Novel amino substituted tetracyclic imidazo[4,5-b]pyridine derivatives: Design, synthesis, antiproliferative activity and DNA/RNA binding study
Borka Loncar a, Natasa Perin b, Marija Mioc c, Ida Bocek b, Lea Grgic d, Marijeta Kralj c, Sanja Tomic d, Marijana Radic Stojkovic d, **, Marijana Hranjec b, *
Abstract
A novel series of tetracyclic imidazo[4,5-b]pyridine derivatives was designed and synthesized as potential antiproliferative agents. Their antiproliferative activity against human cancer cells was influenced by the introduction of chosen amino side chains on the different positions on the tetracyclic skeleton and particularly, by the position of N atom in the pyridine nuclei. Thus, the majority of compounds showed improved activity in comparison to standard drug etoposide. Several compounds showed pronounced cytostatic effect in the submicromolar range, especially on HCT116 and MCF-7 cancer cells. The obtained results have confirmed the significant impact of the position of N nitrogen in the pyridine ring on the enhancement of antiproliferative activity, especially for derivatives bearing amino side chains on position 2. Thus, regioisomers 6, 7 and 9 showed noticeable enhancement of activity in comparison to their counterparts 10, 11 and 13 with IC50 values in a nanomolar range of concentration (0.3e0.9 mM). Interactions with DNA (including G-quadruplex structure) and RNA were influenced by the position of amino side chains on the tetracyclic core of imidazo[4,5-b]pyridine derivatives and the ligand charge. Moderate to high binding affinities (logKs ¼ 5e7) obtained for selected imidazo[4,5-b]pyridine derivatives suggest that DNA/RNA are potential cell targets.
Keywords: Amines
imidazo[4,5-b]pyridines Antiproliferative activity
DNA/RNA binding
1. Introduction
Due to the structural similarity of imidazo-pyridine heterocyclic system with naturally occurring purines and great therapeutic potential and significance, suchlike derivatives nowadays play an important role in medicinal chemistry and drug discovery [1]. Being one of the chosen privileged building motifs in medicinal chemistry, they showed a wide range of different biological features playing thus an important role in the prevention of numerous diseases. Imidazo-pyridine derivatives foregrounded their importance in the prevention of proper functioning of cancerous cells, diseases related to the central nervous system, inflammation, etc. Some of derivatives are known as GABAA receptor positive allosteric modulators [2], aromatase inhibitors [3], metalloproteinase inhibitors [4], DNA/RNA intercalators [5], proton pump inhibitors [6], Aurora kinase inhibitors [7,8], serine/threonine-protein kinase inhibitors [9] or blockers of the gastric proton pump (Fig. 1) [10]. Among all possible isomers forms, the most promising biological potential was shown by imidazo[4,5-b]pyridine and imidazo[4,5-c] pyridine derivatives. The majority of published papers regarding the antitumor activity are describing the inhibition of the different protein kinases caused by mentioned derivatives.
Recently, a group of authors has confirmed the inhibition of Aurora A, B and C kinases, caused by 2-arylimidazo[4,5-b]pyridine derivatives and derivatives substituted with 1-benzyl-1H-pyrazol4-yl side chain [11,12]. Additionally, some imidazo[4,5-b]pyridines were identified as potent and selective JAK1 and B-Raf kinase inhibitors [13,14]. Furthermore, 2,6-disubstituted imidazo[4,5-b] pyridines were identified as highly potent TAM inhibitors [15]. A group of authors has synthesized novel imidazo[4,5-b]pyridine derivatives which showed promising anticancer activity against either breast or colon cancer cell lines [16]. Imidazo[4,5-b]pyridine derivatives were identified as well as promising c-Met kinase inhibitors with activity against human lung cancer cells [17]. Also, Ncyclopentyl substituted 2-aryl/heteroarylimidazo[4,5-b]pyridines were shown to be very efficient agents with good microsomal stability [18].
Since many anticancer drugs base their activity on non-covalent interaction with DNA and RNA, one of our aims was to investigate interactions of prepared ligands with these biological targets. Nucleic acids are due to involvement in many crucial processes in cells (repository and transfer of genetic information, carcinogenesis, gene expression, transcriptional and translational regulation, cell death) molecular targets for a large number of drugs [21,22].
Hence, the assessment of binding strength and specificity is essential to elucidate the mode of biological action of drugs. The most common and effective non-covalent modes of binding of small molecules to DNA include intercalative and groove binding. Commonly, groove binders are crescent-shaped molecules, consisting of more aromatic rings while intercalators possess planar aromatic surfaces suitable for insertion between DNA bases [21]. However, it is also possible for small molecules (with planar structures and positively charged pendants) to bind to nucleic acids in multiple binding modes (mixed binding mode).
Results of our previously prepared imidazo[4,5-b]pyridine derivatives showed that intercalation is a dominant binding mode to DNA. One of these derivatives triggered apoptosis in cancer cells which could be associated with its ability to intercalate into DNA (Fig. 1) [5]. In a more recent series of substituted benzimidazo[1,2a]quinolines, it was found that one derivative intercalated in the DNA helix and localized in the nucleus [19]. In continuation of this study, some of amino-substituted benzimidazo[1,2-a]quinoline DNA intercalators showed weak topoisomerase I poisoning activity [22]. Pyrimidinyl-imidazo[1,2-a]pyridines, dual inhibitors of bacterial DNA gyrase and topoisomerase IV, exhibited antibacterial activity against Gram-positive pathogens, wild-type and methicillin-resistant Staphylococcus, and wild-type and fluoroquinolone-resistant Streptococcus [23]. A combination of imidazo[1,5-a]pyridine and pyrrolobenzodiazepine scaffolds showed enhanced DNA binding ability (minor groove) and significant antitumor activity. These compounds also induced the expression of proteins involved in apoptosis and DNA damage like p53, p21 and g-H2AX (Fig. 1) [24].
Taking into account a great biological potential and the fact that imidazo[4,5-b]pyridine scaffold is among the most privileged and important building blocks in medicinal chemistry, we have designed and synthesized novel tetracyclic derivatives as novel and potent antiproliferative agents. Tetracyclic derivatives were substituted with chosen amino side chains which have significantly enhanced the antiproliferative activity of tetracyclic benzimidazole analogues and are placed at different position of skeleton [19,20]. Additionally, the impact of the N atom position in pyridine nuclei on biological activity was studied.
