The effects of dimethyl sulfoxide and retinoic acid on the cell growth and the phenotype of ovarian cancer cells

Open Access
Publisher Website
Google Scholar
Abstract
We have compared the in vitro effects of the differentiation inducers dimethyl sulfoxide (DMSO) and retinoic acid (RA) on a polyclonal human ovarian cancer cell line (HOC-7). Density gradient fractionation of untreated cells reveals that a proportion of rapidly growing, polygonal cells with medium density is capable of spontaneous reversion into a slowly growing low-density phenotype with flattened morphology similar to non-transformed human ovarian surface epithelial cells. Clonal expansion of these low-density cells proves that the observed characteristics are stable for prolonged culture periods. Exposure of HOC-7 cells to DMSO and RA or removal of the serum from the medium is effective in enhancing the proportion of these low-density cells. Application of DMSO causes the cells to become flattened and elongated, and to develop rodlike protrusions. In these cytoplasmic extensions thick filament bundles are dominant. Immunofluorescence studies demonstrate that both untreated low-density subclones and DMSO-treated polyclonal cells are much more reactive for cytokeratin than medium-density subclones or untreated parental cells. Furthermore, immunocytochemistry and fixedcell ELISA reveal 2-to 5-fold greater amounts of desmoplakins I and II and of fibronectin in low-density subclones and in DMSO-treated cells as compared to medium-density subclones and control cultures. RA exerts weaker effects on the phenotype of the cells. Both inducers reduce DNA synthesis and inhibit the anchorage-dependent and the anchorageindependent cell growth in a dose- and time-dependent manner. The restoration of the original morphology and growth rate after removal of the differentiation-inducing agents proves that the observed changes are reversible; this indicates that the cells do not become terminally differentiated. The described phenotypic and growth kinetic reactions should prove useful for the future recognition of induced maturation of ovarian cancer cells.