Six novel complexes based on 5-Acetoxy-1-(6-chloro-pyridin-2-yl)-1H-pyrazole-3-carboxylic acid methyl ester derivatives: Syntheses, crystal structures, and anti-cancer activity

2.1 Synthesis

The reaction between compound a and Cu(NO₃)₂·3H₂O was conducted by the solvothermal method in methanol/acetonitrile (1:1) in order to synthesize new pyridyl-pyrazol-3-one derivatives (Scheme 1). As a result, the dinuclear copper complexes [Cu₂(L¹)₂]·(CH₃CN)₂ (2) was obtained. Herein, HL¹, which has two chiral centres, was synthesized in-situ reaction. Maybe copper ions play a catalytic role in the pyridyl-prazol-3-one derivatives synthesis system.

Close

Scheme 1

Synthesis routes for 1–6.

Export to PPT

To study the role of the copper ions in the reaction, Ni²⁺ and Co²⁺ ions were used to replace Cu²⁺ ions in the reaction. However, these substitutions led only to the production of unknown white precipitates which have not been characterized because the white precipitates were mixture and were impossible characterized.

Using the same in-situ synthesis conditions but without copper ion, compound 1 was produced in high purity (Scheme 1). The possible formation mechanism of compound 1 is shown in Scheme S1. Furthermore, compound 2 was produced in progressively lower yields when the molar ratio of Cu(NO₃)₂·3H₂O to a was decreased stepwise from 1:1 to 1:9. Compound 1 was also formed when the molar ratio was decreased to 1:10. The results indicated that copper ions play a critical role in the formation of HL¹. It must be noted that the divalent copper was reduced to the monovalent copper during the reaction.

Using the same in-situ synthesis conditions but the mixed solution methanol/acetonitrile (1:1) was replaced by the mixed solution ethanol/acetonitrile (1:1), compound 3 was produced in high purity (Scheme 1). Herein, we found that there exists transesterification between HL¹ and ethanol and formed two new ligands HL² and HL³. Furthermore, 3,4,6-tricarboxylic acid trimethyl ester of HL¹ changed 3,4,6-tricarboxylic acid 4,6-diethyl ester 3-methyl ester of HL² and 3,4,6-tricarboxylic acid 4-ethyl ester 3,6-dimethyl ester of HL³. we decreased reaction time, complex 6 was prepared while the degree of ester group exchange was significantly reduced. This result indicates that the lipid exchange reaction is related to the reaction time. In order to understand transesterification, we used ethanol and n-propanol mixed dissolution (1:1) and n-propanol and n-butanol (1:1) mixed solvents instead of the methanol in the mixed solvent of methanol and acetonitrile (1:1). Complexes 4 and 5 were prepared, respectively. The results indicated that the transesterification rate is basically the same for ethanol, n-propanol and n-butanol. And transesteration occurs mainly in the 4th position of the HL¹ ligand.

