Altered growth and differentiation of cultured mouse epidermal cells infected with oncogenic retrovirus: contrasting effects of viruses and chemicals.

Mouse epidermal basal cells can be cultured in medium with a Ca2+ concentration below 0.1 mM and induced to terminally differentiate in medium with higher Ca2+. Infection of basal cells with Kirsten or Harvey sarcoma virus causes a marked stimulation of basal cell proliferation. Challenge of infected basal cell cultures with 0.5 mM Ca2+ medium indicates that terminal differentiation is altered, since infected cultures persist in this medium for long periods without stratifying and sloughing from the culture dish. Infected cultures express much higher levels of protein with a molecular weight of 21,000 than controls, and studies with temperature-sensitive mutants indicate that persistent function of protein with a molecular weight of 21,000 is required for the observed effects of the sarcoma viruses on terminal differentiation. In response to 0.5 mM Ca2+, the proliferation rate is markedly reduced in both control and virally infected cells. However, the proliferation rate of the sarcoma virus-infected cultures is reduced less in response to elevated levels of calcium. Virus-infected cells as well as controls also demonstrate a rise in the differentiation-associated enzymatic activity of epidermal transglutaminase and will terminally differentiate when the multiplicity of infection is below one. These results suggest that sarcoma virus-infected cells are not blocked completely in their program of terminal differentiation in response to 0.5 mM Ca2+ and differ in this respect from epidermal cells altered by exposure to chemical carcinogens and from malignant keratinocytes. Infection with Kirsten or Harvey sarcoma virus and expression of protein with a molecular weight of 21,000 therefore are closely linked to epidermal proliferation and retard, but do not completely block, the program of terminal differentiation. The effects of ras oncogenes on NIH 3T3 cells are discussed relative to the effects of ras oncogenes observed on the primary mouse epidermal cells in the context of multistep models for naturally occurring epithelial cancers.