No abstract

Deposits of abnormally phosphorylated tau protein are found in numerous neurodegenerative disorders; the 'tauopathies', which include Alzheimer's and Pick's diseases, but tau pathology is also found in the ageing brain. Variation in tau pathology in brain ageing and its relationship to development of tauopathies and cognitive impairment remains unclear. We aimed to determine the extent and pattern of spread of tau pathology in the hippocampus, a susceptible region important in dementia and milder states of memory impairment, using hippocampal samples from the elderly population-based Medical Research Council Cognitive Function and Ageing Study neuropathology cohort. Tau deposition was assessed in hippocampal anatomical sub-regions using the AT8 antibody to phosphorylated tau and isoform-specific antibodies to 3 and 4-repeat tau (RD3 and RD4). Abeta pathology was also assessed. In this population sample, which includes the full ageing spectrum from individuals with no cognitive impairment to those with dementia satisfying clinico-pathology criteria for Alzheimer's disease, we have demonstrated a high prevalence at death of tau pathology. AT8, Abeta, RD3 and RD4 showed similar regional distribution and increased RD3 was noted in late-stage ghost tangles. Abeta was shown to be a poor explanatory variable for tau pathology. Tau deposition progressed in a hierarchical manner. Hippocampal input regions and projection zones (such as lateral entorhinal cortex, CA1/subiculum border and outer molecular layer of dentate) were initially affected, with anterograde progression though the hippocampal circuitry. Six hippocampal tau anatomical stages were defined, each linking projectionally to their adjacent stages, suggesting spread of tau malfunction through neuroanatomical pathways in hippocampal ageing. These stages were significantly associated with dementia, and may provide a clinically useful tool in the clinico-pathological assessment of dementia and mild cognitive impairment.

These findings suggest that the changes in oxidative stress and the astrocyte DNA damage response are not accounted for as an accumulating effect due to established Alzheimer-type pathology. We hypothesize that astrocyte damage, leading to impaired function, may contribute to the development of ageing brain pathology in some individuals.

We here describe the study-design major findings from the neuropathological component of the Medical Research Council Cognitive Function and Aging Study (MRC CFAS). MRC CFAS is a population-representative study of aging and health including more than 18000 participants at baseline. More than 500 brain donations were accrued to date and have been subjected to comprehensive pathological assessment. This resource enables a thorough epidemiological description of the neuropathology associated with dementia in the UK. Results to date reveal a high prevalence of mixed Alzheimer and vascular pathology, a significant population who die with dementia but with a more limited pathological burden than is traditionally associated with dementia, and a group who die with a significant pathological burden yet remained cognitively intact until death. This dissociation between pathology and dementia increases with increasing age. Further studies have described the distribution and etiology of neurodegenerative disease in the population, and determined pathological correlates of cognitive impairment and dementia. Brain donation programs linked to epidemiological studies provide an invaluable resource for describing the pathological correlates of dementia in a way that is representative of the population, thereby identifying targets for and assessing the likely effect of therapeutic and preventive interventions.

Background: Glial tau pathology is seen in certain tauopathies and in ageing. We determined its frequency in ageing mesial temporal lobe and its relationship to other tau pathologies in the MRC-CFAS population-representative neuropathology cohort. Methods: Mesial temporal lobe, including hippocampus, amygdala, entorhinal cortex and white matter, was examined using immunohistochemistry with the AT8 antibody to phospho-tau and RD3 and RD4 antibodies to 3R and 4R tau isoforms. Gallyas silver stain was used to detect fibrillar aggregates. Results: Thorn-shaped astrocytes (TSA), positive with AT8, RD4 and Gallyas, were present in 49% of cases. They were particularly prevalent in subpial, periventricular and white matter perivascular locations and were less frequent in grey matter. Coiled bodies were seen in 18.8%. TSA were not related to Braak neurofibrillary tangle or hippocampal tangle pathology stages. TSA in grey matter were associated with coiled bodies (p = 0.011) and argyrophilic grains (p = 0.048), which were identified in 11.5% of cases. They did not correlate with dementia. Conclusions: Astrocyte tau pathology is common in the ageing mesial temporal lobe. Its formation is independent of Alzheimer-type pathology. It is a 4R tauopathy, which may form part of a mesial temporal age-related 4R tauopathy that includes oligodendroglial tau and argyrophilic grains.