Tau aggregates: where, when, why and what consequences?
With the recognition that tangles are formed primarily from hyperphosphorylated tau, a microtubule‐binding protein, it also became clear that tau aggregates can take on forms other than traditional neurofibrillary tangles. In PSP and CBD, a significant portion of the aggregated tau occurs in the form of distinct types of glial inclusions: tufted astrocytes, astrocytic plaques and coiled bodies in oligodendrocytes. Better understanding of the structure of tau and its gene (MAPT) has allowed for the recognition that tau aggregates can contains mixtures of the protein with 3 (3R) and 4 (4R) copies of the microtubule‐binding domain (as in AD), predominantly 4R tau (PSP, CBD) or predominantly 3R tau (Pick disease). Electron microscopy provided additional early evidence for structural differences between tau aggregates in different settings, with identification of paired helical filaments as well as straight filaments. Finally, it has been shown that the tau aggregates in distinct settings are characterized by different patterns of phosphorylation and overall conformation as assessed by immunological methods.
In the context of AD, a wide range of studies have validated the Braak and Braak scheme of spatial progression of neurofibrillary tangles from the entorhinal cortex through the hippocampal formation and related areas to the neocortex with eventual involvement of the primary motor and sensory cortices. This propagation across the nervous system through anatomic regions with functional connections has been more recently validated through the demonstration that aggregates of tau can be taken up by cells and result in the development of de novo aggregates. This prion‐like behaviour has been demonstrated in a range of model systems, with recent studies in mouse models demonstrating that the release of tau aggregates from one neuron with impact on another happens early in the process of disease development, well in advance of irreversible cellular injury. Additionally, there has been mounting evidence that, just as prion strains can propagate with transmission from cell to cell, tau aggregate characteristics are similarly maintained when an existing aggregate seeds the development of new aggregates. Studies so far have found that propagation occurs in an anterograde direction or in local regions, suggesting that synaptic release of tau aggregates may be involved.
While the early entorhinal cortex involvement with tangles has been long established, other brain regions have also been found to have accumulation of tangles with associated neuronal loss. The recognition of these changes in the cholinergic neurons of the nucleus basalis of Meynert was associated with the ‘cholinergic hypothesis’ of the aetiology of AD. Part of the attraction of this hypothesis was the wide cortical projection from this nucleus – a mechanism for widespread dysfunction driven by a localized cellular lesion.