Quantitative cell dispersion analysis: new test to measure tumor cell aggressiveness

Abstract

Tumor progression requires the dispersion of epithelial cells from neoplastic clusters and cell invasion of adjacent stromal connective tissue. Aiming at demonstrating the precise relationships between cell dispersion and cell invasion, related respectively to expression of E-cadherin/catenin complex and matrix metalloproteinases (MMPs), we developed an original in vitro model of cell dispersion analysis. Our study reports the validation of this model that allowed us to analyze and quantify the cell cohesion level by means of time-lapse videomicroscopy and computer analysis based on the observation of spatial and temporal cell distribution. Our model was able to distinguish 2 groups among different human bronchial and mammary epithelial cells previously characterized for the expression of E-cadherin/catenin complex and MMPs and their invasive capacity in the Boyden chamber assay. The first group (16HBE14o(-), MCF-7, T47D) that expressed membranous E-cadherin and beta-catenin, and was negative for MMP-2 expression and non-invasive, displayed a highly cohesive pattern corresponding to a cluster spatial distribution. The second group (Beas2B, BZR, BZR-T33, MDA-MB-231, MDA-MB-435, BT549 and HS578T) that was invasive and showed lack of expression of E-cadherin and a cytoplasmic redistribution of beta-catenin, displayed a dispersed pattern corresponding to a random spatial distribution. Downregulation of E-cadherin by a blocking antibody induced a more random distribution. Conversely, expression of E-cadherin by cDNA transfection induced a cluster distribution. Moreover, tumor cell lines that co-expressed MT1-MMP and MMP-2 (Beas2B, BZR, BZR-T33, MDA-MB-435, BT549 and HS578T) showed a more dispersed pattern than tumor cell lines that did not express MMP-2 (MDA-MB-231). In conclusion, we demonstrated that the spatial group behavior of cell lines, i.e., their cohesion/dispersion ability, reflects their invasive properties. Thus, this model of cell dispersion analysis may represent a new test to measure tumor cell aggressiveness.