Topical silicone gel or silicone cream with occlusive dressing has proved to be an efficacious method for the treatment and prevention of hypertrophic scars and keloids, but how this action is triggered remains unknown. Hydration of the epidermis and/or the cellular effects of the released low-molecular-weight silicone oil have been suggested as possible mechanisms. In order to further elucidate the mechanism, we used an in vitro keratinocyte-fibroblast coculture model to investigate the cellular effects of silicone and hydration. In this model, the condition of clinical usage of topical silicone gel or cream or the condition of hydration exerted by occlusive dressing could be mimicked. The model consisted of two chambers separated by a semipermeable membrane, in which a fully differentiated stratified epithelium is present in the upper chamber and medium and monolayer fibroblasts are located in the lower chamber. The keratinocytes were nourished from the basal side only, while the apical surface was submerged in silicone oil, paraffin, Hanks' balanced salt solution, or medium (hydration); or it was exposed to air (control). In the hydration-treated group, the proliferation of fibroblasts measured as [3H]thymidine incorporation and their collagen, glycosaminoglycan production was significantly inhibited when compared with the controls, but exposure of the keratinocyters to silicone oil or paraffin did not influence fibroblast behavior. The results suggest that hydration, not silicone, modulates the in vitro keratinocyte-fibroblast interaction. This may be one possible mechanism by which topical silicone or occlusive dressing treatment may affect the development of hypertrophic scars and keloids.