Fluorescent Small Molecules Are BIG Enough To Sense Biomacromolecule: Synthesis of Aromatic Thioesters and Understanding Their Interactions with ctDNA.

The visible fluorescent chromophoric moiety present in the water-soluble photoactive yellow protein (PYP) of Ectothiorhodospira halophila is p-hydroxycinnamic acid linked to the cysteine residue (Cys-69) by a thioester bond and it controls the key photoinduced biological processes of the host organism. In the present work, we have synthesized and characterized three structurally different thiophenyl esters [viz., p-hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), and S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3)] in addition to a novel (to the best of our knowledge) stilbene-type olefinic compound, N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4), under the same reaction condition. All of these four compounds showed characteristic and distinguishable chromophoric/fluorophoric behavior in ethanol and also at pH 7.4. However, we have observed that the intrinsic chromophoric/fluorophoric activities of (1) and (2) were greatly influenced during their interactions with calf-thymus DNA, studied by a range of spectroscopic and physicochemical measurements. We have also applied density functional theory [B3LYP, 6-311G+(d,p)]-based method to get optimized structures of (1) and (2), which were explored further for molecular docking studies to understand their mode of interaction with DNA. The present study opens up their possible applications as fluorescence probes for biomacromolecules like DNA in future.

Introduction

Small molecules have wide spectrum of biological activities. As a cell signaling chemical entity or as a drug, a small molecule can regulate a specific biological process by binding to a particular biotarget, like a specific protein or nucleic acid.[1] Small molecules, e.g., organic dyes or drugs having fluorophoric/chromophoric activity show intense change in their spectral behavior when they bind to a specific biomacromolecule, which can be a protein, enzyme, or DNA. This change of spectral patternpan> of the small molecules is often inpan>dicative of their mode of inpan>teractionpan> with that particular biomacromolecule.[2−10] Hence, this type of small molecules are beinpan>g extensively explored as important research tools to probe a biotarget (i.e., biomacromolecule regulatinpan>g a particular biological process) and also for the development of novel therapeutic drugs as well as effective diagnostic protocols.[1,11] It is a highly cherished research goal for the scientists to develop novel, nonpan>toxic, photostable, and easily synpan>thesizable small-molecule-based fluoroprobe for a specific biotarget. Plethora of organic compounpan>ds is put inpan>to trials for this,[1,11] and inpan> this conpan>text, fluorescent thioesters, especially α,β-unsaturated thioesters, are highly relevant.[12−17]

In livipan class="Chemical">ng cells, several enzymatic reactions are triggered by such thioester derivatives.[13,18] On the other hand, thioesters are often considered as better choices over oxoesters and used as vital synthetic intermediates in various important organic transformations, viz., biosynthesis of natural products.[13,19,20] Apart from this, α,β-unsaturated thioesters are found in the photoactive yellow protein (PYP) of Ectothiorhodospira halophila.[15,21] The visible fluorescent chromophoric moiety present in this water-soluble small chromoprotein is p-hydroxycinnamic acid, which is linked to the cysteine residue (Cys-69) by a thioester bond and controls the key photoinduced biological processes of the host organism.[15,22] Moreover, photochromic α,β-unsaturated thioesters (e.g., p-hydroxycinnamic-thiophenyl ester) are studied extensively as model compounds to understand the mechanism of various photochemical reactions, and such compounds are explored nowadays for the development of memory and optical devices.[16]

In the present work, we have synthesized three structurally different thioester derivatives, p-hydroxycinnamic-thiophenyl ester [or (E)-S-phenyl-3-(4-hydroxyphenyl)prop-2-enethioate] (1), p-N,N-dimethylaminocinnamic-thiophenyl ester [or (E)-S-phenyl-3-(4-(dimethylamino)phenyl)prop-2-enethioate] (2), and S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3) in addition to a novel (to the best of our knowledge) olefinic type of compound, substituted with amine and thiophenyl groups, i.e., N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4) under the same reaction condition (Figure and Scheme ). Each of these compounds showed characteristic chromophoric/fluorophoric behavior in ethanol and also at pH 7.4 (physiological pH) (Tris–HCl buffer medium) [results are shown in Figure S1 (Supporting Information, pp S3 and S4)].

Figure 1

Single-crystal X-ray diffraction (XRD) structures of the synthesized compounds, p-hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

Scheme 1

Synthesis of (1–4)

p-Hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3), and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

Single-crystal X-ray diffractiopan class="Chemical">n (XRD) structures of the synthesized compounds, p-hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

n class="Chemical">p-Hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3), n class="Chemical">and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

We have explored the characteristic chromophoric/fluorophoric behavior of compounds (1) apan class="Chemical">nd (2) for probing the biomacromolecule, DNA, the funpan>damental genclass="Chemical">pan>etic material regulating the life processes. The inpan>trinpan>sic fluorescence activities of p-hydroxycinnamic-thiophenyl ester (1) and p-N,N-dimethylaminocinnamic-thiophenyl ester (2) were found to be strongly dependent on their interactions with calf-thymus DNA (ctDn class="Chemical">NA). We have optimized the structures of (1) and (2) by applying a density functional theory (DFT) [B3LYP, 6-311G+(d,p)]-based method, and their mode of interaction was studied with the help of various spectroscopic and physicochemical techniques and molecular docking studies, using DFT-based optimized structures of (1) and (2).

General Experimental Section

Materials

All of the reagents used for the synthesis of compounds (1–4) were of AR grade (E Merck). Solvents were dried and distilled prior to their use. In all of the experiments, Milli-Q (Milli-Q academic with 0.22 mm Milli pack-40) water was used as per the requirement. Calf-thymus DNA (ctDNA) of molecular weight 8.4 MDa (SRL, India) was used as the model DNA for studying DNA-binding interactions of the synthesized compounds (1) and (2). Reagents used for the biocompatibility assay are of molecular biology (MB) grade, and details of these reagents are given in the Supporting Information (p S21).

For this study, a stock solution of ctDpapan class="Chemical">n class="Chemical">NA was prepared by dissolvinpan>g solid ctDn class="Chemical">NA (n class="Chemical">sodium salt) in Tris–HCl buffer (10 mM, pH 7.4) (SRL, India) containing 0.1 mol L–1 NaCl solution and stored at 4 °C for further use. The purity of ctDNA was checked spectrophotometrically by measuring the ratio of its maximum absorbance at λmax values 260 and 280 nm, i.e., A260/A280. A value of A260/A280 ≈ 1.87 was observed, which denoted that ctDNA used was sufficiently free from protein. The concentration of ctDNA stock solution calculated from its maximum absorbance at 260 nm (A260) and εctDNA (6600 L mol–1 cm–1) data was found to be 3.95 × 10–3 M.[7,8]

All other reagents for the Dpapan class="Chemical">n class="Chemical">NA-binpan>ding studies were of molecular biology (MB) grade (SRL, India), and these regents were used as they were purchased without any further purificationpan>. However, purity of all of the reagents used for the present study was routinpan>ely checked either spectrophotometrically or by thinpan>-layer chromatography (TLC) onpan> n class="Chemical">silica gel GF254 precoated plates using appropriate eluting solvent systems.

