Study of the effect of molecular structure and alkyl groups bound with tin(IV) on their cytotoxicity of organotin(IV) 2-phenyl-4-selenazole carboxylates

Abstract

Five organotin(IV) compounds Ph₃SnL (1), (R₂Sn)₄O₂L₄ [R = n-Bu (2), n-Oct (3)], (R₂Sn)₄O₂L₂Cl₂ [R = n-Bu (4), Me (5)], have been synthesized from the reactions of 2-phenyl-4-selenazole carboxylic acid (HL) with the corresponding organotin(IV) oxide or chlorides. These compounds have been characterized by elemental analysis, IR, ¹H, ¹³C and ¹¹⁹Sn NMR spectroscopy, and single crystal X-ray diffraction analysis. Structural studies reveal that compound 1 exhibits a mononuclear four-coordinated tetrahedral geometry. Differently, the crystal structures of compounds 2–5 reveal the formation of the tetranuclear species containing a planar Sn₄O₄ core. All compounds were screened for their in vitro cytotoxic activities toward three cancer cell lines (Caco-2, A549, and HCT-116) and one normal rat hepatocyte cell line (BRL). The results indicate that both di-n-butyltin(IV) and triphenyltin(IV) derivatives not only show excellent cytotoxic activities on cisplatin-sensitive lung cancer cell line A549 and but also exhibit good cytotoxicity against cisplatin-insensitive colon cancer cell lines HCT-116 and Caco-2. Whereas, dimethyltin(IV) and di-n-octyltin(IV) complexes exhibit lower or no cytotoxic activity. The structure–activity relationship of the cytotoxicity of the title complexes has also been discussed.

---

1. Introduction

After the discovery of cisplatin by Rosenberg as an effective anticancer drug, the search for alternative metal based drugs has been an important area of interest for researchers. However, platinum compounds suffer from two main disadvantages: inefficiency against platinum-resistant tumors and severe side effects. Furthermore, as a consequence of its particular chemical structure, cisplatin in particular offers little possibility for rational improvements to increase its tumor specificity and thereby reduce undesired side effects.¹ Organotin(IV) complexes have attracted much interest because of their bioactivities, in particular as potential biocidal (e.g., antifungal, anti-microbial)² and anticancer agents.³ Among main-group metal compounds, they appear to exhibit the most potent antitumor activities, in some cases being more effective than cisplatin in vitro tests.⁴ Organotin complexes showed promising antitumor activities against a wide panel of tumors or tumor cell lines. Many organotin complexes were screened against a variety of cell lines and were found to be active both in vitro and in vivo. In general, the biochemical activity of organotin(IV) complexes is influenced greatly by the organic groups bound with tin centers, molecular structure and coordination number of the tin atoms.⁵

At the same time, a wide range of heterocyclic ring systems has been studied for the development of novel chemical entities as a lead molecule in the drug discovery paradigm. Selenium is one of the necessary microelements for vital movement. It plays a significant role in anti-oxidation, anti-aging, protecting the heart, tumour prevention and treatment, relieving side reaction caused by chemotherapy drugs, increasing drug tolerance, reducing cisplatin nephrotoxicity and cytotoxicity, maintaining normal endocrine function and so on.^6,7 Since the 1980's, a large number of organic selenium compounds with biological activity have been synthesized. Among them, selenazole derivatives show good anti-tumor and antibacterial action, which can be used as potential drugs and drug intermediates. For example, Srivastava and Boritzki^8,9 found selenazofurin was a highly efficient antiviral and antitumor drug, which exhibited significant in vitro inhibitory activity against lymphoblastic leukemia diseased cells P338 and L1210, and Kumar¹⁰ also found some selenazole derivatives exhibited in vitro antiproliferative activity against cell L1210.

In view of the above considerations, in the present work we have selected 2-phenyl-4-selenazole carboxylic acid (HL) as the ligand for the synthesis of metallic compounds, which bears a phenyl group and a selenazole ring. According to the literatures, there are a lot of transition-metal compounds coordinating with this ligand being reported, such as Co(II), Ni(II), Cu(II), Zn(II) and Cd(II) compounds.¹¹ But to the best of our knowledge, there are no reports focusing on the in vitro cytotoxic activities of the organotin(IV) derivatives with selenazole.

Herein, we have been investigating the coordination behavior of HL towards organotin(IV), owing to the possible biological applications and their structural diversity in the crystalline state (Scheme 1). Structural analyses reveal that complex 1 exhibits a mononuclear four-coordinated organotin compound, complexes 2–5 reveal the formation of the tetranuclear species contains a planar Sn₄O₄ core. And we checked their in vitro cytotoxic activities against three different cancer cell lines [human lung carcinoma cell line (A549), human colon carcinoma cell line (HCT-116) and human colon adenocarcinoma cell line (Caco-2)]. The target compounds are mainly based on the selenazole ligand and the R_nSn⁽⁴⁻ⁿ⁾⁺ active site for inhibiting the growth of cancer cells. In a word, the synthesis, structural characterization and in vitro cytotoxicity investigation are reported here, and studies demonstrate that this type of complexes has a huge potential as anticancer chemical drugs in medicinal field.

Scheme 1 Synthetic routes of compounds 1–5.

