Synthesis, characterization, and cytotoxic activity of triphenyltin complexes of N-(5-bromosalicylidene)-α-amino acids

Spectroscopic analysis

The infrared spectra of 1–5 show the strong broad band at 3000–3400 cm⁻¹ assigned to ν(OH) of the phenolic O-H group with intramolecular O-H···N=C hydrogen bond and at ~1640 cm⁻¹ and ~1400 cm⁻¹ assigned to the asymmetric stretching vibration, ν_as(COO), and symmetric stretching vibration, ν_s(COO), of the carboxylate, respectively. In organotin carboxylates, IR spectroscopy can provide useful information concerning the coordination mode of the carboxylate group. Generally, the difference between the ν_as(COO) and ν_s(COO) bands, Δν(COO), of bidentate carboxylate group is below 200 cm⁻¹, while unidentate carboxylate is above 200 cm⁻¹ (Deacon and Phillips, 1980). The magnitudes (256–268 cm⁻¹) of Δν(COO) in 1~5 indicate that the carboxylate group has a monodentate coordination to tin in the solid state, which is in agreement with the X-ray structures of 3 and 5 shown below.

In the CDCl₃ solution, a broad peak appearing at ~13.20 ppm is the resonance signal of the phenolic OH proton because the action of the intramolecular O-H···N=C hydrogen bond makes an H shift to the N atom. The signal at ~8.20 ppm is assigned to the azomethine (CH=N) proton. The absence of spin-spin coupling between the azomethine proton and the tin nucleus, ³J(¹¹⁹Sn-¹H), confirmed that the azomethine nitrogen atom is not coordinated to the tin atom in 1–5. The signal of the proton of =NCHR appears in the range of 3.86–4.48 ppm. The signals of the carboxyl carbon (C=O) and imine carbon (C=N) appear at ~176 and ~165 ppm, respectively. The carbon chemical shift value of =NCHR is in the range of 59–78 ppm, depending on the nature of the substituent R.

The ¹¹⁹Sn chemical shifts primarily depend on the coordination number and the nature of the donor atom directly bound to the central tin atom (Davies et al., 2008). The ¹¹⁹Sn chemical shifts of 1–5 occur in the range of δ −95 to −99 ppm, suggesting that the tin atoms in the complexes have four coordinate environments in CDCl₃ solution (Holeček et al., 1983). Thus, in the CDCl₃ solution, phenolic O of the ligand in these complexes is not coordinated to the tin atom of the adjacent molecule, and all the complexes are monomeric with four coordinated tin atoms.

Structure analysis

The molecular structures of 3 and 5 are shown in Figures 1 and 2, and the selected bond lengths and bond angles are listed in Table 1. Complex 3 crystallizes in monoclinic system with space group C2/c and is a zigzag chain polymer associating via a carboxylate oxygen O(1) and a phenolic oxygen O(3) in the ligand with the distance of 7.592(2) Å between two tin atoms (Figure 3). The Sn atoms in this polymeric structure exist in a distorted trans-C₃SnO₂ trigonal bipyramidal environment with a trigonal plane defined by the three phenyl groups. The C-Sn-C angles are in the range of 115.2(3)–123.8(3)°. The axial positions are occupied by the carboxylate oxygen O(1) and phenolic oxygen O(3)^#1 (Symmetry code #1: −x, y−1/2, −z+1/2) of the ligand of an adjacent molecule with the 168.6(2)° of O(1)-Sn(1)-O(3)^#1 angle. The Sn(1)-O(3)^#1 (2.268(5) Å) bond is slightly longer than the Sn(1)-O(1) (2.201(6) Å) bond, so that the Sn atom is only displaced out of the C₃ trigonal plane of the trans-C₃SnO₂ trigonal bipyramidal polyhedron in the direction of the O(1) atom by 0.033(2) Å. The O(1)-Sn(1)-C angles are in the range of 84.6(3)° to 97.8(3)°, and deviated from the ideal 90°. The dihedral angles of three phenyls bound to the Sn atom are 21.96(3)°, 45.73(3)°, and 67.50(3)°, respectively. The ligand is coordinating in the form of a zwitterion, and the carboxylate group is monodentate, which is reflected in the long separation of Sn(1)···O(2) (3.203(6) Å) and the disparity C(19)-O(1) and C(19)-O(2) bond lengths (1.254(10) and 1.211(11) Å).

