Mortality and morbidity of prostate cancer result from extracapsular invasion and metastasis. This tumor progression depends on active cell motility. Previous studies have shown that calpain-regulated rear detachment enabling forward locomotion is required for cell migration initiated by growth factor and adhesion receptors. Therefore, we asked whether calpain would be a target for limiting tumor progression, using as our model the PA DU-145 human prostate carcinoma cell line and a highly invasive subline, wild-type DU-145, derived from it. In vitro, the calpain-specific inhibitor CI-I (ALLN) and the preferential-but-less-specific inhibitor leupeptin decreased transmigration of both cell lines across a Matrigel barrier. These calpain inhibitors limited epidermal growth factor-induced motility but did not alter the growth rate of the tumor cells, as expected. Antisense down-regulation of the growth factor-activated calpain-2 (m-calpain) isoform also reduced transmigration and cell motility. These in vitro findings were then buttressed by in vivo studies, in which i.p. DU-145 tumor xenografts were treated with leupeptin. Tumor invasion into the diaphragm was reduced by leupeptin treatment for both the PA and wild-type DU-145 cells (from 1.7 to 0.78 for the parental line and 2.3 to 1.2 for the invasive derivative, respectively). Tumor cells of both types engineered to express calpain-2 antisense constructs also demonstrated a similar 50% reduced invasiveness in vivo. Finally, we found by gene expression survey of 53 human prostate tumors and 23 normal prostates that calpain was not up-regulated in relationship to invasiveness or metastatic activity, consistent with expectation from the biological role of this effector. Taken together, these results strongly suggest that epigenetic activation of calpain plays an important role in the invasion of human prostate cancer and that it can be targeted to reduce tumor progression.