2. Results and discussion

2.1. Syntheses

The synthesis procedure of 2-phenyl-4-selenazole carboxylic acid is given in Scheme S1,† which was prepared according to the literatures.^11–13 The synthesis procedures of complexes 1–5 are shown in Scheme 1. Complexes 1 and 2 were obtained by the reaction of the ligand/Ph₃SnCl [or (n-Bu)₂SnCl₂]/EtONa in a 1 : 1 : 1 ratio in methanol under refluxing. On the contrary, under the same condition complexes 4 and 5 were obtained by the reaction of the ligand/(n-Bu)₂SnCl₂ [or Me₂SnCl₂]/EtONa in a 1 : 2 : 1 molar ratio. Complex 3 was prepared from the reaction of (n-Oct)₂SnO with the ligand in 1 : 1 molar ratio in methanol. All complexes were recrystallized from a mixture of dichloromethane and petroleum ether (1 : 1).

2.2. IR spectra

The IR data provide detailed support for the molecular constitution of the title compounds and some useful information concerning the coordination of the carboxyl groups in organotin carboxylates.¹⁴ In IR spectrum of the ligand 2-phenyl-4-selenazole carboxylic acid, there exists an obvious absorption peak at 3441 cm⁻¹ assigned to ν_OH of the carboxyl group. But this is absent in the IR spectra of compounds 1–5. The new absorption peaks appear in the 407–445 cm⁻¹, which should be assigned to the Sn–O vibration absorption according to the literature.^15,16 These results prove that the carboxyl group has been deprotonated and coordinates to the central tin atom. The band at 536–544 cm⁻¹ may belong to the ν(Sn–C) vibration. In addition, the moderate bands appearing around 680 cm⁻¹ indicate the presence of ν(Sn–O–Sn) group in the spectra of complexes 2–5. It indicates that compounds 2–5 have Sn–O–Sn bridged structure, this is also consistent with the crystal structure described below.

2.3. NMR spectra

As shown in ESI,† the ¹H NMR spectrum of 2-phenyl-4-selenazole carboxylic acid ligand shows two types of peaks, that is a singlet assigned to selenazole methenyl group (δ = 8.98) and a set of multiplet for the phenyl group (δ = 7.95–7.42). Upon interaction of ligand with organotin(IV) acceptors, for all compounds studied, no significantly large shifts were evident for the proton atoms of the 2-phenyl-selenazole rings: this observation implies a lack of direct bond involvement of these rings with the metal. Differently, there is no signal of –COOH in ¹H NMR spectra of compounds 1–5, which suggests that the carboxyl group is deprotonated. For di-n-butyltin compounds 2 and 4 as well as di-n-octyltin compound 3, a set of multiplet around 1.80–0.70 ppm is assigned to butyl or octyl group. The methyl group in compound 5 show a triplet at 0.95–0.48 ppm.

Compared with the ligand, the ¹³C NMR spectra of compounds 1–5 show a down field shift. This phenomenon should be due to the electron transfer from ligand to tin atom, which is consistent with the previous reports.^17,18 The value of δ(¹¹⁹Sn) in CDCl₃ for compound 1 is δ = −100.47 ppm, and is similar to those found for triphenyltin(IV) carboxylates, such as, [SnPh₃(OOCC₆H₄NH₂-4)] and [SnPh₃(OOCC₆H₄NH₂-2)] (δ = −122.6 and −116.8 ppm),¹⁹ [Sn(C₆H₅)₃-{OOCC₆H₃(NH₂)₂-3,4}] and [Sn(C₆H₅)₃{OOC-2-C₆H₄N=NC₆H₄N(CH₃)₂-4}] (δ = −118.9 and −108.6 ppm),²⁰ as well as triphenyltin(IV) 3-maleimidopropionato (δ = −115.8 ppm).²¹ Based on the literary evidences, this spectroscopic data confirm a four-coordinated tetrahedron geometry of compound 1 in inert solvents. The ¹¹⁹Sn NMR spectra of tetraorganodistannoxanes 2 and 3 show two signals around −172 and −220 ppm, confirming the ladder hydrolysis products. This is similar to those ladder compounds {[R₂Sn(O₂CR′)]₂O reported previously.²² For diorgano-dichloro-distannoxanes 4 and 5, there also exhibit two signals around −143 and −190 ppm due to endocyclic and exocyclic tin atoms, respectively, and suggests the retention of dimeric structure in solution. These shifts of compounds 2–5 are all in the range of reported for related five-coordinated diorganotin(IV) carboxylates with similar distannoxane structure.²³

2.4. X-ray crystallography

Diffraction data for the title compounds were obtained on a Bruker Smart 1000 CCD diffractometer (graphite monochromized Mo Kα radiation, λ = 0.71073 Å). All data were corrected using SADABS method and the final refinement was performed by mull-matrix least-square methods with anisotropic thermal parameters for non-hydrogen atoms on F² using SHELX-97 program.²⁴ The hydrogen atoms were added theoretically, riding on the concerned atoms and refined with fixed thermal factors. The relevant crystallographic data of all compounds are given in Tables S1 and S2.†