2.2 Crystal structures

2.2.1

2.2.1 Crystal structure of 1

The structure of 1 is shown in Fig. 1. Single-crystal X-ray diffraction analysis revealed that 1 crystallises in the triclinic with space group P $\overline{1}$ . All bond lengths in the pyrazole rings (II and III) show partial double-bond character, which indicates the delocalization of the π-electronic system throughout the pyrazole rings (Huang et al., 1813). The bond angles and bond lengths are within the normal ranges for a pyrazole ring (Table S1), and are similar to those reported for 5-(acetyloxy)-1-(6-bromo-2-pyridinyl)-1H-pyrazole-3-carboxylate (Huang et al., 1813) and the literature values (Allen et al., 1991). The C25-O9 and C15-O6 bond lengths are 1.336(5) and 1.330(4) Å, respectively. The values of the bond lengths are similar to that reported for a pyrazole-OH bond (1.329 Å) (Huang et al., 1813; Ciolkowski et al., 2009). However, C9 is sp³ hybridised in ring I. The C9-C11, C9-C10, C9-C21, and C9-C6 bond lengths are 1.510(4), 1.533(5), 1.528(5), and 1.518(5) Å, respectively, which are consistent with a single bond of a sp³-sp² hybrid orbital (the average length of a C-C sp³-sp² single bond is 1.51 Å (Allen et al., 1991). The C10-O3 bond length is 1.195(4) Å, which clearly indicates a carbonyl group (the average length of a C = O double bond is 1.20 Å (Carey et al., 2007), while the C6-N3 bond length is 1.269(5) Å, which clearly indicates a C = N double bond (Xiao et al., 2015; Xiao et al., 2015). Thus, there are no protons on the N3 and O3 atoms. Of cause, an electron density on the fourier maps were not found around the N3 and O3 atoms. The rings position in the same plane of pyridyl-pyrazole (II and V rings) with the substituents (O6 and ester carbonyl group) at II and Cl2 at V ring, which is the same as the reported compound 5-(acetyloxy)-1-(6-bromo-2-pyridinyl)-1H- pyrazole-3-carboxylate (Huang et al., 1813). The maximum deviation of the least-squares plane (10.586 x  + 2.666 y + 8.981 z = 14.1672) is from + 0.0968 Å (C14) to − 0.1068 Å (O4). It should be noted that ring I bearing the substituents (C7, C8, O1, O2 of the ester carbonyl group and O3) is co-planar (−1.240 x - 3.991 y + 13.769 z = 5.450). The dihedral angle between rings I and IV is 67.4°. Compound 1 forms a dimer through weak intermolecular Cl···Cl interactions (Cl2···Cl3a, 3.673 Å, Cl3···Cl2a, 3.673 Å, symmetry code: (a) - x, 2 - y, 2 - z, Fig. S1), which form a two-dimensional network through Cl···O interactions (Cl1···O7b, 3.569 Å, symmetry code: (b) × , y − 1, z; Fig. S2). It must be noted that the three methyl of the ester base groups in 1 is disordered over two sets of sites in a 0.36:0.64 ratio.

Close

Fig. 1

Molecular structure of 1. Displacement ellipsoids are drawn at 30% probability level. Part H atoms are omitted for clarity.

Export to PPT

2.2.2

2.2.2 Crystal structures of 2–6

Complexes 2–6 (Figs. 2-6) are also dinuclear copper complexes with the different crystal system and the different space group (2,4,5 belong to triclinic crystal system with P $\overline{1}$ space group but 3 and 6 do monoclinic crystal system with P2₁/c space group) and the different ester group substituents of the HL ligands. Therefore only complex 2 is analyzed here.

Close

Fig. 2

Molecular structure of 2 (omitted acetonitrile molecules). Displacement ellipsoids are drawn at 30% probability level. All H atoms are omitted for clarity.

Export to PPT

Close

Fig. 3

Molecular structure of 3. Displacement ellipsoids are drawn at 30% probability level. H atoms and solvent molecules are omitted for clarity.

Export to PPT

Close

Fig. 4

Molecular structure of 4. Displacement ellipsoids are drawn at 30% probability level. H atoms and solvent molecules are omitted for clarity.

Export to PPT

Close

Fig. 5

Molecular structure of 5. Displacement ellipsoids are drawn at 30% probability level. H atoms are omitted for clarity.

Export to PPT

Close

Fig. 6

Molecular structure of 6. Displacement ellipsoids are drawn at 30% probability level. H atoms are omitted for clarity.