Method of Synthesis of Thioester Derivatives (1–4)

n class="Chemical">p-Hydroxycinnamicpan> acid (600 mg, 1 equiv, 3.65 mmol) was dissolved in 5 mL of dichloromethane (DCM). The temperature of this reaction mixture was maintained at 0 °C, and n class="Chemical">oxalyl chloride (0.93 mL, 3 equiv, 10.97 mmol) was added to it in a dropwise manner. To this reaction mixture, dimethylformamide (DMF) (0.28 mL, 1 equiv, 3.65 mmol) was added at 0 °C and mixed thoroughly. After that, the reaction mixture was stirred at room temperature (r.t.) for 2 h. The progress of the reaction was monitored by TLC of the reaction mixture at regular intervals of time. After the completion of the reaction, the reaction mixture was subjected to evaporation under reduced pressure to get a concentrated mass, which was diluted with a little amount (∼1.5 mL) of DCM, and triethylamine (Et3N) (1.53 mL, 3 equiv, 11.85 mmol) and thiophenol (PhSH) (0.807 mL, 2 equiv, 7.90 mmol) were added to it simultaneously and mixed properly by keeping the temperature of the reaction mixture at 0 °C. This reaction mixture was stirred for 12 h at r.t. After the completion of the reaction (as monitored by TLC), the reaction mixture was diluted with DCM and workup was done with aqueous NaHCO3 solution. The organic part was separated, and the solvent was evaporated under reduced pressure, whereby a gummy mass was obtained. This was subjected to column chromatography over silica gel (mesh 100–200), and the column was eluted with ethyl acetate (EA) and petroleum ether (PE, 60–80 °C) solvent mixture, whereby we have got four products: p-hydroxycinnamic-thiophenyl ester (1) (30–40% EA/PE), p-N,N-dimethylaminocinnamic-thiophenyl ester (2) (15–20% EA/PE), S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3) (75–90% EA/PE), and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4) (2–5% EA/PE).

Structures of (1)−(4) were confirmed opan class="Chemical">n the basis of physicochemical studies, spectroscopy (1H and npan> class="Chemical">13C NMR, IR, liquid chromatography–mass spectrometry (LC–MS)), and single-crystal X-ray diffraction (XRD) [for (1), (2), and (4) only] data. Single-crystal XRD patterns of (1), (2), and (4) are shown in Figure . NMR spectra were recorded on a 400 MHz Bruker AVANCE-400 NMR spectrometer with CDCl3 or dimethyl sulfoxide (DMSO)-d6 as solvents and tetramethylsilane as the internal standard. XRD experiments were done as per standard procedure, and relevant experimental details are given in Supporting Information (pp S5–S9).

Characterization Data for the Synthesized Products (1–4)

p-Hydroxycinnamic-thiophenyl Ester or (E)-S-Phenyl-3-(4-hydroxyphenyl)prop-2-enethioate (1)

Light yellow crystal, 264 mg, 28.16% yield, mp 134–136 °C, n class="Chemical">1H NMR (400 MHz, n class="Chemical">CDCl3): δ (ppm) 7.60 (d, 1H, J = 15.6 Hz), 7.48–7.51 (m, 2H), 7.42–7.44 (m, 5H), 6.80 (d, 2H, J = 8.8 Hz), 6.65 (d, 1H, J = 15.6 Hz); 13C NMR (100 MHz, CDCl3): δ (ppm) 189.30, 158.62, 141.93, 134.86, 130.70, 129.64, 129.38, 127.69, 126.56, 121.56, 116.21; IR (KBr, cm–1) 3371, 3058, 2936, 1649, 1579, 1511, 1435, 1277, 1040; LC–MS (M – H+) 255.[12,16]

Single-Crystal X-ray Crystallography for (1)

Empirical formula C15H12O2S, FW 256.31, n class="Mutation">T 296 Kpan>, monoclinpan>ic, space group = P21/n, a = 16.6461(6) Å, b = 7.6854(3) Å, c = 20.8254(8) Å, α = 90°, β = 98.583(2)°, γ = 90°, V = 2634.40(17) Å3, Z = 8, λ (Mo Kα) = 0.71073 Å, μ = 0.236 mm–1, θmin = 1.695°, θmax = 28.452°, R = 0.0663(5065), wR2 = 0.2219(6619). Crystallographic data for the structures of 1 reported in the present work have been deposited at the Cambridge Crystallographic Data Centre with CCDC No. 1523138.

p-N,N-Dimethylaminocinnamic-thiophenyl Ester or (E)-S-Phenyl-3-(4-(dimethylamino)phenyl)prop-2-enethioate (2)

Yellow crystal, 88 mg, 8.5% yield, mp 166–168 °C, n class="Chemical">1H NMR (400 MHz, n class="Chemical">CDCl3): δ (ppm) 7.61 (d, 1H, J = 15.6 Hz), 7.47–7.50 (m, 3H), 7.41–7.45 (m, 4H), 6.70 (d, 2H, J = 8.0 Hz), 6.57 (d, 1H, J = 15.6 Hz); 13C NMR (100 MHz, CDCl3): δ (ppm) 187.77, 152.15, 142.50, 134.85, 130.56, 129.23, 129.18, 128.52, 121.90, 118.91, 112.06, 40.30; IR (KBr, cm–1) 3059, 2914, 2806, 1654, 1590, 1529, 1433, 1373, 1022; LC–MS (M + H+) 284.

Single-Crystal X-ray Crystallography for (2)

Empirical formula C17H17NOS, FW 283.37, papan class="Chemical">n class="Mutation">T 296 K, monpan>oclinic, space group = n class="Gene">P21, a = 6.2317(5) Å, b = 7.5812(6) Å, c = 15.8318(13) Å, α = 90°, β = 97.343(3)°, γ = 90°, V = 741.82(10) Å3, Z = 2, λ (Mo Kα) = 0.71073 Å, μ = 0.213 (mm–1), θmin = 1.297°, θmax = 27.605°, R = 0.0672(2160), wR2 = 0.1995(3290). Crystallographic data for the structures of 2 reported in the present work have been deposited at the Cambridge Crystallographic Data Centre with CCDC No. 1523157.

S-Phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3)

Off-white crystal, 65 mg, 4.6% yield, mp 98–102 °C, n class="Chemical">1H NMR (400 MHz, n class="Chemical">CDCl3): δ (ppm) 7.33–7.36 (m, 5H), 7.27–7.31 (m, 5H), 7.15–7.17 (dd, 2H, J = 2.0 Hz, J = 6.8 Hz), 6.73–6.75 (dd, 2H, J = 2.0 Hz, J = 6.4 Hz), 4.69–4.73 (t, 1H, J = 7.6 Hz), 3.23–3.25 (d, 2H, J = 7.6 Hz); 13C NMR (100 MHz, DMSO-d6): δ (ppm) 194.46, 157.18, 134.67, 134.41, 132.09, 130.20, 130.05, 129.82, 129.55, 129.34, 127.83, 127.50, 115.56, 49.40, 48.13; IR (KBr, cm–1) 3055, 2935, 1693, 1591, 1514, 1439, 1231, 1043; LC–MS (M + H+) 385.

