Tau immunophenotypes in chronic traumatic encephalopathy recapitulate those of ageing and Alzheimer's disease

Abstract

Traumatic brain injury (TBI) is a risk factor for neurodegenerative disease, including chronic traumatic encephalopathy (CTE). Preliminary consensus criteria define the pathognomonic lesion of CTE as patchy tau pathology within neurons and astrocytes at the depths of cortical sulci. However, the specific tau isoform composition and post-translational modifications in CTE remain largely unexplored. Using immunohistochemistry, we performed tau phenotyping of CTE neuropathologies and compared this to a range of tau pathologies, including Alzheimer's disease, primary age-related tauopathy, ageing-related tau astrogliopathy and multiple subtypes of frontotemporal lobar degeneration with tau inclusions. Cases satisfying preliminary consensus diagnostic criteria for CTE neuropathological change (CTE-NC) were identified (athletes, n = 10; long-term survivors of moderate or severe TBI, n = 4) from the Glasgow TBI Archive and Penn Neurodegenerative Disease Brain Bank. In addition, material from a range of autopsy-proven ageing-associated and primary tauopathies in which there was no known history of exposure to TBI was selected as non-injured controls (n = 32). Each case was then stained with a panel of tau antibodies specific for phospho-epitopes (PHF1, CP13, AT100, pS262), microtubule-binding repeat domains (3R, 4R), truncation (Tau-C3) or conformation (GT-7, GT-38) and the extent and distribution of staining assessed. Cell types were confirmed with double immunofluorescent labelling. Results demonstrate that astroglial tau pathology in CTE is composed of 4R-immunoreactive thorn-shaped astrocytes, echoing the morphology and immunophenotype of astrocytes encountered in ageing-related tau astrogliopathy. In contrast, neurofibrillary tangles of CTE contain both 3R and 4R tau, with post-translational modifications and conformations consistent with Alzheimer's disease and primary age-related tauopathy. Our observations establish that the astroglial and neurofibrillary tau pathologies of CTE are phenotypically distinct from each other and recapitulate the tau immunophenotypes encountered in ageing and Alzheimer's disease. As such, the immunohistochemical distinction of CTE neuropathology from other mixed 3R/4R tauopathies of Alzheimer's disease and ageing may rest solely on the pattern and distribution of pathology.

Keywords: TBI; ageing-related tau astrogliopathy; chronic traumatic encephalopathy; tau; traumatic brain injury.

Figure 1

Sulcal depth astrocytic and neuronal tau pathologies in CTE. (A and B) PHF1 immunohistochemistry reveals NFT and TSA pathology concentrated at the depths of cortical sulci of a former American football player (Case 3) and (C and D) chronic survivor of a single severe TBI (Case 11). (B and D) High magnification images from the same sulci showing perivascular pathology composed of mixed neuronal and astrocytic populations consistent with the preliminary diagnostic criteria for CTE. Scale bars = 100 µm.

Figure 2

3R versus 4R tau immunoreactivity in CTE, Alzheimer’s disease, PART and Pick’s disease. (A) Sulcal depth CTE neuropathology with prominent subpial TSAs in addition to patchy and perivascular TSAs and NFTs within the deeper layers of cortex (Case 10; PHF1 staining). (B) Higher magnification of box in A displaying perivascular astrocytic and neuronal tau pathologies. (C) Immunoreactivity-specific for 3R tau in the same region as in B showing perivascular NFTs, but an absence of immunoreactivity within astrocytes. (D) In contrast, immunohistochemistry specific for 4R tau identified cells with both neuronal and astrocytic morphologies in the same region. (E) Consistent with previous descriptions, cases meeting diagnostic criteria for Alzheimer’s disease (AD; Case 25) and PART (Case 15) displayed neurofibrillary tangles in the cortex that were immunoreactive for 3R and 4R tau. In contrast, Pick bodies of Pick’s disease (PiD) within the dentate granule cells of Case 29 were composed of only 3R tau. Scale bars = 100 µm in A–D, 50 µm in E.

Figure 3

3R versus 4R tau immunoreactivity within the astrocytic pathologies of CTE, ARTAG, PSP, CBD and Pick’s disease. Representative examples of serial sections showing immunoreactivity specific for 4R tau, but not 3R tau, in the TSAs in cases of both CTE (Case 6) and ARTAG (Case 21), tufted astrocytes in a case with PSP (Case 35) and astrocytic plaques within a case of CBD (Case 41). The ramified astrocytes observed in just one case of Pick’s disease (PiD; Case 29) were immunoreactive for 3R and 4R tau. Scale bars = 50 µm.

Figure 4

Post-translational modifications of tau within astrocytes and neurons in CTE, Alzheimer's disease, PART, ARTAG, PSP, CBD and Pick’s disease. (A) Representative examples of serial sections showing the TSAs of CTE (Case 3) and ARTAG (Case 22), astrocytic plaques of CBD (Case 41) and ramified astrocytes in a case of Pick’s disease (PiD; Case 29) all displayed robust immunoreactivity to phospho-epitope antibodies PHF1, CP13, AT100 and pS262, but not Tau-C3 (truncation at D421). In contrast with all other astrocytic tau pathologies examined, the tufted astrocytes of PSP in a subset of cases (Case 35 pictured here) were labelled with Tau-C3, indicative of truncation as described. (B) Representative examples of serial sections showing the neurofibrillary tangles of CTE (Case 5), Alzheimer’s disease (AD; Case 28) and PART (Case 15) with immunoreactivity to the phospho-epitope antibodies PHF1, CP13, AT100 and pS262, as well as Tau-C3 indicating truncation at D421. In addition, while Pick bodies (Case 29) also demonstrated the same post-translational modifications, immunoreactivity to p262 and Tau-C3 was noted in just one case as shown here within the dentate granule cells, where immunoreactivity was notably less robust. Scale bars = 50 µm.

Figure 5

Neurofibrillary tangles, but not astrocytes, are immunoreactive for antibodies that detect a conformation-dependent epitope of tau in Alzheimer’s disease. PHF1 immunohistochemistry revealed sulcal depth astrocytic and neuronal tau pathology in CTE (Case 8), including prominent clusters of perivascular and subpial TSAs (top left; black arrows). However, GT-7 and GT-38 antibodies labelled NFTs in CTE, but not TSAs, on serial tissue sections (top middle and top right). GT-7 and GT-38 reliably detected NFTs in Alzheimer’s disease (Case 28), but failed to label the characteristic pathologies of ARTAG (Case 19), Pick’s disease (PiD; Case 29), PSP (Case 35) or CBD (Case 41). Scale bars = 100 µm in CTE (low power); = 50 µm in CTE (high power), Alzheimer’s disease (AD), Pick’s disease, ARTAG, PSP and CBD.