Differential modulation of epidermal keratinization in immortalized (HaCaT) and tumorigenic human skin keratinocytes (HaCaT-ras) by retinoic acid and extracellular Ca2+.

The growth and differentiation response to retinoic acid (RA) was studied in the human keratinocyte line HaCaT and tumorigenic clones transfected with c-Ha-ras oncogene (HaCaT-ras). Differentiation (mainly keratin synthesis) was evaluated and correlated to cell proliferation in vitro but also growth behaviour in vivo (tumorigenicity). Comparable to normal keratinocytes, HaCaT cells and ras clones showed increased expression of the epidermal suprabasal keratins K1 and K10 upon RA depletion of the media (delipidized serum), while simple epithelial type keratins K7, K8 and K18 as well as K19 and K13 (typical of internal stratified epithelia) were almost completely suppressed. The cell density-dependent increase of K1 and K10 at intermediate RA levels (as in regular media with untreated serum) was also observed at Ca2+ levels below 0.1 mM, thus being clearly unrelated to stratification, whereas K13 synthesis was Ca(2+)-dependent and initiated with stratification. The effects on keratins were fully reversed by increasing RA concentrations. There was only mild stimulation of proliferation at RA doses (10(-10) to 10(-8) M) not directly corresponding to suppression of keratinization. Thus, the negative RA influence on K1 and K10, opposed to the effect on simple keratins, substantiates the preserved regulatory capacity rendering these cells appropriate models for biological testing. Among the various tumorigenic HaCaT-ras clones highly and moderately differentiating ones could be distinguished, accordingly induction in vitro led to a comparable spectrum of differentiation markers (K1 and K10 appearing early, and filaggrin late) as growth in vivo. These in vitro results demonstrate that, in spite of some differences in RA sensitivity, virtually all clones possess the epidermal differentiation repertoir which is regulated according to the same principles. Finally, this confirms our in vivo data that differentiation potential is not inversely related to the state of transformation or tumorigenicity.