N1,N1,N2,N2-Tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4)

White crystal, 272 mg, 22.5% yield, mp 100–104 °C, n class="Chemical">1H NMR (400 MHz, n class="Chemical">CDCl3): δ (ppm) 7.24–7.26 (m, 4H), 7.13 (s, 1H), 2.66 (s, 3H), 2.53 (s, 3H); 13C NMR (100 MHz, DMSO-d6): δ (ppm) 138.00, 136.98, 125.62, 133.67, 129.35, 129.29, 129.19, 127.85, 126.42, 125.85, 42.63, 42.21; IR (KBr, cm–1) 3060, 2914, 2789, 1565, 1460, 1318, 1072; LC–MS (M + H+) 331; Anal. calcd for C18H22N2S2: C, 65.41; H, 6.71; N, 8.48. Found: C, 65.48; H, 6.67; N, 8.53%.

Single-Crystal X-ray Crystallography for (4)

Empirical formula n class="Chemical">C18H22N2S2, FW 330.47, n class="Mutation">T 296 K, monoclinic, space group = P2/c, a = 16.6147(13) Å, b = 11.1160(13) Å, c = 11.1363(11) Å, α = 90°, β = 121.806(7)°, γ = 90°, V = 1747.9(3) Å3, Z = 4, λ (Mo Kα) = 0.71073 Å, μ = 0.303 mm–1, θmin = 2.33°, θmax = 28.27°, R = 0.0454(1757), wR2 = 0.1452(2160). Crystallographic data for the structures of 4 reported in the present work have been deposited at the Cambridge Crystallographic Data Centre with CCDC n class="Chemical">No. 1004693.

Instrumentation and Methods

Molecular Docking Study

Molecular docking studies were dopan class="Chemical">ne to understand the nature of intermolecular interactions between the thioester derivatives p-hydroxycinnamic-thiophenyl ester (1) or p-N,N-dimethylaminocinnamic-thiophenyl ester (2) and ctDNA. We have used Molecular Graphics Laboratory (MGL) and AutoDock Vina tools to set up and perform blinded docking studies between (1) or (2) and ctDNA.[23] The crystal structure of B-DNA (protein data bank (PDB) ID: 1BNA) was used for the docking study. This was downloaded from protein data bank (PDB) prior to the docking of either (1) or (2) with ctDNA. Water molecules were subtracted from the DNA PDB file and polar hydrogen atoms were added to the DNA molecule. Receptor (ctDNA) and ligand files [for (1) and (2)] were prepared using AutoDock. Before that, we have done DFT-based optimization of the structures of the ligands [(1) and (2)] using Gaussian 09 package. B3LYP exchange correlation functional together with standard 6-311G+(d,p) basis sets[24,25] was used for DFT calculations (Supporting Information, pp S13–S16).

In each case, the Dpapan class="Chemical">n class="Chemical">NA was fixed inpan> a grid box of the dimensionpan> 74 × 52 × 92. The grid spacinpan>g was set at 0.375 Å. The center of the grid was fixed up at 15.63, 19.75, and 10.08 Å. Lamarckian genetic algorithm modules available with AutoDock were employed to perform the ligand dockinpan>g calculationpan>s. The default settinpan>gs available with AutoDock program were chosen to assign other parameters.[8,26] In each case of the dockinpan>g of (1) or (2) with ctDn class="Chemical">NA, the lowest energy docked conformation according to the AutoDock scoring function was selected as the binding mode.

Python Molecular Viewer software (Stefano Forli, Olson Molecular Graphics Laboratory, The Scripps Research Institute, CA) was used to observe the output of the AutoDock studies.

Steady-State UV–Vis Absorption Spectroscopic Studies

UV–vis absorption spectra of all of the sypan class="Chemical">nthesized compounds (1–4) in ethanol and Tris–HCl buffer (10 mM, pH 7.4) medium were recorded on a dual-beam UV–vis spectrophotometer (PerkinElmer, Lambda 35) in the wavelength range of 200–600 nm. The interactions of either (1) or (2) with ctDNA were also studied with the help of UV–vis spectroscopy. Tris–HCl buffer (10 mM, pH 7.4) was used as the reference solution in all of the studies of ctDNA–compound interaction. In each case of the test compounds [(1) or (2)], the stock solution (1 mM) was prepared in EtOH. In a typical interaction study, specific volume of this stock solution [0.04 and 0.16 mL for (1) and (2), respectively] was added to 2.5 mL of Tris–HCl buffer in a 1.0 cm quartz cuvette and mixed thoroughly. In each case of (1) and (2), this test solution was allowed to stand for 5 min. After this time interval, the respective test solution was titrated spectrophotometrically by the successive addition of 3.95 mM of ctDNA solution. In case of (1), the volume of ctDNA solution added each time was 2 μL, whereas 5 μL of ctDNA was added each time to (2). The final concentration of ctDNA in test solution varied from 0 to 0.07 mM for (1) and 0 to 0.08 mM for (2). The concentrations of (1) and (2) or ctDNA used for the present study were optimized after several trials. In all of the cases, the volume effect was considered to be negligible. All of the UV–vis absorption measurements were performed at 25 °C. The final concentration of EtOH in the test solution was negligible (<1%).

Steady-State Fluorescence Spectroscopic Measurements

Fluorescence spectra of compoupan class="Chemical">nds (1) and (2) in both EtOH and Tris–HCl buffer (10 mM, pH 7.4) media were recorded using PerkinElmer spectrophotometer of model no. LS55. Fluorescence emission spectra of all of the experimental solutions were recorded in the wavelength range of 300–540 nm with an excitation wavelength (λex) of 280 nm, and slit width was fixed at 10 nm for both excitation and emission beams.

We have also used steady-state fluorescence spectroscopic method to study the ipan class="Chemical">nteraction between ctDNA and the thioester derivative (1) or (2). For this study, we have used the same concentrations of ctDNA, (1) or (2) and Tris–HCl buffer, as discussed in the previous section. All of the fluorescence measurements were done at 25 °C, and we have observed that spectra of all of the test solutions remained unchanged for a long time during which the experiments were done. Hence, we can safely rule out the possibility of photodecomposition of the experimental samples, which may give errors in the results.

Viscometric Studies

Complexation/ipan class="Chemical">nteraction of ctDNA with a binpan>dinpan>g liganpan>d [like n class="Chemical">thioester derivatives (1) and (2)] changes its viscosity. Hence, we have done viscometric measurements of ctDn class="Chemical">NA solution in the presence of (1) or (2) to study the interaction of ctDNA with this type of ligands.

A Brookfield DV-II+ Pro viscometer, thermostated at 25 °C was used for the present study. papan class="Chemical">n class="Chemical">Tris–npan> class="Chemical">HCl buffer (2.5 mL, 10 mM, pH 7.4) was used as reference solution. Different volumes of (1) or (2) were added separately into the viscometer by keeping ctDNA concentration fixed at 0.08 mM, and for each addition, viscosity of the experimental solution was measured. The concentration of (1) was varied within the range of 0.01–0.08 mM in the experimental solution. The same procedure was followed for (2).

The results were expressed as relative viscosity (η/η0) versus the ratio of concepan class="Chemical">ntrations of ligand [(1) or (2)] to ctDNA (i.e., [(1)]/[ctDNA] or [(2)]/[ctDNA]). Here, η and η0 denote the viscosities of ctDNA solution in the presence and absence of ligand [either (1) or (2)], respectively.

