The purpose of these studies was to directly compare the pharmacologic effects of retinoids on cutaneous cells in vitro and in vivo. Previously, it was demonstrated that all-trans-retinoic acid stimulates the proliferation of growth-arrested human keratinocytes and fibroblasts in culture. In the present studies, all-trans-retinoic acid was compared to three other retinoids--13-cis-retinoic acid, P-[(E)-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-napthalenyl-1-p ropenyl] benzoic acid (TTNPB) and M-[(E)-2-(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-napthalenyl-1-p ropenyl] benzoic acid (meta-carboxy-TTNPB]--for growth stimulation using a cultured human squamous epithelial cell line (UM-SCC-1) and human dermal fibroblasts. These four retinoids were also evaluated for their effects on reduction of horn-filled utriculi when topically applied to the skin of rhino mice. All-trans-retinoic acid stimulated proliferation of both fibroblasts and epithelial cells over concentrations ranging from 0.01 to 1.0 micrograms/ml. In fibroblasts, 13-cis-retinoic acid was less potent than all-trans-retinoic acid, whereas, in epithelial cells these two retinoids were equipotent. In contrast, TTNPB was more potent than all-trans-retinoic acid at stimulating the growth of both fibroblasts and epithelial cells. The analog, meta-carboxy-TTNPB was essentially inactive as a growth stimulator of both cell types. In the rhino mouse utriculus reduction model, the rank order of potency for the retinoids was the same as that for in vitro cell growth stimulation (TTNPB greater than all-trans-retinoic acid greater than 13-cis-retinoic acid). Meta-carboxy-TTNPB was inactive at reducing utriculi at a dose of 5,000 times the ED50 of TTNPB.(ABSTRACT TRUNCATED AT 250 WORDS)