Efficient adenovirus infection requires the presence of coxsackie-adenovirus receptor (CAR) and αv integrin on the surface of cells. Previously, we showed that treatment of several cancer cell lines with a low concentration of the histone deacetylase inhibitor FK228 (FR901228, depsipeptide) (1 ng/ml) caused an increase in the RNA levels of CAR and αv-integrin. FK228 pre-treatment was associated with a 5–10 fold increase in adenoviral transgene expression following adenovirus infection. The levels of CAR and αv integrin RNA were not increased in cultured normal cells from breast, liver or kidney following similar FK228 treatment. These results suggest that FK228, a drug currently in phase II clinical trials for the treatment of patients with peripheral or cutaneous T-cell lymphoma, may result in preferential enhancement of adenoviral transgene expression in cancer cells. To further evaluate this differential sensitivity between normal and cancer cells, we examined the effect of FK228 in athymic mice bearing advanced-stage subcutaneous LOX IMVI and UACC-62 human melanoma xenografts and MDA-MB-231 human breast cancer xenografts. Mice with melanoma xenografts were treated with FK228 iv q4d × 3 (days 1, 5, and 9) with one of three doses- 0.7, 1.6, or 3.6 mg/kg/dose. The highest dose was effective in causing tumor regression. Animals were sacrificed 6 h, 24 h, or 48 h following the last dose of FK228 and tissues were harvested. The levels of CAR and αv integrin RNA were monitored in mouse liver, kidney and lung and in the human xenografts using semi-quantitative RT-PCR analysis. The levels of CAR RNA were increased in the xenografts. The time of maximum increase in the levels of CAR RNA in the xenografts was 6 h after the administration of 3.6 mg/kg/dose FK228. The levels of αv integrin RNA in the xenografts were unchanged as were the levels of both CAR and αv integrin, in the livers, kidneys and lungs from the same animals. In order to define an optimal dose and schedule of FK228 administration for CAR induction, mice with LOX IMVI xenografts were treated with FK228 one, two, or three times. The tissues were then harvested either 6 or 24 hours following each dose and the levels of CAR and αv integrin RNA were determined. The results showed that the optimal dose for CAR induction was 24 hours after a single treatment with 3.6 mg/kg FK228 where CAR RNA levels were increased by 14-fold. Accordingly, CAR protein levels increased by 3–8 fold. In mice with MDA xenografts treated with a single dose of 1.6, 3.6 or 5.4 mg/kg FK228, CAR protein levels increased after 48 hours by 4-fold following the 3.6 mg/kg dose, whereas αv integrin levels remained unchanged. Because we have previously shown a correlation between the level of CAR and the extent of adenovirus infection in vitro, these results suggest that FK228 may preferentially increase the levels of adenovirus infection in cancer cells in vivo as it did in vitro. Experiments are in progress to determine if FK228 pre-treatment can increase the efficiency and selectivity of adenovirus gene therapy in vivo.