In one clever but still unproved scheme for blocking the progression of AIDS, a so-called "intrakine" is to be transduced into lymphocytes to prevent surface expression of a molecule required for HIV entry. Infection by HIV depends in part on the surface protein CXCR4, a receptor whose endogenous ligand is the chemokine SDF-1. Earlier work showed that an engineered SDF-1 variant that is retained in the endoplasmic reticulum has a dominant effect on CXCR4, preventing the receptor from reaching the cell surface and thus rendering the cell resistant to HIV. Clinical tests of this approach are ongoing, but meanwhile Zeelenberg and colleagues have used the same system to examine other cellular effects of loss of CXCR4 surface expression. They show here that T lemphocyte-derived hybridomas--usually an invasive cell type that can cross multiple barriers to reach target organs-- lose their ability to migrate in response to SDF-1. These transfected hybridomas fail to cross cell monolayers in culture or to exit the bloodstream when introduced in vivo. A control intrakine that blocks surface expression of a different chemokine receptor, CCR4, does not affect the ability of these cells to disseminate or generate malignant tumors, indicating that SDF-1 signaling is specifically required for invasive behavior by these cells. Whether normal T cells also depend on the SDF-1/CXCR4 interaction to reach their target tissues is not clear, but if so, the SDF-1 intrakine strategy for treating AIDS may prove to be immunosuppressive in its own right.