1}{A_0}+\frac{1}{K\times A_0\times c_{DNA}}}$

Where A₀ is absorption of the free substance; A is absorption in the presence of DNA; and C_DNA is the DNA concentration (Buraka et al., 2014).

Fourier transform infrared (FTIR) spectroscopy

For the FTIR analyses AV-153-Li and AV-153-K were dissolved in distilled water (H2O), but AV-153-Ca, AV153-Mg, and AV-153-Rb were dissolved in dimethyl sulfoxide (DMSO). Aliquots containing from 5 to 10 µL of 7.5 mM DNA solutions, AV-153 salts or a mixture of DNA and the salts in a ratio of 1:5 were dried at T<50 °C on a 384-well silicone plate. The FTIR absorption spectra were recorded on a VERTEX 70 coupled with an HTS-XT microplate reader extension (BRUKER, Germany) over a 4,000–6,000 cm⁻¹ range with a four cm⁻¹ resolution, with 64 scans co-added. The baseline was corrected using the rubber band method, and CO₂ bands were excluded. Spectra with absorption limits between 0.25 and 0.80 were used for data analyses in accordance with the Lambert-Bouger-Beer law, that is, the concentration of a component should be proportional to the intensity of the absorption band. The data were processed using OPUS 6.5 software.

Interaction with single-stranded and guanine quadruplex forming oligonucleotides

Oligonucleotide corresponding to human telomere repeat (5′-AGGGTTAGGGTTAGGGTTAGGG-3′) was ordered in Metabion. To induce the formation of the antiparallel G4, the oligonucleotide was heated at 90° and allowed to slowly cool down for 1 h at room temperature and then for 3 d in 10 mM sodium phosphate buffer, 0.3 mM EDTA, 100 mM NaCl, pH 7.2 at +4 °C in a 10 mm path-length quartz cell at room temperature. The G4 formation was checked with circular dichroism spectroscopy (CD). To observe the interactions of AV-153 salts with the oligos, 5 µM of the oligonucleotide were titrated in 10 mM sodium phosphate buffer, 0.3 mM EDTA, 100 mM NaCl, pH 7.2 with 10 µM of AV-153 salt at each step, CD spectra were recorded after each stabilization of the solution (5 min) in a 220–320 nm range at room temperature. In a parallel binding of the oligo in G4 and single-stranded form (before the 3rd day of incubation in the presence of salt) was assayed with UV/VIS spectroscopy, as described above. A solution of AV-153 salts of monovalent metals (25 µM) in the buffer was titrated with either G4 or single-strand oligonucleotides, 50 nM each time.

Statistics

All measurements were performed in triplicate. Numerical data are presented as mean ± S.E.M.

Electrochemical study of the interactions between AV-53 salts and DNA

Cyclic voltammograms of 5 mM AV-153-K in the absence and presence of various DNA concentrations in 0.1 M Tris–HCl buffer, pH = 7.4 are shown in Fig. 2. The peak potential shifted to a more positive value (0.018 V) in the presence of DNA. A slight increase in the peak current upon the addition of increasing concentrations of DNA was also observed. The compound exhibited a single well defined anodic peak, which corresponds to the oxidation of the dihydropyridine ring (Augustyniak et al., 2010). No peak was observed in a reverse scan, indicating that the oxidation of the compound is an irreversible process. The peak current (Ip) of the oxidation wave AV-153-K was proportional to the square root of the scan (v^1∕2). The cyclic voltammograms of other AV-153 salts were similar (not shown). The differences in the binding constants of the AV-153 salts were calculated using cyclic voltammetry data followed the same trend as affinity to DNA formerly determined spectrofluorimetrically (Leonova et al., 2019) (see Table 1).

Fluorescent intercalator displacement assay

To assess the mode of the DNA interaction with the AV-153 salts we have performed a fluorescent intercalator displacement assay. The method is based on the quenching of the enhanced fluorescence of the DNA-EtBr complex by a second ligand (Ghosh et al., 2010). As presented in Fig. 1A, all AV-153 salts quenched the EtBr fluorescence up to 70%, indicating that the compounds competed with EtBr for DNA intercalation sites. The Stern-Volmer quenching constant calculations indicated that Ca and Mg salts displaced the EtBr from intercalation sites more intensively than Na, Li and K salts, and Rb salt produced the weakest effect. (Fig. 1B, Table 2).