For chosen counterparts of regioisomers, the DNA (including Gquadruplex structure)/RNA binding affinity was studied by using spectroscopic methods (fluorescence, circular dichroism (CD)) and thermal denaturation experiments. In addition, ligand-DNA/RNA interactions were investigated by molecular docking.
2. Results and discussion
2.1. Chemistry
The synthesis of all newly prepared compounds was conducted according to the two main experimental procedures shown in Schemes 1 and 2 using either conventional methods of organic synthesis as well as microwave-assisted synthesis. Acyclic precursors 4, 16 and 17 were prepared in the reaction of aldol condensation from 2-chloro-4-fluorobenzaldehyde 3 or benzoyl chlorides 14 and 15 with 2-(1H-imidazo[4,5-b]pyridine-2-yl) acetonitrile 2. Compound 2 as the main precursor for the synthesis of acyclic derivatives, was obtained in the reaction of cyclocondensation from 2,3-diaminopyridine within the heating with 2cyanoacetamide at 185 C. Compound 4 underwent a thermal cyclization in sulfolane to afford cyclic compounds 5a and 5b as a mixture of two regioisomers in the ratio 1:1. Regioisomers were successfully separated by column chromatography on SiO2 using CH2Cl2/CH3OH as eluent. On the other hand, the acyclic compounds 16 and 17 underwent a thermic cyclization in DMF under basic conditions to give the corresponding 18a/18b and 19a/19b as a mixture of regioisomers which at this synthetic point were not separated.
Subsequent treatment of oxo derivatives with POCl3 and PCl5 gave the mixture of regioisomers 20:22 (ratio 1:5) and 21:23 (ratio 1:4) which were separated by column chromatography on SiO2 using CH2Cl2/CH3OH as eluent. Finally, amino (6e13) and diamino (24e37) substituted pyrido[30,2’:4,5]imidazo[1,2-a]quinoline-6carbonitriles and pyrido[20,3’:4,5]imidazo[1,2-a]quinoline-6carbonitriles obtained in uncatalyzed microwave-assisted amination from compounds 20e23 by using power 800 W, at 170 C and 40 bar in acetonitrile with five to seven-fold excess of the appropriate amine. The structures of all prepared compounds were determined by NMR spectroscopy (1H and 13C) based on the analysis of HeH coupling constants as well as chemical shifts and by elemental analysis.
The structures of regioisomers 24, 31, 30 and 37 were additionally confirmed by using 2D NMR spectroscopy. In the NOESY spectra of compounds 31 and 37, NOE interactions between the H1 and H11 protons of pyridine and benzene nuclei (indicated with green color) confirmed the structure of the regioisomer (Fig. 2).
Additionally, the interactions between the protons of the aliphatic part of the molecule, namely the N,N-.dimethylaminopropyl side chain with the protons of phenyl ring could be also observed (Figs. 2 and 3).
2.2. Antiproliferative activity in vitro
All newly prepared compounds were first tested against HCT116, MCF-7 and H 460 cancer cell lines to assess their antiproliferative activity in vitro. The results are presented in Table 1 and are compared to known antiproliferative agent etoposide. In addition, we selected 12 representative derivatives, which either showed the most prominent antiproliferative activities, or evaluated their cytotoxic activity on non-cancerous cells using human embryonic kidney cell line HEK 293. The prepared tetracyclic derivatives were designed with the aim of systematically studying structural effects on antiproliferative features in order to find the positions of amino substituents and N atom in pyridine ring as well as the type of amino substituents which showed the most significant impact on the improvement of the antiproliferative activity.
The obtained results indicated that the majority of tested compounds showed enhanced antiproliferative effect towards cancer cells but without significant selectivity between cancer cells. The exception was fluoro substituted compound 5a, N,N-dimethylaminopropyl substituted derivatives 10, 24 and 31, piperazine substituted derivative 13 and chloro substituted derivative 22 with selective activity towards HCT116 cell in comparison to other cancer cells. Several compounds showed improved activity in comparison to standard drug etoposide, especially towards HCT116 cells. Interestingly, obtained results revealed that the position of N nitrogen in the pyridine ring had a very strong impact on the enhancement of antiproliferative activity. Regioisomers 6, 7 and 9 bearing amino substituents on the position 2 of tetracyclic skeleton, showed significant improvement of activity in comparison to their counterparts 10, 11 and 13 having IC50 values in a nanomolar range of concentration (0.3e0.9 mM). Regarding the 5-amino substituted regioisomers, it could be observed that the position of N atom did not significantly influence the antiproliferative activity. Furthermore, only 2,5-dipiperazinyl substituted regioisomer 30 showed better activity compared to its regioisomer 37 which surprisingly did not show any inhibitory effect at all. The type of amino side chains was chosen based on previously published results for benzimidazole-derived tetracyclic derivatives. As it could be observed, there was no significant difference in the influence of N,N-dimethylaminopropyl, N-isobutyl and piperazine side chain on antiproliferative activity with the exception of mono-piperidine substituted derivatives 8 and 12 which showed the lowest antiproliferative effect.
Additionally, we have also tested halogen substituted derivatives with compound 22 bearing chloro group at position 5 being the most active one (IC50(HCT116) 0.1 mM). The results obtained from the testing against non-tumour cells showed that 12 tested compounds had a relatively similar cytotoxic profile in tumour cells in comparison to non-tumour cells with the exception of N-isobutyl substituted derivative 32 which showed to be non-cytotoxic (IC50 > 100 mM). Interestingly, compound 22 showed significantly higher activity (~16 times) against HCT116 cells in comparison to non-tumour cells, while compound 29 showed higher and selective activity against MCF-7 cells being non-cytotoxic against HEK 293 cells.
Considering the lipophilicity of imidazo[4,5-b]pyridine derivatives, the clogP values increased upon the introduction of Nisobutyl side chain at all positions on the tetracyclic skeleton. A significant decrease in lipophilicity was observed upon the introduction of the piperazine ring. Piperidine substituent caused slight decrease of clogP value in comparison to N-butyl side chain and an increase compared to compounds bearing N,N-dimethylaminopropyl side chain.
