Synthesis and Biological Evaluation of Benzo [4,5]- and Naphtho[2′,1′:4,5]imidazo[1,2-c]pyrimidinone Derivatives

Azacarbazoles have attracted significant interest due to their valuable properties, such as anti-pathogenic and antitumor activity. In this study, a series of structurally related tricyclic benzo[4,5]- and tertacyclic naphtho[2′,1′:4,5]imidazo[1,2-c]pyrimidinone derivatives with one or two positively charged tethers were synthesized and evaluated for anti-proliferative activity. Lead tetracyclic derivative 5b with two amino-bearing arms inhibited the metabolic activity of A549 lung adenocarcinoma cells with a CC50 value of 3.6 μM, with remarkable selectivity (SI = 17.3) over VA13 immortalized fibroblasts. Cell-cycle assays revealed that 5b triggers G2/M arrest without signs of apoptosis. A study of its interaction with various DNA G4s and duplexes followed by dual luciferase and intercalator displacement assays suggests that intercalation, rather than the modulation of G4-regulated oncogene expression, might contribute to the observed activity. Finally, a water-soluble salt of 5b was shown to cause no acute toxic effects, changes in mice behavior, or any decrease in body weight after a 72 h treatment at concentrations up to 20 mg/kg. Thus, 5b is a promising candidate for studies in vivo; however, further investigations are needed to elucidate its molecular target(s).

tures.In particular, the chosen planar aromatic system, supposed to stabilize duplex and G4 DNA structures via stacking interactions, was expanded to naphtho[2′,1′:4,5]imidazo [1,2-c]pyrimidinone in order to enhance stabilization efficiency [20,21].Moreover, basic side chains were introduced into the compounds to provide electrostatic interaction with the negatively charged DNA sugar-phosphate backbone [20,21].In addition to the unmodified amino group, guanidino-and dimethylamino-containing substituents were proposed due to their higher basicity.The choice of the aforementioned modifications is supported by their successful application in the design of carbazole-and γ-carbolinebased derivatives with antitumor activity [5,12,16].Thus, we synthesized a series of benzo [4,5]-and naphtho[2′,1′:4,5]imidazo [1,2-c]pyrimidinones with one or two positively charged tethers.Their cytotoxicity and selectivity of action was evaluated on cell lines of cancerous and non-cancerous etiology, and possible DNA-related mechanisms of action were investigated.A,E) and carbazole (B-D) .An example of synthesis of the adduct of 2′-deoxycytidine with 1,4-benzoquinone (F).Blue scaffolds are γ-carboline and carbazole.

Materials and Methods
1,8-Diazabicyclo [5.4.0]undec-7-ene (DBU) and 37% aqueous CH2O were purchased from Thermo Fisher Scientific (Madison, WI, USA).1,4-Benzoquinone, 1,4-naphthoquinone, methyl 2-bromoacetate, ethane-1,2-diamine, 1H-pyrazole-1-carboxamidine hydrochloride, and sodium cyanoborohydride were obtained from Sigma-Aldrich (St. Louis, MO, USA).Solvents were purchased from commercial sources and used without further purification, except for CH2Cl2, which was distilled over calcium hydride.Thin layer chromatography (TLC) was performed on plates (Merck, Darmstadt, Germany) pre-coated More than 30 years ago, an adduct of 2 -deoxycytidine with 1,4-benzoquinone was synthesized during the investigation of the cancerogenic properties of benzene [19] (Figure 1F).However, a biological evaluation of this heteroaromatic scaffold was not performed.Taking into account the intriguing antitumor activity of γ-carbolines, we decided to evaluate this property for a set of compounds with a benzo [4,5]imidazo [1,2-c]pyrimidinone scaffold.Since DNA fragments are considered the main target of most γ-carbolines, we designed compounds in a way that allows them to effectively interact with DNA structures.In particular, the chosen planar aromatic system, supposed to stabilize duplex and G4 DNA structures via stacking interactions, was expanded to naphtho [2 ,1 :4,5]imidazo [1,2c]pyrimidinone in order to enhance stabilization efficiency [20,21].Moreover, basic side chains were introduced into the compounds to provide electrostatic interaction with the negatively charged DNA sugar-phosphate backbone [20,21].In addition to the unmodified amino group, guanidino-and dimethylamino-containing substituents were proposed due to their higher basicity.The choice of the aforementioned modifications is supported by their successful application in the design of carbazole-and γ-carboline-based derivatives with antitumor activity [5,12,16].Thus, we synthesized a series of benzo [4,5]-and naphtho [2 ,1 :4,5]imidazo [1,2-c]pyrimidinones with one or two positively charged tethers.Their cytotoxicity and selectivity of action was evaluated on cell lines of cancerous and non-cancerous etiology, and possible DNA-related mechanisms of action were investigated.

