Synthesis and Biological Properties of Bioreductively Targeted Nitrothienyl Prodrugs of Combretastatin A-4

2886

Synthesis and biological properties of bioreductively targeted nitrothienyl prodrugs of combretastatin A-4

Peter Thomson,1 Matthew A. Naylor,1 cytoskeleton of the endothelial cells lining the tumor Steven A. Everett,1 Michael R.L. Stratford,1 vasculature (6–8). When this tubulin structure is disrupted, Gemma Lewis,1 Sally Hill,1 Kantilal B. Patel,1 the endothelial cells change shape from flat to round, Peter Wardman,1 and Peter D. Davis2 impeding blood flow through the capillary, starving the tumor of nutrients, and causing tumor cell death (8–10). 1University of Oxford, Gray Cancer Institute, Mount Vernon Preclinical animal model studies and subsequent clinical Hospital, Northwood, Middlesex, United Kingdom and 2 trials have shown that the drug drastically reduces blood Angiogene Pharmaceuticals Ltd., The Magdalen Centre, Oxford flow in tumors (11). Phase I human cancer clinical trials of Science Park, Oxon, United Kingdom the sodium CA4 phosphate prodrug (2; Fig. 1) have been successfully completed (12, 13) and the drug is currently Abstract undergoing phase II trials. 1 Nitrothienylprop-2-yl ether formation on the 3¶-phenolic The discovery of prompted the synthesis of many position of combretastatin A-4 (1) abolishes the cytotox- structural variations as improved vascular targeting agents icity and tubulin polymerization-inhibitory effects of the (14–17). However, the combretastatins also have powerful drug. 5-Nitrothiophene derivatives of 1 were synthesized antiproliferative activity against cancer cell lines in vitro followingmodel kinetic studies with analogouscoumarin (3, 14, 18–25), and it is thought that this is due to an derivatives, and of these, compound 13 represents a antimitotic action brought about by inhibition of tubulin promisingnew lead in bioreductively targetedcytotoxic polymerization. In contrast to the antivascular activity, this anticancer therapies. In this compound, optimized gem- antimitotic activity requires prolonged exposure of the cells dimethyl A-carbon substitution enhances both the aerobic to the compounds. Therefore, it is probable that the latter metabolic stability and the efficiency of hypoxia-mediated activity is generally not expressed in vivo because rapid drugrelease. Only the gem-substituted derivative 13 elimination of the compounds preclude this exposure at released 1 under anoxia in either in vitro whole-cell nontoxic doses. We have sought to exploit tumor hypoxia experiments or supersomal suspensions. The rate of and examine whether combretastatin analogues delivered release of 1 from the radical anions of these prodrugs is by a hypoxia-driven fragmentation strategy may offer the enhanced by greater methyl substitution on the A-carbon. potential to deliver prolonged tumor exposure that may be Cellular and supersomal studies showed that this antimitotic while minimizing host toxicity. This strategy A-substitution pattern controls the useful range of oxygen may also, via a bystander effect, retain vascular targeting concentrations over which 1 can be effectively released activity. It is thus the object of this study to synthesize by the prodrug. [Mol Cancer Ther 2006;5(11):2886–94] and evaluate prodrugs that on bioreductive activation break down to release an antimitotic stilbene compound 1 Introduction (i.e., CA4; ; Fig. 2). Although the phenolic 3¶-hydroxyl group of 1 is not Combretastatin A-4 (CA4; 1; Fig. 1) is an antineoplastic and essential for tubulin binding, large bulky groups in this vascular targeting stilbene that was isolated from the South region inhibit binding (9). This position was therefore Combretum caffrum African bush willow tree (1–4). This considered a good candidate for prodrug derivatization. new class of therapeutic compounds are known primarily Bioreductive targeting of phenolic compounds by indole- as vascular targeting agents, which have potential use in quinones has been shown, and much work has been disease conditions or pathologies, such as cancer, where an documented on the factors that control this process abnormal growth of blood vessels is an essential compo- (26–28). Nitroaromatic compounds have also long been nent to the disease and its progression (5). The mechanism known to exhibit similar redox properties and have been of action is through microtubule disruption, affecting the studied as potential bioreductively activated prodrug delivery systems for a variety of drugs (29–33). However, despite the body of work about compounds that break Received 7/21/06; revised 9/1/06; accepted 9/25/06. down selectively under low oxygen tensions to release an The costs of publication of this article were defrayed in part by the anticancer agent, no such compound is yet in clinical use. payment of page charges. This article must therefore be hereby marked Several problems have been encountered, including a lack advertisement in accordance with 18 U.S.C. Section 1734 solely to of stability of the prodrugs toward nonbioreductive indicate this fact. processes. Thus, carbonate-linked Taxol prodrugs were Requests for reprints: Peter Thomson, Gray Cancer Institute, University of Oxford, Mount Vernon Hospital, Northwood, Middlesex, United Kingdom reported to be unstable toward enzymatic hydrolysis in HA6 2JR. Phone: 441923828611. E-mail: [email protected] cellular assays, thereby releasing Taxol by a nonreductive Copyright C 2006 American Association for Cancer Research. process (34). Hypoxia-activated nitroheterocyclic phos- doi:10.1158/1535-7163.MCT-06-0429 phoramidates have also been reported, which were