2.3. DNA/RNA binding study
2.3.1. Interactions with double-stranded (ds-) polynucleotides
Based on the results of antiproliferative activity, we have chosen six compounds for study with nucleic acids. Compounds 10, 6, 37, 30, 32 and 25 were soluble in water (c ¼ 3 103 mol dm3). The absorbancies of buffered aqueous solutions of studied compounds were proportional to their concentrations up to c ¼ 3 105 mol dm3. Such behavior suggests that studied compounds do not aggregate by intermolecular stacking at experimental conditions used. Absorption maxima and corresponding molar extinction coefficients (ε) were given in Fig. 4 and Table S1 (Supporting information, SI).
Fluorimetric measurements were performed in an area where the emission and excitation spectra do not overlap. Chemicalize software was used for the prediction of pKa values of imidazo[4,5b]pyridine derivatives (or simply IP derivatives). At pH 7.0,10 and 6 possess one positive charge, piperazine derivatives (37, 30) two charges while isobutylamino derivatives 25 and 32 were not charged [25].
The binding study with DNA and RNA was carried out with ctDNA, AT-DNA (poly(dAdT)2) and GC-DNA (poly(dGdC)2), as models for classical B-helix and AU-RNA (poly A e poly U), as a model for A-helix [26]. Titration with GC-DNA yielded fluorescence decrease/quenching of all studied compounds. Addition of AU-RNA and AT-DNA to buffer aqueous solution of 37, 30, 32 and 25 resulted in an emission increase of those compounds (Table 2, Fig. 5, SI). In contrast to that, 10 and 6 emission decreased in titrations with AU and AT polynucleotides. The position of amino side chains on the tetracyclic ring most likely influenced the fluorescence changes since, in the case of 10 and 6, the addition of AT- and AU-sequences did not cause an increase in their emission (Table 2, SI). The position and the nature of the substituent on a heterocyclic ring have a significant impact on the electron density of aromatic systems which in turn may influence its binding orientation and fluorescence properties upon interaction with nucleic acids [27e29]. In this case, C-2 (in 10 and 6) and C-5 (in 37, 30, 32 and 25) amino side pendants cause redistribution in electron densities of the aromatic core which results in either a decrease in fluorescence (10 and 6 with all polynucleotides and 37, 30, 32 and 25 with GC sequences) or its increase (37, 30, 32 and 25 with AT/AU sequences). Also, it can be noticed that the addition of ctDNA caused a decrease in fluorescence of the majority of compounds which can be probably ascribed to the mixed basepair composition of this natural polynucleotide (about 42% GC base pairs). Similar opposite fluorescence responses with AT and GC sequences, noticed with 37, 30, 32 and 25, were previously observed in few compounds (acridine and phenanthridine derivatives, 4,9-diazapyrenium cations) [30e32]. Such changes have been associated with intercalation of the abovementioned hetero/polycyclic compounds to ds-DNA/RNA since only p-p stacking interactions with the most electrondonating nucleobase, guanine result in efficient fluorescence quenching [33].
The binding constants Ks and ratios n[bound compound]/[DNA/RNA] were calculated by the processing of fluorimetric titration data with the Scatchard equation [35] (Table 2). Fluorescence changes of 25 with some polynucleotides were too small or linear for the accurate calculation of binding constants. While fluorescence changes depended mostly on the position of amino side chains on the tetracyclic ring, the strength of the interactions with DNA/RNA was influenced by both, the position of the amino side chain and the ligand charge.
Thus 10 and 6 with amino side pendant (dimethylaminopropylamino) at C-2 position with one positive charge and 37 and 30 with piperazine at C-2 and C-5 positions with two positive charges showed greater affinities towards ds-DNA and ds-RNA compared to 32 and 25 with isobutylamino at C-5 position possessing no net positive charges (Table 2).
Thermal melting (Tm) value is an important factor in the characterization of compound/polynucleotide interactions. DTm value is a difference between the Tm value of free polynucleotide and complex with a small molecule [36]. In general, the intercalative way of binding stabilizes ds-nucleic acids and yields positive DTm values while minor groove binders may cause stabilization (positive DTm values) or destabilization of ds-DNA/RNA (negative DTm values). 10, 6, 37 and 30 showed moderate to big stabilization effects of ctDNA and AT/AU polynucleotides (Table 3, Fig. 6, SI).
32 and 25 showed a small or no stabilization effect of polynucleotides. Like with binding affinities, such negligible stabilization effect of 32 and 20 can be related to different positions of the amino side chains on the tetracyclic skeleton and the ligand charge.
Higher DTm values displayed by 10, 6, 37 and 30 towards AT/AU polynucleotides in comparison to ctDNA can be attributed to the mixed basepair composition of ctDNA. Also, it was noted that regioisomer 6 showed higher values of DTm for AT-DNA and AURNA than the other regioisomer 10 (Table 3, SI).
To obtain information on the conformation of nucleic acids and its change due to interaction with small molecules, we used CD spectroscopy. CD spectroscopy can also provide insight into the modes of interactions based on the mutual orientation of the small molecule and the nucleic acid chiral axis [37].
Achiral small molecules such as IP derivatives may also produce induced CD spectrum (ICD) after binding to nucleic acids. Since these compounds possess UV/Vis spectra in the region from 240 to 500 nm, the observed ICD spectra at the wavelengths >300 nm, where nucleic acids do not absorb, can be attributed to them (Fig. 7, SI).
Generally, the addition of 10, 6, 37 and 30 caused greater changes (increase/decrease) in CD spectra of DNA/RNA polynucleotides compared to 32 and 25 (Fig. 7, SI). Besides, significant ICD bands appeared in titrations of 10, 6, 37 and 30 with studied polynucleotides. With AT-DNA, 10 and 6 caused the increase in the intensity of CD signal around 273 nm (where both ligands and DNA absorb) and positive ICD signals positioned at 305 nm. Also, 6 induced the appearance of bisignate signals in the area from 370 to 490 nm while 10 induced negative ICD signals positioned around 445 nm. These changes could be the consequence of mixed binding mode – minor groove binding and binding of aggregated 10 and 6 molecules along the polynucleotide backbone and possibly inside the hydrophobic major groove [38].
ICD signals like those induced with AT-DNA but smaller in intensity (probably due to guanine amino group protruding into the minor groove and hindering the interaction) were noticed in CD spectra of GC-DNA upon addition of 10 and 6.
