Synthesis of new bisaryl cyclopentathiophene and thieno-cyclopentoxazolidine derivatives as potential cytotoxic agents

Abstract

The synthesis of new bisaryl thienocyclopentoxazolidine derivatives was achieved through a Suzuki cross-coupling procedure with the aim to enhance the previously reported cytotoxicity of the series. The biological activity, evaluated in the NCI's in vitro human disease-oriented tumor cell line screening panel, was however partially conserved by the pharmacomodulations.

Keywords::

• Thiophene
• cyclopentane
• oxazolidine
• Suzuki cross-coupling
• cytotoxicity

Introduction

We have previously reported the synthesis and potent cytotoxicity measured in the NCI's in vitro human disease-oriented tumor cell line screening panel of some cyclopenta[c]thiophene derivatives like compounds 1 and 3, especially against leukaemia cell lines (Figure 1)1-5. Some of them exerted further an in vivo antitumor activity assessed in the hollow fiber assay and standard xenograft testing developed by the NCI[6]. In a view to enhance this activity, we have performed further pharmacomodulations in the series and herein we report the study of the replacement of the thiophene bromine atoms of 1,3 by various aromatic rings, using Suzuki cross-coupling.

Figure 1 Structure of cytotoxic cyclopentathiophene derivatives 1–3.

Materials and methods

Chemistry

Instrumentation

Melting points were determined on a Kofler melting point apparatus and are uncorrected. IR spectra were recorded on a Genesis series FTIR spectrometer using KBr pellets. The ¹H (400 MHz) and ¹³C (100 MHz) NMR spectra were obtained on a Jeol Lambda 400 spectrometer using DMSO-d₆ or CDCl₃ as solvent and TMS as internal standard. The chemical shifts (δ) are reported in ppm, and the coupling constants are in Hertz. Electron impact mass spectra (EIMS) were obtained using a Jeol JMS GCMate spectrometer. Reactions were monitored by thin-layer chromatography (TLC) using 0.2 mm Polygram Sil silica gel G/UV 254 precoated plates with visualization by irradiation with a short-wavelength UV light. Silica gel chromatography was performed using 63–200 mM Kieselgel Merck 60 silica gel.

Synthesis

General procedure for the Suzuki cross-coupling reaction

To a solution of 1 mmol of the starting material (3 or 8) in DMF (15 mL), boronic acid (4 mmol), [1,1′-bis(diphenylphosphino)ferrocene]dichloropalladium (160 mg, 0.2 mmol) and triethylamine (2mL, 15 mmol) were successively added under argon atmosphere. The mixture was stirred at 65°C for 12 h. 150 mL of distilled water and 30 mL of ethyl acetate were then added. The organic layer was separated, washed three times with distilled water and dried over MgSO₄. After filtration, the solvent was removed under reduced pressure. The crude bisaryl compounds thus obtained were purified by column chromatography on silica gel (CH₂Cl₂/MeOH 99:1).

(+/−) 2,2,2-trifluoro-N-(6-oxo-1,3-diphenyl-5,6-dihydro-4H-cyclopenta[c]thien-4-yl) acetamide (12)

Yield: 10%. Mp >230°C. IR (cm^{− 1}): 3270 (NH), 1698 (CO), 3100, 1560, 1208, 1179, 697. ¹H NMR (DMSO-d₆) δ: 9.95 (d, J = 8.5 Hz, 1H, NH), 8.01 (d, J = 7.4 Hz, 2H, H-2′ and H-6′), 7.45 (m, 8H, H-aromatics), 5.37 (dt, J = 3.5, 8.5 Hz, 1H, H-4), 3.55 (dd, J = 8.5, 18.3 Hz, 1H, H-5b), 2.82 (dd, J = 3.5, 18.3 Hz, 1H, H-5a). ¹³C NMR (DMSO-d₆) δ: 193.90, 155.50 (q, J = 36.2 Hz), 148.51, 143.02, 138.97, 135.01, 131.10, 131.06, 129.60, 128.90, 128.86, 128.35, 127.59, 126.86, 115.53 (q, J = 287.0 Hz), 50.80, 43.75. MS (EI⁺) m/z: 401.4 (M⁺).

