Chii M. Lin

Combretastatin A-4 (CS-A4), 3,4,5-trimethoxy-3'-hydroxy-4'-methoxy-(Z)-stilbene, and combretastatin A-2 (CS-A2), 3,4-(methylenedioxy)-5-methoxy-3'-hydroxy-4'-methoxy-(Z)-stilbene, are structurally simple natural products isolated from the South African tree Combretum caffrum. They inhibit mitosis and microtubule assembly and are competitive inhibitors of the binding of colchicine to tubulin [Lin et al. (1988) Mol. Pharmacol. 34, 200-208]. In contrast to colchicine, drug effects on tubulin were not enhanced by preincubating CS-A4 or CS-A2 with the protein. The mechanism of their binding to tubulin was examined indirectly by evaluating their effects on the binding of radiolabeled colchicine to the protein. These studies demonstrated rapid binding of both compounds to tubulin even at 0 degrees C (binding was complete at the earliest times examined), in contrast to the relatively slow and temperature-dependent binding of colchicine. Although the binding of the C. caffrum compounds to tubulin was quite tight, permitting ready isolation of near-stoichiometric amounts of drug-tubulin complex even in the absence of free drug, both CS-A4 and CS-A2 dissociated rapidly from tubulin in the presence of high concentrations of radiolabeled colchicine. Apparent rate constants for drug dissociation from tubulin at 37 degrees C were 3.2 x 10(-3) s-1 for CS-A4, 4.8 x 10(-3) s-1 for CS-A2, and 2.9 x 10(-5) s-1 for colchicine (half-lives of 3.6, 2.4, and 405 min, respectively). Thus, the effectiveness of the C. caffrum compounds as antimitotic agents appears to derive primarily from the rapidity of their binding to tubulin.(ABSTRACT TRUNCATED AT 250 WORDS)

A series of stilbenes has been prepared and tested for cytotoxicity in the five human cancer cell lines A-549 non-small cell lung, MCF-7 breast, HT-29 colon, SKMEL-5 melanoma, and MLM melanoma. The cis stilbenes 6a-f proved to be cytotoxic in all five cell lines, with potencies comparable to that of combretastatin A-4. These cytotoxic compounds were all potent inhibitors of tubulin polymerization. The corresponding trans stilbenes 7b-f were inactive as tubulin polymerization inhibitors and were significantly less cytotoxic in the five cancer cell lines. In the dihydro series, 8b, 8c, and 8f were inactive as tubulin polymerization inhibitors, while 8a, 8d, and 8e were less active than the corresponding cis compounds 6a, 6d, and 6e. The lack of tubulin polymerization inhibitory activity and cytotoxicity displayed by the phenanthrene 23b, which was synthesized as a conformationally rigid analogue of the lead compound 1, indicates that the activity of the stilbenes is not due to a totally planar conformation. Similarly, inactivity of the conformationally restricted analogue 26 suggests that the biologically active conformation of 1a resembles that of the cis alkene 1. Additional inactive compounds prepared include the benzylisoquinoline series 28-32 as well as the protoberberines 38 and 39. Shortening the two-carbon bridge of 1a to a one-carbon bridge in the diphenylmethane 20 resulted in a decrease in cytotoxicity and tubulin polymerization inhibitory activity. Although the corresponding benzophenone 18 was as active as 1a as a tubulin polymerization inhibitor, it was less cytotoxic than 1a, and the benzhydrol 19 was essentially inactive. With the exception of the amide 15c, which displayed low antitubulin activity, all of the phenylcinnamic acid derivatives 14a-c and 15a-f were inactive in the tubulin polymerization inhibition assay. The acid 14b and the ester 15a were cytotoxic in several of the cancer cell cultures in spite of their inactivity as tubulin polymerization inhibitors.

A series of 1,6,7,8-substituted 2-(4'-substituted phenyl)-4-quinolones and related compounds have been synthesized and evaluated as cytotoxic compounds and as antimitotic agents interacting with tubulin. The 2-phenyl-4-quinolones (22-30) with substituents (e.g. F, Cl, and OCH3) at C-6, C-7, and C-8 show, in general, potent cytotoxicity against human lung carcinoma (A-549), ileocecal carcinoma (HCT-8), melanoma (RPMI-7951), and epidermoid carcinoma of the nasopharynx (KB) and two murine leukemia lines (P-388 and L1210). Introduction of alkyl groups at N-1 or C-4 oxygen led to inactive compounds (35-43 and 50). In addition, compounds 24, 26, and 27 were evaluated in the National Cancer Institute's 60 human tumor cell line in vitro screen. These compounds demonstrated the most marked effects in the screen on two colon carcinoma cell lines (COLO-205 and KM-20L2) and on a central nervous system tumor cell line (SF-539) with compound 26 the most potent of the three agents. Compounds 24, 26, and 27 were potent inhibitors of tubulin polymerization, with activity nearly comparable to that of the potent antimitotic natural products colchicine, podophyllotoxin, and combretastatin A-4. The three agents also inhibited the binding of radiolabeled colchicine to tubulin, but this inhibition was less potent than that obtained with the natural products.

Data generated in the new National Cancer Institute drug evaluation program, which are based on inhibition of cell growth in 60 human tumor cell lines, were probed with nine known antimitotic agents using the COMPARE algorithm. Cytotoxicity data were available on approximately 7000 compounds at the time of the analysis, and, based on the criteria used, 82 compounds were selected as positive by the computer search. Nine were the probe compounds themselves, and 41 were analogues of known antimitotic agents. Among the remaining 32 compounds there were 19 distinct chemical species. Agents in ten of these groups (containing 20 compounds) were effective inhibitors of in vitro tubulin polymerization and caused the mitotic arrest of cells grown in culture. Two compounds were related natural products binding in the Vinca domain of tubulin, and the others were synthetic agents which interfered with colchicine binding. The remaining 12 agents (one natural product, the remainder synthetic) fell into several groups: two compounds were weak inhibitors of tubulin polymerization, inhibited colchicine binding, and caused mitotic arrest; one compound weakly inhibited tubulin polymerization but did not cause an increase in the number of cells arrested in mitosis; two compounds caused mitotic arrest at micromolar concentrations, but thus far no in vitro interaction with tubulin has been observed; the remainder neither inhibited tubulin polymerization nor caused a rise in the number of cultured cells arrested in mitosis. Tubulin-dependent GTP hydrolysis was stimulated or inhibited by all agents which inhibited tubulin polymerization with the exception of one compound. The analysis of differential cytotoxicity data thus appears to have great promise for the identification of new antimitotic agents with antineoplastic potential.