The addition of 10 and 6 to poly A-poly U solution caused a small decrease in intensity of CD spectra of RNA and the formation of weak negative ICD signals around 305 nm (SI). Such changes in ICD signals usually imply an intercalative binding where the ligand with its longer axis is positioned parallel to the longer axis of the adjacent base pair [38]. Rather similar changes in ICD spectra were observed in CD titrations of 37 and 30 with all studied dspolynucleotides. In most titrations, a weak negative ICD signal was formed at ratio r 0.1 and stronger negative signals from 300 to 450 nm were noticed at ratios higher than r > 0.1. Changes at lower ratios could be associated with the intercalation while at higher ratios they were most probably the result of binding of aggregated molecules along the polynucleotide backbone (Fig. 7, SI). Most likely, this type of binding is dominant at higher ratios due to the bulkiness of piperazine groups that hampers the insertion of the tetracyclic core between nucleobases.
However, since these compounds induced different fluorescence responses, increase with AT-DNA and quenching with GC base pairs, it is possible that at lower ratios they partially intercalate only in GC-rich DNA. 32 and 25 induced observable changes (slight increase) only in CD spectra of AT-DNA (Fig. 7, SI). The addition of 25 in ctDNA solution caused an appearance of small negative CD signals around 305 nm which may suggest an intercalative mode of binding. Nevertheless, since these compounds do not stabilize ds-DNA/RNA, an intercalative binding mode can be excluded. A small decrease in the intensity of CD spectra of GC-DNA could be a result of partial intercalation or groove binding where the long axes of the tetracyclic ring and adjacent base pairs are positioned at an angle to each other resulting in the abolition of positive and negative contributions (SI) [39]. These negligible changes observed in most CD titrations of 32 and 25 agree well with the calculated binding affinities and absence of thermal stabilization of DNA and RNA.
Some of the suggested ways of binding were also examined by molecular docking. Interactions of 6 and 32 with alternating double-stranded copolymer, poly(dA-dT)$poly(dA-dT) are shown in Fig. 8. Modes of binding for 6 and 32 with AT-DNA suggested by the spectroscopic methods were consistent with the results obtained by molecular docking.
Binding free energies, calculated by the program MMPBSA.py [44], for orientations of 6, shown in Fig. 8a) and b) are 43 ± 4 kcal/ mol and 46 ± 3 kcal/mol, respectively. Apparently, both orientations are similarly stable.
2.3.2. Interactions with G-quadruplex
We have also investigated the binding of ligands with G-quadruplex DNA quadruplex, a target thought to be associated with significant biological processes, among else cancer growth and progression [45]. IP derivatives possess structural characteristics (flat aromatic surfaces, positively charged side chains with the exception of 25 and 32) that could allow good recognition of Gquadruplex and potentially selectivity for quadruplex over duplex DNA. As a model for the human telomere sequence, we have used 22 nt sequence (d[AGGG(TTAGGG)3], Tel 22). Depending on the applied methodology (NMR, X-ray crystallography, biophysical studies) and the variants of the human telomere sequence, different hybrid forms exist in Kþ environment [45e47]. Xu and others implied that 22 nt telomere sequence in Kþ solution exists as a mixture of mixed-parallel/antiparallel and chair-type G-quadruplex [48]. Titration with Tel22 yielded fluorescence decrease of the majority of studied compounds (10, 6, 37 and 30), except with 32 and 25 whose emission increased (Table 4, Fig. 9, SI). Interestingly, the emission of 32 and 25 decreased upon interaction with ds-GC polynucleotide. The obtained result could indicate the importance of the length of p-stacked DNA systems where quenching of emission of the fluorophore with guanine can be accomplished only within longer DNA molecules [33].
The data from fluorimetric measurements were used for the calculation of the binding constants (log Ks) and stoichiometries of complexes in the concentration range that correspond to 20e80% complex formed (Table 4, Fig. 9). The experimental data of 37 and 30 were best fitted to a binding stoichiometry of ligand-Tel22 of 1:1 and 1:2. In addition, the highest constants were obtained for 1:2 37/ 30-Tel22 complexes. 1:1 stoichiometries were found for complexes of Tel22 and 10, 6, 32 and 25 with rather similar log Ks values (Table 4). The smallest log Ks values to Tel22 showed 32 and 25 possessing isobutylamino side pendant at C-5 position and no charges. Studied ligands exhibited a moderate stabilization effect of Tel22. The greatest stabilization effect was shown by 37 with piperazine groups at C-2 and C-5 positions (Table 4).
A positive band at 290 nm and a shoulder at 260 nm, characteristic of mixed-parallel/antiparallel arrangement, were visible in CD spectrum of Tel 22 in Kþ solution (SI) [49e51]. Intensities of the CD band of Tel 22 were not greatly modified upon the addition of studied ligands. All compounds caused a small increase of CD band of Tel22 at 290 nm (SI) and a small to moderate (especially 10 and 6) decrease of the shoulder at 250 nm. Similar changes obtained in CD titrations of Tel22 with all compounds suggest a similar way of binding (for 1:1 complexes), most probably stacking on one of the terminal quartets of Tel22 [52].
On the other hand, 37 and 30 in 1:2 complexes with Tel22 most probably bind between terminal quartets at the interface of the two quadruplexes. However, to determine an accurate and preferred topology for Tel22-ligand complex [53], additional experiments (NMR, x-ray crystal structure, molecular modeling) should be performed, that goes beyond the scope of this paper. Although the selectivity for quadruplex over duplex DNA was not achieved, due to the good recognition potential of Tel 22, 37 and 30 structures could be further optimized through additional functionalization of piperazine side chains at C-2 and C-5 positions.
3. Conclusions
Within this work, we have described the design and synthesis of tetracyclic systems based on the imidazo[4,5-b]pyridine nuclei which were substituted with amino substituents chosen according to the previously published results for benzimidazole analogues. The antiproliferative activity of regioisomers was studied against human cancer cells and non-tumour cells in order to get insight into the impact of the different position and type of amino side chains and particularly, and the influence of the N atom position in pyridine nuclei.
The majority of compounds showed improvement of antiproliferative activity on HCT116 and MCF-7 cancer cells when compared to a standard drug etoposide. From the obtained results it could be noticed that the position of N nitrogen in the pyridine ring has a strong impact on the antiproliferative activity, while the type of amino substituent did not influence activity significantly. Thus, regioisomers 6, 7 and 9 substituted with amino substituents at position 2 showed noticeable enhancement of activity in comparison to their counterparts 10, 11 and 13 having IC50 values from 0.3 mM to 0.9 mM against all three cancer cells.