Mol Cancer Ther 2006;5(11). November 2006

0.1% trifluoroacetic acid (TFA); B: 100% acetonitrile; gradient 20% to 50% or 50% to 100% B, 4 minutes, or isocratically at 100% acetonitrile, at a flow rate of 0.5 mL/ min. Reverse-phase chromatography was conducted on Varian C18 Bond Elut straight barrel columns (sorbent mass, 1 g; volume, 6 mL; and particle size, 40 Am). Analytic TLC was done on precoated silica gel plates (60 F254, 0.2 mm Figure 1. Structures of CA4 (1) and its phosphate prodrug (2). thick, VWR). Visualization of the plates was accomplished using UV light and/or potassium permanganate staining. Solutions in organic solvents were dried by standard unstable in vivo, displaying rapid metabolism and conse- procedures, and dichloromethane, benzene, dimethylfor- quent elimination half-lives of only a few minutes (35, 36). mamide, and tetrahydrofuran were anhydrous commercial Similarly, nitroheteroaryl quaternary salts have been grades. Solvents used for chromatography were HPLC synthesized as bioreductive prodrugs of mechlorethamine, grade and obtained from Sigma-Aldrich Chemical Co. but the compounds were too unstable with regard to (Dorset, United Kingdom) 7-Hydroxy-4-methylcoumarin nonspecific release of mechlorethamine to be of use as (3), diethyl azodicarboxylate (DEAD), diisopropyl azodicar- bioreductive agents (29, 30). Thus, prodrugs showing boxylate (DIAD), and 1,1-(azodicarbonyl)dipiperidine improved stability toward nonreductive processes are (ADDP) were all obtained from Sigma-Aldrich Chemical desirable. A further consideration is the rate of release of Co. CA4 (1) was synthesized according to the procedure of the active drug under hypoxic conditions. To be effective, Pettit and Rhodes (40). 2-Hydroxymethyl-5-nitrothiophene the bioreductively activated prodrug needs to deliver the (4; ref. 41), 2-(1-hydroxyethyl)-5-nitrothiophene (6; ref. 42), drug at a rate that competes with clearance of the prodrug and ethyl 2-hydroxy-2-(5-nitrothien-2-yl)propanoate (10; and diffusion of the drug out of the solid tumor. Prodrugs ref. 43) were prepared by literature methods. that fragment faster or that fragment more efficiently at 2-Bromomethyl-5-Nitrothiophene (5) (44). Compound 4 oxygen tensions commonly found in solid tumors are (6.0 g, 38 mmol) was dissolved in dichloromethane (30 mL) desirable. and the solution was cooled to 0jC. Phosphorus tribromide 5-Nitrothiophenes have the required reduction potential (2.5 g, 50 mmol) in dichloromethane (30 mL) was then to be of use in this respect (37, 38), and we have sought to added dropwise and the solution was stirred for a further exploit this moiety for the reductive delivery of the 0.5 hour. Dichloromethane (250 mL) was added and the 1 phenolic stilbene in this manner. We report herein the solution was washed with water (2 250 mL) and brine synthesis of combretastatin nitrothiophene ether-linked (100 mL), dried, and evaporated in vacuo. The product a 1 conjugates, optionally substituted on the -carbon atom. was used without further purification (6.0 g, 72%). HNMR Substitution at this position was carried out to facilitate y (60 MHz, CDCl3) 4.61 (s, 2H), 7.02 (d, J = 4.2 Hz, 1H), 7.74 manipulation of rates of reductive elimination and meta- (d, J = 4.2 Hz, 1H) ppm. bolic stability (39). Initial compounds incorporated 7- 2-(1-Bromoethyl)-5-Nitrothiophene (7). Compound 6 (5.0 3 hydroxy-4-methylcoumarin ( ) as a model because of facile g, 29 mmol) was dissolved in dichloromethane (60 mL) detection of its release in bioreductive conjugates (28). The and the solution was cooled to 0jC. Phosphorus tribromide analogous CA4 derivatives were then synthesized for (2.5 g, 50 mmol) in dichloromethane (5 mL) was then added further chemical and biological evaluation. dropwise and the solution was stirred for a further 2 hours. Dichloromethane (150 mL) was added and the solution was Materials and Methods washed with water (2 250 mL) and brine (100 mL), dried, Chemistry and evaporated in vacuo. The residue was purified on silica Nuclear magnetic resonance (NMR) spectra were (25% ethyl acetate/hexane) to give a yellow oil (4.0 g, 58%). 1 y obtained at 500 MHz using a Bruker AVC500, at 250 H NMR (60 MHz, CDCl3) 2.11 (d, J = 6.6 Hz, 3H), 5.32 MHz using a Bruker AVC250, and at 60 MHz using a (q, J = 6.6 Hz, 1H), 7.03 (d, J = 4.2 Hz, 1H), 7.74 (d, J =4.2 Jeol MY60 spectrometer with tetramethylsilane as internal Hz, 1H) ppm. Liquid chromatography-retention time, 6.42 standard. Melting points were obtained using an Electro- minutes (TFA 50–100%); MS (m/z, %) 237 (M +, 1), 235 (M+, thermal IA9100. High-resolution mass spectra (HRMS) on 1), 156 (100), 141 (6), 125 (8), 110 (12). chromatographically homogeneous compounds were 2-Thien-2-yl-Propan-2-ol (8). 2-Acetylthiophene (8.8 g, recorded on a VG Trio 2000 mass spectrometer. Elemental 70 mmol) was dissolved in anhydrous diethyl ether (200 mL) analyses were carried out by Medac Ltd., Brunel Science Centre, Egham, Surrey, United Kingdom. Silica gel for flash chromatography was Merck Kieselgel 60H grade (230–400 mesh). High-performance liquid chromatography (HPLC)- mass spectrometry (MS) was done using a Waters Integrity System with electron effect ionization. Chromatography was carried out using a Hichrom (Theale, United Kingdom) RPB column (100 3.2 mm) with eluents A: 5% acetonitrile, Figure 2. One-electron reduction of prodrug and elimination of CA4 (1).

