Seeding properties of amyloid-beta and tau in the cerebrospinal fluid

Alzheimer’s disease (AD) is the most common neurodegenerative disorder, and its prevalence is still increasing. However, currently there are no treatment options available, nor reliable presymptomatic biomarkers for its diagnosis. Neuropathologically, AD is characterized by aggregates (plaques and cerebrovascuar deposits; and tangles) composed of two different proteins: amyloid-β (Aβ) and tau, respectively. Brain derived Aβ and tau exhibit prion-like properties, as based on recent studies.
Small and soluble Aβ species have been identified as the most potent Aβ seeds, and they may be present in the CSF which could be of a diagnostic value. Injection of CSF obtained from AD patients into APP23 transgenic mice however did not result into induction or acceleration of amyloidosis, indicating that human CSF Aβ is not seed competent, in contrast to brain Aβ (Fritschi et al. 2014). In order to test this hypothesis further, we injected susceptible mice with APP23 mice’ CSF, containing Aβ at higher amounts than the human CSF, and sacrificed the mice after longer seeding time. As a result, we did not see a significant seeding effect, confirming that CSF Aβ does not show relevant prion-like properties in vivo.
Next, we looked at the prion-like properties of tau in the CSF. Since tau concentration in the CSF increases with progression of AD, unlike Aβ levels, it is possible that seeding competent tau species might reach the CSF compartment. For this purpose, we collected CSF from aged, tangle-bearing P301S mice and injected it into young, pretangle stage P301S mice. As a result, we observed significantly higher number of hyperphosphorylated tau inclusions in the injected hippocampus, indicating that CSF tau can induce tau hyperphosphorylation in the host mice. This could be indicative of an early prion-like seeding response. We then investigated whether CSF obtained from human tauopathy patients could induce tau aggregation using the same methodology. The collection of human CSF samples as well as the analysis of human CSF seeded mice is still ongoing.
Tau protein fragments have been identified in both the CSF and the brain of AD patients, however it is not clear what is their role in the progression of the disease. Recent studies suggest that tau fragmentation, rather than aggregation per se, may play an important role in neurodegeneration. To study the relation between truncated and full-length tau in vivo and shed more light on this question, we used an inducible mouse line expressing truncated 3R tau, and crossed it with full-length tau-expressing mice (3R, or 4R with and without a mutation). As a result, mice exhibited severe neuronal loss and motor palsy in the absence of tau aggregation. However, they recovered once the expression of fragmented tau was ceased, except for the 3R-expressing mice. This shows the importance of fragmented tau for toxicity and points at new therapeutic targets in the treatment of tauopathies.
Taken together, the results presented here point at the possible use of CSF tau in the development of future AD diagnostic essays, and implicate tau truncation as a potential pharmacological target in tauopathies.