2.4.1. X-ray crystallography of complex 1. The molecular structure of complex 1 is illustrated in Fig. 1. Selected bond lengths and bond angles are given in Table S3.† From Fig. 1, it can be seen that the tin atom in 1 presents in a four-coordinated environment, binding with three phenyl groups and the monodentate deprotonated carboxylate, which is common to those mononuclear-monoligand triphenyltin(IV) carboxylates.^20,21 The C11–Sn1–C23 and C17–Sn1–C23 angles are 111.49(15) and 110.31(15)°, respectively, and the C11–Sn1–C17 angle is 117.38(15)°. The O(1)–Sn(1)–C(17) angle is 113.49(13)°, while the O(1)–Sn(1)–C(11) and O(1)–Sn(1)–C(23) angles are 104.31(13) and 98.08(12)°, respectively. These results demonstrate that compound 1 adopts a distorted tetrahedral structure. The distance of Sn1–O1 (carboxylate) 2.087(3) Å is shorter than the sum of the covalent radii of Sn and O (2.56 Å) in the literature.²⁵ However, the Sn1⋯O2 bond length 2.793(4) Å is longer these, but is significantly shorter than the sum of the van der Waals radii of tin and oxygen (3.68 Å). Such, Sn1⋯O2 bond should be considered as weak bonding interaction.

Fig. 1 Crystal structure of compound 1. All H atoms are omitted for clarity.

2.4.2. X-ray crystallography of complexes 2 and 3. Complexes 2 and 3 have similar molecular structures, as illustrated in Fig. 2 and 3. In addition, due to the bad quality of crystals of compounds 2 and 3, the diffraction for it is quite weak. Also, the long chain structure of n-butyl or n-octyl group and the bulky ligand made the molecules instability. Therefore, the R factor of compounds 2 and 3 is bad in crystallography. From the structural descriptions of the complexes, it can be seen that all the tin atom is rendered five-coordinate with a distorted trigonal bipyramidal configuration. This tetraorganodistannoxane ladder structure were very common for carboxylate derivatives^26,27 by the reaction of carboxylic acid ligand and organotin salts with the ratio of 1 : 1.

Fig. 2 Crystal structure of compound 2. All H atoms and n-butyl carbon (except Sn–C) atoms are omitted for clarity.

Fig. 3 Crystal structure of compound 3. All H atoms and n-octyl carbon (except Sn–C) atoms are omitted for clarity.

Selected bond lengths and angles are listed in Tables S4 and S5.† The detailed structure was described taking complex 2 for example. As shown in Fig. 2, complex 2 exhibits a centrosymmetric tetranuclear structure containing a planar Sn₄O₂ core, in which two μ₃-oxo O atoms connect an Sn₂O₂ ring to two exocyclic Sn-atoms to give a (n-Bu)₈Sn₄O₂ central unit. In complex 2, the coordinate environment of all tin atoms involved can be classified into two sorts, the endocyclic tin [Sort I: Sn1 and Sn1A] in the Sn₂O₂ ring and the exocyclic tin [Sort II: Sn2 and Sn2A] of the three-fold ladder bridged by one carboxylate oxygen of the ligand. While the second carboxylate is monodentate on Sn2 via the carboxyl oxygen atom.

Two n-butyl groups completed the five coordination to the metal ion with a distorted trigonal bipyramidal geometry; the equatorial positions are occupied by two n-butyl carbon atoms (C21 and C23) and one oxygen (O3) and the axial positions O4 and O3#1 for Sn1 (#1 − x + 1, −y, −z + 1), and for Sn2 by C29, C33 and O3 in the equator, while in axial positions are O2 and O4.

The greatest deviations within the trigonal plane are seen in the C–Sn–C angles [C25–Sn1–C21 128.0(10)° and C33–Sn2–C29 139.8(11)°] and in the axial O–Sn–O angles [O3#1–Sn1–O4 151.6(6)° and O2–Sn2–O4 146.8(7)°]. The Sn–O and Sn–C distances in the molecules are in the expected value. The ligands have weak interactions with the Sn atoms (Sn1⋯O5 3.398 Å, Sn1⋯O2#1 3.179 Å and Sn2⋯O1 2.898 Å) as it has been observed in {(CH₃)₂[(pyS)₂CHCO₂]SnOSn[(pyS)₂CHCO₂](CH₃)₂}₂,^23a [(Me₂SnO₂CC₆H₄-o-NH₂)₂]₂ and [(Me₂SnO₂CC₆H₄-p-NH₂)₂]₂.²²

2.4.3. X-ray crystallography of complexes 4 and 5. Complexes 4 and 5 were obtained by the reaction of carboxylic acid ligand and organotin dichloride with the ratio of 1 : 2. As illustrated in Fig. 4 and 5, both compounds have similar molecular structures. Selected bond lengths and angles are listed in Tables S6 and S7.† The detailed structure was described taking complex 4 for example. In addition, due to the bad quality of crystals of compound 4, the diffraction for it is quite weak. Also, the long chain structure of n-butyl group and the bulky ligand made the molecules instability. Therefore, the R factor of compounds 4 is bad in crystallography.

Fig. 4 Crystal structure of compound 4. All H atoms and n-butyl carbon (except Sn–C) atoms are omitted for clarity.

Fig. 5 Crystal structure of compound 5. All H atoms are omitted for clarity.