Circular Dichroism (CD) Studies on the Binding Interaction of ctDNA with p-Hydroxycinnamic-thiophenyl Ester (1) or p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2)

A JASCO J-815 spectropolarimeter (Jasco Interpan class="Chemical">national Co., Ltd., Hachioji, Japan) was used to record the CD spectra of all of the sample solutions. A rectangular quartz cuvette of path length 1.0 cm was used in all of the measurements. All of the CD measurements were done under inert and dry atmosphere, and for this, the optical chamber of the CD spectropolarimeter was kept in the dry and nitrogen atmosphere. The conpan>centrationpan> of ctDNA was kept constant at 0.15 mM, and the concentration of test compound [either (1) or (2)] was varied. For both (1) and (2), the concentration range was 0–0.037 mM. CD spectra were recorded in the wavelength range of 200–320 nm at a scan speed of 100 nm min–1 and a band width of 1.0 nm. The scans were collected and automatically averaged by the instrument. All experiments were performed in Tris–HCl buffer (10 mM, pH 7.4) medium at 25 °C.

Cytotoxicity Assays of (1) and (2)

[3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (papan class="Chemical">n class="Chemical">MTT) assays of (1) anpan>d (2) were done to determinpan>e their cytotoxicity in human embryonic kidney cells (n class="CellLine">HEK293). Furthermore, we have also tested the in vivo cytotoxicity of these compounds using Saccharomyces cerevisiae 699 (MATa ade2-1 ura3-1 trp1-1 leu2-3, 112 his3-11, 15 ceul-100) cells. Detailed methods for these assays are discussed in Supporting Information (p S21).

Results and Discussion

Synthesis of p-Hydroxycinnamic-thiophenyl Ester (1), p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2), S-Phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3) and N1,N1,N2,N2-Tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4) and Studies on Their Spectral Behavior

Usefulness of papan class="Chemical">n class="Chemical">thioesters either as synpan>thetic intermediates or photoactive compounpan>ds is very much dependent onpan> their structures and therefore the synpan>thetic organic chemists are engaged inpan> the development of synpan>thetic protocols for the thioesters having task-specific structures. One of the most common methods for the synthesis of thioesters involves the reaction of carboxylic acid or acid chloride with thiol in the presence of a coupling agent, like dicyclohexyl carbodiimide (DCC)/N,N-dimethylaminopyridine.[27,28] Various Lewis acid-catalyzed acylations of thiols are also common methods for the synthesis of thioesters.[29−31] In this context, solid-supported of S-acylation reactions are noteworthy.[32,33] Bandgar et al. reported Dess–Martin periodinane (n class="Chemical">DMP)-mediated reaction of aldehyde and thiol to produce corresponding thioester derivatives.[34] TfOH-catalyzed direct thioesterification of carboxylic acids using thiol was done by Iimura et al.[35] Synthesis of thioesters via ruthenium-catalyzed olefin cross-metathesis with thioacrylate was reported by van Zijl et al.[13] These two methods were also effective in the synthesis of α,β-unsaturated thioesters.[13,35] However, the synthesis of α,β-unsaturated thioesters by the direct thioesterification of the α,β-unsaturated acid, e.g., p-hydroxycinnamic acid, which itself has functional groups susceptible toward the self-esterification, is troublesome and protection of such functional group(s) (e.g., −OH group) is required to get better yield.[17]

PYP chromophore analogue havipan class="Chemical">ng an α,β-unsaturated thioester moiety, i.e., p-hydroxycinnamic-thiophenyl ester (1), was prepared by the coupling of p-hydroxycinnamic acid with thiophenol in the presence of DCC as the coupling agent in DMF medium.[15] Yoya et al. synthesized p-hydroxycinnamic-thiophenyl ester (1) by reacting (COCl)2/DMF-treated p-hydroxycinnamic ester with thiophenate lithium salt at −30 °C in tetrahydrofuran medium.[12] Duran et al. have also reported similar synthetic protocol for thioester derivatives.[36] Unlike these methods, the present method is very simple and cost-effective, and most importantly, we have obtained four structurally and photochromically different S compounds under the same reaction condition (Scheme ). Among these, three are thioester analogues [p-hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), and S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3)] and the fourth one, N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4), is not a thioester compound but dimethylamine and thiophenyl group-substituted olefinic (stilbene-type) compound (Scheme and Figure ).

Single-crystal XRD patterpan class="Chemical">ns of (1), (2), and (4) are shown in Figure , and Tables S1–S3 (Supporting Information, pp S5–S9) represent the corresponding single-crystal XRD data and related experimental dn class="Gene">etails for (1), (2), and (4). Selected bonpan>d lengths and bonpan>d angles of the sinpan>gle-crystal XRD structures of (1), (2), and (4) are shownpan> inpan> Tables S4–S6, respectively (Supportinpan>g Informationpan>, pp S10–S12).

The plausible mechanism of formatiopan class="Chemical">n of these four products (1–4) is shown in Scheme . Reaction of oxalyl chloride with DMF generates N,N-dimethylchloromethylene iminium chloride (b), which on reaction with p-hydroxycinnamic acid produces carboxymethylene iminium chloride intermediate (c). The thiophenate anion (d) (may be generated by the reaction of thiophenol and triethylamine) reacts with (c) to give (1) (Scheme ).[12] However, (1) may also react with (b) at its phenolic-OH nucleophilic center to give intermediate (e). This intermediate (e) may be decomposed by path (A) (shown by red line) or by path (B) (shown by blue line) to give (f) or (2), respectively. There is another possibility that, (f) on reaction with another thiophenate moiety (d) gives (3). During these conversions, DMF may be regenerated, which on reaction with excess oxalyl chloride present in the reaction medium produces carboxymethylene iminium chloride intermediate (b) again. This iminium chloride (b) besides reacting with p-hydroxycinnamic acid can also react with unreacted (d) ions present in the reaction medium to generate (g), which on self-coupling may produce (4).

Scheme 2

Plausible Mechanism for the Formation of (1–4)

p-Hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3), and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

n class="Chemical">p-Hydroxycinnamic-thiophenyl ester (1), p-N,N-dimethylaminocinnamic-thiophenyl ester (2), S-phenyl-3-(4-chlorophenyl)-3-(phenylthio)propanethioate (3), n class="Chemical">and N1,N1,N2,N2-tetramethyl-1,2-bis(phenylthio)ethene-1,2-diamine (4).

Thus, this present simple apan class="Chemical">nd low-cost method was found to be useful in the synthesis and isolation of three different thioester analogues with an S- and N-substituted olefinic compound simultaneously under the same reaction condition. In future, this method may be suitability scaled up to get better yield of each individual compound (1–4).

Among compoupan class="Chemical">nds (1–3), (1) and (2) were α,β-unsaturated thiophenyl esters having electron-releasing groups (−OH and −NMe2, respectively) at p-positionpan> of the cinpan>namic–aromatic rinpan>g, which can conpan>trol the chromophoric/fluorophoric behavior of these compounpan>ds. On the other hanpan>d, (3) has a n class="Chemical">chlorine substituent at p-position of cinnamic–aromatic ring, and it has no olifinic unsaturation, rather the n class="Chemical">olefinic double bond of cinnamic moiety is substituted here by the thiophenyl group. Thus, the conjugation structure was lost in case of compound (3), which imposes a different chromophoric character to this compound. On the other hand, the S-analogue of stilbene type of compound (4) may have future application as model compound for studying photochemical (cis–trans) isomerization processes.

