Dear Editor:In the development of more effective therapeutic strategies for the prevention and treatment of various cancers, it is important to focus on biological characteristics that differentiate benign and malignant neoplasms. A key difference between benign and malignant tumors appears to be related to cellular motility and invasiveness. Since cancer cell movement is an important component of metastatic potential, various in vitro assays have been developed and used to study general cell movement, including chemoinvasion of coated membranes and phagokinetic tracks (1,8). To better define the relationship between cell motility and metastatic potential, the present study used time-lapse videomicroscopy and image analysis to compare and characterize the differences in motile behavior between two human breast cancer cell lines that have been used in numerous in vitro studies. These were MDA-MB-231 (MDA) cells, which display a high metastatic potential and MCF-7 cells, which have a low metastatic potential. Similarly, other investigators have used time-lapse videomicroscopy to quantify the motility of other cancer cells (4,6). MCF-7 human breast cancer cells were provided by Dr. Sam Brooks of the Michigan Cancer Foundation (Detroit, MI), MDA-MB-231 cells were obtained from ATCC (Rockville, MD) and the cells were grown as monolayer cultures in RPMI 1640 media as previously described (3). Cells were plated sparsely in T-25 flasks and their motility was recorded with phase-contrast optics at 400 X with a color video camera (Sony, Model CCD-IRIS/RGB). The microscope stage was maintained at a constant temperature (37 ° C) by a thermostat-controlled stage heater (Fryer, Model A-50-IN). The video signal was fed to a time-lapse videocassette recorder with a time compression ratio of 240:1 (Panasonic, Model AG-6730 S-VHS, Secaucus, N J). Phase-contrast images of each cell line were recorded for 9-10 h in 3-5 replicate experiments. In each recording the movement of 5-10 cells in a microscopic field was analyzed. At least 30 cells from each cell line were analyzed in this study.The video images, at different time intervals (10-60 rain) over a period of 4-8 h, were captured and digitized with video capture software (Apple Video Player, Videocapture) on a Macintosh computer (Quadra 630-AV). Computer-assisted image analysis was accomplished with the NIH Image 1.59 program. Cell contours of the digital images were manually outlined and their x,y coordinates, area, and perimeter were calculated with the NIH Image program.Parameters of cell motility and morphology that were calculated for each cell line included rate of movement, total path length, and perimeter index. Rate of cell movement was determined as the distance traveled by the centroid (geometric center) of individual cells divided by time (gm/h) and plotted at the midpoint between each time interval. Total path length was computed as the total path traveled by the cell centroid (in p.m) during the experimental period. Perimeter index (PI) was caiculated as the perimete...

The motile behavior of tumor cells is regulated in part by growth factors, cytokines, and other endogenous factors. In some instances, stromal tissue surrounding the tumor cells produces these growth factors which interact with tumor cells and thus may play an important role in tumor proliferation and progression. We and others have shown that conditioned media from NIH 3T3 fibroblasts (3T3-CM) increases the invasiveness of breast cancer cells. The present study characterized the influence of 3T3-CM on breast cancer cell motility and examined the hypothesis that antiestrogens inhibit this 3T3-CM-induced effect. In this study we observed that 3T3-CM added to MCF-7 cells produced an immediate cell-scattering effect as determined by time-lapse videomicroscopy, scanning electron microscopy, and F-actin labeling. The results of this study indicate that keratinocyte growth factor in 3T3-CM is largely responsible for the 3T3-CM-induced scattering motility of MCF-7 cells. These results emphasize the importance of stromal-tumor cell (epithelial-mesenchymal) interactions in the motility of breast cancer cells. Further, our results demonstrate that antiestrogens (tamoxifen, ICI-182,780 and Analog II) inhibit 3T3-CM-induced motility of MCF-7 breast cancer cells. Antiestrogen treatment reduced membrane movements and the motile morphology of MCF-7 cells induced by 3T3-CM. These results suggest that antiestrogens inhibit breast cancer cell motility and that antiestrogen treatment may be used to reduce the metastatic spread of breast cancer.