As shown in Fig. 4, the geometry of 4 composed by two types of cyclic atoms is similar to the 1,3-disubstituted distannoxane compound with the ‘ladder’ structure.²⁸ In the structure of 4, the part of dichlorodistannoxane adopts the most common structural type found for compounds of the general formula [R₂(Cl)SnOSn(μ₃-O)R₂]₂ and exists as ladder structure that contains two endo- and two exo-cyclic Sn atoms. The n-butyl groups are the same for both the endo- and the exo-Sn atoms, and of the center Sn₂O₂ ring, the bridge-oxo atom serves as tridentate and the internal angles [O3–Sn2–O4: 74.9(5)°; Sn2–O3–Sn3: 121.2(6)°] are different with those in other centrosymmetric distannoxane systems,²⁸ such as 72.09°, 107.9° of [(PhCH₂)₂(Cl)SnOSn(OEt)(CH₂Ph)₂]₂, 73.16°, 106.84° of [(PhCH₂)₂(Cl)SnOSn(Cl)(CH₂Ph)₂]₂ (ref. 28a) and 72.3(6)°, 107.7(6)° of [(n-Bu)₂(Cl)SnOSn(Cl)(n-Bu)₂]₂.^28b The distances of μ₃-O to the endo- and exo-cyclic Sn atoms exhibits a certain difference: Sn1–O3 2.127(12) Å, Sn2–O3 2.064(15) Å and Sn3–O3 1.993(13) Å, reflecting the strong coordination of bridge-oxo with tin atoms in the compound.

Both the endo- and exo-cyclic tin atoms are penta-coordinated and arranged in distorted trigonal bipyramids. The trigonal plane is defined by two carbon atoms of n-butyl groups and O3. The axial angles O2–Sn1–Cl1 [155.6(4)°] and O2–Sn2–O4 [143.9(5)°] for compound 4, are all smaller than the ideal value (180°) of trigonal bipyramid, which is possible due to the presence of weak interaction between the chlorine atom and the endo-cyclic tin atom. The distance of Cl1⋯Sn3 is 3.334 Å, which is less than the sum of the van der waals radii of Sn and Cl (4.0 Å), so there is weak interaction between the two atoms. In addition, the disorder phenomenon of the ligand has been observed for compound 5.

2.5. In vitro cytotoxicities activity

Complexes were tested for cytotoxic activity on three tumor cell lines: human lung cancer cell line (A549), human colon cell line (HCT-116) and colon adenocarcinoma (Caco-2), and one normal cell line (rat hepatocytes cell line, BRL). This study has been carried out in order to understand the possible structure–activity relationship between the different tin moieties (bearing methyl, n-butyl, phenyl or n-octyl groups), the coordination geometry and their in vitro cytotoxicity.

Comparison of inhibition rate for complexes 1, 2, 4 and 5 with the concentration of 10 μg mL⁻¹ is shown in Fig. 6. The IC₅₀ values, calculated from the dose survival curves, obtained after 48 h of drug treatment in the MTT test, are summarized in Table 1. And the inhibition effects of complexes 1, 2, 4 and 5 on the four cell lines at different concentrations are successively shown in Fig. 7–10. In addition, due to the poor in vitro cytotoxicity, the activity response of compound 5 on the different dosage is shown in the higher concentration range. From the derived data, we can see that although with the same ligand, triphenyltin compound 1 and di-n-butyltin compounds 2 and 4 show higher in vitro cytotoxic activities than cisplatin against all three cancer cell lines and one normal cell line. Furthermore, organotin(IV) 2-phenyl-4-selenazole carboxylates exhibit better in vitro cytotoxicity than their corresponding organotin(IV) salt precursor.

Fig. 6 Inhibition [%] of compounds 1, 2, 4 and 5 [dose level of 10.0 μg mL⁻¹] against three human tumor cell lines and one normal cell line.

Table 1 IC₅₀ (μM) of all compounds against three human tumour cell lines and one normal cell line

Compound	IC50
	HCT-116	A-549	Caco-2	BRL
1	0.08 ± 0.02	0.29 ± 0.16	1.42 ± 0.61	0.15 ± 0.07
2	0.35 ± 0.15	0.22 ± 0.08	0.13 ± 0.05	0.12 ± 0.07
3	>100	>100	>100	>100
4	0.16 ± 0.08	1.67 ± 0.29	0.63 ± 0.23	0.29 ± 0.05
5	41.60 ± 2.38	33.00 ± 0.61	40.80 ± 1.27	8.30 ± 0.97
HL	83.57 ± 1.25	96.78 ± 2.21	>100	>100
Ph2SnO	0.12 ± 0.15	2.42 ± 0.23	0.12 ± 0.11	0.29 ± 0.22
n-Bu2SnCl2	0.50 ± 0.25	2.67 ± 0.75	0.38 ± 0.15	0.21 ± 0.14
Me2SnCl2	>100	>100	>100	>100
n-Oct2SnO	>100	>100	>100	>100
Cisplatin	>100	28.50 ± 4.61	>100	54.88 ± 3.75

---

Fig. 7 The inhibition effects of compound 1 on the four cells lines at different concentration (0.01, 0.1, 0.5, 1, 10 μg mL⁻¹).

Fig. 8 The inhibition effects of compound 2 on the four cells lines at different concentration (0.01, 0.1, 0.5, 1, 10 μg mL⁻¹).

Fig. 9 The inhibition effects of compound 4 on the four cells lines at different concentration (0.01, 0.1, 0.5, 1, 10 μg mL⁻¹).

Fig. 10 The inhibition effects of compound 5 on the four cells lines at different concentration (5, 10, 20, 40, 50 μg mL⁻¹).