Export to PPT

Single-crystal X-ray diffraction analysis revealed that complex 2 belongs to the triclinic with space group P $\overline{1}$ (Fig. 2). The dinuclear complex 2 is constructed by two (L¹)^- ligands, two Cu^I ions, and two acetonitrile molecules. In 2, the Cu1 atom is coordinated by four nitrogen atoms from two different L¹ ligands, forming a distorted tetrahedral geometry. The Cu1-N1, Cu1-N3, Cu1-N16, and Cu1-N18 distances are 2.045(4), 1.989(4), 2.011(4), and 2.066(4) Å, respectively (Table S2). Cu2 is also tetra-coordinated by four nitrogen atoms from two different L¹ ligands, forming a distorted tetrahedral geometry. The Cu2-N12, Cu2-N7, Cu2-N10, and Cu2-N9 distances are 1.996(4), 1.994(4), 2.038(4), and 2.095(4) Å, respectively. Two pyrazole-pyridyl rings of the L¹ ligand link the two Cu atoms to construct a dinuclear complex. Therein, two copper ions and two L¹ ligands form a pore with dimensions of approximately 6.146 × 9.275 Å while the Cu1···Cu2 distance is 6.146 Å and the C12···C43 distance is 9.275 Å. It should be noted that the four ester carbonyl groups control the entrance and exit of small molecules to and from the pore. It is interesting to note that the L¹ ligand presents as a monovalent anion and exhibits a η¹:η¹:η¹:η¹-μ₂ coordination mode. It should also be pointed out that the L¹ ligand is formed in-situ reaction and forms two chiral centers (C11(R), C12(R) or C42(R), C43(R)). However, complex 2 is not a chiral molecule and possesses a symmetry centre because the two chiral L¹ ligands form mesomers. The dinuclear complex 2 forms a one-dimensional chain through intermolecular Cl···Cl and Cl···O halogen bonds (Cl6···Cl6a, 3.330 Å, Cl3···Cl3b, 3.577 Å, O6···Cl5a, 3.139 Å, symmetry codes: (a) -x, 1-y, -z; (b) -x, 2-y, 1-z, Fig. S3).

Additionally, the atomic valences for two copper ions of 2–6 can be determined by the bond valence model (Brown and Altermatt, 1985; Brese and O'keeffe, 1991; Brown, 2009). According to this model, the sum of all the bond valences around any ion is equal to its ionic charge or valence. Herein, bond valences (s) are calculated as s = exp[r₀-r]/B; B = 0.37, r₀ = 1.61 for Cu(II)-N pairs from the reference (Brown and Altermatt, 1985; Brese and O'keeffe, 1991). The calculated results are listed in Table S2. Evidently, the calculated values of bond valence sum are in good agreement with the values of expected atomic valence, that is to say, the two copper cations of 2–6 have a valence state of + 1 (Table S3).

For 1–6, the end group is disordered. For example, the three methyl of the ester base groups in 1 is disordered over two sets of sites in a 0.36:0.64 ratio. The methyl (C48, C48′ C46, C46′) of 2, ethyl (C16-C17, C16′-C17′, C48-C49, C48′-C49′, C60, C60′-C61′) of 3 and C54-C55, C54′-C55′) of 4, n-propyl (C21-C22-C23, C21′-C22′-C23′) of 4 and C13-C14-C15, C13′-C14′-C15′) of 5, n-butyl (C46-C47-C48-C49, C46′-C47′-C48′-C49′) of 5 in the ester base groups are disordered over two sets in 0.49:0.51, 0.37:0.63, 0.43:0.57, 0.43:0.57, 0.29:0.71, 0.29:0.71 ratio, respectively. In compound 6, there were methyl ester and ethyl ester in the same position, and their occupations were 0.68 and 0.32, respectively.

2.3 Qutantum calculation for 1

In order to further characterize the energies and conformations of the OH and NH groups in 1, density functional theory analysis was performed (Fig. S4). The geometric optimization was performed at the b3lyp/cc-pVDZ level of theory, followed by single-point of energy calculations at the MP2/cc-pVTZ level. All the calculations were performed using ORCA 2.9.1(Neese, 2012) together with the RIJCOSX approximation(Kossmann and Neese, 2010). According to the results; the conformation of OH for 1 is more stable than that of NH for 1 by approximately 8.4 kcal/mol. Closer study revealed that an O-H···O intramolecular hydrogen bond in OH contributes to the stabilization of the molecule. The short H···O distance (1.653 Å) and the almost linear configuration of the O-H···O moiety confirms the formation of the hydrogen bond. It must be noted that there does not exist O-H···O hydrogen bond in 1 of the crystal structure.

