The antineoplastic constituents of Combretum caffrum (Eckl. and Zeyh) Kuntze (Combretaceae family), a species indigenous to South Africa, have been investigated. Subsequently we isolated a series of closely related bibenzyls, stilbenes, and phenanthrenes from C. caffrum. Some of the stilbenes proved to be potent antimitotic agents which inhibited both tubulin polymerization and the binding of colchicine to tubulin. Combretastatin A-4 has been shown to be the most potent cancer cell growth inhibitor of the series. Presently this cis-stilbene is the most effective inhibitor of colchicine binding to tubulin and the simplest natural product yet described with such potent antitubulin effects. Combretastatin A-4, A-5, and A-6 were also found to inhibit growth of Neisseria gonorrhoeae. Details of the isolation and syntheses of combretastatins A-4 (2a), A-5 (2c), and A-6 (3a) have been described.

Biofilms are at the root of many infections largely because they are much more antibiotic resistant than their planktonic counterparts. Antibiotics that target the biofilm phenotype are desperately needed, but there is still no standard method to assess biofilm drug susceptibility. Staphylococcus epidermidis ATCC 35984 biofilms treated with eight different approved antibiotics and five different experimental compounds were exposed to the oxidation reduction indicator Alamar blue for 60 min, and reduction relative to untreated controls was determined visually and spectrophotometrically. The minimum biofilm inhibitory concentration was defined as <50% reduction and a purplish well 60 min after the addition of Alamar blue. All of the approved antibiotics had biofilm MICs (MBICs) of >512 g/ml (most >4,096 g/ml), and four of the experimental compounds had MBICs of <128 g/ml. The experimental aaptamine derivative hystatin 3 was used to correlate Alamar blue reduction with 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT) reduction and viable counts (CFU/ml) for S. epidermidis ATCC 35984, ATCC 12228, and two clinical isolates. For all four strains, Alamar blue results correlated well with XTT (r ‫؍‬ 0.83 to 0.97) and with CFU/ml results (r ‫؍‬ 0.85 to 0.94). Alamar blue's stability and lack of toxicity allowed CFU/ml to be determined from the same wells as Alamar blue absorbances. If the described method of microplate Alamar blue biofilm susceptibility testing, which is simple, reproducible, cost-effective, nontoxic, and amenable to high throughput, is applicable to other important biofilm forming species, it should greatly facilitate the discovery of biofilm specific agents.Given the tremendous clinical importance of biofilms, it is somewhat surprising that there is no standard method for investigating the drug susceptibility of bacterial biofilms. Several methods are available but are limited by long processing times, incompatibility with high throughput, expensive reagents or equipment, or the fact that the method measures mass instead of viability (2,4,7,13,14,24,25). For bacteria, a common method of assessing susceptibility is to quantitate the mass of biofilms by crystal violet or safranin staining, followed by extraction of bound dye with a solvent and measurement of absorption (6,24). This method provides no information about viability. Another common method of assessing bacterial biofilm susceptibility is to disrupt the biofilm by sonication, vortexing, or scraping, followed by dilution plating for determination of CFU/ml (27,28,31). This method has serious limitations; biofilm clumps can be difficult to dissociate into single-cell suspensions for plate counting, it is extremely laborious, and antibiotic carryover is a concern. For fungi, the most common method is an XTT [2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide] reduction assay (1,8,25), and this method has also been used for bacterial biofilms (1). While XTT reduction does measure metabolic activity,...

A structure-activity relationship (SAR) study of the South African willow tree (Combretum caffrum) antineoplastic constituent combretastatin A-4 (1b) directed at maintaining the (Z)-stilbene relationship of the olefin diphenyl substituents led to synthesis of a potent cancer cell growth inhibitor designated phenstatin (3b). Initially phenstatin silyl ether (3a) was unexpectedly obtained by Jacobsen oxidation of combretastatin A-4 silyl ether (1c --> 3a), and the parent phenstatin (3b) was later synthesized (6a --> 3a --> 3b) in quantity. Phenstatin was converted to the sodium phosphate prodrug (3d) by a dibenzyl phosphite phosphorylation and subsequent hydrogenolysis sequence (3b --> 3c --> 3d). Phenstatin (3b) inhibited growth of the pathogenic bacterium Neisseriagonorrhoeae and was a potent inhibitor of tubulin polymerization and the binding of colchicine to tubulin comparable to combretastatin A-4 (1b). Interestingly, the prodrugs were found to have reduced activity in these biochemical assays. While no significant tubulin activity was observed with the phosphorylated derivative of combretastatin A-4 (1d), phosphate 3d retained detectable inhibitory effects in both assays.

A series of cis- and trans-stilbenes related to combretastatin A-4 (1a), with a variety of substituents at the 3'-position of the aryl B-ring, were synthesized and evaluated for inhibitory activity employing six human cancer cell lines (NCI-H460 lung carcinoma, BXPC-3 pancreas, SK-N-SH neuroblastoma, SW1736 thyroid, DU-145 prostate, and FADU pharynx-squamous sarcoma) as well as the P-388 murine lymphocyte leukemia cell line. Several of the cis-stilbene derivatives were significantly inhibitory against all cell lines used, with potencies comparable to that of the parent 1a. All were potent inhibitors of tubulin polymerization. The corresponding trans-stilbenes had little or no activity as tubulin polymerization inhibitors and were relatively inactive against the seven cancer cell lines. In terms of inhibition of both cancer cell growth and tubulin polymerization, the dimethylamino and bromo cis-stilbenes were the most potent of the new derivatives, the latter having biological activity approaching that of 1a. As part of the present study, the X-ray crystal structure of the 3'-O-phosphate of combretastatin A-4 (1b) was successfully elucidated. Compound 1b has been termed the "combretastatin A-4 prodrug", and it is currently undergoing clinical trials for the treatment of human cancer patients.