Based on this, possible structure–activity relationships could be recognized as follows:

(i) Triphenyl-, di-n-butyl- or dimethyltin(IV) 2-phenyl-4-selenazole carboxylates compounds display obvious inhibitory activity on different cell lines, and this toxicity is concentration-dependent and is related to the lipophilic properties of the complexes as well as the number of tin atoms included in the complex.^4,5

(ii) It is well known that triorganotin compounds exhibit strong cytotoxic activity most probably due to the free coordination position on tin atom.⁵ And this has been proved again here that triphenyltin(IV) compound show higher cytotoxicity, especially for HCT-116 cell line with the least IC₅₀ value of 0.08 μM.

(iii) For the same ligand, the same structure, but with different organic groups, the difference is very significant in vitro activity. In general the activity of organotin order is n-butyl ≥ phenyl > methyl > n-octyl group, which is consistent with previous literature reports.^29–32 These results reveal that alkyl groups bound with tin center play a key role on their cytotoxicity.

(iv) For the hydrocarbonyl-tin(IV) compounds, the length of straight chain alkyl group bound with tin center plays an important role on their in vitro cytotoxicity, which should be related with their moderate lipophilic properties. As reported herein, n-butyl group compounds show the highest in vitro cytotoxicity and the shorter chain methyltin compound exhibit lower activity. But when the length of the chain is too long, it is bad for the activity. For example, di-n-octyltin compound 3 show non-active absolutely.

(v) In this report, the structure and ligand of the organotin(IV) compounds have little effect on their in vitro cytotoxicity, which are mainly dependent on the organic groups bound with tin center and unrelated to some other factors like coordination number, geometry and hydrophilicity and so on.^33,34 For instance, although diorganotin(IV) compounds 2 and 3 have the same tetraorganodistannoxane molecular structure each other. Di-n-butyltin compound 2 display better activity than di-n-octyltin compound 3. Similar phenomenon was observed for 4 and 5. However, although with different structure for 2 and 4, they exhibit similar activity for all cell lines.

(vi) Except for di-n-octyltin(IV) compound 3, other four organotin(IV) compounds exhibit enhanced cytotoxicity compared with 2-phenyl-4-selenazole carboxylic acid ligand and four di- or tri-organotin(IV) chloride precursors, which clearly implied a positive synergistic effect against all four cell lines.

3. Conclusions

In this article, we described the synthesis, crystal structure and in vitro cytotoxic activity of five new 2-phenyl-4-selenazole carboxylic acid based organotin(IV) complexes. Structural characterization shows that one mononuclear coordination compound is formed when the ligand reacts with triphenyltin chloride. When the ligand reacts with the dialkyltin precursors, four ladder-like structure is generally formed. The central tin atoms for all compounds are five-coordinated trigonal bipyramidal geometries or six-coordinated octahedrons.

What's more, the results of in vitro cytotoxic activity screening for three cancer cell lines HCT-116, A549 and Caco-2 indicate that di-n-butyltin(IV) and triphenyltin(IV) derivatives show much higher activity than cisplatin the anticancer drug used clinically. In the other word, which have some potential medicinal values. But unfortunately, they exhibit poor specificity for cancer cell lines and also have obvious toxic effects on normal cells.

4. Experimental details

4.1. Materials and measurements

Selenium powder, benzonitrile, sodium borohydride, 1,3-dichloroacetone, potassium bichromate and sodium hydrate, triphenyltin chloride, dimethyltin dichloride, di-n-butyltin dichloride, di-n-octyltin oxide were commercially available and they were used without further purification. Analytical grade solvents used in this work were undried. Elemental analyses were performed on a PE-2400-II elemental analyzer. IR spectra were recorded on a Nicolet-5700 spectrophotometer using KBr discs. ¹H, ¹³C and ¹¹⁹Sn NMR spectra were recorded on a Varian Mercury Plus-400 NMR spectrometer. Chemical shifts were given in ppm relative to Me₄Si (¹H, ¹³C) and Me₄Sn (¹¹⁹Sn) in CDCl₃ or DMSO solvent. X-ray measurements were made on a Bruker Smart-1000 CCD diffractometer with graphite monochromated Mo Kα (λ = 0.71073 Å) radiation.

4.2. Preparation of the ligand

2-Phenyl-4-selenazole carboxylic acid prepared by the modified methods reported in the literature.^11–13 The detailed synthesis procedure and Experimental section were shown as Scheme S1 in ESI.†

4.3. Syntheses of complexes 1–5

4.3.1. Synthesis of Ph₃SnL (1). The reaction was carried out under atmospheric conditions. 2-Phenyl-4-selenazole carboxylic acid (0.252 g, 1 mmol), EtONa (0.068 g, 1 mmol) were added to a stirred solution of methanol (30 mL) in a round bottomed flask and stirred for 0.5 h. Triphenyltin chloride (0.385 g, 1 mmol) was then added to the reactor. The reaction mixture was refluxed for 8 h more, and then filtrated. The filtrate was evaporated in vacuum. The obtained solid was recrystallized from dichloromethane/petroleum (1 : 1). Colourless block crystals of complex 1 were slowly formed at room temperature. Yield: 72%, m.p.: 152–155 °C. Anal. calc. for: C₂₈H₂₁NO₂SeSn: C, 56.04; N, 2.33; H, 3.36. Found: C, 56.32; N, 2.45; H, 3.38%. IR (KBr, cm⁻¹): 1637 ν(OCO)_asym, 1650 ν(C=N), 1347 ν(OCO)_sym, 544 ν(Sn–C), 445 ν(Sn–O). ¹H NMR (CDCl₃, ppm): δ = 8.95 (s, 1H, Se–CH), 7.96–7.40 (m, 20H, Ph-H). ¹³C NMR (400 MHz, CDCl₃, ppm): δ = 174.88 (COO), 167.25 (C=N); 127.71–138.05 (aromatic carbons). ¹¹⁹Sn NMR (CDCl₃, ppm): −100.47.