Furthermore, chosen compounds had a relatively similar cytotoxic profile in tumour cells in comparison to non-tumour cells. The exception was compound 32 substituted with N-isobutylamino group being non-cytotoxic (IC50 > 100 mM). DNA (including G-quadruplex structure)/RNA binding strength was affected by the position of the amino side chain on the tetracyclic core of imidazo[4,5-b]pyridine derivatives and the ligand charge. At lower compound to polynucleotide ratios (many data points at r 0.1 in fluorescence titrations and data at r 0.1 in CD titrations, Table 2 and Fig. 7), obtained results (1e10 mM, positive ICD signals in the region from 273 to 305 nm noticed for AT- and GC-DNA and weak negative ICD signals around 305 nm for AU-RNA, enhanced thermal stability of polynucleotides) suggest that 10 and 6 bind inside the minor groove of AT- and GC-DNA and intercalate into AU-RNA. Derivatives 37 and 30 with piperazine most probably bind by intercalative binding mode (negative ICD signals, moderate to big DTm values) while lower affinities of 32 and 25 in comparison to 10, 6, 37 and 30, and absence of thermal stability, as well as negligible ICD signals, point to non-intercalative binding mode, most likely aggregation of 32 and 25 molecules along the polynucleotide backbone and inside the grooves. At excess of 10, 6, 37 and 30 over polynucleotide binding sites, compounds form aggregates along DNA double helix. Moderate to high binding affinities (logKs ¼ 5e7) obtained for selected imidazo[4,5-b]pyridine derivatives suggest that DNA/RNA are potential cell targets. Nevertheless, since high binding affinities were noticed even in the case of low antiproliferative activity (like with 37), targets other than DNA and RNA could also be involved in the mechanism of biological action.
4. Experimental part
4.1. General methods
All chemicals and solvents were purchased from commercial suppliers Aldrich and Acros. Melting points were recorded on SMP11 Bibby and Büchi 535 apparatus. All NMR spectra were measured in DMSO‑d6 solutions using TMS as an internal standard. The 1H and 13C NMR spectra were recorded on a Varian Bruker Avance III HD 400 MHz/54 mm Ascend. Chemical shifts are reported in ppm (d) relative to TMS. All compounds were routinely checked by TLC with Merck silica gel 60F-254 glass plates. The microwave-assisted synthesis was performed in a Milestone start S microwave oven using quartz cuvettes under the pressure of 40 bar. Elemental analysis for carbon, hydrogen and nitrogen were performed on a PerkinElmer 2400 elemental analyzer. Where analyses are indicated only as symbols of elements, analytical results obtained are within 0.4% of the theoretical value. Mass spectra were recorded on Agilent 1200 Series HPLC system with an Agilent 6420 DAD mass detector and a triple quadrupole mass spectrometer.
4.2. Synthesis
The experiments were carried out on three human cell lines: HCT 116 (colon carcinoma), H 460 (lung carcinoma), MCF-7 (breast carcinoma) and HEK 293 (human embryonic kidney cells), according to the previously published experimental procedure. Briefly, the cells were grown in DMEM medium with the addition of 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/mL penicillin and 100 mg/mL streptomycin, and cultured as monolayers at 37 C in a humidified atmosphere with 5% CO2. Cells were seeded at 2 103 cells/well in standard 96-well microtiter plates and left to attach for 24 h. The next day, the test compound was added in five serial 10-fold dilutions. The cell growth rate was evaluated after 72 h of incubation, using MTT assay. Obtained results are expressed as IC50 value which stands for the concentration of the compound necessary for 50% of growth inhibition. The IC50 values are calculated from the concentration-response curve using linear regression analysis by fitting the test concentrations that give PG values above and below the reference value (i.e. 50%). Each test was performed in quadruplicate in at least two individual experiments.
4.4. DNA/RNA binding study
The UV/vis spectra were recorded on a Varian Cary 100 Bio spectrophotometer, CD spectra on JASCO J815 spectrophotometer and fluorescence spectra on a Varian Cary Eclipse spectrophotometer at 25 C using appropriate 1 cm path quartz cuvettes.
Materials. Polynucleotides were purchased as noted: poly Aepoly U, calf thymus ctDNA, poly(dAdT)2 and poly(dGdC)2 (Sigma-Aldrich). Polynucleotides were dissolved in Na-cacodylate buffer, I ¼ 0.05 mol dm3, pH ¼ 7. The calf thymus ctDNA was additionally sonicated and filtered through a 0.45 mm filter [54]. Polynucleotide concentration was determined spectroscopically as the concentration of phosphates [55].
Spectrophotometric titrations were performed at pH ¼ 7 (I ¼ 0.05 mol dm3, sodium cacodylate buffer) by adding portions of polynucleotide solution into the solution of the studied compound for fluorimetric experiments and CD experiments were done by adding portions of the compound stock solution into the solution of a polynucleotide. In fluorimetric experiments excitation wavelength of lexc ¼ 364/365,397,408 and 428/429 nm was used to avoid the inner filter effect caused due to increasing absorbance of the polynucleotide. Emission was collected in the range lem ¼ 400e650 nm. Values for Ks obtained by processing titration data using the Scatchard equation (Table 2), all have satisfactory correlation coefficients (0.99). Thermal melting curves for DNA, RNA and their complexes with studied compounds were determined as previously described by following the absorption change at 260 nm as a function of temperature. The absorbance of the ligands was subtracted from every curve and the absorbance scale was normalized. Tm values are the midpoints of the transition curves determined from the maximum of the first derivative and checked graphically by the tangent method. The DTm values were calculated by subtracting Tm of the free nucleic acid from Tm of the complex. Every DTm value here reported was the average of two measurements. The error in DTm is ±0.5 C. 50- AGGG(TTAGGG)3e30(Tel22) was obtained from IDT (Integrated DNA Technologies), USA. Tel22 was dissolved in 0.1 M potassium phosphate buffer. The starting Tel22 oligonucleotide solution was first heated up to 95 C for 10 min and then slowly cooled to 10 C at the cooling rate of 1 C/min to allow DNA oligonucleotide to adopt a G-quadruplex structure [56]. The G-quadruplex structure was confirmed by thermal melting and CD spectra [41]. The concentration of G quadruplex was expressed in terms of oligonucleotide structure. In fluorimetric titrations, aliquots of Tel 22 solution were added to the solution of ligands (c ¼ 5 107 M).