MTT Assay
The cytotoxicity of the substances was tested using the MTT (3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide) Mosmann assay [23] with some modifications.A total of 2500 cells per well for the MCF7 , HEK293T, and A549 cell lines or 4000 cells per well for the VA13 cell line were plated out in 135 µL of DMEM-F12 medium (Gibco, USA) in a 96-well plate and incubated in a 5% CO 2 incubator for 20 h without treatment.Then, 15.8 µL of the corresponding medium-DMSO solutions of the substances being tested was added to the cells in triplicate for each dilution (eight concentrations with three-times dilutions with final concentrations in cells ranging from 0.05 to 100 µM; the final DMSO concentrations in the medium were 0.5% or less) and incubated for 72 h.Doxorubicin was used as a control substance.Then, the MTT reagent (Paneco LLC, Moscow, Russia) was added to the cells up to the final concentration of 0.5 g/L (10× stock solution in PBS was used) and incubated for 1.5 h (MCF7 , HEK293T and A549 cells) or 3 h (VA13 cells) at 37 • C in the 5% CO 2 incubator.Then, the MTT solution was discarded, and 140 µL of DMSO (PharmaMed LLC, Moscow, Russia) was added.The plates were agitated on a shaker (120 rpm) to dissolve the formazan.Absorbance was measured using a Victor X5 microplate reader (PerkinElmer, Waltham, MA USA) at a wavelength of 555 nm.The results were used to construct dose-response curves by non-linear regression approximation of the normalized data and to estimate IC 50Abs values (IC 50Abs is the concentration resulting in two-fold decrease in the number of cells compared to untreated cells) with GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA, USA).

Apoptosis/Necrosis Assay
A total of 3 × 10 5 cells per well of the A549 cell line were seeded in 2 mL of DMEM-F12 medium (Gibco, USA) in 6-well plates, and then incubated in the 5% CO 2 incubator for 20 h without treatment.After that, the medium-DMSO solutions of the compounds under test in amounts of 1-33 µL were added (the DMSO concentrations in the wells were 1.5% or less) and the cells were incubated for 24 h.The cells were collected (a sedimentation procedure was performed at 200-300 g) in 1.5 mL tubes, flushed with PBS, and resuspended in 1X Annexin V binding buffer consisting of 0.01 M HEPES (pH 7.4), 0.14 M NaCl, and 2.5 mM CaCl 2 solution.Then, 5 µL of Annexin V-FITC (A13201, ThermoFisher, Waltham, MA, USA) was added per 100 µL of cell solution in a 1× binding buffer, and the cells were incubated for 15 min at room temperature.Afterwards, the cells were sedimented, and the solution was replaced with 1X Annexin V binding buffer containing propidium iodide (2.5 µL of PI (1 mg/mL, ThermoFisher, USA) per 100 µL of buffer), incubated for 15 min at room temperature.Then, the cells were analyzed with a LongCyte C3090 flow cytometer (Challenbio, Beijing, China) after staining for 2 h.
Caspase 3/7 activation was analyzed with the Muse ® Caspase-3/7 Kit (Cytek Biosciences, Fremont, CA, USA).The cells were prepared in the same way as in the previous assay, but they were resuspended in 500 µL of growth medium, followed by staining and analysis of 100 µL of the suspension in accordance to the manufacturer's protocol.