Mol Cancer Ther 2006;5(11). November 2006

and the solution was cooled to 0jC. MeMgBr (3.0 mol/L (d, J = 5.0 Hz, 1H), 7.80 (d, J = 5 Hz, 1H) ppm. LC-RT 3.85 + in diethyl ether, 30 mL, 90 mmol) was then added via minutes (100% CH3CN); MS (m/z, %) 471 (M , 34), 425 (14), syringe under nitrogen. The solution was warmed to 20jC 316 (79), 301 (71), 252 (93), 141 (76), 125 (100); HRMS found and stirring was continued for 3 hours. The solution was 472.1426 C24H25NO7S requires 472.1424 (M+H). poured onto ice/water (25 g) and 0.1 mol/L HCl (250 mL) 1-(4-Methoxy-3-(2-(5-Nitrothien-2-yl)Prop-2-Yloxy))- was added. The solution was extracted with diethyl ether Phenyl-2-(3,4,5-Trimethoxy)Phenyl-Z-Ethene (13). 9 (200 (3 70 mL), dried, and evaporated. The residue was mg, 1.07 mmol) was dissolved in benzene (2.5 mL) together purified on silica (20% ethyl acetate/hexane) and then a with 1 (320 mg, 1 mmol) and ADDP (250 mg, 1 mmol) and second silica column (dichloromethane) to give a colorless the solution was maintained under argon with stirring. 1 y oil (3.5 g, 35%). H NMR (60 MHz, CDCl3) 1.65 (s, 6H), Tributylphosphine [200 mg, 1 mmol, dissolved in benzene 2.16 (s, 1H), 7.0 (m, 3H) ppm. LC-RT 3.09 minutes (TFA 50– (0.5 mL)] was then added via syringe and under argon. 100%); MS (m/z, %) 142 (M+, 12), 124 (100), 109 (93). The solution was stirred for 24 hours at 20jC and then 2-(5-Nitrothien-2-yl)Propan-2-ol (9) (45). Compound partitioned with ethyl acetate/water (100 mL) and the 8 (3.0 g, 16 mmol) was dissolved in acetic anhydride (50 organic layer was washed with brine (50 mL), dried, and mL) and the solution was cooled to 70jC. Fuming nitric evaporated. The residue was purified on silica (33% acid (1.2 mL, 20 mmol) was then added gradually with ethyl acetate/hexane) and then on a second silica column vigorous stirring. The solution was stirred for 1 hour at (dichloromethane) to give a pale yellow oil (150 mg, 31%). j j 1 y 70 C and then allowed to warm to 40 C and stirred for H NMR (500 MHz, CDCl3) 1.60 (s, 3H), 1.63 (s, 3H), 3.75 further 1 hour at this temperature. Ice/water (300 g) was (s, 3H), 3.76 (s, 3H), 3.85 (s, 3H), 3.89 (s, 3H), 6.475 (d, J =5 then added and the slurry was stirred for 30 minutes before Hz, 4H), 6.74 (s, 1H), 6.81 (s, 1H), 6.86 (d, J = 5 Hz, 1H), extracting with ethyl acetate (3 75 mL). The organic layer 7.05 (d, J = 5 Hz, 1H), 7.775 (d, J = 5 Hz, 1H) ppm. LC-RT + was washed with sodium bicarbonate (saturated, 100 mL) 4.34 minutes (100% CH3CN); MS (m/z, %) 485 (M , 43), 316 and brine (50 mL), dried, and evaporated. The residue was (100), 301 (56); HRMS found 486.1582 C25H27NO9S requires purified on silica (20% ethyl acetate/hexane) to give an 486.1581 (M+H). 1 y Ethyl 2-(2-Methoxy-5Z-[2-(3,4,5-Trimethoxyphenyl)Vinyl]- orange wax (1.5 g, 50%). H NMR (60 MHz, CDCl3) 1.67 (s, 6H), 2.1 (br, 1H), 6.88 (d, J = 4 Hz, 1H), 7.8 (d, J = 4 Hz, Phenoxy-2-(5-Nitrothien-2-yl)Propanoate (14). DIAD (128 1H) ppm. LC-RT 3.26 minutes (TFA 50–100%); MS mg, 0.63 mmol) was added dropwise to a solution of (m/z, %) 187 (M+, 8), 172 (100), 157 (9), 142 (11), 127 (13). 10 (54 mg, 0.22 mmol), 1 (100 mg, 0.32 mmol), triphenyl- 1-(4-Methoxy-3-(5-Nitrothien-2-yl)Methyloxy)Phenyl- phosphine (166 mg, 0.63 mmol), and tetrahydrofuran 2-(3,4,5-Trimethoxy)Phenyl-Z-Ethene (11). Compound 4 (500 (1 mL). The reaction was stirred for 16 hours then adsorbed mg, 3.14 mmol) was dissolved in tetrahydrofuran (5 mL) onto flash silica in vacuo. The residue was purified on silica together with triphenylphosphine (1.68 g, 6.28 mmol) and (25% ethyl acetate/hexane) then on a second silica column 1 (1.98 g, 6.28 mmol). To this solution was added DEAD (3% ethyl acetate/dichloromethane) to give a yellow oil j 1 (1.09 g, 6.28 mmol) and the solution was heated at 50 C (50 mg, 42%). TLC Rf = 0.2, 30% ethyl acetate/hexane; H y for 3 hours and evaporated to dryness and the residue NMR (250 MHz, CDCl3) 1.27 (t, J = 7.2 Hz, 3H), 1.78 was purified on silica (25% ethyl acetate/hexane) to give a (s, 3H), 3.76 (s, 6H), 3.85 (s, 3H), 3.89 (s, 3H), 4.26 (q, J =7.3 pale yellow solid (mp 88–90jC, 810 mg, 57%) after Hz, 2H), 6.49 (s, 4H), 6.85 (d, J = 8.3 Hz, 1H), 7.02 (d, J =4.3 recrystallization from ethyl acetate/hexane. 1HNMR Hz, 1H), 7.06 (m, J = 8.4 Hz, 2H), 7.82 (d, J = 4.3 Hz, 1H) y (60 MHz, CDCl3) 3.69 (s, 6H), 3.85 (s, 3H), 3.94 (s, 3H), ppm; LC-RT 5.75 minutes (TFA, 50–100%); MS (m/z, %) 5.05 (s, 2H), 6.47 (bs, 4H), 6.88 (bs, 3H), 7.24 (bs, 1H), 7.79 543 (M+, 9), 497 (8), 470(2), 316 (100), 301 (90), 283 (59), 252 (bs, 1H) ppm. LC-RT 9.98 minutes (TFA, 50–100%); MS (90), 241 (27), 226 (12), 213 (15), 197 (15), 183 (16), 168 (11), + (m/z, %) 457 (M , 85), 316 (61), 301 (47), 252 (100). Anal. C; 154 (16), 139 (11); HRMS found 544.1632 C27H29NO9S 60.47, H; 5.11, N; 2.88% C23H23NO7S requires C; 60.38, H; requires 544.1636 (M+H). 5.07, N; 3.06%. (4-Methylcoumarin-7-yl)Oxymethyl-5-Nitrothiophene (15). 1-(4-Methoxy-3-(1-(1-(5-Nitrothien-2-yl))Ethoxy))Phenyl- 3 (1.0 g, 5.68 mmol) was dissolved in chloroform (15 mL) 2-(3,4,5-Trimethoxy)Phenyl-Z-Ethene (12). DEAD (357 mg, together with silver (I) oxide (1.0 g, 4.24 mmol). 5 (1.0 g, 2.05 mmol) was added dropwise to a solution of alcohol 6 4.5 mmol) was then added in five portions over 6 hours, (55 mg, 0.32 mmol), 1 (648 mg, 2.05 mmol), triphenylphos- and the solution was stirred for a further 48 hours. The phine (288 mg, 1.10 mmol), and tetrahydrofuran (3 mL). mixture was filtered, evaporated and then dissolved in a The reaction was stirred for 16 hours at ambient temper- minimum amount of chloroform, and purified by column ature and then partitioned (ethyl acetate and brine), chromatography on silica (dichloromethane followed by a aqueous phase extracted (ethyl acetate), organic phase second column with 10% then 50% ethyl acetate/hexane) 1 washed (H2O and brine), dried (MgSO4), and concentrated to give a yellow oil (13 mg, 4%). H NMR (60 MHz, CDCl3) in vacuo. The residue was purified on silica (33% then 50% y 1.84 (s, 6H), 2.37 (s, 3H), 6.14 (s, 1H), 6.8 (m, 2H), 6.96 ethyl acetate/hexane and finally 100% ethyl acetate) to give (d, J = 4.2 Hz, 1H), 7.35 (m, 3H), 7.79 (d, J = 4.2 Hz, 1H). LC- 1 y a yellow oil (15 mg, 10%). H NMR (500 MHz, CDCl3) 1.82 RT 6.214 minutes (TFA, 50–100%); MS (m/z, %) 317 (d, J = 5.0 Hz, 3H), 3.75 (s, 6H), 3.91 (s, 3H), 3.95 (s, 3H), 5.28 (M+, 4), 271 (16), 176 (42), 142 (100). Anal. C; 56.82, H; 3.54, (q, J = 5.0 Hz, 1H), 6.50 (d, J = 5.0 Hz, 4H), 6.85 (m, 3H), 6.91 N; 4.40% C17H15NO5S requires C; 56.78, H; 3.49, N; 4.41%.