References
[1] J. Joule, K. Mills, Heterocyclic Chemistry, fifth ed., Blackwell Publishing Ltd, 2010.
[2] A.C. Foster, J.A. Kemp, Glutamate- and GABA-based CNS therapeutics, Curr. Opin. Pharmacol. 6 (2006) 7e17, https://doi.org/10.1016/j.coph.2005.11.005.
[3] M. Dowsett, D. Smithers, J. Moore, P.F. Trunet, R.C. Coombes, T.J. Powles, R. Rubens, I.E. Smith, Endocrine changes with the aromatase inhibitor fadrozole hydrochloride in breast cancer, Eur. J. Canc. 30 (1994) 1453e1458, https://doi.org/10.1016/0959-8049(94)00281-9.
[4] N. Ando, S. Terashima, Synthesis and matrix metalloproteinase (MMP)-12 inhibitory activity of ageladine A and its analogs, Bioorg. Med. Chem. Lett 17 (2007) 4495e4499, https://doi.org/10.1016/j.bmcl.2007.06.005.
[5] M. Hranjec, B. Lucic, I. Ratkaj, S.K. Pavelic, I. Piantanida, K. Pavelic, G. Karminski-Zamola, Novel imidazo[4,5-b]pyridine and triaza-benzo[c]fluorene derivatives: synthesis, antiproliferative activity and DNA binding studies, Eur. J. Med. Chem. 46 (2011) 2748e2758, https://doi.org/10.1016/j.ejmech.2011.03.062.
[6] M. Bamford, 3 Hþ/Kþ ATPase inhibitors in the treatment of acid-related disorders, Prog. Med. Chem. 47 (2009) 75e162, https://doi.org/10.1016/ S0079-6468(08)00203-8.
[7] P. Lan, W.N. Chen, W.M. Chen, Molecular modeling studies on imidazo[4,5-b] pyridine derivatives as Aurora A kinase inhibitors using 3D-QSAR and docking approaches, Eur. J. Med. Chem. 46 (2011) 77e94. j.ejmech.2010.10.017.
[8] V. Bavetsias, J.M. Large, C. Sun, N. Bouloc, M. Kosmopoulou, M. Matteucci, N.E. Wilsher, V. Martins, J. Reynisson, B. Atrash, A. Faisal, F. Urban, M. Valenti, A. de Haven Brandon, G. Box, F.I. Raynaud, P. Workman, S.A. Eccles, R. Bayliss, J. Blagg, S. Linardopoulos, E. McDonald, Imidazo[4,5-b]pyridine derivatives as inhibitors of Aurora kinases: lead optimization studies toward the identification of an orally bioavailable preclinical development candidate, J. Med. Chem. 53 (14) (2010) 5213e5228, https://doi.org/10.1021/jm100262j.
[9] B.J. Newhouse, S. Wenglowsky, J. Grina, E.R. Laird, W.C. Voegtli, L. Ren, K. Ahrendt, A. Buckmelter, S.L. Gloor, N. Klopfenstein, et al., Imidazo[4,5-b] pyridine inhibitors of B-Raf kinase, Bioorg. Med. Chem. Lett 23 (2013) 5896e5899, https://doi.org/10.1016/j.bmcl.2013.08.086.
[10] C. Scarpignato, R.H. Hunt, Proton pump inhibitors: the beginning of the end or the end of the beginning? Curr. Opin. Pharmacol. 8 (2008) 677e684, https:// doi.org/10.1016/j.coph.2008.09.004.
[11] V. Bavetsias, C. Sun, N. Bouloc, J. Reynisson, P. Workman, S. Linardopoulos, E. McDonald, Hit generation and exploration: imidazo[4,5-b]pyridine derivatives as inhibitors of Aurora kinases, Bioorg. Med. Chem. Lett. 17 (2007) 6567e6571, https://doi.org/10.1016/j.bmcl.2007.09.076.
[12] V. Bavetsias, Y. Perez-Fuertes, P.J. McIntyre, B. Atrash, M. Kosmopoulou, L. O’Fee, R. Burke, C. Sun, A. Faisal, K. Bush, S. Avery, A. Henley, F.I. Raynaud, S. Linardopoulos, R. Bayliss, J. Blagg, 7-(Pyrazol-4-yl)-3H-imidazo[4,5-b]pyridine-based derivatives for kinase inhibition: Co-crystallisation studies with Aurora-A reveal distinct differences in the orientation of the pyrazole N1substituent, Bioorg. Med. Chem. Lett 25 (2015) 4203e4209, https://doi.org/ 10.1016/j.bmcl.2015.08.003.
[13] M.M. Vasbinder, M. Alimzhanov, M. Augustin, G. Bebernitz, K. Bell, C. Chuaqui, T. Deegan, A.D. Ferguson, K. Goodwin, D. Huszar, et al., Identification of azabenzimidazoles as potent JAK1 selective inhibitors, Bioorg. Med. Chem. Lett. 26 (2016) 60e67, https://doi.org/10.1016/j.bmcl.2015.11.031.
[14] B.J. Newhouse, S. Wenglowsky, J. Grina, E.R. Laird, W.C. Voegtli, L. Ren, K. Ahrendt, A. Buckmelter, S.L. Gloor, N. Klopfenstein, J. Rudolph, Z. Wen, X. Li, B. Feng, Imidazo[4,5-b]pyridine inhibitors of PIN1 inhibitor API-1 B-Raf kinase, Bioorg. Med. Chem. Lett. 23 (2013) 5896e5899, https://doi.org/10.1016/j.bmcl.2013.08.086.
[15] T. Baladi, J. Aziz, F. Dufour, V. Abet, V. Stoven, F. Radvanyi, F. Poyer, T.-D. Wu, J.-L. Guerquin-Kern, I. Bernard-Pierrot, S. Marco Garrido, S. Piguel, Bioorg.Med. Chem. 26 (2018) 5510e5530, https://doi.org/10.1016/ j.bmc.2018.09.031.
[16] N.M. Ghanema, F. Farouka, R.F. Georgeb, S.E.S. Abbasb, O.M. El-Badrya, Design and synthesis of novel imidazo[4,5-b]pyridine based compounds as potent anticancer agents with CDK9 inhibitory activity, Bioorg. Chem. 80 (2018) 565e576, https://doi.org/10.1016/j.bioorg.2018.07.006.