Cell Cycle Assay
A total of 1 × 10 5 cells per well of the A549 cell line were seeded in 1 mL of DMEM-F12 medium (Gibco, USA) in a 12-well plate and incubated in the 5% CO 2 incubator for 20 h without treatment.Then, 2 µL of medium-DMSO solutions of the substances under test (the final DMSO concentrations in the medium were 0.5% or less) was added to the cells and incubated for 24 h.The cells were collected (a sedimentation procedure was performed at 200-300 g) in 1.5 mL tubes, flushed with PBS, and stained with the staining solution (500 µL, RPMI1640 medium (Gibco, USA), Hoechst-33342 (10 µg/mL, Invitrogen, Waltham, MA, USA) with TritonX100 (0.1%) and HEPES-KOH (10 mM pH 7.3)) for 15 min at room temperature.The cells were analyzed with a LongCyte C3090 (Challenbio, Beijing, China) or Amnis FlowSight (Luminex, Austin, TX, USA) flow cytometer after staining for 0.5 h.

Oligonucleotide-Based Assays 2.3.1. FRET-Melting Assay
Oligodeoxyribonucleotides (ODNs) labeled with 6-carboxyfluorescein (FAM) and Black Hole Quencher 1 (BHQ1) at 5 -and 3 -termini, respectively, and unlabeled oligodeoxynucleotide ds26 were purchased from Litekh (Moscow, Russia) (purity > 95%, HPLC).The ODN sequences are presented in Table S1.Solutions of the ODNs (1 µM) were prepared in 20 mM sodium phosphate buffer, pH 7.4, supplemented with 10 mM KCl (buffer 1), in all cases except VEGF, for which a 5 mM sodium phosphate buffer, pH 7.4, supplemented with 25 mM LiCl (buffer 2) was used.The compounds under test were added to ODN solutions to final concentrations of 1-20 µM (1:1, 1:2, 1:5, 1:10, or 1:20 ODN:compound ratio).The mixtures were heated to 95 • C for 5 min and then cooled on ice to facilitate the intra-molecular folding of ODN structures.FRET melting curves were obtained using a QuantStudio 5 PCR system (Thermo Fisher Scientific, USA).Changes in fluorescence at 520 nm were recorded every 0.3 • C during stepwise heating of the samples at an average rate of 1.5 • C/min.Melting temperatures were determined from the maxima of first derivatives of the melting curves.For selectivity analysis, labeled G4-forming oligonucleotides were mixed with pre-annealed unlabeled ds26 (G4 and ds26 concentrations: 0.5 µM and 10 µM, respectively); then, the compound under test was added to a final concentration of 10 µM, and the melting experiments were performed as described above.

MST Assay
ODNs labeled with hexachlorofluorescein (HEX) at the 5 -terminus were purchased from Litekh (Russia) (purity > 95%, HPLC).The ODN sequences are presented in Table S2.The pre-annealed 100 nM solution of 5 -HEX-labeled cKit1 ODN in buffer 1, STAT ODN in buffer 1, or VEGF ODN in buffer 2 was mixed 1:1 (v/v) with two-fold dilutions of the compound under test in the corresponding buffer supplemented with 5% DMSO and 0.5% Tween 20.The final concentrations of the compounds being tested ranged from 0 to 240 µM.The mixtures were incubated at room temperature for 10 min prior to measurements.Microscale thermophoresis (MST) curves were recorded using a Monolith NT.115 instrument and standard capillaries (NanoTemper, Germany) at 22 • C in GREEN mode.To calculate K d values, the dependence of MST data on the concentration of the compound under test was analyzed using MO.Affinity Analysis software (NanoTemper, Germany).

