The pathogenicity of Trichosporon dermatis isolated from skin lesions of a patient has been examined in mice. Balb/c mice were treated with two intraperitoneal injections of 100 mg/kg cyclophosphamide on days 4 and 1 and one subcutaneous injection of 10 mg/kg dexamethasone on day 1 pre-inoculation, and then challenged with 0.2 ml T. dermatis inoculum (1 × 10(8) CFU/ml) by topical application on an abrasive wound in the dermabrasive group and by hypodermic injection in the subcutaneous group. In the intravenous group, 0.2 ml of high (1 × 10(8) CFU/ml) or low (1 × 10(7 )CFU/ml) inoculum was injected into the tail vein. Histopathology and inverse fungal culture were performed on the skin lesion and viscera, and renal fungal burden was also determined. Inoculated sites developed localized infections after dermabrasive and subcutaneous challenge in all mice, but the maximum area of skin lesions, and number of positive cultures from the lesions, were higher for immunocompromised mice. In the intravenous group, all immunocompetent animals survived during the four-week period, whereas 100 and 70% of immunocompromised animals died by 3 and 5 days in the high and low-inoculum groups, respectively. The incidence of disseminated infection and the renal fungal burden of immunocompromised mice were higher than those of immunocompetent mice. Our results demonstrate that subcutaneous and intravenous injection of T. dermatis can successfully establish cutaneous and systemic infection models in immunocompromised mice, with the kidney and lung being most susceptible.