So, all of these four compounds have distipan class="Chemical">nguishable structural features, for which each of these compounds may show characteristic UV–vis and fluorescence spectroscopic behavior. To further confirm this assumption, we have done UV–vis and fluorescence spectroscopic studies of these compounds (1–4) in EtOH. We have also studied their spectral behavior at pH 7.4 (which is conpan>sidered to be physiological pH) usinpan>g Tris–HCl buffer. This was helpful to ascertain their usefulness as probe for biomacromolecules in the in vivo system. The results are shown in Figure S1a(i–iv),b(i–iv) (Supporting Information, pp S3 and S4).

It is noteworthy that each of these four compoupan class="Chemical">nds showed distinct and strong UV–vis and fluorescence behavior in EtOH as well as inpan> Tris–HCl buffer medium. This made us interested to explore their chromophoric/fluorophoric responses in Tris–HCl buffer medium of pH 7.4 in the present study of their interaction with biomacromolecule, like DNA.

Before going through the dpapan class="Chemical">n class="Gene">etails of the experimenclass="Chemical">pan>tal study on their inpan>teractionpan> with ctDn class="Chemical">NA, we have performed a preliminary screening based on theoretical study, i.e., molecular docking.

Molecular Docking Studies To Investigate the Interaction of Thiophenyl Esters [(1) or (2)] with ctDNA

Among the four sypan class="Chemical">nthesized thio compounds, the core structure of (1) and (2) resembles that of PYP chromophore anpan>d shows characteristic UV–vis and fluorescence behavior at pH 7.4. So, these two compounpan>ds were subjected to molecular dockinpan>g studies to inpan>vestigate their inpan>teractionpan> with ctDNA. However, we did not get any significant results with compounds (3) and (4). So, these two compounds were not further selected for other experimental studies on the ctDNA binding interaction (discussed in a later section).

Results of the docking study betweepan class="Chemical">n (1) and ctDNA or (2) and ctDNA are shown in Figure a,b respectively. The present molecular docking study gives an idea about the two important parameters involved in this binding interaction of either (1) and ctDNA or (2) and ctDNA: (i) structure of the binding ligand [i.e., (1) or (2)] and the binding site(s) of the ligand and (ii) binding sites of the biomacromolecule (i.e., ctDNA). These parameters not only show strong influence on the binding profile and activity of (1) and (2), but also control the binding nature of the biomacromolecule, DNA. These are actually the factors determining the efficacy of (1) or (2) as fluorescent probe for biomacromolecules like DNA or as bioactive compounds (e.g., drugs).

Figure 2

(a) Docking models of p-hydroxycinnamic-thiophenyl ester (1) with ctDNA showing its binding at the minor groove of the DNA: (i) atomic sphere model, (ii) ribbon model, and (iii) close-up view (shown in line model); (iv) DFT [B3LYP 6-311G+(d,p)]-optimized structure of (1). (b) Docking models of p-N,N-dimethylaminocinnamic-thiophenyl ester (2) with ctDNA showing its binding at the minor groove of the DNA: (i) atomic sphere model, (ii) ribbon model, and (iii) close-up view (shown in line model); (iv) DFT [B3LYP 6-311G+(d,p)]-optimized structure of (2).

(a) Docking models of papan class="Chemical">n class="Chemical">p-hydroxycinnamic-thiophenclass="Chemical">pan>yl ester (1) with ctDn class="Chemical">NA showing its binding at the minor groove of the Dn class="Chemical">NA: (i) atomic sphere model, (ii) ribbon model, and (iii) close-up view (shown in line model); (iv) DFT [B3LYP 6-311G+(d,p)]-optimized structure of (1). (b) Docking models of p-N,N-dimethylaminocinnamic-thiophenyl ester (2) with ctDNA showing its binding at the minor groove of the DNA: (i) atomic sphere model, (ii) ribbon model, and (iii) close-up view (shown in line model); (iv) DFT [B3LYP 6-311G+(d,p)]-optimized structure of (2).

Results of docking studies ipan class="Chemical">ndicate huge deviation of structural geometry of both thiophenyl esters (1) and (2) onpan> inpan>teraction with DNA from their respective crystal structures and also from their DFT-optimized structures, and the corresponding molecular docking profiles are shown in Figure .

Docking studies revealed that both papan class="Chemical">n class="Chemical">thiophenyl esters (1) anpan>d (2) adopted fully nonpan>planar geometry to accommodate itself inpan>to the G–C-rich regionpan> of ctDNA and thus covalent interaction, e.g., hydrogen-bonding interaction, played a vital role in the binding of ctDNA and the thioester analogue (1) or (2), as shown in Figure .

We have also observed drastic changes in some of the dihedral angles and slight changes in bond lengths and bond angles in the ctDn class="Chemical">NA-docked structures of (1) and (2) compared to their free/unbound structures (Scheme and Figure ). These data are summarized in Tables S7a,b [for (1)] and S8a,b [for (2)] (Supporting Information, pp S17–S20).

The bindipan class="Chemical">ng free energies (−ΔG) of the ctDNA-docked structures of both (1) and (2) were calculated as 6.4 kcal mol–1. These results support the existence of stronpan>g bindinpan>g (possibly via nonpan>covalent inpan>teractionpan>s) between (1) and ctDNA and also between (2) and ctDNA, as shown in Figure and Tables S7a,b, and S8a,b, respectively (Supporting Information, pp S17–S20). Although, it is apparent from Figure a,b that these thiophenyl esters, either (1) or (2), bind to the minor groove of the DNA, but there is an inherent drawback associated with AutoDock software, which identifies most of the small molecules as minor groove binders for DNA.[23] Hence, to further understand the binding interaction between the synthesized thiophenyl esters [(1) or (2)] and ctDNA, we have done a series of experiments and the results are discussed below.

UV–Vis Spectral Responses of p-Hydroxycinnamic-thiophenyl Ester (1) and p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2) in the Presence of ctDNA

Chromophoric/fluorophoric small molecules show intepan class="Chemical">nse change in the spectral behavior when their microenvironment is changed, and this can happen when biomacromolecules, like proteins and DNA, are present inpan> their vicinpan>ity. These spectral changes of small molecules can be monpan>itored to explore their usefulness as molecular probe for that biomacromolecule (inpan> the present case, it is ctDNA). Hence, to explore the possibility of the use of thiophenyl esters (1) and (2) as probe for DNA, we have studied their UV–vis spectral responses individually in the presence of different concentrations of ctDNA.

The absorption spectra of (1) and (2) recorded in the absence and presence of different concentrations of ctDn class="Chemical">NA are shown in Figure a,b respectively.

Figure 3

UV–vis spectra of (a) p-hydroxycinnamic-thiophenyl ester (1) (0.015 mM) in the absence (dashed line) and presence of different concentrations (0.003–0.076 mM) of ctDNA and (b) p-N,N-dimethylaminocinnamic-thiophenyl ester (2) (0.06 mM) in the absence (dashed line) and presence of different concentrations (0.014–0.087 mM) of ctDNA in Tris–HCl buffer (10 mM) at pH 7.4. Directions of arrows indicate the increase of Absmax at the corresponding λmax value.

UV–vis spectra of (a) n class="Chemical">p-hydroxycinnamicpan>-thiophenyl ester (1) (0.015 mM) inpan> the absence (dashed linpan>e) and presence of different conpan>centrationpan>s (0.003–0.076 mM) of ctDNA and (b) p-N,N-dimethylaminocinnamic-thiophenyl ester (2) (0.06 mM) in the absence (dashed line) and presence of different concentrations (0.014–0.087 mM) of ctDNA in Tris–HCl buffer (10 mM) at pH 7.4. Directions of arrows indicate the increase of Absmax at the corresponding λmax value.