Dual Luciferase Reporter (DLR) Assay
The HEK293 cells were cultured in a 25-cm 2 flask until 70-80% confluency was achieved.Once the required density was reached, the cells were transfected with the pC-Kit1 plasmid (Addgene plasmid #118983) using the GenJect-40 reagent (Molecta).After transfection for 4 h, the cells were plated into a 96-well plate at a seeding density of 4 × 10 4 cells/well.Immediately after seeding, various concentrations of 5b (5, 20, and 50 µM) were added to the cells.Luciferase activity was measured 24 h after the treatment using the Dual-Glo ® Luciferase Assay System (Promega, Madison, WI, USA).All assays were performed in triplicate.

Fluorescent Intercalator Displacement (FID) Assay
An Ethidium bromide (EtBr) displacement assay was performed in a 40 µL volume in 384-well black, flat-bottom, polystyrene microtiter plates.Solutions of EtBr in Tris buffer (0.05 M Tris•Cl, pH 7.5, 0.1 M KCl) at a concentration of 6 µM, 0.016 g/L calf thymus DNA that corresponds to 24 µM of nucleotide pairs and the compound with final concentrations of 4, 16, 62.5, and 250 µM dissolved in DMSO (final concentration up to 3%) were prepared.The compound and blank solutions contained the same percentage of DMSO.Once the ethidium bromide solution had been added, the plates were protected from light.The mixtures were incubated for 15 min, and EtBr fluorescence (545 nm for excitation and 595 nm for emission) was measured using Victor X5 (PerkinElmer).

Synthesis of the Compounds
To construct tri-and tetracyclic heteroaromatic systems, we used a reported reaction of 2 -deoxycytidine with 1,4-benzoquinone in sodium acetate buffer (pH 4.5) [19], but started from N1-substituted cytosine 1 (Scheme 1).The treatment of 1 with 1,4benzoquinone afforded tricyclic derivative 2a.For the preparation of tetracyclic derivative 2b, 1,4-naphthoquinone was used with an addition of DNA that corresponds to 24 µM of nucleotide pairs and the compound with final concentrations of 4, 16, 62.5, and 250 µM dissolved in DMSO (final concentration up to 3%) were prepared.The compound and blank solutions contained the same percentage of DMSO.
Once the ethidium bromide solution had been added, the plates were protected from light.The mixtures were incubated for 15 min, and EtBr fluorescence (545 nm for excitation and 595 nm for emission) was measured using Victor X5 (PerkinElmer).

Cytotoxicity Assays
The cytotoxicity of benzo-and naphtho-imidazopyrimidinyl derivatives and their intermediates was evaluated via an MTT assay [21], using human breast cancer cell line MCF7 (fast-growth subclone), human lung epithelial carcinoma cell line A549, and noncancerous lung fibroblast cell line VA13 (Table 1).The cytotoxicity for immortalized human embryonic kidney cell line HEK293T as control non-cancerous fast-growth cells was also studied.Most of the compounds were non-or weakly cytotoxic both for cancerous and control cell lines.The introduction of an additional benzene ring into the heteroaromatic scaffold appears to determine the activity throughout the naphtho series b.However, only 5b with two amino-bearing tethers demonstrated significant cytotoxicity, with the selectivity index (SI) being 3.0 and 17.3 for the VA13/MCF7 and VA13/A549 pairs, respectively (Table 1, Figure 2A).Its congeners carrying dimethylamino and guanidino moieties exhibited weak activity.For some heterocyclic scaffolds, the lower activity of dimethylaminoand guanidino-substituted compounds compared to derivatives with amino groups has been reported [27][28][29].Of note, 5b outperformed commonly used chemotherapy drugs, such as cisplatin, doxorubicin, and 5F-uracil [30,31], in terms of selectivity.
Table 1.Anti-proliferative effects of benzo-and naphtho-imidazopyrimidinone derivatives against HEK293T, MCF7 , A549, and VA13 cell lines, and selectivity indexes after 72 h incubation and viability measurement using the Mosmann assay (MTT).VA13 and HEK293T are slow-and fastgrowth non-cancerous cell lines, respectively; MCF7 and A549 cell lines are of breast and lung origin, respectively.The treatment with 5b did not cause complete cell death even at high concentrations (Figure 2A), and apoptosis induction is not the main mechanism of action of the compound (Table S3, Figures S1 and S2).We checked if 5b had cytostatic activity and found that it caused G2 arrest in A549 cells (Table 2, Figure 2B).This effect was significant, though less prominent than after the treatment of cells with the well-established cytostatic drug paclitaxel with G2-arresting activity [36,37].

