In several brain injury models, zinc accumulates in degenerating neuronal somata. Suggesting that such zinc accumulation may play a causal role in neurodegeneration, zinc chelation attenuates neuronal death. Because histochemically reactive zinc is present in and released from synaptic vesicles of glutamatergic neurons in the forebrain, it was proposed that zinc translocation from presynaptic terminals to postsynaptic neurons may be the mechanism of toxic zinc accumulation. To test this hypothesis, kainate seizure-induced neuronal death was examined in zinc transporter 3 gene (ZnT3)-null mice, a strain that completely lacks histochemically reactive zinc in synaptic vesicles. Intraperitoneal injection of kainate induced seizures to a similar degree in wild type and ZnT3-null mice. Staining of hippocampal sections with a zinc-specific fluorescent dye, N-(6-methoxy-8-quinolyl)-p-carboxybenzoylsulfonamide, revealed that zinc accumulated in degenerating CA1 and CA3 neurons in both groups, indicating that zinc originated from sources other than synaptic vesicles. Injection of CaEDTA into the cerebral ventricle almost completely blocked zinc accumulation in ZnT3-null mice, suggesting that increases in extracellular zinc concentrations may be a critical event for zinc accumulation. Arguing against the possibility that zinc accumulation results from nonspecific breakdown of zinc-containing proteins, injection of kainate into the cerebellum did not induce zinc accumulation in degenerating granule neurons. Taken together, these results support the existing idea that zinc is released into extracellular space and then enters neurons to exert a cytotoxic effect. However, the origin of zinc is not likely to be synaptic vesicles, because zinc accumulation robustly occurs in ZnT3-null mice lacking synaptic vesicle zinc.