2.4 In vitro anti-tumour activity assays

The cytotoxic activity of 1–4, Cu(NO₃)₂·3H₂O, cisplatin complexes was examined in HepG2, Capan-2, A549, Hela, NP69, MGC803 and HUVECs cell lines. Results of MTT assay were presented as IC₅₀ (μM) values for each compound after 48 h of treatment (Table 2).

Compounds 1 and 2 were incubated with the cell lines at concentrations of 4, 8, 16, 32, and 64 μM, and compounds 3 and 4 was incubated with the cell lines at concentrations of 1, 2, 4, 8, 16 μM for 48 h under the same experimental conditions. It was found that compounds 1–4 exhibit different activities towards normal cells and cancer cells (Figs. S5-S8). With the increase in concentration, the inhibitory rate of compounds 1–4 on all selected cell lines increased, except for compound 1 on NP69.

The IC₅₀ values are shown in Table 2 to further illustrate the results. Therefore, compound 2 showed greater selective cytotoxicity towards Hep-G2,Capan-2, Hela cancer cells after 48 h treatment, in comparison with the standard drug cisplatin (Qin et al., 2018; Qin et al., 2018; Qin et al., 2015; Zhang et al., 2019). The IC₅₀ values of compounds 3 and 4 are lower than that of cisplatin. The results indicate that Compounds 3 and 4 have broad anti-cancer activity. However, The cytotoxic effect of 4 on normal HUVEC cells is significantly reduced and is lower than the cytotoxicity of cisplatin on normal HUVEC cells. These results indicate that compound 2–4 may be a good substitute for platinum-based drugs as a potential antitumor drugs (Psomas and Kessissoglou, 2013; Brown et al., 1980).

2.5 Cell apoptosis assay

Compound 2 induces significantly more apoptosis in HeLa cells compared with 1 (Fig. 7). After incubated with 2 (5 μg/mL, 10 μg/mL,15 μg/mL) and 1 (20 μg/mL) for 24 h, the Hela cancer cells were stained by Annexin V-FITC (FITC is fluorescein isothiocyanate) and PI (propidium iodide) and examined by flow cytometry. As shown in Fig. 3, the percentage of 2 (15 μg/mL) and 1 (20 μg/mL), promoting cell apoptosis were 69.1% and 3.5%, respectively. These results also suggested that 2 (15 μg/mL) was more effective in promoting cell apoptosis than 1 (20 μg/mL). As the concentration of compound 2 increased from 5 μg/mL to 15 μg/mL, the apoptosis rate of Hela cell lines increased from 7.3% to 69.1%. These results indicate that apoptosis in HeLa cell lines by 2 presents a dose–response relationship.

Close

Fig. 7

Induced apoptosis in HeLa by 1 and 2 for 24 h.

Export to PPT

2.6 Morphology assay

The cell apoptosis in NP69, A549, Capan-2, Hela, HepG2, and HUVECs cell lines induced by compound 2 was further affirmed by cellular morphology observations (Fig. 8). After the NP69, A549, Capan-2, Hela, HepG2, and HUVECs cells were incubated with compound 2 at a concentrations gradient of 8–16 μg/mL for 24 h, the changes in cell morphology were observed with an inverted fluorescent microscope (10 × ). As shown in Fig. 4, most of the NP69, A549, Capan-2, Hela, HepG2, and HUVECs cells in the control group present a normal outline. However, those cell lines exposed to compound 2 shows a clear change in cell morphology. The densities of the tumour cells treated with 2 are lower than those of the control group, indicating that 2 plays a key role in inhibiting cell proliferation. In addition, with an increase in the concentration of 2, morphological changes in the cells are observed with the cells beginning to become round and some of the cells losing cohesion (some even become suspensions), confirming that 2 induces apoptosis in the tested tumour cells. At the same concentration of 2, the degree of apoptosis in Hela cells was greater than that in the other cell lines. However, for the normal NP69 line (Fig. 4a), the shape of the cells in the treatment and control groups are the same, indicating that the cytotoxicity of 2 towards the normal cell NP69 is very low. Thus, these results indicate that 2 can induce apoptosis in Capan-2, HeLa, HePG2, HUVECs, and A549 cells in a dose-dependent manner. The results are consistent with those of the MTT assays above.