Verification of DNA Targets: FRET-Melting and Microscale Thermophoresis (MST) Assays
In an attempt to elucidate the mechanism of action underlying the anti-proliferative properties, we investigated the ability of the most active tetracyclic derivatives to interact with a panel of biologically relevant DNA secondary structures.The non-active tricyclic analogs were used as negative controls.Compound 5b contains a planar condensed aromatic core and positively charged tethers, allowing for interactions with the DNA duplex leading to DNA damage.In addition, carbazole-based derivatives can stabilize G4s that regulate oncogene transcription and inhibit telomere elongation [38].Telomeric and c-Myc G4s have previously been reported as the main targets of carbazole-based G4 ligands [3].

Verification of DNA Targets: FRET-Melting and Microscale Thermophoresis (MST) Assays
In an attempt to elucidate the mechanism of action underlying the anti-proliferative properties, we investigated the ability of the most active tetracyclic derivatives to interact with a panel of biologically relevant DNA secondary structures.The non-active tricyclic analogs were used as negative controls.Compound 5b contains a planar condensed aromatic core and positively charged tethers, allowing for interactions with the DNA duplex leading to DNA damage.In addition, carbazole-based derivatives can stabilize G4s that regulate oncogene transcription and inhibit telomere elongation [38].Telomeric and c-Myc G4s have previously been reported as the main targets of carbazole-based G4 ligands [3].
Here, we used a FRET-melting assay with FAM/BHQ-labeled ODNs to evaluate the stabilizing effects of the synthesized compounds on the DNA duplex and G4s.The 23mer hairpin Hair was chosen as a model duplex [39].In addition to telomeric 22AG [40] and c-Myc [41] G4s, three G4s from the promoters of oncogenes cKit1 [42], STAT [43], and VEGF [44], as well as three imperfect G4s (Ct1, BclT, and 22CTA) [45,46], were also included.To verify a topological preference, if any, the G4 set encompassed various topologies: antiparallel (22CTA and STAT), parallel (cMyc, cKit1, and VEGF), hybrid (22AG) and mixed hybrid/antiparallel (Ct1 and BclT) G4s.Furthermore, three of them, namely, cKit1, STAT, and VEGF, contained long loops that can be involved in the interactions with G4 stabilizers.In order to determine the effective concentration range and roughly evaluate the binding affinity of the compounds to G4 targets, the concentration dependence of the stabilizing effects was analyzed in a series of titration assays (Tables S4 and S5).
The G4/hairpin-stabilizing effects of all compounds are summarized in Figure 3.Both tricyclic and tetracyclic compounds increased the T m of G4s, but the stabilization of the hairpin was observed only in the naphtho series b.The compounds demonstrated no apparent selectivity for particular types of G4 topologies.The smallest stabilizing effect was observed for c-Myc G4, which has the highest intrinsic thermal stability (the highest T m value in the absence of the ligand), in all cases except for the case of 6b.Weak to moderate stabilizing effects were predominantly observed for telomeric (22AG) and imperfect (BclT, 22CTA, and Ct1) G4s, except for the highly efficient stabilizers 5b and 7b.Both tri-and tetracyclic derivatives gave the highest increase in melting temperature for cKit1, STAT, and VEGF G4s.Taking into account that the last three G4s have the longest loops, the efficacy of G4 stabilization could depend on loop-ligand interactions.
The introduction of an additional benzene ring into the benzoimidazopyrimidinone scaffold had a positive impact on the stabilization of the G4s and the hairpin, presumably due to additional stacking contacts with outer G tetrads or canonical Watson-Crick base pairs.Mono-substituted compounds were the weakest stabilizers in both series, suggesting that the number of positively charged terminal groups influences stabilization efficacy, presumably due to electrostatic interactions with the sugar-phosphate backbone.The superiority of the guanidylated derivatives can be explained by the high pKa value of the guanidino group [47].The lower efficacy of N-dimethylated derivatives can result from charge shielding and steric hindrance [27].