Mol Cancer Ther 2006;5(11). November 2006

2-(1-(4-Methylcoumarin-7-yl)Oxy)Ethyl-5-Nitrothio- Ethyl 2-(4-Methylcoumarin-7-yl)Oxy-2-(5-Nitrothien-2- phene (16). 3 (37 mg, 0.212 mmol) was dissolved in yl)Propanoate (18). DEAD (100 mg, 0.57 mmol) was added chloroform (1 mL) together with silver (I) oxide (50 mg, dropwise to a solution of 10 (50 mg, 0.22 mmol), 3 (120 mg, 0.212 mmol). 7 (50 mg, 0.212 mmol) was then added, and 0.68 mmol), triphenylphosphine (100 mg, 0.38 mmol), the solution was stirred for a further 48 hours. The mixture and tetrahydrofuran (2 mL). The reaction was stirred for was filtered, evaporated, and then dissolved in a minimum 72 hours and then concentrated in vacuo. The residue was amount of chloroform and purified by column chromatog- dissolved in a minimum amount of acetone and purified raphy on silica (dichloromethane followed by a second a by column chromatography on silica (dichloromethane column with dichloromethane) to give a pale yellow solid followed by a second column with 25% ethyl acetate/ recrystallized from diethyl ether (15 mg, 21%, mp hexane) to give a yellow oil (10 mg, 11%). 1HNMR j 1 y y 146–148 C). H NMR (60 MHz, CDCl3) 1.84 (s, 6H), (60 MHz, CDCl3) 1.33 (t, J = 7.2 Hz, 3H), 2.02 (s, 3H), 2.38 2.37 (s, 3H), 6.14 (s, 1H), 6.8 (m, 2H), 6.96 (d, J = 4.2 Hz, 1H), (s, 3H), 4.22 (q, J = 7.3 Hz, 2H), 6.15–7.80 (m, 7H) ppm; LC- 7.35 (m, 3H), 7.79 (d, J = 4.2 Hz, 1H). LC-RT 6.214 minutes RT 2.724 minutes (100% MeCN); MS (m/z, %) 404 (M+, 5), (TFA, 50–100%); MS (m/z, %) 331 (M +, 2), 176 (100), 357 (10), 330 (22), 283 (16), 228 (90), 213 (12), 200 (45), 176 156 (42). Anal. C; 57.76, H; 4.04, N; 4.15% C17H15NO5S (100), 154 (38); Anal. C; 56.48, H; 4.37, N; 3.44% C19H17NO7S requires C; 58.00, H; 3.95, N; 4.23%. requires C; 56.57, H; 4.25, N; 3.47%. 2-(1-Methyl-1-(4-Methylcoumarin-7-yl)Oxy)Ethyl-5- Pulse Radiolysis Nitrothiophene (17). 9 (187 mg, 1 mmol) was dissolved in Prodrug radicals were formed by reduction of the parent tetrahydrofuran (2 mL) together with triphenylphosphine nitro compounds (typically 50 Amol/L) by the 2-propanol 3 (472 mg, 1.8 mmol) and (580 mg, 3.3 mmol). DEAD radical generated radiolytically in a N2O-saturated (574 mg, 3.3 mmol) was then added and the solution was 2-propanol/water mixture (50% v/v) with 4 mmol/L heated at 105jC for 3.5 hours, cooled, and evaporated. The potassium phosphate buffer (pH 7.4–9). Experiments were residue was dissolved in a minimum amount of acetone done using a 6 MeV linear accelerator to generate an and purified by column chromatography on silica electron pulse (typically f500 ns) as described previously (dichloromethane followed by a second column with 25% (46). The absorbed radiation dose per electron pulse ethyl acetate/hexane) to give a yellow oil. This material (typically 5–35 Gy, equivalent to 3–23 Amol/L reducing was purified by preparative HPLC (10% H2O/CH3CN) radicals) was determined by the thiocyanate dosimeter. to give a white solid [13 mg, 4%, mp 156–158jC (dec.)] 1H Changes in absorbance were measured using a tungsten y NMR (60 MHz, CDCl3) 1.84 (s, 6H), 2.37 (s, 3H), 6.14 lamp and photodiode detector preceded by a single-pass (s, 1H), 6.8 (m, 2H), 6.96 (d, J = 4.2 Hz, 1H), 7.35 (m, 3H), monochromator. 7.79 (d, J = 4.2 Hz, 1H). LC-RT 2.29 minutes (100% CH3CN); Steady State ;-Radiolysis MS (m/z, %) 176 (100), 170 (88), 148 (36). Anal. C; 58.99, Prodrug solutions in 50% IPA/buffer (typically 40 Amol/L H; 4.48, N; 4.07% C17H15NO5S requires C; 59.12, H; 4.38, or below depending on solubility) were saturated with 60 N; 4.06%. N2O in gas-tight syringes and then irradiated in a Co

Figure 3. Synthesis of nitrothiophene triggers. Reagents: (i) NaBH4, MeOH, 0jC; (ii) PBr3, DCM, 0jC; (iii) MeMgBr, Et2O, 0jC; (iv)Ac2O, f.HNO3, 40jC; (v) ref. 43.

Mol Cancer Ther 2006;5(11). November 2006

Figure 4. Synthesis of combretastatin prodrugs and 7-hydroxy-4- methyl-coumarin analogues. Reagents: for 11, 12, 14 – 16, and 18: either (i) DEAD (or DIAD), triphenylphosphine, THF, or (ii)Ag2O, CHCl3 (with 5 or 7; for 13 and 17. ADDP, tributylphosphine, benzene.