Absorption spectrum of (1) (ipan class="Chemical">n Tris–HCl buffer, pH 7.4) is characterized by the appearance of a prominent band at 343 nm and two overlapping small humps at 223 and 236 nm (denoted by the dotted curve in Figure a). In case of (2), a prominent band at 354 nm was observed in addition to another small peak at 221 nm and a broad hump at 280 nm (denoted by the dotted curve in Figure b). On gradual addition of ctDNA solution to (1), humps at 223 and 236 nm became more sharp and shifted to λmax values 212 and 258 nm, respectively, at saturation.

In case of (2), gradual addition of ctDn class="Chemical">NA solution chanclass="Chemical">pan>ged its spectral pattern. The band at 354 nm became more broad, whereas the two overlapping bands at 221 and 280 nm became distinctly separated. At the same time, these bands became more sharp and shifted to 215 and 258 nm, respectively. The increase of absorbance value at 223 nm was associated with blue shift or hypsochromic shift for (1). A similar phenomenon was observed in case of (2). But in case of the band at λmax 236 nm, the bathochromic or red shift was observed for (1). Hypsochromic or blue shift of the λmax values (from 221 to 215 nm and from 280 to 258 nm) were observed for (2). Intensities of these bands were also gradually increased with the increasing concentration of ctDn class="Chemical">NA solution.

Small molecules can ipan class="Chemical">nteract with DNA through two types of binpan>dinpan>g modes: intercalationpan> and groove binpan>dinpan>g. Electrostatic mode of binpan>dinpan>g is also observed dependinpan>g uponpan> the availability of the charged structure of the small molecule, but this does not resemble the present case. In case of inpan>tercalationpan> mode of binpan>dinpan>g, pronpan>ounpan>ced hypochromic shift associated with broadening of the band envelope and bathochromic or red shift of the λmax value are observed.

A large shifting of λmax value is gepan class="Chemical">nerally observed in the case of groove binding small molecules.[2,3,8,9,37−39,40a] However, no such prominent characteristic features of either intercalation or groove binding were observed in the present case (Figure ). But prominent changes in the UV–vis spectral characteristics of (1) and (2) in the presence of ctDNA surely inpan>dicated the stronpan>g nonpan>covalent inpan>termolecular inpan>teractionpan>s between (1) [or (2)] and ctDNA, leading to the formation of ground-state (1)–ctDNA or (2)–ctDNA complexes, respectively.

The formation copan class="Chemical">nstants KB of the (1)–ctDNA or (2)–ctDNA complex were evaluated spectrophotometrically by applying the Benesi–Hildebrand equation[40b] as shown belowwhere A0 and A are the absorbance values of free and bound ligands [(1) or (2)], respectively, and A∞ is the absorbance of the final complex. The Benesi–Hildebrand plots of (1) and (2) are shown in the inset of Figure a,b, respectively. The straight-line trend in the Benesi–Hildebrand plot suggests the possible formation of 1:1 complex between (1) [or (2)] and ctDNA.[6] From the plot of 1/(A – A0) to 1/[ctDNA], the ratio of the intercept to the slope gives the binding constant KB. The binding constants (KB) were found to be 1.63 × 104 and 0.10 × 103 M–1 for (1) and (2), respectively.

Steady-State Fluorescence Studies on the Interaction of Thiophenyl Esters [(1) or (2)] with ctDNA

The fluorescence emissiopan class="Chemical">n spectra of (1) and (2) in the absence and presence of different concentrations of ctDn class="Chemical">NA were recorded inpan>dividually and the results are shownpan> inpan> Figures a and 5a, respectively.

Figure 4

(a) Fluorescence emission spectra of p-hydroxycinnamic-thiophenyl ester (1) (0.037 mM) in the absence (dashed line) and presence of different concentrations (0.030–0.527 mM) of ctDNA solution in Tris–HCl buffer (10 mM, pH = 7.4) at 25 °C. The excitation wavelength (λex) was 280 nm. The direction of arrow indicates the fluorescence quenching on gradual increase of concentration of ctDNA. (b) Stern–Volmer plot for the quenching of the fluorescence emission intensity of (1) in the presence of ctDNA as fluorescence quencher. (c) Plot of log[(F0 – F)/F] vs log[ctDNA] to calculate binding constant (KB) and the number of binding sites (n) of (1) at 25 °C in Tris–HCl buffer medium.

Figure 5

(a) Fluorescence emission spectra of p-N,N-dimethylamino cinnamic-thiophenyl ester (2) (0.074 mM) in the absence (dashed line) and presence of different concentrations (0.029–1.067 mM) of ctDNA solution in Tris–HCl buffer (10 mM, pH = 7.4) at 25 °C. The excitation wavelength (λex) was 280 nm. The direction of arrow indicates the fluorescence quenching on gradual increase of concentration of ctDNA. (b) Stern–Volmer plot for the quenching of fluorescence emission intensity of (2) in the presence of ctDNA as fluorescence quencher. (c) Plot of log[(F0– F)/F] vs log[ctDNA] to calculate binding constant (KB) and the number of binding sites (n) of (2) at 25 °C in Tris–HCl buffer medium.

(a) Fluorescence emissiopan class="Chemical">n spectra of p-hydroxycinnamic-thiophenyl ester (1) (0.037 mM) inpan> the absence (dashed linpan>e) and presence of different conpan>centrationpan>s (0.030–0.527 mM) of ctDNA solution in Tris–HCl buffer (10 mM, pH = 7.4) at 25 °C. The excitation wavelength (λex) was 280 nm. The direction of arrow indicates the fluorescence quenching on gradual increase of concentration of ctDNA. (b) Stern–Volmer plot for the quenching of the fluorescence emission intensity of (1) in the presence of ctDNA as fluorescence quencher. (c) Plot of log[(F0 – F)/F] vs log[ctDNA] to calculate binding constant (KB) and the number of binding sites (n) of (1) at 25 °C in Tris–HCl buffer medium.

(a) Fluorescence emissiopan class="Chemical">n spectra of p-N,N-dimethylamino cinnamic-thiophenyl ester (2) (0.074 mM) in the absence (dashed line) and presence of different concentrations (0.029–1.067 mM) of ctDNA solution in Tris–HCl buffer (10 mM, pH = 7.4) at 25 °C. The excitation wavelength (λex) was 280 nm. The direction of arrow indicates the fluorescence quenching on gradual increase of concentration of ctDNA. (b) Stern–Volmer plot for the quenching of fluorescence emission intensity of (2) in the presence of ctDNA as fluorescence quencher. (c) Plot of log[(F0– F)/F] vs log[ctDNA] to calculate binding constant (KB) and the number of binding sites (n) of (2) at 25 °C in Tris–HCl buffer medium.

The distinct chapan class="Chemical">nge of fluorophoric behavior of (1) [and also that of (2)] was observed in the presence of ctDNA. The dashed curves inpan> Figures a and 5a denote the fluorescence spectra of (1) anpan>d (2) in the absence of ctDNA, respectively. Upon photoexcitation at 280 nm (λex), (1) showed three emission band maxima (λem) at 312, 368, and 494 nm, whereas (2) showed two emission band maxima at λem 494 and 516 nm.