In summary, the tetracyclic molecules exhibited a more pronounced stabilization of both G4s and duplex DNA, that is mostly consistent with their anti-proliferative activity.In contrast, the tricyclic derivatives that were inactive in MTT assays demonstrated lower G4-stabilizing effects and no impact on duplex thermal stability.Tricyclic derivative 3a with a positively charged amino-containing tether demonstrated no stabilizing effect.Among the di-substituted tricyclic derivatives, guanidylated 7a exhibited the highest stabilizing effect, while 6a with dimethylamino-bearing arms showed the lowest one.A similar dependence of the effect on the type and number of tethered groups (3b < 5b < 6b < 7b in most cases) was observed among the tetracyclic compounds.
The introduction of an additional benzene ring into the benzoimidazopyrimidinone scaffold had a positive impact on the stabilization of the G4s and the hairpin, presumably due to additional stacking contacts with outer G tetrads or canonical Watson-Crick base pairs.Mono-substituted compounds were the weakest stabilizers in both series, suggesting that the number of positively charged terminal groups influences stabilization efficacy, presumably due to electrostatic interactions with the sugar-phosphate backbone.The superiority of the guanidylated derivatives can be explained by the high pK a value of the guanidino group [47].The lower efficacy of N-dimethylated derivatives can result from charge shielding and steric hindrance [27].
In summary, the tetracyclic molecules exhibited a more pronounced stabilization of both G4s and duplex DNA, that is mostly consistent with their anti-proliferative activity.In contrast, the tricyclic derivatives that were inactive in MTT assays demonstrated lower G4-stabilizing effects and no impact on duplex thermal stability.
To assess G4 vs. duplex selectivity, we chose five G4 targets that were stabilized most effectively in the previous assay, and performed FRET melting in the presence of derivatives and a competitor-unlabeled hairpin ds26 at a 20-fold excess relative to G4 (Figure 4).The mono-substituted derivative 3a was excluded due to the lack of effect.For all investigated compounds, except for the mono-substituted tetracyclic compound 3b, the melting point of VEGF decreased significantly in the presence of ds26.Regarding tricyclic molecules, the stabilizing effect on 22AG, 22CTA, STAT, and cKit1 G4s remained almost unchanged, with the derivatives demonstrating a preference for these G4 targets over duplex DNA.The stabilizing effects of di-substituted tetracyclic compounds mainly decreased in the presence of the competitor.Surprisingly, a high stabilizing effect was observed for STAT in the presence of ds26, which might be explained by the formation of a triple G4/ds26/derivative complex.cKit1 G4 can be highlighted as a preferred target for tetracyclic derivatives, but, upon entering the cells, they can target both telomeric G4 and duplex DNA due to their high abundance.
vestigated compounds, except for the mono-substituted tetracyclic compound 3b, the melting point of VEGF decreased significantly in the presence of ds26.Regarding tricyclic molecules, the stabilizing effect on 22AG, 22CTA, STAT, and cKit1 G4s remained almost unchanged, with the derivatives demonstrating a preference for these G4 targets over duplex DNA.The stabilizing effects of di-substituted tetracyclic compounds mainly decreased in the presence of the competitor.Surprisingly, a high stabilizing effect was observed for STAT in the presence of ds26, which might be explained by the formation of a triple G4/ds26/derivative complex.cKit1 G4 can be highlighted as a preferred target for tetracyclic derivatives, but, upon entering the cells, they can target both telomeric G4 and duplex DNA due to their high abundance.For the most promising compounds, the dissociation constants (Kd) of their complexes with proto-oncogenic G4s STAT, VEGF, and cKit1 were determined using microscale thermophoresis (MST) assays with 5′-HEX-labeled G4 ODNs.The results are presented in Figure 5.The Kd values of 3b/cKit, 5b/cKit, 3b/STAT, and 6a/STAT were in the submicromolar range, and the rest of the complexes tested showed micromolar Kd.Despite causing only a moderate increase in G4 Tm, 3b turned out to be the most