source. An absorbed dose of 1 Gy = 0.62 Amol/L 2-propanol NADPH (10 mmol/L), and 2.4 mL potassium phosphate radicals in N2O-saturated 2-propanol/water mixture buffer [250 mmol/L (pH 7.4)] to give a final prodrug (50% v/v) as determined by ferricyanide reduction. A dose concentration of 5 Amol/L and incubated at 37jC. For f 1 rate of 3.9 Gy min was used, as determined by Fricke anoxic experiments, the mixture was degassed with N2 for dosimetry. 20 minutes before prodrug addition and then overgassed A HPLC with N2 during incubation. Samples (100 L) were added to HPLC analysis was done using a Waters 2695 separations acetonitrile (100 AL), mixed, and then centrifuged at 14,300 module, Waters 2996 photodiode array detector, and Waters rpm for 2 minutes before HPLC analysis. 474 fluorescence detector. Data were collected using Waters Later experiments, showing comparative release of CA4 Millennium software. The column was a C18 Hichrom RPB (1) over a range of oxygen tensions, were carried out by (100 3.2 mm) and the eluents were A: 10% acetonitrile, dissolving prodrugs in DMSO to a concentration of 90% water; B: 75% acetonitrile, 25% water. A gradient from 52 Amol/L, and 60 AL were added to a mixture of 10 AL 65% to 100% B was used over 3 minutes, with a 1-minute Supersomal P450R, 10 AL NADPH (10 mmol/L), and 1.17 hold at 100%, and the flow rate was 1 mL/min. Detection mL potassium phosphate buffer [250 mmol/L (pH 7.4)] was by absorbance at 292 nm (prodrugs) and by fluores- to give a final prodrug concentration of 2.5 Amol/L and cence (320 nm excitation and 390 nm emission; CA4). incubated at 37jC. Anoxic and hypoxic experimental Supersome Experiments mixtures were degassed with either N2, 0.02%, 0.04%, Prodrugs were dissolved in DMSO to a concentration of 0.06%, 0.1%, 0.2%, 0.3%, 0.5%, 1%, 2%, or 5% O2, or air for 625 Amol/L, and 20 AL of this stock solution were added 20 minutes before prodrug addition and then overgassed to a mixture of 60 AL Supersomal P450 reductase (P450R; with the appropriate gas during incubation. Samples Gentest, BD Biosciences, Oxford, United Kingdom), 20 AL (100 AL) were added to acetonitrile (80 AL), mixed, and

Table 1. Characterization of radical fragmentation and effector release efficiency using radiolytic production of radicals

Compound R1 R2 LG Radical % Fragmentation half-life/ms* efficiencyc

11 H H CA4 ndb 39 12 H Me CA4 1,400 49 13 Me Me CA4 130 54 14 b Me CO2Et CA4 nd 75 15 H H HMC 1,700 49 16 H Me HMC 90 57 17 Me Me HMC 2 100 18 Me CO2Et HMC 2 72

* Pulse radiolysis done in N2O-saturated 50% 2-propanol/water (pH 9.2) using a 6 MeV linear accelerator. c 1 Steady-state g-radiolysis determined in N2O-saturated 50% 2-propanol/water (pH 7.4); dose rate, 3.26 Gy min ; percentage fragmentation efficiency = G(drug) / G(Me2C OH) 100. bNot measurable.

Mol Cancer Ther 2006;5(11). November 2006

were taken at regular intervals, added to an equivalent volume of acetonitrile, mixed, and centrifuged at 14,300 rpm for 2 minutes before HPLC analysis. Tubulin Binding A tubulin binding assay kit (Totam Biologicals Ltd., Peterborough, United Kingdom) was used to determine the relative binding efficiencies of the prodrugs. Composition A (100 AL) was added to a vial containing 250 Ag bovine tubulin protein and either 2 AL DMSO for control or 2 AL of the appropriate concentration of drug in DMSO. The sample was mixed well and transferred to a spectropho- tometer cuvette in a cell holder previously equilibrated at 37jC and the absorbance was read immediately at 340 nm. Absorbance readings were continued at 2-minute intervals for 1 hour.

Results and Discussion The primary and secondary 5-nitrothienyl alcohols, 4 and 6, were prepared by sodium borohydride reduction of commercially available 5-nitrothiophene-2-carboxaldehyde and 2-acetyl-5-nitrothiophene, respectively (Fig. 3). Prepa- ration of the gem-dimethylalcohol 9 involved Grignard addition of methylmagnesium bromide to 2-acetylthiophene and then careful nitration of the intermediate tertiary alcohol 8 Figure 5. CA4 (1) production in P450R supersomes under air (A) and with fuming nitric acid and acetic anhydride. a-Hydroxy anoxia (B). Prodrugs were dissolved in DMSO to a concentration of 625 ester 10 was synthesized via a vicarious nucleophilic Amol/L, and 20 AL were added to a mixture of 60 AL Supersomal P450R substitution reaction as described by Lawrence et al. (43). (Gentest), 20 AL NADPH (10 mmol/L), and 2.4 mL potassium phosphate buffer (pH 7.4) to give a final prodrug concentration of 5 Amol/L and Two different sets of Mitsunobu conditions (47) were incubated at 37jC. For anoxic experiments, the mixture was degassed used to couple the alcohols to stilbene 1 (Fig. 4). The more with N2 for 20 min before prodrug addition and then overgassed with N2 commonly used reagent mixture of DEAD (or DIAD)/ A A during incubation. Samples (100 L) were added to acetonitrile (100 L), triphenylphosphine was used to synthesize primary and mixed, and then centrifuged at 14,300 rpm for 2 min before HPLC 11 12 analysis. Points, single observation at each time. secondary alkyl-aryl ethers, and . Surprisingly, ether- ification of the hindered a-hydroxy ester 14 was effected using the same conditions, whereas a combination of ADDP/tributylphosphine was necessary for successful then centrifuged at 14,300 rpm for 2 minutes before HPLC preparation of the hindered tertiary alkyl-aryl ether, 13. analysis. The oxygen content in the samples was measured Synthesis of the analogous coumarins (Fig. 4) was carried using an Oxy Lab fitted with an oxygen-monitoring probe out in the same manner or by treating the bromides 5 and 7, (Oxford Optronix, Oxford, United Kingdom). A549Whole-Cell Experiments Prodrugs were dissolved in DMSO to a concentration of 625 Amol/L, and 80 AL were added to 10 mL A549 cells in Eagle’s MEM (f106 cells/mL) to give a final prodrug concentration of 5 Amol/L and incubated at 37jC. For anoxic experiments, the mixture was degassed with N2 +5% CO2 for 30 minutes before prodrug addition and then overgassed with N2 during incubation. Hypoxic experimental mixtures were degassed with either 0.1% or 0.3% O2 (5% CO2, balance N2) for 30 minutes before prodrug addition and then overgassed with the appropriate gas during incubation. Samples (1 mL) were added to 100 mmol/L hydrochloric acid (0.2 mL), mixed, and then extracted (via solid-phase extraction) before HPLC analysis. Figure 6. Prodrug metabolism in oxic liver homogenates. Metabolism of Liver Metabolism 5 Amol prodrug in air was done with 0.5 mL mouse liver homogenate Metabolism of 5 Amol prodrug in air was done with (f4 mg protein by the Bradford assay) with 100 Amol/L NADPH in j 0.5 mL mouse liver homogenate [4 mg protein (Bradford 50 mmol/L potassium phosphate buffer at pH 7.4 incubated at 37 C. A A Samples were taken at regular intervals, added to an equivalent volume of assay)] with 100 mol/L NADPH in 50 mol/L potassium acetonitrile, mixed, and then centrifuged at 14,300 rpm for 2 min before phosphate buffer at pH 7.4 incubated at 37jC. Samples HPLC analysis. Points, single observation at each time.