(+/−) 2,2,2-trifluoro-N-(6-oxo-1,3-di-2-thienyl-5,6-dihydro-4H-cyclopenta[c]thien-4-yl) acetamide (13)

Yield: 35%. Mp >230°C. IR (cm^{− 1}): 3283 (NH), 1698 (CO), 3099, 1550, 1207, 1180, 695. ¹H NMR (DMSO-d₆) δ: 10.10 (d, J = 8.3 Hz, 1H, NH), 8.01 (d, J = 4.2 Hz, 1H, H-3′), 7.75 (d, J = 4.2 Hz, 1H, H-5′), 7.69 (d, J = 4.3 Hz, 1H, H-5″), 7.29 (d, J = 4.3 Hz, 1H, H-3″), 7.20 (t, J = 4.2 Hz, 1H, H-4′), 7.14 (t, J = 4.3 Hz, 1H, H-4″), 5.56 (dt, J = 2.2, 8.3 Hz, 1H, H-4), 3.62 (dd, J = 8.3, 19.0 Hz, 1H, H-5b), 2.80 (dd, J = 2.2, 19.0 Hz, 1H, H-5a). ¹³C NMR (DMSO-d₆) δ: 193.88, 155.85 (q, J = 37.0), 147.01, 138.03, 134.84, 132.91, 132.44, 129.47, 129.15, 128.70, 128.34, 127.83, 127.51, 126.16, 118.98 (q, J = 286.3 Hz), 51.27, 43.96. HRMS: calculated (412.9831); found (412,9826).

(+/−) 2,2,2-trifluoro-N-(6-oxo-1,3-di-2-furyl-5,6-dihydro-4H-cyclopenta[c]thien-4-yl) acetamide (14)

Yield: 8%. Mp >230°C. IR (cm^{− 1}): 3300 (NH), 1700 (CO), 3090, 1550, 1200, 1180, 695. ¹H NMR (DMSO-d₆) δ: 10.07 (d, J = 8.3 Hz, 1H, NH), 7.88 (d, J = 2.4 Hz, 1H, H-5′), 7.74 (d, J = 4.3 Hz, 1H, H-5″), 7.65 (d, J = 3.7 Hz, 1H, H-3′), 6.74 (dd, J = 2.4, 3.7 Hz, 1H, H-4′), 6.66 (dd, J = 3.3, 4,3 Hz, 1H, H-4″), 6.61 (d, J = 3.3 Hz, 1H, H-3″), 5.60 (dt, J = 2.4, 8,8 Hz, 1H, H-4), 3.58 (dd, J = 8.8, 18,6 Hz, 1H, H-5b), 3.00 (dd, J = 2.4, 18.6 Hz, 1H, H-5a). ¹³C NMR (DMSO-d₆) δ: 193.98, 155.55 (q, J = 37.0), 146.61, 140.33, 137.91, 136.44, 132.83, 132.11, 131.40, 131.10, 123.53, 123.32, 121.61, 120.26, 118.99 (q, J = 286.3 Hz), 51.29, 43.09. MS (EI⁺) m/z: 381.1 (M⁺).

(+/−) 6-amino-1,3-di-2-thienyl-5,6-dihydro-4H-cyclopenta[c]thiophen-4-one hydro-chloride (15)

A suspension of 13 (200 mg) in a mixture of ethanol (5 mL) and 6N aqueous solution of hydrochloric acid (15 mL) was refluxed for 7 days. The mixture was then evaporated to dryness under reduced pressure. The residue was then thoroughly washed with acetone, filtered and dried to give fluorescing red crystals. Yield: 20%. Mp >230°C. IR (cm^{− 1}): 3428 (NH), 1703 (CO), 2922, 1083, 798, 697. ¹H NMR (DMSO-d₆) δ: 8.75 (s, 3H, NH₃), 7.60 (m, 6H, H-aromatics), 5.56 (m, 1H, H-6c), 3.62 (m, 1H, H-5b), 2.94 (dd, J = 2.2, 18.9 Hz, 1H, H-5a). ¹³C NMR (DMSO-d₆) δ: 192.47, 144.59, 137.39, 135.77, 133.71, 132.67, 131.93, 129.87, 129.45, 128.96, 128.77, 127.96, 127.17, 49.01, 43.82. MS (EI⁺) m/z: 381.1 (M⁺ -HCl).