Circular dichroism

Circular dichroism spectra of ct-DNA in the presence of increasing concentrations of AV-153-Rb and AV-153-Ca are shown in Fig. 3. The DNA manifested a negative band at 245 nm due to helicity and a positive band at 270 nm because of base stacking which is characteristic of the B form of DNA. Adding AV-153 salts to DNA increased the negative band intensity and decreased the positive band intensity. A 2 nm red shift of crossover point was also observed. These data clearly indicate interactions of the compound with DNA, although changes in spectra are not typical for any binding mode.

Interaction with bases

The above results indicate that the binding between AV-153-Na and DNA partly occurs via intercalation. The influences of DNA bases, C, G, A and T, on the UV/VIS absorption spectra of AV-153 were used to evaluate the possible base-specificity of binding (Fig. 4, Table 3). The absorption intensity gradually increased with an increase in the concentration of all the four bases. Affinity to G, C and T is greater than that to A. The results indicate that AV-153-Na can interact with the four types of bases, with somewhat different affinities. Titration was performed in solutions of 1M NaCl and 8M urea to evaluate the role of ionic and hydrogen bonds in AV-153-Na interactions with bases. During this process, the media affinity of AV-153-Na to bases was weakened, especially for G, and the shape of the spectra was also changed. However, interactions were not abolished.

FTIR spectroscopy of AV-153 salts and their complexes with DNA

The FTIR spectra of AV-153 salts showed several strong absorption bands (Fig. 5). In the spectra of all AV-153 salts, the most intensive were broad absorption bands with maximums in the wavenumber 1,220–1,243 cm⁻¹ range, assigned to N-H bending and C-N stretching vibrations (Zhang et al., 2013), and 1,665–1,684 cm⁻¹, assigned to CO, C=O, C=N stretching vibrations (Varsanyi, 1974). The shape of the absorption bands in the wavenumber region of 1,570–1,670 cm⁻¹was quite different: (i) two sharp separate bands in spectra of AV-153 salt with K, Rb, Ca; (ii) two partly overlapping bands at 1,593 and 1,666 cm⁻¹ in AV-153- Li; (iii) a broad strongly overlapping band with maximum at 1,684 cm⁻¹. All the AV-153 salt spectra showed absorption bands at ∼1,025, ∼1,045, ∼1,101 and ∼1,128 cm⁻¹, with variable mutual intensities. The broad and strong absorption bands in the 900–1,200 cm⁻¹ region were assigned to the stretching vibrations of C-O, C-C bonds and C-O, C-C, C-O-O deformation vibrations, though many other functional groups have bands in this region as well (Naumann & Meyers, 2000). Also, a separate strong absorption band in the 1,497–1,505 cm⁻¹ region with slightly varying intensity (the lowest in AV-153- Rb) and the band at ∼1,498 cm⁻¹ in spectra of Li, CA and Mg AV-153 salts while at 1,500 and 1,505 cm⁻¹ in spectra of AV-153 salts with Rb and K correspondingly.

The evaluation of AV-153 salt spectral band intensities and shape showed major variations in the bands at ∼1,675 cm⁻¹, assigned to the C=O and/or C=C bonds and ∼1,234 cm⁻¹, assigned to the N-H and/or C-N bonds. These variations indicate the different metals binding with the AV-153.

The FTIR DNA spectrum was recorded to identify the specific absorption bands assigned to the DNA bases and PO₂⁻ groups. In the FTIR spectrum of sonicated ct-DNA (Fig. 6) we identified the specific absorption bands of DNA-bases: guanine (G) at 1,697 cm⁻¹, thymine (T) at 1,655 cm⁻¹, adenine (A) at 1,608 cm⁻¹, and cytosine (C) at 1,488 cm⁻¹ (Tajmir-Riahi, N’Soukpoé-Kossi & Joly, 2009). The specific absorption bands of PO₂⁻ groups were detected at 1,241 cm⁻¹ and assigned to the symmetric stretching vibrations, and at 1,093 cm⁻¹, which was assigned to the asymmetric stretching vibrations (Banyay, Sarkar & Gräslund, 2003; Cherng, 2009).