Mol Cancer Ther 2006;5(11). November 2006

Figure 7. CA4 (1) production from 13 in A549 whole cells. ANG704 (13) was dissolved in DMSO to a concentration of 625 Amol/L, and 80 AL Figure 9. Oxygen concentration dependence of CA4 reductive release were added to 10 mL A549 cells in Eagle’s MEM (f106 cells/mL) to give a fromprodrugs 11 to 13. Prodrugs were dissolved in DMSO to a final prodrug concentration of 5 Amol/L and incubated at 37jC. For anoxic concentration of 52 Amol/L, and 60 AL were added to a mixture of 10 AL experiments, the mixture was degassed with N for 30 min before prodrug 2 Supersomal P450R, 10 AL NADPH (10 mmol/L), and 1.17 mL potassium addition and then overgassed with N during incubation. Samples (1 mL) 2 phosphate buffer [250 mmol/L (pH 7.4)] to give a final prodrug were added to 100 mmol/L hydrochloric acid (0.2 mL), mixed, and then concentration of 2.5 Amol/L and incubated at 37jC. Anoxic and hypoxic extracted (via solid-phase extraction) before HPLC analysis. Points, single experimental mixtures were degassed with either N ,0.02%,0.04%, observation at each time. 2 0.06 %, 0.1 %, 0.2 %, 0.3 %, 0.5 %, 1 %, 2%, or 5% O2, or air for 20 min before prodrug addition and then overgassed with the appropriate gas during incubation. Samples (100 AL) were added to acetonitrile synthesized by phosphorus tribromide bromination of (80 AL), mixed, and then centrifuged at 14,300 rpm for 2 min before HPLC the alcohols, with silver (I) oxide and 7-hydroxy-4- analysis. Points, mean of three replicates of one experiment; bars, SD. methylcoumarin (3). Bioreductive activation of prodrugs may involve obligate and monitoring of the prodrug radical (Eqs. 1–4) in one-electron donors, typically flavoproteins, such as microseconds and subsequent radical fragmentation or NADPH:cytochrome P450R, or two-electron reduction, reaction with O , usually in milliseconds. typically diaphorases, such as NQ01. In the present work, 2 þ ! ; ; ð Þ we have emulated reduction by one-electron processes H2 O radiation eðaqÞ OH H 1 in the hypoxic environments of solid tumors, and the þ þ þ ! þ ð Þ inhibition of this process in the normoxic environments of eðaqÞ N2O H N2 OH 2 normal tissues, by generating prodrug radical anions ð Þþ ! ð Þþ ð Þ radiolytically. This enables us to evaluate the relative OH H Me2CHOH H2O H2 Me2C OH 3 abilities of bioreductively activated prodrugs to release the þ ! þ þ þ ð Þ active drug after one-electron reduction. Pulse radiolysis Me2C OH ArNO2 Me2CO H ArNO2 4 with spectrophotometric detection allows the generation where ArNO2 is the prodrug radical anion.

Figure 8. Release of CA4 (1)from13 under different oxygen concentrations in A549 cells. ANG704 (13) was dissolved in DMSO to a Figure 10. Cytotoxicity studies of 1 and 13 in A549 cells. A549 cells concentration of 625 Amol/L, and 80 AL were added to 10 mL A549 cells were seeded in Eagle’s MEM supplemented with 10% FCS and in Eagle’s MEM (f106 cells/mL) to give a final prodrug concentration of nonessential amino acids at 103 per well on a 96-well plate and allowed 5 Amol/L and incubated at 37jC. Anoxic and hypoxic experimental to attach for 24 h. Compounds were dissolved in DMSO and diluted with mixtures were degassed with either N2 or 0.1% or 0.3% O2 for 30 min cell culture medium before addition. The cells were exposed to test before prodrug addition and then overgassed with the appropriate gas compound (0 – 2 Amol/L) for 6 h and then incubated for a further 72 h. during incubation. Samples (1 mL) were added to 100 mmol/L HCl (0.2 Viable cells were determined using the CellTiter 96 Aqueous One Solution mL), mixed, and then extracted (via solid-phase extraction) before HPLC Cell Proliferation Assay kit (Promega Corp., Southampton, United analysis. Columns, mean of three replicates of one experiment; bars, SD. Kingdom). Points, mean of eight wells of one experiment.

Mol Cancer Ther 2006;5(11). November 2006

drugs 15 to 18 showed the fragmentation rate-enhancing effect of a-substitution. Fragmentation is expected to be in competition with the reversal of reduction by oxygen (Eq. 5).

