Brain aging is becoming a critically important issue to consider as human longevity increases, because aging is a key risk factor for several neurodegenerative diseases, particularly AD. To understand the underlying pathogenesis of cognitive decline with aging and distinguish those changes associated with aging from those associated with dementia, it is important to identify the morphologic, age-related changes that occur in the brain. Age-related neuronal dysfunction involves many subtle changes within various brain regions. These changes include alterations in receptors, loss of dendrites and spines, and myelin dystrophy, as well as alterations in synaptic transmission. Pathological processes such as the abnormal aggregation of specific proteins are seen in many neurodegenerative diseases. Together, these multiple factors likely constitute the substrate for age-related loss of cognitive function.11–13
To assess these age-related changes in cortical cellular integrity, stereological analyses were performed on a subset of pyramidal neurons known to be vulnerable in AD and characterized by high somatodendritic levels of unphosphorylated neurofilament protein.14,15
In the neocortex these large pyramidal neurons reside in layers III and V and form long corticocortical association pathways. A comparison of prefrontal cortex area 9 in elderly individuals (controls) and individuals with different degrees of cognitive dysfunction demonstrated that large pyramidal neurons decrease dramatically in cases of definite dementia, correlating strongly with the severity of cognitive dysfunction, to a nearly complete loss (> 90%) in the end stages of AD. Furthermore, these neurons are far more likely to develop neurofibrillary tangles (NFT) and do so at a faster rate than other pyramidal cells; they also shrink considerably during NFT formation, and the largest among them are preferentially affected. Thus, these unphosphorylated neurofilament, protein-enriched neurons (the large pyramidal neurons) emerge as a strikingly vulnerable subpopulation of neurons that provide specific corticocortical connections between association areas.16,17
The loss of presynaptic markers is thought to represent a strong pathologic correlate of cognitive decline in AD. For example, assessed by immunoreactivity for the spine marker spinophilin, numbers of spines in the CA1 and CA3 fields of the hippocampus and area 9 are decreased in elderly individuals with various degrees of cognitive decline; reduced immunoreactivity was significantly related to both NFT staging and clinical severity. In addition, the total number of spines in the CA1 field and area 9 were predictive of variability in cognitive scores, and those in area 9 in particular significantly related to the cognitive outcome of the cases. These data suggest that neocortical dendritic spine loss is an independent parameter to consider in clinicopathologic correlations.18
Importantly, whereas neuronal loss in normal aging has not been demonstrated, alterations in the perforant path—which projects from the layer II neurons of the entorhinal cortex to the outer molecular layer of the dentate gyrus and is critically involved in learning and memory formation—was observed in a macaque monkey model. Expression of N
-aspartate receptor subunit 1 (NR1) was significantly decreased in the outer molecular layer of aged macaque monkeys, while no differences in the number of layer II neurons were found.13,19
These results suggest that the circuit-specific decrease in NR1 expression occurs in the absence of structural deficits of the perforant path and is due to age-dependent changes in the functional properties of this circuit.
There were also age-related morphologic alterations in pyramidal neurons that contribute to working memory circuits in the macaque monkey superior temporal cortex and that form “long” projections to the prefrontal cortex. Global dendritic mass homeostasis, assessed by three-dimensional scaling analysis, was conserved with aging in these neurons. Spine densities, dendrite diameters, dendritic lengths, and branching complexity were all significantly reduced in their apical dendrites.20,21
The passive electrotonic structure in apical dendrites of the projection neurons was significantly reduced in aged monkeys and simulated, passive back-propagating action potential efficacy was significantly higher in the apical dendrites of old neurons. These effects, in turn, increase the excitability of pyramidal neurons in aging, thus compromising the precisely tuned activity required for working memory, and ultimately result in age-related cognitive dysfunction.
In summary, while a modest disruption of memory occurs frequently in normal aging in humans and in animal models, significant neuron death does not appear to be the cause of such age-related memory deficits. Evidence from rodent and nonhuman primate models reveals that the same hippocampal and cortical circuits affected in AD exhibit subtle age-related changes in neurochemical phenotype, dendritic and spine morphology, and synaptic integrity that correlate with impaired function. Molecular alterations of synapses, such as shifts in expression of excitatory receptors, also contribute to these deficits. As such, integrity of spines and synapses may reflect age-related memory decline, whereas the loss of select cortical circuits seems to be a crucial event for functional decline in AD. Likewise, these cortical components provide important targets for the development of preventive and protective therapeutic strategies.