On gradual addition of ctDn class="Chemical">NA solution to (1), a significant decrease in the fluorescence intensity was observed. A similar phenomenon was also observed for (2). A prominent quenching of fluorescence intensity at λem 494 nm of (1) was observed, whereas the fluorescence intensity of other two bands at λem 312 and 368 nm remained almost unchanged. A gradual blue shift of the band at λem 494 nm occurred to 487 nm (Figure a). On the other hand, for (2), a prominent quenching of the bands at λem 494 and 516 nm was observed (Figure a), but interestingly, the λem at 516 nm of (2) began to disappear at saturation, accompanying with a gradual blue shift of the bands at λem 494–486 nm (Figure a).

Mode of Interaction between Thiophenyl Ester [p-Hydroxycinnamic-thiophenyl Ester (1) or p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2)] and ctDNA

Quenching (as in the present case) or enhancement of fluorescence of a fluorophore during its interaction with either a fluorescence quencher or an enhancer may be the outcome of several processes like excited-state energy-transfer reactions or ground-state complex formation[8,41] which are mechanistically classified into two categories: dynamic and static fluorescence quenching or enhancement.[37,41,42] To understand the mechanistic pathway involved in the case of fluorescence quenching of either (1) [or (2)] during its interaction with ctDNA, we have applied the classical Sternpan>–Volmer equationpan> for quenchinpan>g[7,42] as shownpan> belowwhere F0 and F represent the fluorescence inpan>tensities of (1) [or (2)] inpan> the absence and presence of different conpan>centrationpan>s of the quencher, i.e., ctDNA, respectively. The concentration of ctDNA is denoted by [E], and KSV represents the Stern–Volmer constant.

Stern–Volmer plots (F0/F vs [E]) for (1) and (2) are shown in Figures b and 5b, respectively, and we have calculated the Stern–Volmer fluorescence quenching constant n class="Chemical">KSV as 8.75 × 103 M–1 for (1) and 8.49 × 103 M–1 for (2).

We have further exploited the fluorescence ipan class="Chemical">ntensity quenching data of (1) [and also that of (2)] in the presence of ctDNA to measure their binpan>dinpan>g conpan>stants (KB) and number of binpan>dinpan>g sites (n). The change of fluorescence inpan>tensity at λem 494 nm for (1) and also for (2) was used to calculate the corresponpan>dinpan>g KB and n values by applyinpan>g the followinpan>g equationpan>[8,43,44]where F0 and F are the fluorescence inpan>tensities of (1) and (2) at λem 494 nm inpan> the absence and presence of different conpan>centrationpan>s of ctDNA, respectively. In each case of (1) or (2), log KB was calculated from the linear plot of log[(F0 – F)/F] vs log[ctDNA] [r2 = 0.99 for both (1) and (2)], as shown in Figures c and 5c, respectively. The values of KB and n were obtained as 1.38 × 103 M–1 and 1.07 ≈ 1.0 for (1) and 2.40 × 10 M–1 and 1.12 ≈ 1.0 for (2), indicating the weak binding between (1) and (2) with ctDNA.

These n class="Chemical">KSVpan> values are comparable to those of the small-molecule fluorophores, such as n class="Chemical">3-hydroxyflavones, n class="Chemical">isoxazolcurcumin, and curcumin derivatives[7,8] which showed static fluorescence enhancement/quenching phenomena associated with ground-state complex formation. These small molecules are also known to be DNA groove binders. So, in the present case of (1) and (2), we can assume a similar type of process of fluorescence quenching of these compounds in the presence of a quencher, like DNA. This may be originated due to the ground-state complex formation of (1) or (2) with DNA possibly via groove binding mode of interaction. However, to further verify these facts, we have done viscometric measurements of ctDNA in the presence of (1) or (2).

Viscosity Measurements To Elucidate the Mode of Interaction between Thiophenyl Esters [p-Hydroxycinnamic-thiophenyl Ester (1) or p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2)] and ctDNA

Change of viscosity of Dpapan class="Chemical">n class="Chemical">NA solutionpan> in the presence of various conpan>centrationpan>s of small-molecule-based binpan>dinpan>g ligands is an inpan>dicationpan> of their inpan>termolecular inpan>teractionpan>/complexationpan> with ctDNA. In the presence of a perfect intercalator ligand, the DNA base pairs became separated and thus the chain length of DNA increases, which ultimately increases the viscosity of the DNA solution.[8,45] However, decrease of viscosity of DNA solution is observed when the ligand is a partial or nonclassical intercalator.[8,37,45] In case of ligands binding at DNA grooves or ligands binding DNA through electrostatic interactions, no change of viscosity of DNA solution can be observed.[8,37,45]

The change of relative viscosity (η/η0) of ctDpapan class="Chemical">n class="Chemical">NA solutionpan> (0.08 mM) on gradual additionpan> of different conpan>centrationpan> of (1) [or (2)] inpan> Tris–HCl buffer (10 mM, pH = 7.4) is shown in Figure .

Figure 6

Effects of p-hydroxycinnamic-thiophenyl ester (1) (black diamond line) and p-N,N-dimethylaminocinnamic-thiophenyl ester (2) (red triangle line) (0.01–0.08 mM) on the relative viscosity of ctDNA (0.08 mM) at 25 °C.

Effects of n class="Chemical">p-hydroxycinnamic-thiophenyl esterpan> (1) (black diamond linpan>e) and p-N,N-dimethylaminocinnamic-thiophenyl ester (2) (red triangle line) (0.01–0.08 mM) on the relative viscosity of ctDNA (0.08 mM) at 25 °C.

As shown ipan class="Chemical">n Figure , a gradual increase in viscosity of ctDNA solutionpan> onpan> inpan>teractionpan> with (1) was noticed. A similar result was also observed with (2). So, these viscosity measurement data obviously conpan>firmed the binpan>dinpan>g activities of (1) or (2) with ctDNA through intercalation mode of intermolecular interaction. This was further established with the help of circular dichroism (CD) studies of ctDNA in the presence of (1) [or (2)].

Circular Dichroism (CD) Studies To Confirm the Mode of Interaction between Thiophenyl Esters [p-Hydroxycinnamic-thiophenyl Ester (1) or p-N,N-Dimethylaminocinnamic-thiophenyl Ester (2)] and ctDNA

DNA upan class="Chemical">ndergoes conformational changes on binding with small-molecule ligands. This can be monitored by measuring its CD spectra in the absence and presence of the respective binding ligand. These changes in the CD pattern of DNA often indicate the mode of binding of that particular ligand to DNA. For a small-molecule ligand-based classical intercalator, a significant decrease of intensities of both the positive and negative bands of CD spectrum of the bound DNA is observed, but the intensities of these two CD bands of DNA remain unchanged on its binding with a small molecule, which prefers groove binding, more precisely minor groove binding mode of interaction.[8,46]

CD spectra of ctDNA ipan class="Chemical">n the absence and presence of different concentrations of (1) [or (2)] with ctDNA are shown in Figure .

Figure 7

Circular dichroism (CD) spectra of ctDNA (0.15 mM) in the absence (red line) and presence of different concentrations (0.01–0.037 mM) of p-hydroxycinnamic-thiophenyl ester (1) and p-N,N-dimethylamino cinnamic-thiophenyl ester (2) in Tris–HCl buffer (10 mM, pH = 7.4) medium at 25 °C. The dashed line shows the CD spectrum of Tris–HCl buffer as the reference. Benesi–Hildebrand plots of 1/Δθ vs 1/[(1)] or 1/Δθ vs 1/[(2)] are shown in a(i) and b(i), respectively.