Figure 1:

The molecular structure of 3.

Symmetry code A: −x, y−1/2, −z+1/2.

Figure 2:

The molecular structure of 5.

Symmetry code A: −x+1/2, y−1/2, −z+1/2.

Figure 3:

The 1D zigzag chain of 3 and 5.

All hydrogen atoms, R group, and partial carbons of phenyl on the Sn atoms are omitted for clarity.

Complex 5 crystallizes in a monoclinic system with the space group C2/c and is very similar to 3 and exhibits the same structural motif of a polymeric chain with 7.797(4) Å of the separation between two tin atoms (Figure 3). The structural features of 3 and 5 are consistent with those of the reported triphenyltin complexes such as 2-HOC₆H₄C(H)=NCH₂COOSnPh₃ (Basu Baul et al., 2002), 2-HOC₆H₄C(H)=NCH(CH₂Ph)COOSnPh₃ (Basu Baul et al., 2007), and 2-HOC₆H₄C(CH₃)=NCH(CH₂CH(CH₃)₂)COOSnPh₃ (Basu Baul et al., 2009).

In vitro cytotoxicity

In order to evaluate the cytotoxicity of the synthesized triphenyltin complexes, we test the activity of 1, 2, 3, and 5 against three human tumor cell lines HeLa, CoLo205, and MCF-7. In the prepared stock solutions, 1, 2, 3, and 5 are stable (after a week, using CHCl₃ to extract the solution, and the ¹¹⁹ Sn NMR shift in CDCl₃ does not change obviously) (see Supplementary data online). The results of the cytotoxic assay are shown in Table 2. These compounds displayed the high in vitro antitumor activities, which were more active than the clinically widely used cisplatin did. The activity of the four compounds against MCF-7 is better than that against CoLo205. The activity of 5-Br-2-HOC₆H₃C(H)=NCH(CH₃)COOSnPh₃ is the highest, indicating that the R group of the amino acid seems to have an effect on the activity. Compared with the reported triphenyltin analogs, the activity against MCF-7 of the complexes is similar to that of 2-HOC₆H₄C(H)=NCH(CH₂Ph)COOSnPh₃ (IC₅₀=0.115 μg mL⁻¹) (Basu Baul et al., 2007), but less active than 2-HOC₆H₄C(R)=NCH(CH₂CH(CH₃)₂)COOSnPh₃ (IC₅₀: R=H, 0.076 μg mL⁻¹; R=CH₃, 0.034 μg mL⁻¹) (Basu Baul et al., 2009). Thus, further structure modification of triphenyltin compounds of the Schiff base derived from α-amino acid are warranted, which may modulate the cytotoxicity.

General

All chemicals were commercial grade and were used without purification. Glycine, l-alanine, l-valine, l-leucine, l-isoleucine, and 5-bromosalicylaldehyde were purchased from Sinopharm Chemical Reagent Company Limited (Shanghai, China), cisplatin was purchased frome Mayne Pharma Pty Ltd (Australia), and other chemicals were from Shanghai Darui Finechemical Company Limited (Shanghai, China). Carbon, hydrogen, and nitrogen analyses were performed using a Perkin Elmer 2400 Series II elemental analyzer (Perkin Elmer, Waltham, MA, USA). IR spectra were recorded on a Nicolet Nexus 470 FT-IR spectrophotometer using KBr discs in the range 4000–400 cm⁻¹ (Thermo Nicolet Corporation, Madison, WI, USA). ¹H and ¹³C NMR spectral data were collected using a Bruker Avance DPX300 NMR spectrometer (Bruker Corporation, Switzerland) with CDCl₃ as the solvent and tetramethylsilane (TMS) as the internal standard. ¹¹⁹Sn NMR spectra were recorded in CDCl₃ on a Varian Mercury Vx300 spectrometer using Me₄Sn external reference (Varian Corporation, USA).