þ ! þ ð Þ ArNO2 O2 ArNO2 O2 5

This suggests that such substitution might extend the range of oxygenation status within the cells that could be targeted. It was thus predicted, based on these model studies, that the geminally substituted 13 would exhibit the broadest range of oxygen tensions that could be Figure 11. Inhibition of tubulin polymerization by 1 and 13. Done with targeted, and Fig. 8 does indeed indicate that significant 250 Ag pure bovine brain tubulin (TotamBiologicals) in G-PEM buffer comprising 80 mmol/L piperazine-N,N¶-bis(2-ethanesulfonic acid) sesqui- CA4 release was achieved up to a 0.3% oxygen concen- sodium salt; 0.5 mmol/L magnesium chloride; 0.5 mmol/L ethylene glycol- tration in A549 cells with this compound. The range of bis(h-aminoethyl ether)-N,N,N¶,N¶,-tetraacetic acid; 1 mmol/L GTP (pH oxygen concentrations over which CA4 was released in 6.8). Tubulin polymerization was observed by measuring the absorbance 11 13 j Points, supersomal suspensions by compounds to is shown (340 nm) of the solution in a quartz cuvette at 37 C over time. 11 single observation. in Fig. 9. The unsubstituted analogue only released CA4 over a narrow range of low oxygen concentrations (<0.01% O2) with complete inhibition above this level. The Table 1 shows radical anion half-lives of conjugates of monosubstituted (12) and geminal (13)-substituted com- nitrothiophenes with both CA4 (1) and coumarin deriva- pounds, on the other hand, efficiently released the drug tives. The data clearly show that geminal a-dimethyl over a much broader range of oxygen concentrations, 13 substitution increases both the rate and the efficiency of with being >50% inhibited (compared with N2) only reductive elimination from reduced nitrothiophenes, above 0.5% O2. with a 10-fold decrease in radical half-life for the a-gem- The prodrug 13 did not inhibit the growth of A549 cells substituted CA4 conjugate compared with its a-monosub- up to concentrations of 2 Amol/L (Fig. 10) and similarly stituted analogue. The related coumarin derivatives had no effect on tubulin polymerization (Fig. 11), further exhibited a similar, although more pronounced trend, with showing the prodrug status of the compound and the relative half-lives of 850:45:1 for the series, a-unsubstituted, successful masking of the activity of 1 through ether a-methyl, and a-gem-dimethyl. The efficiency of reductive linkage to a nitrothiophene moiety. elimination (fragmentation efficiency) was also increased In conclusion, the presence of two a-substituents bestows by geminal substitution within this series. beneficial properties on the compounds by inhibiting The selection of 13 over the related lesser-substituted aerobic metabolism. The increased steric bulk provided prodrugs was supported following studies of reduction by by these substituents may thus stabilize the compounds cytochrome P450R supersomes (Fig. 5). These data show against release of the cytotoxic or cytostatic drug moiety that 1 is selectively produced by 13 in P450R supersomes by chemical or enzymatic processes other than the desired under anoxia, whereas 11 and 12 were ineffective in this bioreductive processes. Furthermore, the absence of a respect. hydrogen atom a to the aromatic group is likely to prevent In oxic liver homogenates, 13 was metabolically stable metabolic oxidation at this position, which can lead to over an incubation period of 16 hours, whereas the release of the effector outside hypoxic regions. The other, lesser-substituted analogues 11 and 12 were not. electronic effects of substituents R1 and R2 clearly enhance These data are presented in Fig. 6 and suggest that the rate of reductive elimination and leads to an extended such substitution has also successfully inhibited unde- range of hypoxic oxygen tensions at which the cytotoxic or sired aerobic metabolism, which was suffered by 11 cytostatic moiety is released (up to 1% oxygen for and 12 during this incubation period. Benzylic ethers, compound 13). This may ultimately provide effective and with at least one hydrogen atom on the a-carbon, are selective delivery of the cytotoxic or cytostatic compound prone to oxidation in oxic liver metabolites to give the to a solid tumor. corresponding carbonyl compound (i.e., an ester), which can then be cleaved by cellular esterases thus producing References 1. Compound 13 was also stable when incubated in 1 1. Pettit GR, Singh SB, Niven ML, Hamel E, Schmidt JM. Isolation, aerobic A549 whole-cell suspensions but released structure, and synthesis of combretastatins A-1 and B-1, potent new efficiently under anoxic conditions (Fig. 7). The efficiency inhibitors of microtubule assembly, derived from Combretum caffrum. of this drug release in supersomal suspensions over a J Nat Prod 1987;50:119 – 31. range of clinically relevant oxygen concentrations was 2. Pettit GR, Singh SB, Niven ML, Hamel E, Schmidt JM. Isolation, structure and synthesis of combretastatins A-1 and B-1, potent new determined and the results are presented in Fig. 8. inhibitors of microtubule assembly, derived from Combretum caffrum. The radiolysis studies with the coumarin model pro- J Nat Prod 1987;50:119 – 31.

