Cell bodies of neurons at risk of death in Alzheimer disease (AD) have increased lipid peroxidation, nitration, free carbonyls, and nucleic acid oxidation. These oxidative changes are uniform among neurons and are seen whether or not the neurons display neurofibrillary tangles and, in fact, are actually reduced in the latter case. In consideration of this localization of damage, in this review, we provide a summary of recent work demonstrating some key abnormalities that may initiate and promote neuronal oxidative damage. First, mitochondrial abnormalities might be the source of reactive oxygen species yielding perikaryal oxidative damage. The common 5-kb deletion mitochondrial (mt)DNA subtype was greatly increased in the AD cases, but only in neurons at risk. The importance of such mitochondrial abnormalities to oxidative stress was indicated by a high correlation coefficient between the extent of the mtDNA increase and RNA oxidative damage (r2 = 0.87). Nonetheless, because mitochondria in AD do not show striking oxidative damage, as one would expect if they were the direct producer of free radical species, we suspected that abnormal mitochondria supply a key reactant that, once in the cytoplasm, releases radicals. One such reactant, hydrogen peroxide, (H2O2), abundant in mitochondria, can react with iron via the Fenton reaction to produce.OH. To demonstrate this directly using a modified cytochemical technique that relies on the formation of mixed valence iron complexes, we found that redox-active iron is associated with vulnerable neurons. Interestingly, removal of iron was completely affected by using deferroxamine, after which iron could be rebound to re-establish lesion-dependent catalytic redox reactivity. Characterization of the iron-binding site suggests that binding is dependent on available histidine residues and on protein conformation. Taken together with our previous studies showing abnormalities in the iron homeostatic system including heme oxygenase, iron regulatory proteins 1 and 2, ceruloplasmin, and dimethylargininase, our results indicate that iron misregulation could play an important role in the pathogenesis of AD and therefore chelation therapy may be a useful therapeutic approach. Finally, we wanted to determine the proximal cause of mitochondrial abnormalities. One interesting mechanisms involves re-entry into the cell cycle, at which point organellokinesis and proliferation results in increased mitochondria. Supporting this, we have considerable in vivo and in vitro evidence for mitotic disturbances in AD and its relationship with the pathogenesis of AD.