4.3.2. Synthesis of (n-Bu₂Sn)₄O₂L₄ (2). The reaction was carried out under atmospheric conditions. 2-Phenyl-4-selenazole carboxylic acid (0.252 g, 1 mmol), EtONa (0.068 g 1 mmol) were added to a stirred solution of methanol (30 mL) in a round bottomed flask and stirred for 0.5 h. Di-n-butyltin dichloride 0.303 g (1 mmol) was then added to the reactor. The reaction mixture was refluxed for 8 h more, and then filtrated. The filtrate was evaporated in vacuum. The obtained white solid was recrystallized from dichloromethane/petroleum (1 : 1) to give colourless block crystals of complex 2. Yield: 68%, m.p.: 135–137 °C. Anal. calc. for: C₁₄₄H₁₉₂N₈O₂₀Se₆Sn₈: C, 43.94; N, 2.85; H, 4.92. Found: C, 43.79; N, 2.95; H, 4.81%. IR (KBr, cm⁻¹): 1604 ν(OCO)_asym, 1652 ν(C=N), 1355 ν(OCO)_sym, 687 ν(O–Sn–O), 544 ν(Sn–C), 407 ν(Sn–O). ¹H NMR (400 MHz, CDCl₃, ppm): δ = 8.85 (s, 8H, Se–CH), 8.07–7.48 (m, 40H, Ph-H), 1.86–0.71 (m, 144H, –CH₂CH₂CH₂CH₃). ¹³C NMR (400 MHz, CDCl₃, ppm): δ = 174.62 (COO), 166.06 (C=N), 127.71–135.81 (aromatic carbons), 29.91–26.11 (–CH₂CH₂CH₂Sn), 14.17 (δ-CH₃). ¹¹⁹Sn NMR (CDCl₃, ppm): −171.04, −219.69.

4.3.3. Synthesis of (n-Oct₂Sn)₄O₂L₄ (3). 2-Phenyl-4-selenazole carboxylic acid (0.252 g, 1 mmol) and di-n-octyltin oxide (0.361 g, 1 mmol) were added to a stirred solution of methanol (60 mL) in a round bottomed flask. The reaction mixture was refluxed for 8 h and then filtrated. The filtrate was evaporated in vacuum. The white solid was recrystallized from dichloromethane/petroleum (1 : 1) to give colourless block crystals of complex 3. Yield: 83%, m.p.: 123–125 °C. Anal. calc. for: C₁₁₄H₁₆₇N₅O₁₂Se₅Sn₄: C, 51.30; N, 2.62; H, 6.31. Found: C, 51.37; N, 2.45; H, 6.29%. IR (KBr, cm⁻¹): 1611 ν(OCO)_asym, 1636 ν(C=N), 1363 ν(OCO)_sym, 686 ν(O–Sn–O), 543 ν(Sn–C), 455 ν(Sn–O). ¹H NMR (400 MHz, CDCl₃, ppm): δ = 8.83 (s, 5H, Se–CH), 8.06–7.47 (m, 25H, Ph-H), 1.82–0.91 (m, 112H, Sn–(CH₂)₇–), 0.87 (t, 24H, –CH₃). ¹³C NMR (400 MHz, CDCl₃, ppm): δ = 175.92 (COO), 168.31 (C=N); 126.71–151.34 (aromatic carbons). ¹¹⁹Sn NMR (CDCl₃, ppm): −174.13, −220.10.

4.3.4. Synthesis of (n-Bu₂Sn)₄O₂L₂Cl₂ (4). Compound 4 was obtained by the reaction of the ligand/n-Bu₂SnCl₂/EtONa in a 1 : 2 : 1 molar ratio in methanol. The preparing procedure was similar to compound 2. The white solid was recrystallized from dichloromethane/petroleum (1 : 1) to give colourless block crystals of complex 4. Yield: 81%, m.p.: 129–131 °C. Anal. calc. for: C₅₂H₈₄Cl₂N₂O₆Se₂Sn₄: C, 40.34; N, 1.62; H, 5.11. Found: C, 40.32; N, 1.55; H, 5.02%. IR (KBr, cm⁻¹): 1620 ν(OCO)_asym, 1646 ν(C=N), 1371 ν(OCO)_sym, 678 ν(O–Sn–O), 543 ν(Sn–C), 421 ν(Sn–O). ¹H NMR (400 MHz, CDCl₃, ppm): δ = 8.91 (s, 2H, Se–CH), 8.05–7.44 (m, 10H, Ph-H), 1.87–0.70 (m, 36H, –CH₂CH₂CH₂CH₃). ¹³C NMR (400 MHz, CDCl₃, ppm): δ = 175.25 (COO), 164.12 (C=N), 127.52–135.28 (aromatic carbons). ¹¹⁹Sn NMR (CDCl₃, ppm): −142.96, −181.40.