The comparison of DNA and DNA-AV-153 salt complexes spectra was used to identify the intensity and/or wavenumber shift changes that indicate the structural and conformational changes in DNA that is possibly related to interactions with AV-153 salts. The spectra of AV-153 salts mixed with DNA were vector normalized, and thus the intensity of the vibration bands is proportional to the concentration of particular bonds. The AV-153 salts spectra showed many intensive absorption bands (Fig. 5) but DNA-broad spectrum overlapped with absorption bands assigned to DNA-bases and two distinct PO₂⁻ group vibrations bands (Fig. 6). The profile of the AV-153 salt and DNA complex spectrum, being a superposition of both components, showed all specific DNA bands, but with variable intensities and band shapes compared to that of pure DNA (Fig. 7). Spectra cross-comparison was used to find the intensity and/or frequency changes assigned to structural changes in DNA caused by the binding with the AV-153 salt. In the DNA and salt complexes spectra, the position of a specific band of G at ∼1,697 cm⁻¹ was shifted to ∼1,699 cm⁻¹ by Li and K; ∼1,693cm⁻¹ by Mg; ∼1,692 cm⁻¹ by Rb and ∼1,687 cm⁻¹ by Ca. In the AV-153-Li and AV-153-K complex, the intensity of the G band decreased. The T band wavenumber (1,655 cm⁻¹) was shifted by Ca to 1,661 cm⁻¹, Rb to 1,653 cm⁻¹ and Mg to 1,652 cm⁻¹. The intensity of specific T band was affected by all AV-153 salts but most of all by Ca and Mg salts. The A band wavenumber (∼1,608 cm⁻¹) was slightly shifted to 1,606 cm⁻¹in the spectra of all the AV-153 salt complexes with DNA. The intensity was slightly increased by AV-153 salts of Ca, Mg and Rb. The C band intensity (1,488 cm⁻¹) was not affected by AV-153-Rb, but the other salts provoked an increase in intensity and no remarkable frequency shifts were detected. The intensity of the PO₂⁻ groups at 1,241 cm⁻¹ was significantly increased by AV-153-Ca salt only. AV-153-Li produced a frequency shift from 1,241 cm⁻¹ to 1,238 cm⁻¹ compared to that in the DNA spectrum. All salts modified the shaped and intensity of the bands assigned to PO₂⁻ groups at 1,093 and 1,064 cm⁻¹. These PO₂⁻ groups of DNA seem to be similarly affected by AV-153-K and -Li salts, showing two strongly overlapping bands and a decrease in intensities. Mg and Rb salts in complex with DNA significantly increased the band intensity at ∼1,065 cm⁻¹ compared to that in the DNA spectrum. AV-153-Ca caused significant changes in the DNA, as a broad strongly overlapping band with maximums at ∼1,097 and ∼1,093 cm⁻¹ in DNA spectra was split into two distinct bands in AV-153-Ca spectra and DNA complexes.

The second derivative spectra of AV-153 salts, DNA and their complexes were used to evaluate the binding of AV-153 salts with A, T, C or G. The second derivative spectra of DNA complexes with AV-153-K and AV-153-Li showed (i) significant increase of G and C absorption band intensities, (ii) a shift of T absorption band maximum from 1,649 cm⁻¹ to 1,654/55 cm⁻¹, (iii) a band was not significantly changed (Fig. 8A). The second derivative spectra of DNA complexes with AV-153-Ca, AV-153-Mg and AV-153-Rb showed (i) an increase in G and a decrease in C band intensities, (ii) the intensities of A and T bands were unaffected, (iii) a shift in the peak assigned to C with a maximum range of 1,489 to 1,494/95 cm⁻¹ (Fig. 8B).

Interaction with guanine quadruplexes

Incubation of the human telomere repeat oligonucleotide with 100 mM NaCl resulted in the formation of G4. The CD spectroscopy revealed a dominating band at 295 nm and a shallow negative band around 260 nm. Such CD spectra correspond to the two-tetrad, chair type G4. Salts from monovalent metals were tested for interaction with G4. The titration of the G4 with AV-153-Na made the peaks more pronounced (Fig. 9A), as did AV-153-K (Fig. 9B). AV-153-Li made the negative peak deeper, but the intensity of the positive peak decreased (Fig. 9C), while the AV-153-Rb did not induce any changes in the CD spectra (Fig. 9D).

Affinity to G4 and single-strand oligos was further tested using UV.VIS spectroscopy. Changes in the AV-153-Na spectra upon addition of G4 or single-strand oligonucleotides are presented in Fig. 10. Hypochromic and slight bathochromic effects are observed. Affinity to G4 is higher than to single-stranded DNA. Other salts produced similar changes in spectra (not shown), and the binding constants are given in Table 4. AV-153-Na and AV-153-K manifested higher affinity to G4, and AV-153-Li and AV-153-Rb to single-stranded DNA.