Introduction

Thiazolo[4,5-d]pyrimidines, 7-thio analogs of purines are potentially bioactive molecules. In contrast with related 2-thioxo-thiazolo[4,5-d]pyrimidine derivatives, the 2-oxo analogs have not been very well explored in medicinal chemistry. The synthesis and biological evaluation of the substituted 2-oxo-thiazolo[4,5-d]pyrimidines have been the subject of several review articles. They were reported to possess antibacterial, antifungal (Akbari et al., 2008; Habib et al., 2007), and anti-inflammatory activity (CXCR2-receptor antagonists) (Walters et al., 2008), inhibit the growth of HCMV-human cytomegalovirus (Revankar et al., 1998), and be corticotrophin-releasing hormone (CRH-R1) receptor antagonists (display antidepressant activity) (Beck et al., 1999).

In this study, in continuation of our work on thiazolo[4,5-d]pyrimidine derivatives, the synthesis and in vitro cytotoxic evaluation of thiazolo[4,5-d]pyrimidin-2-ones are reported. These designed thiazolo[4,5-d]pyrimidine-2-ones are related to thiazolo[4,5-d]pyrimidine-2-thiones that have been previously reported to be potent antitumor agents (Becan and Wagner, 2008). Thiazolo[4,5-d]pyrimidine derivatives have been extensively studied as potential drug candidates and also have anticancer activity (Rida et al., 1996; Fahmy et al., 2002, 2003). Most of these compounds provided with anticancer activity possess an aromatic rings and electronegative substituent directly linked to the C-17 of the essential core (Fig. 1) or attached at aromatic moieties. The method involved subsequent treatment of the appropriate 3,5-diaryl-2-thioxo-5,6-dihydro-4H-thiazolo[4,5-d]pyrimidin-7-ones (2) and 7-chloro-3,5-diaryl-thiazolo[4,5-d]pyrimidine-2-thiones (3) with diethyl sulfate and water for the replacement of the 2-thioxo group by an oxygen function (Scheme 1). Compounds 2 and 3 were obtained from 4-amino-5-carboxamido-3-substituted-2,3-dihydrothiazole-2-thiones (1) (Gewald, 1966) according to a reported earlier procedure (Becan and Wagner, 2008). Pyrimidine ring formation with appropriate aryl aldehyde, followed by chlorination provided the desired cores 2 and 3, bearing the respective aromatic substituent at position 3 and 5, which could further be treated with diethyl sulfate and hydrolyzed to yield 2-thiazolones 4a–4f and 5a–5f. All synthesized compounds were submitted to the National Cancer Institute (NCI, Bethesda, Maryland) to evaluate their growth inhibitory effects on 60 human cancer cell lines, derived from nine neoplasmatic diseases. Five derivatives 4a, 4b, 5a, 5b, and 5d were selected for a primary in vitro antitumor assay, at 10−5 M concentration. Results were expressed as percent growth of the treated cells, compound 5a showing mean percent growth =71.26 was further tested at five different concentrations.

Fig. 1
figure 1

X-Ray molecular structure of compound 5a with the atom-numbering scheme used in the crystallographic analysis

Full size image
Scheme 1
scheme 1

(i) LiOH; (ii) POCl3/PCl5; (iii) (C2H5)2SO4; (iiii) H2O

Full size image

Results and discussion

Chemistry

A series of the new substituted thiazolo[4,5-d]pyrimidin-2-ones 4a–4f and 5a–5f were synthesized as shown in Scheme 1. The method involved subsequent treatment of the appropriate 3,5-diaryl-2-thioxo-5,6-dihydro-4H-thiazolo[4,5-d]pyrimidin-7-ones (2) and 7-chloro-3,5-diaryl-thiazolo[4,5-d]pyrimidine-2-thiones (3) (Becan and Wagner, 2008) with diethyl sulfate and water for the replacement of the 2-thioxo group with 2-oxo. First, compounds 2 were obtained through the reaction of the corresponding, refluxing aromatic aldehyde with 4-amino-5-carboxamido-3-substituted-2,3-dihydrothiazole-2-thiones 1 (Gewald, 1966), in the presence of bases according to the earlier reported procedure (Becan and Wagner, 2008). Pyrimidine ring formation with aryl aldehydes followed by chlorination with a mixture of phosphorus pentachloride and phosphorus oxychloride gave the desired cores 2 and 3 which were further treated in boiling acetonitrile with diethyl sulfate. The obtained positively charged 2-ethyltiothiazolium salt was hydrolyzed to yield thiazolones-2. Yields of reaction were variable and were higher when R1 and R2 were not substituted. Elemental analysis, IR, 1H and 13C-NMR, and X-ray data evaluated the structure of synthesized substances. In the IR spectra of compounds 4a–4f, the two stretching bands of 6-NH group were detected in the range of 3470–3080 cm−1. These compounds showed the characteristic vibrations of the C=O group at 1690–1670 cm−1. In the 1H-NMR spectra, characteristic signal of compounds 4a–4f was one-proton singlet of 6 N–H resonated at 13.19–13.27 ppm. Aromatic protons have formed multiplet at 7.22–8.20 ppm.