4.3.5. Synthesis of (Me₂Sn)₄O₂L₂Cl₂ (5). Compound 4 was obtained by the reaction of the ligand/Me₂SnCl₂/EtONa in a 1 : 2 : 1 molar ratio in methanol. The preparing procedure was similar to compound 2. The white solid was recrystallized from dichloromethane/petroleum (1 : 1) to give colourless block crystals of complex 5. Yield: 75%, m.p.: 145–146 °C. Anal. calc. for: C₂₈H₃₆Cl₂N₂O₆Se₂Sn₄: C, 28.02; N, 2.73; H, 3.02. Found: C, 28.10; N, 2.75; H, 3.05%. IR (KBr, cm⁻¹): 1619 ν(OCO)_asym, 1639 ν(C=N), 1366 ν(OCO)_sym, 683 ν(O–Sn–O), 545 ν(Sn–C), 426 ν(Sn–O). ¹H NMR (400 MHz, CDCl₃, ppm): δ = 8.88 (s, 2H, Se–CH), 8.01–7.52 (m, 10H, Ph-H), 0.95–0.48 (m, 24H, Sn–CH₃). ¹³C NMR (400 MHz, CDCl₃, ppm): δ = 176.21 (COO), 169.13 (C=N), 127.93–151.37 (aromatic carbons). ¹¹⁹Sn NMR (CDCl₃, ppm): −145.06, −197.62.

4.4. In vitro cytotoxic activity

4.4.1. Materials. Four cell lines, including three cancer cell lines: human lung cancer cell line (A549), human colon cell line (HCT-116) and colon adenocarcinoma cell line (Caco-2), and one normal cell line (rat hepatocytes cell line, BRL), were used for screening. Cell lines were maintained in the logarithmic phase at 37 °C in a 5% carbon dioxide atmosphere using the following culture media containing 10% fetal bovine serum and 1% antibiotics (50 units per mL penicillin and 50 mg mL⁻¹ streptomycin): (i) DMEM (Dulbecco's Modified Eagle Medium) medium for A549 and BRL cell lines. (ii) McCoy's 5A medium for HCT-116 cell line. (iii) IMDM (Iscove's Modified Dulbecco's Medium) medium for Caco-2 cell line.

4.4.2. Preparation of drug solutions. Stock solutions of the studied organoantimony compounds were prepared in dimethyl sulfoxide (DMSO, Sigma-Aldrich) at concentrations of 10 mg mL⁻¹ and diluted by cell culture medium to various working concentration. DMSO was used due to solubility problems. MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide was dissolved (5 mg mL⁻¹) in phosphate buffer saline pH 7.2, and filtered through Millipore filter, 0.22 mm, before use.

4.4.3. MTT assays. Cells were seeded onto 96-well flat-bottom plates for 24 h before treatment to allow attachment of cell to the wall of the plate, then fed with dilutions of each drug. The plates were incubated at 37 °C, 5% CO₂ for a period of 48 h. After 48 h, the medium was removed, 100 mL of a 0.5 mg mL⁻¹ solution of MTT in medium was added, and the plate was incubated for an additional 4 h. The medium/MTT mixture was aspirated, and 100 mL of DMSO was added to dissolve the purple formazan crystals. The plate was shaken for 20 min on a plate shaker to ensure complete dissolution. The absorbance of the plates was read at 490 nm. IC₅₀ values were extrapolated from the resulting curves. The reported IC₅₀ values are the averages from at least three independent experiments, each of which consisted of three replicates per concentration level.

Acknowledgements

We acknowledge the National Natural Foundation of China (21105042), the Science Foundation of China Postdoctor (no. 2014M560572) and the Natural Science Foundation of Shandong Province (ZR2015BM024, ZR2010BQ021) for financial support. And this work was supported by “Tai–Shan Scholar Research Fund of Shandong Province”.