Mol Cancer Ther 2006;5(11). November 2006

3. Pettit GR, Singh SB, Boyd MR, et al. Antineoplastic agents. 291. 26. Naylor MA, Swann E, Everett SA, et al. Indolequinone anti-tumour Isolation and synthesis of combretastatins A-4, A-5, and A-6. J Med Chem agents: reductive activation and elimination from (5-methoxy-1-methyl- 1995;38:1666 – 72. 4,7-dioxo-indol-3-yl)methyl derivatives and hypoxia-selective cytotoxicity in vitro 4. Chaplin DJ, Pettit GR, Hill SA. Anti-vascular approaches to solid . J Med Chem1998;41:2720 – 31. tumour therapy: evaluation of combretastatin A4 phosphate. Anticancer 27. Everett SA, Naylor MA, Nolan J, Patel KB, Wardman P. Indolequinone Res 1999;19:189 – 95. bioreductive drugs: kinetic factors which influence selectivity for hypoxia. 5. Folkman J. Angiogenesis in cancer, vascular, rheumatoid, and other Anticancer Drug Des 1998;13:635 – 53. diseases. Nat Med 1995;1:27 – 31. 28. Everett SA, Swann E, Naylor MA, et al. Modifying rates of reductive 6. Lin CM, Singh SB, Chu PS, et al. Interactions of tubulin with potent elimination of leaving groups from indolequinone prodrugs: a key factor in natural and synthetic analogues of the antimitotic agent combretastatin: controlling hypoxia-selective drug release. BiochemPharmacol a structure-activity study. Mol Pharmacol 1988;34:200 – 8. 2002;63:1629 – 39. 7. Dark GG, Hill SA, Prise VE, et al. Combretastatin A-4, an agent that 29. Tercel MT, Wilson WR, Anderson RF, Denny WA. Hypoxia-selective displays potent and selective toxicity toward tumor vasculature. Cancer antitumor agents. 12. Nitrobenzyl quaternary salts as bioreductive Res 1997;57:1829 – 34. prodrugs of the alkylating agent mechlorethamine. J Med Chem 1996; 39:1084 – 94. 8. Tozer GM, Prise VE, Wilson J, et al. Mechanisms associated with tumour vascular shutdown induced by combretastatin A4 phosphate: 30. Tercel M, Lee AE, Hogg A, et al. Hypoxia-selective antitumor agents. intravital microscopy and measurement of vascular permeability. Cancer 16. Nitroarylmethyl quaternary salts as bioreductive prodrugs of the Res 2001;61:6413 – 22. alkylating agent mechlorethamine. J Med Chem 2001;44:3511 – 22. 9. Woods JA, Hadfield JA, Pettit GR, Fox BW, McGowan AT. The 31. Everett SA, Naylor MA, Patel KB, Stratford MRL, Wardman P. interaction with tubulin of a series of stilbenes based on combretastatin Bioreductively-activated prodrugs for targeting hypoxic tissues: elimina- A-4. Br J Cancer 1995;71:705 – 11. tion of aspirin from2-nitroimidazolederivatives. Bioorg Med ChemLett 1999;9:1267 – 72. 10. Watts ME, Woodcock M, Arnold S, Chaplin DJ. Effects of novel and conventional anti-cancer agents on human endothelial permeability: 32. Hay MP, Wilson WR, Denny WA. Nitrobenzyl carbamate prodrugs of influence of tumour secreted factors. Anticancer Res 1997;17:71 – 5. enediynes for nitroreductase gene-directed enyzyme prodrug therapy (GDEPT). Bioorg Med ChemLett 1999;9:3417 – 22. 11. Chaplin DJ, Hill SA. The development of CA4P as a vascular-targeting agent. Int J Radiat Oncol Biol Phys 2002;54:1472. 33. Hay MP, Sykes BM, Denny WA, Wilson WR. A 2-nitroimidazole carbamate prodrug of 5-amino-1-(chloromethyl)-3-[5,6,7-trimethoxyindol- 12. Dowlatti A, Robertson K, Cooney M, et al. A phase I pharmacokinetic 2-yl)carbonyl]-1,2-dihydro-3H-benz[e]indole (amino-seco-CBI-TMI) for use and translational study of the novel vascular-targeting agent CA4P on a with ADEPT and GDEPT. Bioorg Med ChemLett 1999;9:2237 – 42. single dose intravenous schedule in patients with advanced cancer. Cancer Res 2002;62:3408 – 16. 34. Damen EWP, Nevalainen TJ, Bergh TJMvd, Groot FMHd, Scheeren HW. Synthesis of novel paclitaxel prodrugs designed for 13. Rustin GJS, Galbraith SM, Anderson H, et al. Phase I clinical trial of bioreductive activation in hypoxic tumour tissue. Bioorg Med Chem 2002; weekly combretastatin A4 phosphate: clinical and pharmacokinetic 10:71 – 7. results. J Clin Oncol 2003;21:2815 – 22. 35. Borch RF, Liu J, Schmidt JP, et al. Synthesis and evaluation of 14. Oshumi K, Nakagawa R, Fukuda Y, et al. Novel combretastatin nitroheterocyclic phosphoramidites as hypoxia-selective alkylating agents. analogues effective against murine solid tumours: design and structure- J Med Chem2000;43:2258 – 65. activity relationships. J Med Chem1998;41:3022 – 32. 36. Steinberg G, Borch RF. Synthesis and evaluation of pteroic acid- 15. Maya A, Rey BD, DeClairac RP, et al. Design, synthesis, and cytotoxic conjugated nitroheterocyclic phosphoramidites as folate receptor-targeted activities of naphthyl analogues of combretastatin A-4. Bioorg Med Chem alkylating agents. J Med Chem2001;44:69 – 73. Lett 2000;10:2549 – 51. 37. Wardman P. Reduction potentials of one-electron couples involving 16. Pinney KG, Mejia MP, Villalobos VM, et al. Synthesis and biological evaluation of aryl azide derivatives of combretastatin A-4 as molecular free radicals in aqueous solution. J Phys ChemRef Data 1989;18:1637. probes for tubulin. Bioorg Med Chem2000;8:2417 – 25. 38. Threadgill MD, Webb P, O’Neill P, et al. Synthesis of a series of 17. Pettit GR, Anderson CR, Herald DL, et al. Antineoplastic agents. 487. nitrothiophenes with basic or electrophilic substituents and evaluation as Synthesis and biological evaluation of the antineoplastic agent 3,4- radiosensitizers and as bioreductively activated cytotoxins. J Med Chem methylenedioxy-5,4¶-dimethoxy-3¶-amino-Z-stilbene and derived amino 1991;34:2112 – 20. acid amides. J Med Chem 2003;46:525 – 31. 39. Naylor MA, Thomson P. Recent advances in bioreductive drug 18. McGowan AT, Fox BW. Differential cytotoxicity of the combretasta- targeting. Mini Rev Med Chem2001;1:17 – 29. tins A1 and A4 in two daunorubicin P388 resistant cell lines. Cancer 40. Pettit GR, Rhodes MR. Antineoplastic agents 389. New syntheses Chemother Pharmacol 1990;26:79 – 81. of the combretastatin A-4 prodrug. Anticancer Drug Des 1998;13: 19. El-Zayat AAE, Degan D, Drabek S, et al. In vitro evaluation of the 183 – 91. antineoplastic activity of combretastatin A4, a natural product from 41. Carella A, Castaldo A, Centore R, et al. Synthesis and second order Combretum caffrum (arid shrub). Anticancer Drugs 1993;4:19 – 25. nonlinear optical properties of new chromophores containing 1,3,4- 20. Oshumi K, Hatanaka R, Nakagawa R, et al. Synthesis and antitumour oxadiazole and thiophene rings. J ChemSoc Perk Trans 2002;2:1791 – 5. activities of amino acid prodrugs of amino-combretastatins. Anticancer 42. Newcombe PJ, Norris RK. Substitution reactions of nitrothiophenes. Drug Des 1999;14:539 – 48. III. Radical and ionic reactions of 4- and 5-nitro-2-thienyl-methyl and -ethyl 21. Pettit GR, Minardi MD, Boyd MR, Pettit RK. Antineoplastic agents chlorides and acetates. Aust J Chem1979;32:2647 – 58. 463. Synthesis of combretastatin A-3 diphosphate prodrugs. Anticancer 43. Lawrence NJ, Lamarche O, Thurrub N. The facile synthesis of a-aryl- Drug Des 2000;15:397 – 403. a-hydroxy esters via one-pot vicarious nucleophilic substitution and 22. Pettit GR, Lippert JW III. Antineoplastic agents 429. Syntheses of the oxidation. J ChemSoc ChemCommun1999;689 – 90. combretastatin A-1 and combretastatin B-1 prodrugs. Anticancer Drug 44. Dullaghan ME, Owen LJ, Nord FF. 5-Nitro-2-thenaldehyde. J Am Des 2000;15:203 – 16. ChemSoc 1952;74:2676 – 7. 23. Boehle AS, Sipos B, Kliche U, Kalthoff H, Dohrmann P. Combretas- 45. Newcombe PJ, Norris RK. Substitution reactions of nitrothiophenes tatin A-4 prodrug inhibits growth of human non-small cell lung cancer in a with the 2-nitropropan-2-ide ion. Aust J Chem1978;31:2463 – 76. murine xenotransplant model. Ann Thorac Surg 2001;71:1657 – 65. 46. Candeias LP, Folkes LK, Porssa M, Parrick J, Wardman P. Rates of 24. Pettit GR, Moser BR, Boyd MR, et al. Antineoplastic agents. 460. reaction of indoleacetic acids with horseradish peroxidase compound I Synthesis of combretastatin A2 prodrugs. Anticancer Drug Des 2001;16: and their dependence on the redox potentials. Biochemistry 1996;35: 185 – 93. 102 – 8. 25. Ohno T, Kawano K, Tahara K, et al. Antitumour effect of 47. Mitsunobu O. The use of diethyl azodicarboxylate and triphenylphos- combretastatin A-4 derivative AC7700 against rat liver cancer. Gastroen- phine in synthesis and transformation of natural products. Synthesis terology 2001;120:2831. 1981;1 – 28.

Mol Cancer Ther 2006;5(11). November 2006

Downloaded from mct.aacrjournals.org on September 25, 2021. © 2006 American Association for Cancer Research. Synthesis and biological properties of bioreductively targeted nitrothienyl prodrugs of combretastatin A-4

Peter Thomson, Matthew A. Naylor, Steven A. Everett, et al.

Mol Cancer Ther 2006;5:2886-2894.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/5/11/2886

Cited articles This article cites 44 articles, 5 of which you can access for free at: http://mct.aacrjournals.org/content/5/11/2886.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/5/11/2886.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/5/11/2886. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.