Evaluation of cell morphology by video recording and computer-assisted image analysis

The aim of this study was to evaluate the effects of various cell culture conditions on cell morphology. Cell morphology was estimated by means of video recording and computer-assisted image analysis. Cell contours from the stored images of either live cells or fixed and stained cells were determined automatically, and cellular area, form factor, and average cell brightness were calculated. Using the mouse fibroblastoid L 929 cell line (L-cells) and the rat glioma BT4C cell line, it was found that a number of methodological parameters strongly affected cell morphology. These included confluency of cells before dissociation, dissociation procedure, cell seeding density, cultivation time, and culture substratum. The substratum, particularly collagen type I and fibronectin, profoundly affected cell morphology. Using drugs affecting cytoskeletal organization or cell substratum interactions, it was shown that average cell brightness was a valuable parameter for estimation of cellular attachment. Cytochalasin D, which impairs actin filaments, caused a dramatic increase in the average cell brightness in both cell lines. Nocodazole, which depolymerizes microtubules, mainly affected the L-cells, whereas the BT4C-cells were largely unaffected, indicating that microtubules were morphological determinants for the former cell line but not for the latter. When cells were grown on fibronectin, an RGD-peptide only affected L-cell attachment, indicating that BT4C-cells only expressed low (if any) amounts of RGD recognizing integrins. The interassay precision of the employed procedure depended on culture substratum; the coefficients of variation ranged from 7-24%. Lowest variations in area determination were found for cells grown on fibronectin. The coefficient of variation of form factor determinations was generally around 20%, independent of substrata and culture time.