References

1. G. Gasser, I. Ott and N. Metzler-Nolte, J. Med. Chem., 2011, 54, 3–25.
2. (a) L. Pellerito and L. Nagy, Coord. Chem. Rev., 2002, 224, 111–150; (b) A. Chaudhary, M. Agarwal and R. V. Singh, Appl. Organomet. Chem., 2006, 20, 295–303; (c) A. Chaudhary and R. V. Singh, Main Group Met. Chem., 2008, 31, 107–155.
3. (a) Metal Compounds in Cancer Therapy, ed. A. J. Crowe and S. P. Fricker, Chapman & Hall, London, 1994, pp. 147–179; (b) M. Gielen, Coord. Chem. Rev., 1996, 151, 41–51; (c) Metallotherapeutic Drug and Metal-based Diagnostic Agents: ⁵⁰Sn Tin Complexes and Their Therapeutic Potential, ed. M. Gielen and E. R. T. Tiekink, 2005, pp. 421–435; (d) M. Gielen, Appl. Organomet. Chem., 2002, 16, 481–494.
4. (a) Nato ASI Series, ed. A. Penninks, M. Bol-Schoenmakers, W. Seinen and M. Gielen, Springer, Berlin, 1990, pp. 169–190; (b) C. Pettinari, F. Marchetti, A. Cingolani, A. Lonrenzotti, E. Mundorff, M. Rossi and F. Caruso, Inorg. Chim. Acta, 1997, 262, 33–46.
5. (a) A. K. Saxena, Coord. Chem. Rev., 1989, 95, 109–123; (b) S. K. Hadjikakou and N. Hadjiliadis, Coord. Chem. Rev., 2009, 253, 235–249.
6. (a) B. H. Xu and Y. Sun, Chin. J. New Drugs, 1998, 7, 176–180; (b) L. Lucia, V. Viera and B. Jela, Toxicology, 2006, 227, 1–14.
7. M. Lovell, S. Xiong and G. Lyubartseva, Free Radical Biol. Med., 2009, 46, 1527–1533.
8. P. C. Srivastava and R. K. Robins, J. Med. Chem., 1983, 26, 445–448.
9. T. J. Boritzki, D. A. Berry, J. A. Besserer, P. D. Cook, D. W. Fry, W. R. Leopold and R. C. Jackson, Biochem. Pharmacol., 1985, 34, 1109–1114.
10. (a) Y. Kumar, R. Green, K. Z. Borysko, D. S. Wise, L. L. Wotring and L. B. Townsend, J. Med. Chem., 1993, 36, 3843–3848; (b) Y. Kumar, R. Green, D. S. Wise, L. L. Wotring and L. B. Townsend, J. Med. Chem., 1993, 36, 3849–3852.
11. G. L. Zhao, X. Shi, J. P. Zhang, J. F. Liu, H. D. Xian and L. X. Shao, Sci. Sin.: Chim., 2010, 40, 1525–1535.
12. A. Z. Al-Rubaie, L. Z. Yousifb and A. J. H. Al-Hamadb, J. Organomet. Chem., 2002, 656, 274–280.
13. A. Shafiee, A. Mazloumi and V. I. Cohen, J. Heterocycl. Chem., 1979, 16, 1563–1566.
14. K. Chandra, R. K. Sharma, B. S. Garg and R. P. Singh, J. Inorg. Nucl. Chem., 1980, 42, 187–193.
15. X. Wu, W. Kang, D. S. Zhu, C. G. Zhu and S. R. Liu, J. Organomet. Chem., 2009, 694, 2981–2986.
16. C. H. Switzer and W. Flores-Santana, Biochim. Biophys. Acta, Bioenerg., 2009, 1787, 835–840.
17. T. S. Basu Baul, C. Masharing and A. Linden, J. Organomet. Chem., 2009, 19, 395–400.
18. F. P. Pruchnik, M. Banbula, Z. Ciunik, M. Latocha, B. Skop and T. Wilczok, Inorg. Chim. Acta, 2003, 356, 62–66.
19. J. Holeček, M. Nádvornik, K. Handliř and A. Lyčka, J. Organomet. Chem., 1983, 241, 177–180.
20. F. P. Pruchnik, M. Banbuła, Z. Ciunik, H. Chojnacki, M. Latocha, B. Skop, T. Wilczok, A. Opolski, J. Wietrzyk and A. Nasulewicz, Eur. J. Inorg. Chem., 2002, 3214–3221.
21. M. I. Khan, M. K. Baloch and M. Ashfaq, J. Organomet. Chem., 2004, 689, 3370–3378.
22. V. Chandrasekhar, R. O. Day, J. M. Holmes and R. R. Holmes, Inorg. Chem., 1988, 27, 958–964.
23. (a) F. Benetollo, G. G. Lobbia, M. Mancini, M. Pellei and C. Santini, J. Organomet. Chem., 2005, 690, 1994–2001; (b) A. W. Addison, T. N. Rao, J. Reedijk, J. Rijn and G. C. Verschoor, J. Chem. Soc., Dalton Trans., 1984, 1349–1353; (c) J. Beckmann, M. Henn, K. Jurkschat, M. Schuermann, D. Dakternieks and A. Duthie, Organometallics, 2002, 21, 192–195.
24. J. Schraml, Appl. Organomet. Chem., 2000, 14, 604–610.
25. C. L. Ma, Q. F. Zhang and R. F. Zhang, Chem.–Eur. J., 2006, 12, 420–428.
26. T. S. Basu Baul, W. Rynjah, E. Rivarola and C. Pettinari, J. Organomet. Chem., 2007, 692, 3625–3635.
27. X. M. Shang, Q. S. Li and J. Z. Wu, J. Organomet. Chem., 2005, 690, 3997–4000.
28. (a) C. L. Ma and F. Li, Chin. J. Chem., 2003, 21, 146–149; (b) M. Hong, H. D. Yin, S. W. Chen and D. Q. Wang, J. Organomet. Chem., 2010, 695, 653–662.
29. S. Gomez-Ruiz, G. N. Kaluderovic, S. Prashar and E. Hey-Hawkins, J. Inorg. Biochem., 2008, 102, 2087–2090.
30. M. Gielen, Appl. Organomet. Chem., 2002, 16, 481–494.
31. C. Syng-Ai, B. T. S. Basu and A. Chatterjee, Mutat. Res., 2002, 513, 49–59.
32. C. Pellerito, L. Nagy, L. Pellerito and A. Szorcsik, J. Organomet. Chem., 2006, 691, 1733–1747.
33. M. Cagnoli, A. Alama, F. Barbieri, F. Novelli, C. Bruzzo and F. Sparatore, Anti-Cancer Drugs, 1998, 9, 603–610.
34. T. Mancilla, L. Carrillo, L. S. Z. Rivera, C. C. Camacho, D. D. Vos, R. Kiss, F. Darro, B. Mahieu, E. R. T. Tiekink, H. Rahier, M. Gielen, M. Kemmer, M. Biesemans and R. Willem, Appl. Organomet. Chem., 2001, 15, 593–603.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1048069 (for 1), 1048070 (for 2), 1048071 (for 3), 1048072 (for 4), 1048073 (for 5). For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c5ra18445b

---