Circular dichroism (CD) spectra of ctDNA (0.15 mM) ipan class="Chemical">n the absence (red line) and presence of different concentrations (0.01–0.037 mM) of p-hydroxycinnamic-thiophenyl ester (1) and p-N,N-dimethylamino cinnamic-thiophenyl ester (2) in Tris–HCl buffer (10 mM, pH = 7.4) medium at 25 °C. The dashed line shows the CD spectrum of Tris–HCl buffer as the reference. Benesi–Hildebrand plots of 1/Δθ vs 1/[(1)] or 1/Δθ vs 1/[(2)] are shown in a(i) and b(i), respectively.

We have used these CD spectrum data of ctDNA whepan class="Chemical">n it binds to (1) [or (2)] to confirm the mode of binding of (1) [or (2)] with ctDNA. Being achiral, (1) and (2) are not optically active and no CD spectra were observed for (1) and also for (2) in the wavelength region (200–350 nm) selected for the present experiment. CD spectrum of the free ctDNA, i.e., in the absence of (1) [or (2)] in Tris–HCl buffer medium (10 mM, pH = 7.4), showed two bands: a positive band 274 nm and a negative band 245 nm (Figure a,b) which originates due to π–π stacking of DNA base pairs and right-handed helicity of DNA, respectively. This type of characteristic nature of CD spectrum is often associated with the B-form of DNA.[47] However, on addition of (1) [or (2)] to ctDNA, significant changes in the intensities of negative bands as well as positive bands of CD spectra of DNA were observed (Figure a,b). Similar conclusions can be drawn regarding the mode of binding of (2) with ctDNA (Figure b). This indicates the intercalative binding mode of interaction of (1) [or (2)] with ctDNA. We have further explored the CD responses of DNA at 245 nm to calculate equilibrium constant (KB) by applying the Benesi–Hildebrand equation[8,48]where θ – θ0 = change of CD responses of ctDNA at 245 nm, θ = CD response of ctDNA on gradual addition of ligand, thiophenyl ester [(1) or (2)], θ0 = CD response of ctDNA in the absence of either (1) or (2), and θ1 = final CD response of thiophenyl ester–ctDNA. From the linear plot of 1/(θ – θ0) versus 1/[ligand] (Figure a(i),b(i)), we have calculated the binding constant KB for (1) and (2) as 3.43 × 104 and 1.93 × 104 M–1, respectively, which are in accordance with the results obtained from fluorescence spectroscopic studies. These values suggest the possibility of intercalative binding mode of interaction between (1) and ctDNA and also between (2) and ctDNA.[48] However, the changes in the CD bands in case of (1) [and also for (2)] are not so drastic as it is observed for a perfect intercalator.

Biocompatibilities of (1) and (2)

Previous report on (1) showed its weak activity toward papan class="Chemical">n class="Species">M. tuberculosis.[12] However, npan>o report on bioactivity was founpan>d for (2). In the present case, we have donpan>e an elaborate study onpan> the biocompatibilities of (1) and (2) (detailed experimental methods are discussed in Supporting Information, pp S21 and S22).

The cell n class="Disease">cytotoxicitypan> assay of compounds (1) and (2) performed onpan> the normal human cell line, HEK293, using the MTT assay method showed a concentration- and time-dependent effect. More than 82% of the cells survived after 24 h of exposure to compound (1) at its 30 μM concentration (Figure S2, Supporting Information, p S22). This value was found to be approximately 70 and 45% at 48 and 72 h, respectively. A proportion of 54% of the cells survived in case of compound (2) after 24 h of its exposure at 15 μM concentration (Figure S2, Supporting Information, p S22). The effect of these compounds (at their 40 μM concentration) on S. cerevisiae cells were also evaluated for a period of up to 10 days. No n class="Disease">toxicity of the compounds was observed after checking for viability of the cells at 0, 2, and 10 days.

In case of compoupan class="Chemical">nd (1), cytotoxicity toward normal npan> class="Species">human cell lines (HEK293) was observed after a concentration of 30 μM, but in case of compound (2), this optimal cytotoxic concentration was found to be 15 μM. Lower concentrations of these compounds did not significantly affect the animal cell lines. More than 70% growth of cells was observed for compound (1) up to its 30 μM concentration. However, the same growth rate of cells was noted with compound (2) when its concentration was 12 μM. Furthermore, no such toxicity was observed in case of in vivo cytotoxicity assay using S. cerevisiae cells. Even at 40 μM concentrations of compounds (1) and (2), no loss in viability was observed in cells. This difference in results between the animal cell lines and yeast cells may be due to the difference in the composition of the cell walls of these two different kinds of cells. Microbial cells may be inherently hard as they are more exposed to inclement conditions on their growth. The difference in their bioactivity may be originated from the difference in their substituent pattern (Figure ).

Conclusions

Development of easy apan class="Chemical">nd low-cost synthetic methodologies for α,β-unsaturated thioester type of compounpan>ds havinpan>g prominpan>enclass="Chemical">pan>t photochromic behavior or having potential to be used as synpan>thetic inpan>termediates inpan> various important organic transformationpan>s is a well-appreciated research problem. In this conpan>text, our work may be noteworthy as inpan> the present work, unpan>der the same reactionpan> conpan>ditionpan> usinpan>g (COCl)2/DMF and PhSH/Et3N in DCM medium at ambient temperature. We have synthesized four structurally and photochromically different thiophenyl derivatives, among which (1–3) are thiophenyl esters of p-hydroxycinnamic and (4) is a novel stilbene-type compound (to the best of our knowledge) substituted with dimethylamine and thiophenyl groups.

Furthermore, (1) and (2) showed a drastic change in their intrinsic chromophoric/fluorophoric activities during their interactions with ctDNA. Spectroscopic (UV–vis, fluorescence, and CD) and viscometric measurements inpan>dicated the inpan>tercalationpan> mode of binpan>ding between these thiophenyl esters (1) or (2) and ctDNA. However, results of a preliminary molecular docking study indicate the possibility of minor groove binding mode of hydrogen-bonding interaction of (1) [or (2)] with ctDn class="Chemical">NA. It is interesting to note that structurally these compounds satisfy the characteristics of both intercalator and groove binders for DNA. These two compounds have single-bonded flexible parts, for which torsional rotation of molecule (1) [or (2)] is possible.[8] This may help these molecules to be fit into the shallow minor groove of DNA. On the other hand, (1) and (2) have planar parts (aromatic rings), which may allow them to interact with DNA through the intercalation mode.[9] So, it is quite possible that these compounds are not perfect intercalator or perfect grooved binder of DNA.

It is also worth mentiopan class="Chemical">ning that the change of intrinsic fluorescence activity of (1) [or (2)] in the presence of biomacromolecule like ctDNA opens up the possibility for applicationpan> of this type of small molecules as DNA-based fluorescence biomarkers. Moreover, the change of fluorescence activity of (1) [or (2)] in the presence of biomacromolecule like DNA (monitored under the physiological condition) can be considered as a model to understand the drug–DNA interaction, which is highly essential for the development of novel therapeutic agents. However, (1) will be comparatively more suitable than (2), considering its high biocompatible.