ABSTRACT
In Alzheimer´s disease (AD), hyperphosphorylated tau spreads along the cerebral cortex in a stereotypical pattern that parallels cognitive deterioration. Tau seems to spread transsynaptically along cortico-cotical pathways that, according to synaptic tract-tracing studies in nonhuman primates, have specific laminar patterns related to the cortical type of the connected areas. This relation is described in the Structural Model. In the present article, we study the laminar distribution of hyperphosphorylated tau, labeled with the antibody AT8, along temporal cortical types in postmortem human brains with different AD stages to test the predictions of the Structural Model. Brains from donors without dementia had scant AT8-immunorreactive (AT8-ir) neurons in allo-, meso-, and isocortical types. In early AD stages, the mesocortical dysgranular type, including part of the transentorhinal cortex, had the highest AT8 immunostaining and AT8-ir neurons density. In advanced AD stages, AT8 immunostaining increased along the isocortical types until reaching the auditory koniocortex. Regarding laminar patterns, in early AD stages there were more AT8-ir neurons in supragranular layers in each de novo involved neocortical type; in advanced AD stages, AT8-ir neurons were equally distributed in supra- and infragranular layers. These AT8-ir laminar patterns are compatible with the predictions of the Structural Model. In summary, we show that hyperphosphorylated tau initially accumulates in allo-, meso-, and isocortical types, offer a proof of concept for the validity of the Structural Model to predict synaptic pathway organization in the human cerebral cortex, and highlight the relevance of nonhuman primate tract-tracing studies to understand human neuropathology.
Subject(s)
Alzheimer Disease , Cerebral Cortex , Neural Pathways , tau Proteins , Alzheimer Disease/pathology , Alzheimer Disease/metabolism , Humans , tau Proteins/metabolism , Male , Female , Cerebral Cortex/metabolism , Cerebral Cortex/pathology , Aged , Phosphorylation , Aged, 80 and over , Neural Pathways/metabolism , Neural Pathways/pathology , Neural Pathways/chemistry , Middle Aged , Models, Neurological , Neurons/metabolism , Neurons/pathologyABSTRACT
Synaptic tract-tracing studies in macaques have provided a wealth of data about cortico-cortical connections that have been used to identify regularities and propose models and theories to explain cortical connectivity. The two most relevant of these models are the distance rule model (DRM) and the structural model (SM). They relate the strength and laminar pattern of cortico-cortical connections to two different factors: Euclidean distance (according to the DRM) and cortical type distance (according to the SM). If both predictive factors were correlated, the DRM and the SM would be compatible, but quite often, two cortical areas of similar cortical type are far apart from each other. In the present article, we have performed a conceptual analysis of the DRM and the SM to obtain predictions from each of the two models about strength and laminar pattern of cortico-cortical connections. We then tested the predictive power of each model with analyses of several cortico-cortical connectivity databases to check which of them provide the most accurate predictions. We conclude that the DRM and the SM capture the decrease in connection strength with increasing Euclidean and cortical type distances, respectively; but, for laminar pattern, type distance is a better predictor than Euclidean distance.