The formation of chlorination products 5a–5f was indicated in the IR spectra by the disappearance of stretching bands of 6-NH group. Besides the absorption bands due to C=N and C–S–C functions, the presence of C=O functional group was marked by the appearance of bond ranging from 1690 to 1680 cm−1, which was lacked in the precursor 3. In the 1H-NMR spectra of 7-chloro derivatives 5a–5f we were observed only aromatic protons signal at 7.26–8.22 ppm. The 13C-NMR spectra of the active compounds 5a, 5b, and 5d, given in Table 1, displayed the appropriate number of resonances that exactly fit the number of carbon atoms. The most active compound 5a was recrystallized from a DMF solvent; the block-shaped crystals formed as a result were submitted to X-ray analysis. Data were collected at 100 K from a single crystal. X-ray crystallography of the most active agent 5a confirmed the chemical structure (Fig. 1). Crystallographic data for the structure are depicted in Table 2.

Table 1 13C-NMR data of compounds 5a, 5b, and 5d
Full size table
Table 2 Crystallographic data for compound 5a
Full size table

Anticancer activity assay

All synthesized compounds were submitted for testing at the NCI to evaluate the growth inhibitory effect. Five compounds 4a, 4b, 5a, 5b, and 5d were selected for a primary in vitro antitumor assay (Monks et al., 1991; Boyd and Paull, 1995; Shoemaker et al., 2002). A process beginning with the evaluation of the compound against approximately 60 different human tumor cell lines representing leukemia, melanoma, and cancers of the lung, colon, brain, breast, ovary, prostate, and kidney at 10−5 M concentration was performed. With one dose, compound 4b was devoid of cytotoxic activity (mean growth percent 99.88) and 4a was slightly active against renal cancer CAKI-1 cell line (26.76 % growth).

Compounds 5a, 5b, and 5d which possess electron-withdrawing 7-chloro substituent showed variable antitumor activity, reported as the percentage of growth of treated cells; the preliminary screening results are shown in Table 3. Compounds 5a, 5b, and 5d exhibited antiproliferative effect against cell lines of leukemia, non-small cell lung cancer, colon cancer, melanoma, ovarian cancer, and renal cancer. It is worth to mention that substance 5a showed noticeable cytotoxic activity against renal cancer (UO-31), melanoma (MALME-3M), and non-small cell lung cancer (NCI-H522) while compound 5b was most effective against the last one. This limited data indicate that the replacement of the 7-oxo group with the small, non-polar chloro substituent substantially increased anticancer activity. Remarkable low growth percent values against a minimum number of cell lines (mean growth) was obtained only for compound 5a which was approved for the further screening test to evaluate the growth inhibition (GI), and cytostatic and cytotoxic effects. The selected compound was additionally evaluated at tenfold dilution of five different concentrations, from 10−4 to 10−8 M on approximately 60 human tumor cell lines panels. Three different dose–response parameters, GI50, TGI, and LC50, were calculated for each cell line. GI50 is the molar concentration of the compound required for half GI. Total growth inhibition (TGI) is the molar concentration of the compound resulting in TGI; TGI signifies the cytostatic effect. LC50 is the molar concentration of the compound resulting in a 50 % death of the initial cells; LC50 signifies the cytotoxic effect. The overview of these parameters of compound 5a is reported in Table 4 and compared with log GI50 values of thioguanine (TG), the NCI standard anticancer agent. The log GI50 values lower than −5 showed a notable activity level. It can be noticed that compound 5a proved to be very sensitive toward non-small cell lung cancer NCI-H522 and renal cancer UO-31 log GI50 −5.91 and −5.88, respectively, (MG_MID: log GI50 −5.1, log TGI −4.4, log LC50 −4.09). GI of most cell lines of standard TG is higher than that showed by investigated compound 5a; but against the following cell lines: K-562, NCI-H322M, NCI-H522, SW-620, U251, SK-MEL-28, IGROV1, A498, and HS 578T, compound 5a was more active than TG. TG is a guanine analog and thiazolo[4,5-d]pyrimidines can be considered as 7-thio analogs of the purine bases guanine and adenine. Thiazolo[4,5-d]pyrimidine derivatives may interfere with the synthesis of guanine nucleotides as antimetabolites.

