A new culture model facilitating rapid quantitative testing of mitotic spindle inhibition in mammalian cells.

A new culture model, which facilitated both mass screening of potential anticancer drugs acting on microtubules and quantitative experiments with known "antitubulins," was found to have the following advantages: use of mammalian cells (either transformed or not), simplicity of the techniques (phase-contrast microscopy or simple microscopy after Giemsa staining), and ease with which it lent itself to quantification. The model was based on the uniform multimicronucleation response induced by antitubulins in MO cells. The specificity (towards antitubulins) of this response was ascertained by the use of many substances, including most of the known antitubulins and a number of nonrelated cytostatic or cytotoxic compounds. The uniformity of the response was established with the use of time-lapse observation of large numbers of cells and quantitative approaches. The results obtained in this model with the standard antitubulins (colchicine, vinblastine, vincristine) showed similar effects. The major difference between colchicine and the Vinca alkaloids was that colchicine was less reversible, which might be an indication of stronger intracellular binding. The Vinca alkaloids acted synergistically with colchicine when threshold subactive doses were combined, although it is known that they bind at a different site on tubulins. A number of substances that have been claimed or were suspected to interfere with microtubules were tested. The results showed that the following substances were indeed active with MO cells: colchicine, vinblastine, vincristine, podophyllotoxin, rotenone, griseofulvin, mercaptoethanol, benomyl, methyl benzimidazol-2-yl carbamate, and R 17934. Compounds that were inactive on these mammalian cells in culture included isopropyl carbanilate and melatonin, both of which have been shown to be active in other systems.