Synthesis of triphenyltin complexes (1–5)

In a 100-mL flask, α-amino acid (2 mmol), 5-bromosalicylaldehyde (0.40 g, 2 mmol), triphenyltin hydroxide (0.73 g, 2 mmol), and a 1:1 mixed solvent of CHCl₃/MeOH (60 mL) were added in sequence. Under electromagnetic stirring, the reaction mixtures were refluxed for 8 h and then allowed to cool to room temperature. The solution was filtered, and the solvent was removed by a rotary evaporator. The resulting yellow solid was recrystallized from ethanol. The yield, m.p., and spectral data for compounds 1−5 are as follows (Scheme 3):

Scheme 3:

The numbering scheme for the NMR assignment.

X-ray crystallography

The yellow single crystals of 3 and 5 were obtained from chloroform-methanol (1:1, V/V) by slow evaporation at room temperature. Diffractions measurements were performed on a Bruker Smart Apex imaging-plate area detector fitted with graphite monochromatized Mo-Kα radiation (0.71073 Å) using the φ and ω scan technique. The structure was solved by direct method and refined by a full-matrix least squares procedure based on F² using SHELXL-97 (Sheldrick, 2008). The non-hydrogen atoms were refined anisotropically, and hydrogen atoms were placed at calculated positions. In 3 and 5, the chiral carbon atoms and R groups (i-Pr and s-Bu) were disordered over two conformations. The site occupancies were refined to 0.525(10):0.475(10) and 0.545(11):0.455(11), respectively. Crystal data, collection procedures, and refinement results are summarized in Table 3. Crystallographic data have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication numbers CCDC 1552755 and 1537471.

In vitro screening

The 200 mg/L stock solutions were prepared by dissolving the test compounds in ethanol and by diluting the resultant solutions with serum-free DMEM (Dulbecco’s modified Eagle medium) culture media. In the assays, the final concentration of ethanol was less than 0.1% (this concentration used was found to be non-cytotoxic against tumor cells.). Three human tumor cell lines, HeLa (cervix tumor cell), CoLo 205 (colon carcinoma cell), and MCF-7 (mammary tumor cell), were obtained from Tumour Institute of Zhejiang University. In vitro cytotoxic activity of the compounds was measured by the MTT assay according to the literature (Denizot and Lang, 1986; Tian et al., 2015). All cells cultured in DMEM supplemented with 10% heat-inactivated new-born calf serum at 37°C in a humidified 5% CO₂ incubator, and a total of 1×10⁴ cells were seeded into each well of a 96-well plate and were fixed for 24 h. The following day, different concentrations of the test compounds were added. After incubation with various concentrations of test compounds for 72 h, the inhibition on cell proliferation was measured. Briefly, 100 μL of MTT working solution (1 mg/mL) was added into each well of a 96-well plate. At 37°C, the medium was incubated for 4 h, and then, it was moved. The converted dye formazan was solubilized with 150 μL acidic isopropanol. The results were read on the absorbance microplate readers (Model 680, Bio-Rad) with a wavelength of 570 nm. Different concentrations of cisplatin and 0.1% of ethanol were used as the positive and negative control for the cytotoxicity study of the test compounds in vitro. The experiments were conducted in triplicate for each tested concentration, and the data are expressed as the mean±standard deviation. The cell proliferation inhibition rates are given as supporting information (online Supplementary Tables 1–3). The dose causing 50% inhibition of cell growth (IC₅₀) was calculated by NDST software as previously described (Zheng et al., 2004).