[17] X.D. An, H. Liu, Z.L. Xu, Y. Jin, X. Peng, Y.M. Yao, M. Geng, Y.Q. Long, et al., Discovery of potent 1H-imidazo[4,5-b]pyridine-based c-Met kinase inhibitors via mechanism-directed structural optimization, Bioorg. Med. Chem. Lett 25 (2015) 708e716, https://doi.org/10.1016/j.bmcl.2014.11.070.
[18] A.M. Sajith, K.K.A. Khader, N. Joshi, M.N. Reddy, M. Syed Ali Padusha,H.P. Nagaswarupa, M. Nibin Joy, Y.D. Bodke, R.P. Karuvalam, R. Banerjee, A. Muralidharan, P. Rajendra, Design, synthesis and structure-activity relationship (SAR) studies of imidazo[4,5-b]pyridine derived purine isosteres and their potential as cytotoxic agents, Eur. J. Med. Chem. 89 (2015) 21e31, https://doi.org/10.1016/j.ejmech.2014.10.037.
[19] N. Perin, R. Nhili, K. Ester, W. Laine, G. Karminski-Zamola, M. Kralj, M.H. David-Cordonnier, M. Hranjec, Synthesis, antiproliferative activity and DNA binding properties of novel 5-Aminobenzimidazo[1,2-a]quinoline-6carbonitriles, Eur. J. Med. Chem. 80 (2014) 218e227, https://doi.org/ 10.1016/j.ejmech.2014.04.049.
[20] N. Perin, I. Martin-Kleiner, R. Nhili, W. Laine, M.H. David-Cordonnier, O. Vugrek, G. Karminski-Zamola, M. Kralj, M. Hranjec, Biological activity and DNA binding studies of 2-substituted benzimidazo[1,2-a]quinolines bearing different amino side chains, Med. Chem. Commun. 4 (2013) 1537e1550, https://doi.org/10.1039/C3MD00193H.
[21] M. Demeunynck, C. Bailly, W.D. Wilson, DNA and RNA Binders: from Small Molecules to Drugs, vol. 1, Wiley-VCH, Weinheim, 2002 (Chapter 5).
[22] N. Perin, R. Nhili, M. Cindric, B. Bertosa, D. Vusak, I. Martin-Kleiner, W. Laine,G. Karminski-Zamola, M. Kralj, M.-H. David-Cordonnier, M. Hranjec, Amino substituted benzimidazo[1, 2- a]quinolines: antiproliferative potency, 3D QSAR study and DNA binding properties, Eur. J. Med. Chem. 122 (2016)530e545, https://doi.org/10.1016/j.ejmech.2016.07.007.
[23] J.T. Starr, R.J. Sciotti, D.L. Hanna, M.D. Huband, L.M. Mullins, H. Cai, J.W. Gage,M. Lockard, M.R. Rauckhorst, R.M. Owen, M.S. Lall, M. Tomilo, H. Chen, S.P. McCurdy, M.R. Barbachyn, 5-(2-Pyrimidinyl)-imidazo[1,2-a]pyridines are antibacterial agents targeting the ATPase domains of DNA gyrase and topoisomerase IV, Bioorg. Med. Chem. Lett 19 (2009) 5302e5306, https://doi.org/ 10.1016/j.bmcl.2009.07.141.
[24] A. Kamal, G. Ramakrishna, M. Janaki Ramaiah, A. Viswanath, A.V. Subba Rao, C. Bagul, D. Mukhopadyay, S.N.C.V.L. Pushpavalli, M. Pal-Bhadra, Design, synthesis and biological evaluation of imidazo[1,5-a]pyridineePBD conjugates as potential DNA-directed alkylating agents, Med. Chem. Commun. 4 (2013) 697e703, https://doi.org/10.1039/C2MD20219K.
[25] Chemicalize was used for pKa value prediction. https://chemicalize.com/% 20developed%20by%20ChemAxon, December, 2020. http://www.chemaxon. com.
[26] J. Olmsted, D.R. Kearns, Mechanism of ethidium-bromide fluorescence enhancement on binding to nucleic-acids, Biochemistry (U.S.A) 16 (1977) 3647e3654, https://doi.org/10.1021/bi00635a022.
[27] N.W. Luedtke, Q. Liu, Y. Tor, On the electronic structure of ethidium, Chem. Eur J. 11 (2) (2005) 495e508, https://doi.org/10.1002/chem.200400559.
[28] H. Szatylowicz, O. Stasyuk, T.M. Krygowski, Substituent effects in heterocyclic systems, in: E.F.V. Scriven, C.A. Ramsden (Eds.), Advances in Heterocyclic Chemistry, Elsevier Inc., 2015, pp. 137e192.
[29] M. Radic Stojkovic, I. Piantanida, Tuning ureaephenanthridinium conjugates for DNA/RNA and base pair Recognition, Tetrahedron 64 (2008) 7807e7814, https://doi.org/10.1016/j.tet.2008.05.142.
[30] I. Piantanida, B.S. Palm, M. Zini c, H.J. Schneider, A new 4,9-diazapyrenium intercalator for single- and double-stranded nucleic acids: distinct differences from related diazapyrenium compounds and ethidium bromide, J. Chem. Soc., Perkin Trans. 2 (2001) 1808e1816, https://doi.org/10.1039/ b103214n.
[31] S. Georghiou, Interaction of acridine drugs with dna and nucleotides, Photochem. Photobiol. 26 (1977) 59e68, https://doi.org/10.1111/j.17511097.1977.tb07450.x.
[32] S.O. Kelley, J.K. Barton, Electron transfer between bases in double helical DNA, Science 283 (1999) 375e381, https://doi.org/10.1126/science.283.5400.375.
[33] G. Scatchard, The attractions of proteins for small molecules and ions, Ann. N. Y. Acad. Sci. 51 (4) (1949) 660e672, https://doi.org/10.1111/j.17496632.1949.tb27297.x.
[34] J.D. Mcghee, P.H. von Hippel, Correction, J. Mol. Biol. 103 (3) (1976), https:// doi.org/10.1016/0022-2836(76)90228-X, 679-679.
[35] J.L. Mergny, L. Lacroix, Analysis of thermal melting curves, Oligonucleotides 13 (6) (2003) 515e537, https://doi.org/10.1089/154545703322860825.
