The mechanism by which cells die in Alzheimer disease (AD) is unknown. Several investigators speculate that much of the cell loss may be due to apoptosis, a highly regulated form of programmed cell death. Caspase-3 is a critical effector of neuronal apoptosis and may be inappropriately activated in AD. To address this possibility, we examined cortical and hippocampal brain sections from AD patients, as well as 2 animal models of AD, for in situ evidence of caspase-3 activation. We report here that senile plaques and neurofibrillary tangles in the AD brain are not associated with caspase-3 activation. Furthermore, amyloid beta (A beta) deposition in the APPsw transgenic mouse model of AD does not result in caspase-3 activation despite the ability of A beta to induce caspase-3 activation and neuronal apoptosis in vitro. AD brain sections do, however, exhibit caspase-3 activation in hippocampal neurons undergoing granulovacuolar degeneration. Our data suggests that caspase-3 does not have a significant role in the widespread neuronal cell death that occurs in AD, but may contribute to the specific loss of hippocampal neurons involved in learning and memory.