(+/−)-cis 4,6-dithien-2-yl-3a,7a-dihydro-2H-thieno[3′,4′:4,5]cyclopenta[1,2-d][1,3] oxazole-2,7(3H)-dione (16)

Yield: 35%. Mp >230°C. IR (cm^{− 1}): 3433 (NH), 1755 (CO), 1702 (CO), 2924, 1480, 1261, 1091, 697. ¹H NMR (DMSO-d₆) δ: 8.84 (laberge, 1H, NH), 8.01 (d, J = 3.7 Hz, 1H, H-3′), 7.80 (d, J = 4.9 Hz, 1H, H-5″), 7.73 (d, J = 4.9 Hz, 1H, H-3″), 7.49 (d, J = 3.7 Hz, 1H, H-5′), 7.22 (t, J = 4,9 Hz, 1H, H-4″), 7.19 (t, J = 3.7 Hz, 1H, H-4′), 5.45 (d, J = 7.4 Hz, 1H, H-3a), 5.34 (d, J = 7.4 Hz, 1H, H-7a). ¹³C NMR (DMSO-d₆) δ: 189.86, 157.58, 147.02, 138.60, 134.60, 132.75, 132.16, 130.29, 129.76, 128.88, 128.80, 128.47, 127.68, 126.61, 83.66, 50.35. HRMS: calculated (358.9744), found (358.9734).

(+/−)-cis 4,6-dithien-3-yl-3a,7a-dihydro-2H-thieno[3′,4′:4,5]cyclopenta[1,2-d][1,3] oxazole-2,7(3H)-dione (17)

Yield: 31%. Mp >230°C. IR (cm^{− 1}): 3420 (NH), 1755 (CO), 1693 (CO), 2920, 1490, 1342, 1259, 1064, 797. ¹H NMR (DMSO-d₆) δ: 8.92 (laberge, 1H, NH), 8.62 (d, J = 2.9 Hz, 1H, H-2′), 7.98 (d, J = 2.8 Hz, 1H, H-2″), 7.83 (dd, J = 2.8, 5.1 Hz, 1H, H-4″), 7.80 (dd, J = 2.9, 5.1 Hz, 1H, H-4′), 7.75 (d, J = 5.1 Hz, 1H, H-5′), 7.52 (d, J = 5.1 Hz, 1H, H-5″), 5.53 (d, J = 7.6 Hz, 1H, H-3a), 5.47 (d, = 7.6 Hz, 1H, H-7a). ¹³C NMR (DMSO-d₆) δ: 190.66, 157.92, 147.19, 140.62, 135.06, 132.17, 131.41, 130.60, 128.48, 128.39, 127.09, 126.90, 126.65, 123.61, 83.93, 50.47. HRMS: calculated (358.9744), found (358.9740).

(+/−)-cis 4,6-bis(5-bromo-2-thienyl)-3a,7a-dihydro-2H-thieno[3′,4′:4,5]cyclopenta[1,2-d] [1,3]oxazole-2,7(3H)-dione (18)

A solution of 16 (360 mg, 1 mmol) in dichoromethane (20 mL) was treated with 1 M solution of bromine in dichloromethane (2.2 mL, 2.2 mmol) at room temperature. The reaction mixture was refluxed for 30 min. and then quenched with 5% Na₂S₂O₅ aqueous solution. The organic layer was separated, washed twice with brine, dried over CaCl₂ and filtered. The solvent was removed under reduced pressure. The compound so obtained was purified by column chromatography on silica gel (CH₂Cl₂/MeOH 99:1). Yield: 85%. Mp >230°C. IR (cm^{− 1}): 3435 (NH), 1752 (CO), 1709 (CO), 2925, 1579, 1425, 1096, 796. ¹H NMR (DMSO-d₆) δ: 8.83 (laberge, 1H, NH), 7.72 (d, J = 4.0 Hz, 1H, H-3′), 7.34 (d, J = 4.0 Hz, 1H, H-4′), 7.32 (d, J = 3.9 Hz, 1H, H-3″), 7.29 (d, J = 3.9 Hz, 1H, H-4″), 5.43 (d, J = 7.8 Hz, 1H, H-3a), 5.31 (d, J = 7.8 Hz, 1H, H-7a). ¹³C NMR (DMSO-d₆) δ: 190.00, 157.70, 147.94, 137.53, 135.23, 134.34, 133.84, 132.21, 132.18, 130.34, 127.51, 127.43, 116.08, 113.04, 83.53, 50.20. MS (EI⁺) m/z: 514.7 (M⁺-2), 516.7 (M⁺), 518.7 (M⁺+2).

