Reply: the early pathological process in sporadic Alzheimer’s disease

Abstract

[7] was designated NFT stage I [8]. Gallyas-negative AT8immunopositive pretangles, described by Bancher et al. [5] and Braak et al. [6], were included in a revised staging protocol for AD [13], but subcortical and cortical “pretangle stages” (a–c, 1a, 1b) as such have been proposed only recently [10, 11, 14]. The LC is not routinely examined at autopsy for the presence of aggregated tau or other proteinopathies in children or young and middle-aged persons. We performed such a study on 42 cases under the age of 30 using AT8 immunohistochemistry and found tau aggregates in LC neurons of 38/42 (90 %) individuals [10]. The tau pathology was located only in dorsal portions of the LC that contain cortically projecting neurons [27, 43]. 41 out of 42 (98 %) cases lacked cortical Aβ deposition [10]. In a larger autopsy-based study, elobeid et al. [23] reported AT8-immunopositive tau in the LC of 25/83 (30 %) young and middle-aged cases, all without concomitant cortical Aβ deposits and all without aggregated tau in the transentorhinal region. If, as Mann and Hardy [41] speculate, the LC lesions in some of the young subjects were driven by acute lethal head injury [10], we also should have encountered at least some Aβ deposition in the brains of these individuals, inasmuch as head trauma studies performed in humans have reported the presence of transient Aβ deposits [26, 54]. Yet, this was not the case. performance of 3r and 4r immunohistochemistry can allay (or confirm) the reservations expressed by Mann and Hardy [41] regarding the absence of evidence for the different isoforms of phospho-tau in such cases. However, the presence of AT8-immunopositive tau aggregates in the LC of both relatively young cohorts makes it unlikely that all or most of the lesions seen there represent prodromal (“incidental”) pathology associated with rare non-AD tauopathies, such as corticobasal degeneration (CBD) or We are inclined to think that the riddle of the chicken or the egg (i.e., the chicken egg or simply an egg?) is too problematic to help illustrate the problem of tau and amyloid-β (Aβ). However, one can summarize our respective viewpoints in the form of “either/or” propositions: Either Aβ is present in the extracellular space at cortical predilection sites and, from there, causes tau to aggregate within coeruleus neurons, as postulated by Mann and Hardy [41]— the authors do not mention how extracellular cortical Aβ reaches the locus coeruleus (LC) to influence nerve cells there (e.g., by means of retrograde axonal transport)—“the undiscerning ‘locus chicken’ eagerly pecks at” the “bad egg (Aβ deposition).” Or involved projection cells (i.e., nerve cells containing aggregated tau) in the LC and in other nonthalamic nuclei with diffuse cortical projections [10, 29, 31, 50] cause via anterograde axonal transport the release of Aβ into the cortical extracellular space [46], where it can aggregate at local diffusion zones into diffuse amyloid plaques, as we have recently postulated [12]. In our view, the pathological process underlying sporadic Alzheimer’s disease (AD) develops over the course of a lifetime [11, 14]. The original staging protocol was based on the topographical distribution patterns of (1) Gallyasstained neurofibrillary changes (tangles, neuropil threads, dystrophic neurites of neuritic plaques) across six stages (NFT I–VI) and (2) Campbell–switzer-stained Aβ plaque deposition (stages A–C) in the cerebral cortex [8]. The earliest stage of cortical neurofibrillary pathology in the superficial entorhinal layer pre-α of the transentorhinal region