We devised a new in vitro method to estimate the proportion of dividing cells and the potential doubling time (Tpot) of tumors using the same technique as with the cytokinesis-block micronucleus assay. The usefulness of this methodology was confirmed by comparing the data with those obtained by flow cytometry after bromodeoxyuridine (BrdUrd) incorporation. Xenografted human and murine tumors were excised 0.5–8 h after BrdUrd injection and disaggregated to single cells. A portion of these cells was then plated in dishes to which 1 or 2 µg/ml cytochalasin B were added. These concentrations of cytochalasin B blocked cytokinesis but not karyokinesis with the result that cells became multinucleate after mitoses. At every 12 or 24 h of culture, the proportion of multinucleate cells and the total number of nuclei and cells were scored. The remaining cells were analyzed with a flow cytometer and the BrdUrd-labeling index and Tpot were determined. In all 8 tumor lines studied, the proportion of multinucleate cells reached a plateau within 3–7 days of culture, and we therefore defined the dividing fraction as the plateau value. The dividing fraction ranged between 33 and 98% and clearly tended to be high in rapidly growing tumors. A significant correlation was seen between the dividing fraction and BrdUrd-labeling index (r = 0.74, P < 0.001). The increase in the average number of nuclei per cell also tended to be higher in rapidly growing tumors. The Tpot was estimated as the time for this nucleus/cell ratio to reach 2.0. In 7 of 8 tumor lines, Tpot values estimated by this method compared reasonably with those estimated by the BrdUrd method. Therefore, this simple technique, originally developed for radiosensitivity prediction, would also seem to be useful in estimating tumor proliferative activity.
This work was supported by the Deutsche Forschungsgemeinschaft, SFB102, and the Alexander von Humboldt Foundation.