efficient G4 binder.In order to assess the impact of 5b on cKit expression, a DLR assay with the pC-Kit1 plasmid (Addgene plasmid #118983) that contains two luciferase genes (Renilla/Firefly) was performed.The expression of Renilla is controlled by the c-Kit promoter containing the cKit1 G4-forming sequence, while the expression of Firefly is under control by a non-G4-forming promoter sequence (the HSV TK promoter).After transfection of the plasmid into HEK293T cells for 4 h, followed by a 24 h treatment with 5b, no significant changes in Renilla/Firefly expression ratio compared with the control were observed (Figure S3), suggesting that G4 in the cKit promoter region is not the main target of 5b.For the most promising compounds, the dissociation constants (K d ) of their complexes with proto-oncogenic G4s STAT, VEGF, and cKit1 were determined using microscale thermophoresis (MST) assays with 5 -HEX-labeled G4 ODNs.The results are presented in Figure 5.The K d values of 3b/cKit, 5b/cKit, 3b/STAT, and 6a/STAT were in the submicromolar range, and the rest of the complexes tested showed micromolar K d .Despite causing only a moderate increase in G4 T m , 3b turned out to be the most efficient G4 binder.
unchanged, with the derivatives demonstrating a preference for these G4 targets over duplex DNA.The stabilizing effects of di-substituted tetracyclic compounds mainly decreased in the presence of the competitor.Surprisingly, a high stabilizing effect was observed for STAT in the presence of ds26, which might be explained by the formation of a triple G4/ds26/derivative complex.cKit1 G4 can be highlighted as a preferred target for tetracyclic derivatives, but, upon entering the cells, they can target both telomeric G4 and duplex DNA due to their high abundance.For the most promising compounds, the dissociation constants (Kd) of their complexes with proto-oncogenic G4s STAT, VEGF, and cKit1 were determined using microscale thermophoresis (MST) assays with 5′-HEX-labeled G4 ODNs.The results are presented in Figure 5.The Kd values of 3b/cKit, 5b/cKit, 3b/STAT, and 6a/STAT were in the submicromolar range, and the rest of the complexes tested showed micromolar Kd.Despite causing only a moderate increase in G4 Tm, 3b turned out to be the most efficient G4 binder.In order to assess the impact of 5b on cKit expression, a DLR assay with the pC-Kit1 plasmid (Addgene plasmid #118983) that contains two luciferase genes (Renilla/Firefly) was performed.The expression of Renilla is controlled by the c-Kit promoter containing the cKit1 G4-forming sequence, while the expression of Firefly is under control by a non-G4-forming promoter sequence (the HSV TK promoter).After transfection of the plasmid into HEK293T cells for 4 h, followed by a 24 h treatment with 5b, no significant changes in Renilla/Firefly expression ratio compared with the control were observed (Figure S3), suggesting that G4 in the cKit promoter region is not the main target of 5b.In order to assess the impact of 5b on cKit expression, a DLR assay with the pC-Kit1 plasmid (Addgene plasmid #118983) that contains two luciferase genes (Renilla/Firefly) was performed.The expression of Renilla is controlled by the c-Kit promoter containing the cKit1 G4-forming sequence, while the expression of Firefly is under control by a non-G4-forming promoter sequence (the HSV TK promoter).After transfection of the plasmid into HEK293T cells for 4 h, followed by a 24 h treatment with 5b, no significant changes in Renilla/Firefly expression ratio compared with the control were observed (Figure S3), suggesting that G4 in the cKit promoter region is not the main target of 5b.
According to the FRET-melting assay, there is evidence for interactions of 5b with DNA structures.The FID assay demonstrated that 5b is a DNA intercalator, though an order of magnitude weaker one than EtBr (Figure S4).
The problem with DNA-targeting compounds is adverse effects and common toxicity in combination with a low selectivity of action.The aims of this study were to evaluate the prospects for further study of this class of derivatives; and to determine whether it is necessary to synthesize additional compounds of this class, study their detailed molecular mechanism, and conduct comprehensive studies in animal models.To evaluate the