(+/−)-cis 4,6-bis(2,5-dibromo-3-thienyl)-3a,7a-dihydro-2H-thieno[3′,4′:4,5]cyclopenta [1,2-d][1,3]oxazole-2,7(3H)-dione (19)

A solution of 17 (360 mg, 1 mmol) in dichoromethane (20 mL) was treated with 1 M solution of bromine in dichloromethane (4.4 mL, 4.4 mmol) at room temperature. The reaction mixture was refluxed for 45 min., and then quenched with 5% Na₂S₂O₅ aqueous solution. The organic layer was separated, washed twice with brine, dried over CaCl₂ and filtered. The solvent was removed under reduced pressure. The compound thus obtained was purified by column chromatography on silica gel (CH₂Cl₂/MeOH 98:2). Yield: 77%. Mp >230°C. IR (cm^{− 1}): 3413 (NH), 1790 (CO), 1718 (CO), 2923, 1472, 1360, 1080, 804. ¹H NMR (DMSO-d₆) δ: 8.82 (laberge, 1H, NH), 7.78 (s, 1H, H-4′), 7.59 (s, 1H, H-4″), 5.40 (m, 2H, H-3a and H-7a). ¹³C NMR (DMSO-d₆) δ: 189.81, 157.37, 150.45, 137.62, 136.52, 132.58, 131.87, 131.81, 131.38, 128.82, 112.50, 112.27, 111.27, 110.76, 83.49, 49.82. MS (EI⁺) m/z: 670.9 (M⁺-4), 672.9 (M⁺-2), 674.9 (M⁺), 676.9 (M⁺+2), 678.9 (M⁺+4).

Cytotoxic assays

The cytotoxic activity of tested compounds was evaluated in the NCI's in vitro human disease-oriented tumor cell-line screening panel[7]. The later consists of 60 human tumor cell lines. Nine subpanels represent diverse histologies, i.e. nonsmall cell lung, renal, breast cancers, central nervous system, colon, melanoma, prostate, ovarian, and leukemia. The screening is a two-stage process, beginning with the evaluation of compounds against three-cell lines (lung: NCI-H460; breast: MCF7; CNS: SF-268) at a single dose. The results are expressed as the percent growth at 10^{− 4} M concentration. Compounds which exhibit significant growth inhibition are evaluated against the 60 cell panel at five different concentration levels. Results are evaluated in terms of specificity and potency. The cytotoxic effects of each of these compounds are expressed as the molar drug concentration required for 50% growth inhibition (GI₅₀).

Results and discussion

Chemistry

Suzuki cross-couplings were achieved starting from either compound 3 or a N-protected derivative of 1, the trifluoroacetamide 8 which constitutes an intermediate in the access to 3.The multi-step synthesis of 8 has been reported by us¹ using thiophene-3-carboxaldehyde 4 as starting material that involved a Rodionov-Johnson reaction to give the β-aryl-β-aminoacid 5 which after N-protection and bromination at the alpha positions of thiophene and Friedel-Crafts cyclisation led to the trifluoroacetylamino-cyclopenta[c]thiophenone 8 as a racemic mixture of (+) and ( − ) stereoisomers (Scheme 1). Synthesis of 3 was further diastereoselectively achieved through a sequence of steps involving first a selective mono trans bromination of the free alicyclic methylene group followed by ring closure in alkaline conditions that lead to the cis trifluoromethyloxazole derivative 10; Cleavage of which under acidic conditions retained the cis configuration of the hydroxyl and amino groups (compound 11). Final treatment of the latter with triphosgene in toluene afforded the attempted racemic cis isomers of thienocyclopentoxazolidinone 3.

Scheme 1 Synthesis of compounds 3,8. Reagents: (i) AcONH₄, (CH₂)₂CO₂H, EtOH; (ii) TFA₂O, Et₂O; (iii) Br₂, CH₂Cl₂; (iv) SOCl₂; (v) AlCl₃, CH₂Cl₂; (vi) Br₂, AcOH; (vii) Na₂CO₃, Me₂CO; (viii) HCl gas, Me₂CO; (ix) ambient air; (x) (CCl₃O)₂CO, toluene.

The Suzuki reaction was realized starting from 8 in DMF by treatment with phenyl, 2-thienyl or 2-furylboronic acids (3 or 4 eq) in the presence of TEA and [1,1′-bis(diphenylphosphino)ferrocene]dichloropalladium as catalyst according to the experimental conditions established in our lab (Scheme 2)[8]. It led to bis aryl cyclopentathiophenones 12–14 respectively with poor yields. Variations of the experimental conditions didn't allow the synthesis of single arylsubstituted derivatives.