Table 3 Anticancer activity as growth % in concentration 10−5 M for the compounds 5a, 5b, and 5d
Full size table
Table 4 The result of the in vitro anticancer activity of compound 5a against 60 human cancer cell lines
Full size table

Experimental

Chemistry

Melting points were determined on a Boethius apparatus and were uncorrected. Elemental analyses for the synthesized compounds were performed on a Perkin Elmer 2400 (Waltham, MA, USA) analyzer, and results within ±0.4 % of the theoretical values were obtained for the new compounds. 1H-NMR and 13C-NMR spectra were acquired in d 6-DMSO on a Bruker ARX 300 MHz (Bruker Analytic, Karlsruhe, Germany; Bruker AG, Fallanden, Switzerland) instrument. Tetramethylsilane was used as the internal standard and all chemical shift values were expressed in parts per million (δ, ppm). IR spectra were recorded on a Specord M80 spectrometer using KBr pellets. X-Ray Crystallography: the data were collected using the Bruker KAPPA APEXII ULTRA controlled by APEXII software. Reaction progress and the purity of the obtained compounds were monitored by thin-layer chromatography on Merck silica gel plates (Merck F254, Darmstadt, Germany) using the solvent system dichloromethane: 1-propanol (10:1) for elution. Iodine was used as a developing agent. The chemicals and reagents for syntheses were obtained from Alfa Aesar (Karlsruhe, Germany), Chempur (Piekary Sl. Poland), and Sigma-Aldrich (Steinheim, Germany). Starting compounds are synthesized according to the literature (Gewald et al., 1966; Becan and Wagner, 2008).

General procedures for the synthesis of compounds 4a–4f and 5a–5f

To a solution of appropriate compound 2 or 3 (10 mmol) in acetonitrile (20 ml), diethyl sulfate (4.62 g, 30 mmol) was added, and the reaction mixture was heated under reflux for 1 h at 130 °C. After cooling, 100 ml of water was added and the reaction mixture was refluxed with stirring for 2 h during which the product was precipitated. The solid was filtered and suspended in a hot mixture of methanol and 5 % NaHCO3. The reaction mixture was allowed to cool, and the crude product was filtered and crystallized from appropriate solvent.

3,5-Diphenyl-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4a)

IR (KBr) cm−1: 3450, 3080 (NH), 1680 (C=O), 1530 (C=N), 1260 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.42–7.93 (m, 10H, arom.), 13.19 (s, 1H, NH). Anal. Calcd for C17H11N3O2S: C, 63.54; H, 3.45; N, 13.08. Found: C, 63.44; H, 3.52; N, 13.27.

5-(4-Chlorophenyl)-3-phenyl-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4b)

IR (KBr) cm−1: 3450, 3090 (NH), 1670 (C=O), 1590 (C=N), 1230 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.51–7.94 (m, 9H, arom.), 13.22 (s, 1H, NH). Anal. Calcd for C17H10ClN3O2S: C, 57.39; H, 2.83; N, 11.81. Found: C, 57.56; H, 3.01; N, 11.97.

5-(2-Chlorophenyl)-3-phenyl-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4c)

IR (KBr) cm−1: 3470, 3080 (NH), 1680 (C=O), 1590 (C=N), 1260 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.34–7.99 (m, 9H, arom.), 13.27 (s, 1H, NH). Anal. Calcd for C17H10ClN3O2S: C, 57.39; H, 2.83; N, 11.81. Found: C, 57.59; H, 2.87; N, 11.85.

5-(4-Fluorophenyl)-3-phenyl-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4d)

IR (KBr) cm−1: 3450, 3090 (NH), 1680 (C=O), 1610 (C=N), 1240 (C–S–C), 770 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.31–8.20 (m, 9H, arom.), 13.20 (s, 1H, NH). Anal. Calcd for C17H10FN3O2S: C, 60.17; H, 2.97; N, 12.38. Found: C, 59.98; H, 3.03; N, 12.41.