Figure 2 .
Figure 2. (A) Dose-response dependencies of cytotoxicity for HEK293T, MCF7′, A549, and VA13 cell lines after a 72 h treatment with 5b.(B) Changes in cell cycle distribution of lung cancer cells A549 after a 20 h treatment with 5b, paclitaxel, and 92504.

Figure 2 .
Figure 2. (A) Dose-response dependencies of cytotoxicity for HEK293T, MCF7 , A549, and VA13 cell lines after a 72 h treatment with 5b.(B) Changes in cell cycle distribution of lung cancer cells A549 after a 20 h treatment with 5b, paclitaxel, and 92504.

Figure 3 .
Figure3.Heatmap of the influence of the compounds on the thermal stability of DNA targets.The concentrations of the compounds and the targets were 1 and 20 µM, respectively.For all targets except VEGF, buffer 1 (20 mM sodium phosphate, pH 7.4, 10 mM KCl) was used.For VEGF, buffer 2 (5 mM sodium phosphate, pH 7.4, 25 mM LiCl) was used.

Figure 4 .
Figure 4. Heatmap of changes in the G4-stabilizing effects of the compounds in the presence of duplex DNA.ND-not determined.The concentrations of the compounds, ds26, and the targets were 10 µM, 10 µM, and 0.5 µM, respectively.Buffer 1 (20 mM sodium phosphate, pH 7.4, 10 mM KCl) was used for all targets, and buffer 2 (5 mM sodium phosphate, pH 7.4, 25 mM LiCl) was used for VEGF.

Figure 5 .
Figure 5. Heatmap of the dissociation constants Kd of G4:derivative complexes according to MST.ND-not determined.

Figure 4 .
Figure 4. Heatmap of changes in the G4-stabilizing effects of the compounds in the presence of duplex DNA.ND-not determined.The concentrations of the compounds, ds26, and the targets were 10 µM, 10 µM, and 0.5 µM, respectively.1 (20 mM sodium phosphate, pH 7.4, 10 mM KCl) was used for all targets, and buffer 2 (5 mM sodium phosphate, pH 7.4, 25 mM LiCl) was used for VEGF..

Figure 4 .
Figure 4. Heatmap of changes in the G4-stabilizing effects of the compounds in the presence of duplex DNA.ND-not determined.The concentrations of the compounds, ds26, and the targets were 10 µM, 10 µM, and 0.5 µM, respectively.Buffer 1 (20 mM sodium phosphate, pH 7.4, 10 mM KCl) was used for all targets, and buffer 2 (5 mM sodium phosphate, pH 7.4, 25 mM LiCl) was used for VEGF.

Figure 5 .
Figure 5. Heatmap of the dissociation constants Kd of G4:derivative complexes according to MST.ND-not determined.

Figure 5 .
Figure 5. Heatmap of the dissociation constants K d of G4:derivative complexes according to MST.ND-not determined.

Table 2 .
Cell cycle effects of 5b and G1 and G2 phase-arresting drugs 92504 and paclitaxel, respectively.
Heatmap of the influence of the compounds on the thermal stability of DNA targets.The concentrations of the compounds and the targets were 1 and 20 µM, respectively.For all targets except VEGF, buffer 1 (20 mM sodium phosphate, pH 7.4, 10 mM KCl) was used.For VEGF, buffer 2 (5 mM sodium phosphate, pH 7.4, 25 mM LiCl) was used.