Close

Fig. 8

Morphology observation of six selected cells treated by 8, 16 μg/mL of 2 for 24 h.

Export to PPT

2.7 IR spectrum

The IR spectral data of compounds 1–6 are shown in Fig. S16-S21. All the characteristic groups of a series of compounds exhibited obvious absorption peaks as shown in the IR spectrum. The compounds 1–6 showed a br absorption band at 3414–3450 cm⁻¹ (O − H), because of the stretching vibration adsorption peak of crystal solution molecules in the air or in the complexes (Liang et al., 1994). Bands at 3103, 3133, 3124, 3115, 3133 cm⁻¹ are attributed to the stretching vibration adsorption peak of the group = C-H for compounds 1–4, 6, respectively (Baryshnikov et al., 2013). The bands at 1736–1728 cm⁻¹ were attributed to the ν(C = O) of the ester group on the compounds 1–6. The bands at 1632–1628 cm⁻¹ were attributed to the ν(C = N) of the compounds 1–6 while the bands at 1588–1581 cm⁻¹ were attributed to the skeleton oscillation of pyridine rings of the compounds 1–6. The bands at 798–786 cm⁻¹ were attributed to characteristic absorption peaks of 1,6-disubstituted pyridine in the compounds 1–6 while The bands at 1439–1428 cm⁻¹ were attributed to the in-plane bending vibration of δ_as(C = N) in the compounds 1–6 (Gusev et al., 2017). The bands at 1235, 1242, 1222, 1228, 1225, 1226 cm⁻¹ were attributed to the stretching vibration adsorption peak of the group = C-O- for 1–6, respectively while the bands at 1145, 1123, 1123, 1127, 1128, 1126 cm⁻¹ were attributed to the stretching vibration adsorption peak of the group -C-O- for 1–6, respectively.

3.1 Materials and instrumentation

All chemicals were commercially available and used as received without further purification. methanol (Xilong Chemical Co., Ltd), ethanol (Xilong Chemical Co., Ltd), acetonitrile (Xilong Chemical Co., Ltd), Cu(NO₃)₂·3H₂O (Xilong Chemical Co., Ltd), n-propanol(Xilong Chemical Co., Ltd), n-butanol (Xilong Chemical Co., Ltd), a (Lab homemade). Elemental analyses (CHN) were performed using an Elemental Vario-EL CHN elemental analyzer. FT-IR spectra were recorded from KBr pellets in the range of 4000–400 cm⁻¹ on a Bio-Rad FTS-7 spectrophotometer. The nuclear magnetic resonance spectra of ¹H and ¹³C were recorded on a Bruker AV500 spectrometer using tetramethylsilane (TMS) as an internal standard and CDCl₃ was used as a solvent. On a Bruker HTC aerosol mass spectrometer records an ESI-MS spectrum. Using the Agilent G8910A CCD diffractometer and SHELXS-18, SHELXL-18 software for X-ray crystal structure was determined for structure resolutions and refinements, respectively. Seven human tumor cell lines, HepG2, Capan-2, A549, Hela, MGC803, HUVECs and one normal liver cell lines NP69 were all obtained from the Chinese Academy of Sciences Shanghai Cellular Bank. The geometries optimization energy calculations were performed by ORCA 2.9.1 program at MP2/cc-pVTZ level.

3.2 Synthesis

5-acetoxy-1-(6-chloropyridin-2-yl)-1H-pyrazole-3-carboxylic acid methyl ester (a) was synthesized according to the procedure reported in literature (Huang et al., 1813).