3,5-Bis(4-fluorophenyl)-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4e)

IR (KBr) cm−1: 3470, 3090 (NH), 1690 (C=O), 1570 (C=N), 1240 (C–S–C), 780 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.22–8.03 (m, 8H, arom.), 13.21 (s, 1H, NH). Anal. Calcd for C17H9FN3O2S: C, 57.14; H, 2.54; N, 11.76. Found: C, 57.31; H, 2.55; N, 11.94.

3-(4-Bromophenyl)-5-phenyl-6H-thiazolo[4,5-d]pyrimidine-2,7-dione (4f)

IR (KBr) cm−1: 3450, 3080 (NH), 1680 (C=O), 1590 (C=N), 1260 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.45–8.16 (m, 9H, arom.), 13.19 (s, 1H, NH). Anal. Calcd for C17H10BrN3O2S: C, 51.01; H, 2.52; N, 10.50. Found: C, 51.14; H, 2.60; N, 10.66.

7-Chloro-3,5-diphenyl-thiazolo[4,5-d]pyrimidin-2-one (5a)

IR (KBr) cm−1: 1680 (C=O), 1590 (C=N), 1260 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.46–8.13 (m, 10H, arom.). Anal. Calcd for C17H10ClN3OS: C, 60.09; H, 2.97; N, 12.37. Found: C, 59.98; H, 3.01; N, 12.38.

7-Chloro-5-(4-chlorophenyl)-3-phenyl-thiazolo[4,5-d]pyrimidin-2-one (5b)

IR (KBr) cm−1: 1680 (C=O), 1560 (C=N), 1230 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.52–8.11 (m, 9H, arom.). Anal. Calcd for C17H9Cl2N3OS: C, 54.56; H, 2.42; N, 11.23. Found: C, 54.60; H, 2.49; N, 11.29.

7-Chloro-5-(2-chlorophenyl)-3-phenyl-thiazolo[4,5-d]pyrimidin-2-one (5c)

IR (KBr) cm−1: 1690 (C=O), 1570 (C=N), 1250 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.52–8.11 (m, 9H, arom.). Anal. Calcd for C17H9Cl2N3OS: C, 54.56; H, 2.42; N, 11.23. Found: C, 54.65; H, 2.50; N, 11.33.

7-Chloro-5-(4-fluorophenyl)-3-phenyl-thiazolo[4,5-d]pyrimidin-2-one (5d)

IR (KBr) cm−1: 1690 (C=O), 1600 (C=N), 1240 (C–S–C), 760 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.27–8.15 (m, 9H, arom.). Anal. Calcd for C17H9ClN3OS: C, 57.10; H, 2.73; N, 11.75. Found: C, 57.21; H, 2.86; N, 11.83.

7-Chloro-3,5-bis(4-fluorophenyl)thiazolo[4,5-d]pyrimidin-2-one (5e)

IR (KBr) cm−1: 1690 (C = O), 1590 (C = N), 1250 (C–S–C), 770 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.26–8.22 (m, 8H, arom.). Anal. Calcd for C17H8ClF2N3OS: C, 54.34; H, 2.15; N, 11.18. Found: C, 54.42; H, 2.20; N, 11.26.

3-(4-Bromophenyl)-7-chloro-5-phenyl-thiazolo[4,5-d]pyrimidin-2-one (5f)

IR (KBr) cm−1: 1680 (C=O), 1560 (C=N), 1250 (C–S–C), 770 (phenyl). 1H-NMR (d 6-DMSO) δ: 7.33–8.16 (m, 8H, arom.). Anal. Calcd for C17H9BrClN3OS: C, 48.77; H, 2.17; N, 10.04. Found: C, 48.91; H, 2.25; N, 10.12.

Characteristic data of the new compounds are depicted in Table 5.

Table 5 Characterization data of compounds 4a–4f and 5a–5f
Full size table

Antitumor in vitro screening

The antitumor studies were performed at the NCI (Bethesda, MD, USA). The test agents were preliminarily evaluated in the full panel of 60 different human tumor cell lines derived from nine cancer diseases, namely leukemia, melanoma, non-small cell lung c., colon c., brain c., breast c., ovarian c., prostate c., and renal c., at single 10−5 M concentration. Compounds with significant GI are evaluated at five different concentrations ranging from 10−4 to 10−8 M. The percent growth was evaluated versus controls not treated with tested compounds. Preparation of the tested compounds and the sulforhodamine B (SRB) protein assay which was used to estimate cell viability of growth were described previously (Becan and Wagner, 2008; Monks et al., 1991; Boyd and Paull, 1995; Shoemaker et al., 2002).