The oxygen required to meet metabolic needs of all tissues is delivered by the erythrocyte, a small, flexible cell, which, in mammals, is devoid of a nucleus and mitochondria. Despite its simple appearance, this cell has an important role in its own distribution, enabling the delivery of oxygen to precisely meet localized metabolic need. When an erythrocyte enters in a hypoxic area, a signalling pathway is activated within the cell resulting in the release of ATP in amounts adequate to activate purinergic receptors on vascular endothelium, which trigger secretion of nitric oxide and other factors resulting in vasodilatation. Among other mechanisms, binding of deoxyhemoglobin to the cytoplasmic domain of the anion-exchange protein band 3 is probably involved in this pathway. The present study investigates the effect of amyloid beta peptide exposure on this molecular mechanism. We report that deoxygenated human erythrocytes fail to release ATP following 24 h exposure to amyloid beta peptide. Concurrently, amyloid beta peptide induces caspase 3 activation. Preincubation of amyloid beta peptide treated erythrocytes with a specific inhibitor of caspase 3 prevents amyloid-induced caspase 3 activation and restores the erythrocyte's ability to release ATP under deoxygenated conditions. Since the activity of red cell phosphofructokinase, a key step in glycolytic flux, is not modified within the red cell following amyloid peptide exposure, it is likely that ATP release reduction is not dependent on glycolytic flux alterations. It has also been suggested that the heterotrimeric G protein, Gi, and adenylyl cyclase are downstream critical components of the pathway responsible for ATP release. We show that cAMP synthesis and ATP release are not failed in amyloid-peptide-treated erythrocytes in response to incubation with mastoparan 7 or forskolin plus 3-isobutyl-1-methyl xanthine, agents that stimulate cAMP synthesis. In conclusion, these results indicate that amyloid beta peptide inhibits ATP release from deoxygenated erythrocytes by activating red cell caspase 3, suggesting a pathophysiologic role for vascular amyloid peptide in Alzheimer's disease.