3.3 In vitro cytotoxicity assay

Cell lines: HepG2, Capan-2, A549, HeLa, NP69, HUVECs, MGC803 were from the Chinese Academy of Sciences Shanghai Cellular Bank. Cell lines were germinated in RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum, 100 U/mL penicillin, 2 mM glutamine, and 100

U/mL streptomycinin at 37 °C, in a highly humidifying the atmosphere of 95% air and 5% CO₂. 1 – 4, Cu(NO₃)₂·3H₂O and cisplatin are cytotoxic to HepG2, Capan-2, A549, HeLa, NP69, HUVECs, MGC803 cell lines examined by the microculture tetrozolium (MTT) assay (Alley et al., 1988). The experiment was carried out by the reporting procedure (Chen et al., 2013). The inhibition rate of cell growth was calculated with the data from three repeated trials by (OD_control-OD_test)/OD_control × 100%. Compounds 1, 2 were incubated with cell lines for 48 h with gradient of concentration in 4 μM, 8 μM, 16 μM, 32 μM, 64 μM, while Compounds 3, 4 did with cell lines for 48 h with gradient of concentration in 1 μM, 2 μM, 4 μM, 8 μM, 16 μM.

3.4 Cell apoptosis assay

According to the manufacturer’s instructions, apoptosis was detected with an Annexin V/PI kit purchased from BD Pharmingen (BD Bioscience). HeLa cells were implanted in 6-hole plates, and were allowed to grow into 75–85% confluence. HeLa cell was treated with 2 and 1 (10 and 15 µg/mL) for 24 h. To detect early and late apoptosis, floating and adherent cells were harvested and washed twice with 1 × PBS, and were centrifuged at 1000 r.min-1 for 5 min removed the medium, and were resuspension of binding buffer 200 µL at a density of 106 cells per mL. Subsequently, annexin V-FITC (5 µL) were added to the binding buffer for stain cells. At 37 ^OC in the dark, Cells lines were incubated for 10 min, and were centrifuged at 1000 r.min-1 for 5 min for removing the binding buffer, and were resuspension of binding buffer 200 µL again. PI (propidium iodide, 5 µL) were added to the binding buffer for stain cells, and cells lines were analyzed by a FACS Calibur (BD Bioscience).

3.5 Morphological characterization of cell apoptosis

logarithmic phase cells were seeded into 96-hole plates with cell density of 8 × 10³/hole in a moist incubator (37 °C and 5% CO₂, 95% air). Cell morphology observation further confirmed that the cell apoptosis of NP69, A549, Capan-2, HeLa, HePG2, HUVECs were induced by 2. At gradient concentrations of 8, 16 μg/mL, the NP69, A549, Capan-2, HeLa, HePG2, HUVECs cells were treated with 2 for 24 h, then the cells lines were further treated for 48 h at incubator. The morphology of the cells was imaged using an inverted microscope (10 × ).

3.6 X-ray crystallography

Single-crystal X-ray diffraction data for 1–6 were collected by SuperNova, Single source at offset, Eos with graphite monochromated Mo-Kα radiation (λ = 0.71073 Å) at 293 K using the ω scan mode in the ranges 2.89° ≤ θ ≤ 25.01° (1), 2.88° ≤ θ ≤ 25.10° (2), 2.90° ≤ θ ≤ 25.10°(3), 3.01° ≤ θ ≤ 25.01°(4), 2.84° ≤ θ ≤ 25.01°(5), 2.88° ≤ θ ≤ 25.01°(6), respectively. The raw frame data were integrated with the CrysAlisPro program. The structures were solved directly by using SHELXS-97 and the structures were refined by full-matrix least-squares on F² using SHELXL-2018 (Sheldrick, 2015) within the OLEX-2 GUI (Dolomanov et al., 2009). An empirical absorption correction was performed on the program spherical harmonics, implemented in SCALE3 ABSPACK scaling algorithm. Compound 1 had a disordered solvent water molecule; and the disordered water molecule was processed by the squeeze software. All of the non-hydrogen atoms were refined anisotropically. All of the hydrogen atoms were arranged in the geometry and they were refined for riding. For The selected bond distances and angles of the compounds 1 – 6, are provided in Tables S1, S2. The crystallographic data of the compounds 1 – 6 are reported in Table 3.