Morrison and Hof (2007) review the mechanisms of action for memory decline in both Alzheimer’s diseases (AD) and non-AD individuals. For AD patients, the key factor seems to be neurofibrillary tangles (NFTs), which are correlated with the death of neurons in corticocortical curcuits, and which disconnect regions within the neocortex. Counter-intuitively, the most damaging changes for everyday life activities probably do not occur in the hippocampus but instead in the association neocortex. Preventing neuron death there is probably the best potential strategy to deal with the disease.
Complicating the matter is that non-AD patients still show memory loss with aging but do not have significant neuron losses in either the neocortex or the hippocampus. In the prefrontal cortex of monkeys, dendritic length and numbers of dendrite segments does not change with age, but the number of dendritic spines emerging from the base and apex of pyramidal cells is 35% less in aged monkeys as compared to young ones. This would decrease the probability of excitation in these neurons and could help account for memory decline without neuron loss. Other noted changes from young to old brains with cell death are decreased glutamate receptors in these same pyramidal neurons, changes in the NR1subunit of NMDA receptors of the dentate gyrus, altered synpatic transimisson due to decreased glutamate release in the outer and middle layers of the dentate gyrus as well as CA3, and probable differences in the way estrogen affects dendrite spine density in CA1. Stopping the structural changes would be hard and is not currently a priority but it is interesting to speculate whether it would be feasible and how it would happen.
Morrison JH, Hof PR. 2007 Life and death of neurons in the aging cerebral cortex. International Review of Neurobiology 81:41-57. doi:10.1016/S0074-7742(06)81004-4.