Initiation and spreading of Tau pathology. is β-Amyloid the only key?

Neurodegenerative diseases associated with dementia affect 5-10 % of individuals over
the age of 65 in the Western world and represent one of the main health-related socioeconomic
burdens. The most common form of dementia, Alzheimer’s disease (AD),
affects 98000 people in Switzerland, with 23000 newly diagnosed cases each year
additionally underscoring the importance of understanding its pathogenesis with the
ultimate goal of developing efficient therapeutic strategies.
Neuropathologically, AD is characterized by the extracellular deposition of Amyloid-β
peptide (Aβ) and the intracellular aggregation of hyperphosphorylated Tau protein.
Early-onset AD, which accounts for only 6%-7% of all AD cases, is generally defined as
occurring before the age of 60 and is due to hereditary mutations in genes that promote
the deposition of Aβ. Based on these familial AD cases (FAD), the ‘Amyloid cascade’
hypothesis was formulated, postulating that deposition of Aβ, which triggers subsequent
Tau pathology, is at the heart of AD pathogenesis. However, somewhat contradictory to
this hypothesis, abundant filamentous Tau deposits and neuronal damage can also occur
in the absence of Aβ pathology and are in fact key features of a heterogeneous group of
neurodegenerative disorders termed ‘tauopathies’. In the late nineties, the discovery of
multiple mutations in the Tau gene underlying a particular condition called
frontotemporal dementia with parkinsonism linked to chromosome 17 (FTDP-17)
provided direct evidence that mutant Tau dysfunction per se is sufficient to cause
neurodegeneration. Of note, in AD and many other tauopathies, Tau pathology spreads
intracerebrally following a stereotypical temporo-spatial pattern that correlates well with
the cognitive impairment of affected patients. Intriguingly, however, the mechanisms by
which Tau pathology initiates and spreads within the brain remain largely unknown so
far.
The present research project therefore focused on the experimental induction and the
spreading of Tau pathology. Taking advantage of several transgenic mouse models that
exhibit neuropathological characteristics of AD and tauopathies, we aimed to elucidate
the relative contributions of Aβ and Tau, respectively, to disease pathogenesis. In particular, we employed the APP23 mouse model where a mutated human amyloid
precursor protein (APP), the so-called Swedish mutation, is transgenically expressed.
These mice develop Aβ plaques at 6 months of age in association with the typical ADassociated
pathology including cerebral amyloid angiopathy (CAA), neuron loss, glial
activation, and cognitive impairment.
Moreover, we also studied various Tau transgenic mouse models including the JNPL3
strain, which harbours the human FTDP-17-associated P301L Tau mutation and which
develops NFTs by the age of 6.5 months. We backcrossed this line to the C57Bl/6J
background (B6/P301L). B6/P301L heterozygous females developed a delayed
filamentous Tau pathology, occurring at around 17 months of age, which might represent
a particularly suitable animal model to experimentally modulate the onset of tauopathy.
Surprisingly, heterozygous B6/P301L male mice developed only little Tau pathology by
the age of 26 months, restricted to the spinal cord and the brain stem.
In addition, we took advantage of the P301S mouse line transgenically expressing the
FTDP-17-linked P301S mutation. These mice develop abundant filaments made of
hyperphosphorylated Tau protein and become severely paralyzed by 5-6 months of age
due to a high load of fibrillar Tau pathology in the brain stem and spinal cord. Finally, we
used the ALZ17 mouse model which transgenically expresses the human wild-type Tau
protein. These mice develop abnormal Tau hyperphosphorylation in the absence of Tau
filaments or neurodegeneration, and therefore lend themselves as so-called ‘pre-tangle’
mice. This model is of particular interest for studying the pathological steps leading to
Tau fibrils formation.
The first part of the experimental studies was centred on the in vivo relationship between
Aβ and Tau pathologies. In our initial study, we used lentivirus technology for its ability
to infect nondividing cells and to stably maintain long-term transgene expression in
specific brain regions. Lentivirus-mediated expression of the mutated human Tau protein
(LV-hTauP301S) was assessed in the brains of both wild-type and APP23 mice. Injection
of LV-hTauP301S into the hippocampus of wild-type as well as APP23 mice led to
strong and stable transgene expression for more than one year. In addition,
hyperphosphorylation of Tau was induced in both mouse lines as early as 3 months post injection, while no Tau aggregation was observed in wild-type animals up to 13 months
of mutated Tau transgene expression. In contrast, APP23 mice injected with LVhTauP301S
developed Gallyas-positive neurons indicative of the formation of Tau
filaments. We concluded from these findings that Tau aggregation can only be induced in
an environment rich in Aβ deposits, also supporting the notion that APP and/or Aβ
promote Tau pathology.
In another set of experiments, we tested alternative approaches to study the induction of
Tau pathology. In particular, we performed a series of cortical as well as
intrahippocampal injections of brain extracts bearing Aβ pathology and/or Tau pathology
into young B6/P301L animals. Aβ extracts were prepared from old APP23 mice whereas
Tau extracts were derived from aged B6/P301L animals. Brain extracts from human AD
patients (containing both Aβ and Tau pathology) were also injected into P301L/B6 mice.
Six months after infusion with Aβ-containing extracts, we found an induction of Tau
pathology not only in the injected hippocampus but also in the entorhinal cortex and
amygdala. In contrast, intracerebral injection of Tau-containing extracts produced only
limited Tau deposition. In parallel to these brain extract injection experiments, a breeding
approach was used to study the relationship between Aβ and Tau pathology by mating
B6/P301L with APP23 transgenic mice. Significantly, double-transgenic mice developed
increased fibrillar Tau pathology when compared to single B6/P301L-transgenic mice,
especially in areas with Aβ deposition. Collectively, these results demonstrated that both
injection of Aβ-containing extract and deposition of Aβ fibrils can induce intracellular
aggregation of Tau.
Finally, we investigated the Aβ-independent induction of tauopathy by injecting murine
P301S brainstem extract rich in Tau filaments into the cortex and hippocampus of young
ALZ17 mice expressing wild-type human Tau. We observed filamentous Tau pathology
as early as six months after injection with fibrillar Tau induction occurring not only in
neurons but also in oligodendrocytes. Strikingly, filamentous Tau was not restricted to
injection sites but also stereotypically developed in discrete brains regions over time, a
finding which – for the first time – clearly demonstrated the transmission and spreading
of filamentous Tau pathology in vivo. Altogether, the studies comprised by the present thesis aim to contribute to a better
understanding of Tau pathology induction and its pathogenic relation to Aβ. We have not
only demonstrated that Aβ deposits are able to potentiate Tau pathology but also that
pathological Tau per se can induce Tau aggregation. The latter observation may provide
the basis for future studies of the transmissibility of Tau pathology with special regard to
potential similarities and differences in comparison to classical prion diseases.