Scheme 2 Synthesis of compounds 12–14. Reagents: (i) ArB(OH)₂, PdCl₂(dppf), TEA, DMF.

As with our observation with polyaromatic cyclopentatrifluoroacetamides, the N-protective group of 12–14 was difficult to cleave, the only deprotection achieved was of 13, after seven days of hydrochloric acid hydrolysis, to its ammonium salt 15 with 20% yield (Scheme 3).

Scheme 3 Synthesis of compound 15. Reagents: (i) HCl 6N.

Similarly oxazolidinone 3 treated with 2- or 3-thienylboronic acids, led to the corresponding bisaryl derivatives 16 and 17 respectively in moderate yield (Scheme 4). In view to attempt to recover the benefit of the bromine atoms in the cytotoxicity, compounds 16 and 17 were further treated with bromine in refluxing dichloromethane to yield the bis monobromo 18 and dibromo 19 derivatives respectively (Scheme 5). All these bisaryl oxazolidinones retained the cis configuration of their starting material 3.

Scheme 4 Synthesis of compounds 16–17. Reagents: (i) ArB(OH)₂, PdCl₂(dppf), TEA, DMF.

Scheme 5 Synthesis of compounds 18–19. Reagents: (i) Br₂, CH₂Cl₂.

Biology

Compounds 15–19, synthesized by Suzuki cross-coupling, were first evaluated in the NCI's three-cell line one-dose primary anticancer assay. The results, expressed as percent growth at 10^{− 4} M concentration, are summarized in Table I. The cytotoxicity dramatically decreased in this new series since only the ammonium salt 15 showed a weak activity especially against the lung cell-line NCI H-460 (percent growth = 19%), whereas the oxazolidinones, brominated or not, were totally devoid of such activity.

Table I.  Cytotoxicity of compounds 15–19 in the NCI's three-cell line one-dose primary anticancer assay, expressed as the percent growth at 10⁻⁴ M concentration.

Compd	Lung NCI-H460	Breast MCF7	CNS SF-268
15	19	60	33
16	46	53	70
17	80	74	113
18	112	63	127
19	114	76	128

On the basis of this preliminary test, 15 was evaluated in the in vitro human disease-oriented tumor cell line screening panel.⁶ The log GI₅₀ values (GI₅₀ being the molar drug concentration required for half growth inhibition) obtained with selected cell lines, along with the mean graph midpoint (MGM) values, are summarized in Table II. The MGM is based on a calculation of the average log GI₅₀ for all of the cell lines tested (approximately 60) in which GI₅₀ values below and above the test range (10^{− 4}–10^{− 8} M) are taken as the minimum (10^{− 8} M) and maximum (10^{− 4} M) drug concentrations used in the screening test.

Table II.  Cytotoxicity of compounds 1,3,15 in the NCI's in vitro human disease-oriented tumor cell line screening expressed as the log molar drug concentration required for 50% growth inhibition (log GI₅₀)

Compd	Leukemia RPMI 8226	Lung NCI- H522	Colon SW-620	CNS SF-539	Melanoma UACC-257	Ovarian ovcar-4	Renal CAKI-1	Prostate PC-3	Breast MDA-231	MGM^a
1	− 6.53	− 5.81	− 5.50	− 5.61	− 7.23	− 5.71	− 5.76	− 5.13	− 5.68	− 5.46
3	− 5.37	− 6.61	− 5.68	− 5.11	− 5.74	− 5.78	− 6.47	− 5.02	− 5.74	− 5.53
15	− 4.91	− 4.65	− 5.04	− 4.67	− 4.28	− 4.37	− 4.48	− 4.51	− 4.55	− 4.61

^a Mean Graph Midpoint for all human cancer cell lines tested.

These results indicated a weak cytotoxicity for 15 with a MGM log GI₅₀ value of − 4.61, corresponding to a MGM GI₅₀ value of 24 μM, ie approximatively 8 fold less active than 1 (3.5 μM) and 3 (3 μM). This study confirms again the crucial role played by the dibromothiophene moiety in the cytotoxicity of the cyclopentathiophenone series. The latter however allows further numerous pharmacomodulations which are currently under investigation.