The delivery of proteins across the blood-brain barrier is severely limited by their size and biochemical properties. Numerous peptides have been characterized in recent years that prevent neuronal death in vitro, but cannot be used therapeutically, since they do not cross cell membrane barriers. It has been shown in the 1990s that the HIV TAT protein is able to cross cell membranes even when coupled with larger peptides. It appears, therefore, that TAT fusion proteins may enter the brain, even when used systemically. Indeed, the systemic delivery of a TAT protein linked with glial-derived neurotrophic factor (GDNF) successfully transduced central nervous system (CNS) neurons in mice. When administered after optic nerve transection and focal cerebral ischemia, TAT-GDNF protected retinal ganglion cells and brain neurons from cell death, elevated tissue Bcl-XL levels and attenuated the activity of the executioner caspase-3. These findings demonstrate the in vivo efficacy of fusion proteins in clinically relevant disease models, raising hopes that neuroprotection may become eventually feasible in human patients.