[36] N. Berova, K. Nakanishi, R.W. Woody, Circular Dichroism Principles and Applications, second ed., Wiley-VCH, New York, 2000.
[37] M. Eriksson, B. Norden, Linear and circular dichroism of drug-nucleic acid complexes, Methods Enzymol. 340 (2001) 68e98, https://doi.org/10.1016/ S0076-6879(01)40418-6.
[38] T. Smidlehner, I. Piantanida, G. Pescitelli, Polarization spectroscopy methods in the determination of interactions of small molecules with nucleic acids tutorial, Beilstein J. Org. Chem. 14 (2018) 84e105, https://doi.org/10.3762/ bjoc.14.5.
[39] S.L. Cree, M.A. Kennedy, Relevance of G-quadruplex structures to pharmacogenetics, Front. Pharmacol. 5 (2014) 160, https://doi.org/10.3389/ fphar.2014.00160.
[40] J. Wang, W. Wang, P.A. Kollman, D.A. Case, Automatic atom type and bond type perception in molecular mechanical calculations, J. Mol. Graph. Model. 25 (2) (2006) 247e260, https://doi.org/10.1016/j.jmgm.2005.12.005.
[41] D.A. Case, R.M. Betz, D.S. Cerutti, et al., AMBER16, University of California, San Francisco, CA, 2016.
[42] M. Zgarbova, J. Sponer, M. Otyepka, T.E. Cheatham, R. Galindo-Murillo, P. Jurecka, Refinement of the sugar-phosphate backbone torsion beta for the AMBER force fields improves the description of Z-DNA and B-DNA, J. Chem. Theor. Comput. 11 (12) (2015) 5723e5736, https://doi.org/10.1021/ acs.jctc.5b00716.
[43] J. Wang, R.M. Wolf, J. Caldwell, P.A. Kollman, D.A. Case, Development and testing of a general amber force field, J. Comput. Chem. 25 (9) (2004) 1157e1174, https://doi.org/10.1002/jcc.20035.
[44] B.R. Miller, T.D. McGee, J. Swails, N. Homeyer, H. Gohlke, A. Roitberg, MMPBSA.py: an efficient program for end-state free energy calculations, J. Chem. Theor. Comput. 8 (2012) 3314e3321, https://doi.org/10.1021/ ct300418h.
[45] J.B. Chaires, Human telomeric G-quadruplex: thermodynamic and kinetic studies of telomeric quadruplex stability, FEBS J. 277 (5) (2010) 1098e1106, https://doi.org/10.1111/j.1742-4658.2009.07462.x.
[46] A.T. Phan, V. Kuryavyi, K.N. Luu, D.J. Patel, Structure of two intramolecular Gquadruplexes formed by natural human telomere sequences in Kþ solution, Nucleic Acids Res. 35 (19) (2007) 6517e6525, https://doi.org/10.1093/nar/ gkm706.
[47] A. Ambrus, D. Chen, J. Dai, T. Bialis, R.A. Jones, D. Yang, Human telomeric sequence forms a hybrid type intramolecular G-quadruplex structure with mixed parallel ⁄antiparallel strands in potassium solution, Nucleic Acids Res.19 (9) (2006) 2723e2735, https://doi.org/10.1093/nar/gkl348, 34.
[48] Y. Xu, Y. Noguchi, H. Sugiyama, The new models of the human telomere d [AGGG(TTAGGG)3] in Kþ solution, Bioorg. Med. Chem. 14 (2006) 5584e5591, https://doi.org/10.1016/j.bmc.2006.04.033.
[49] H. Gampp, M. Maeder, C.J. Meyer, A.D. Zuberbuhler, Calculation of equilibrium constants from multiwavelength spectroscopic data–II: SPECFIT: two userfriendly programs in basic and standard FORTRAN 77, Talanta 32 (4) (1985) 257e264, https://doi.org/10.1016/0039-9140(85)80077-1.
[50] R. del Villar-Guerra, R.D. Gray, J.B. Chaires, Characterization of quadruplex DNA structure by circular dichroism, Curr. Protoc. Nucleic Acid Chem. 68 (2017) 17, https://doi.org/10.1002/cpnc.23, 8.1-17.8.16.
[51] G. Zagotto, A. Ricci, E. Vasquez, A. Sandoli, S. Benedetti, M. Palumbo, C. Sissi, Tuning G-quadruplex vs double-stranded DNA recognition in regioisomeric lysyl-peptidyl-anthraquinone conjugates, Bioconjugate Chem. 22 (10) (2011) 2126e2135, https://doi.org/10.1021/bc200389w.
[52] C.I.V. Ramos, S.P. Almeida, L.M.O. Lourenco, P.M.R. Pereira, R. Fernandes, M.A.F. Faustino, J.P.C. Tome, J. Carvalho, C. Cruz, Neves, M. Multicharged, Phthalocyanines as selective ligands for G-quadruplex DNA structures, Molecules 24 (4) (2019) 733, https://doi.org/10.3390/molecules24040733.
[53] N.H. Campbell, G.N. Parkinson, A.P. Reszka, S. Neidle, Structural basis of DNA quadruplex recognition by an acridine drug, J. Am. Chem. Soc. 28 (21) (2008) 6722e6724, https://doi.org/10.1021/ja8016973, 130.
[54] J.B. Chaires, N. Dattagupta, D.M. Crothers, Studies on interaction of anthracycline antibiotics and deoxyribonucleic-acid – equilibrium binding-studies on interaction of daunomycin with deoxyribonucleic-acid, Biochemistry (U.S.A) 21 (17) (1982) 3933e3940, https://doi.org/10.1021/bi00260a005.
[55] T.V. Chalikian, J. Volker, G.E. Plum, K.J. A Breslauer, A more unified picture for the thermodynamics of nucleic acid duplex melting: a characterization by calorimetric and volumetric techniques, Proc. Natl. Acad. Sci. U.S.A. 96 (14) (1999) 7853e7858, https://doi.org/10.1073/pnas.96.14.7853.
[56] M. Boncina, C. Podlipnik, I. Piantanida, J. Eilmes, M.P. Teulade-Fichou, G. Vesnaver, J. Lah, Thermodynamic fingerprints of ligand binding to human telomeric G-quadruplexes, Nucleic Acids Res. 43 (21) (2015) 10376e10386, https://doi.org/10.1093/nar/gkv1167.