Evaluation of tau seeding, spreading, and cytotoxicity using in vitro and in vivo models of tau pathology

Abstract

[eng] Abnormal folding, hyperphosphorylation, aggregation, and subsequent deposition of the microtubule-associated protein tau, is the hallmark of a group of devastating neurodegenerative diseases known as tauopathies, including Alzheimer’s disease. One striking aspect of Alzheimer’s disease is that the presence of tau-related lesions in the brain occurs in a systematic, sequential manner, maintaining a predictable distribution pattern between synaptically connected neurons that varies very little among individuals. Increasing evidence suggests that the progression of tau pathology in the diseased brain behaves like a prion. The “prion-like” hypothesis suggests that “pathological” tau engages in self-seeded fibrillization and propagates through cell-to-cell spreading. However, despite intensive research, the cellular and molecular mechanisms involved and the pathological processes linking neuronal death and tau dysfunction are not fully understood.

Although Alzheimer’s disease was first described in 1906 and has an increasing prevalence in the aging population, there is currently no treatment to prevent or cure this or any other tauopathy. Progress limitations are partially explained by the lack of appropriate models to study human tauopathies. Indeed, tau-targeting therapies that had demonstrated an improvement in the pathology in several models (i.e., in vitro and in vivo) were unable to produce positive results in clinical trials. These incongruences could be related to the fact that most experimental models rely on the over-expression of mutated tau species and the use of recombinant tau fibrils, which do not reproduce the sporadic nature of most human tauopathies.

Through this doctoral thesis, we examined various aspects of tau pathology, including tau seeding, spreading, and cytotoxicity, by implementing experimental approaches that better mimicked sporadic tauopathies. Nevertheless, at the beginning of this work, the reliability of the only commercially available cell line designed to be used as a cell-based assay to detect and report proteopathic seeding in biological samples was questioned in one publication. Given that this cell-based assay was central to the validation of the samples employed in this thesis, we conducted a thorough characterization of this cell line, known as the Tau biosensor cell line, and its ability to produce fluorescent tau aggregates. Our results show that the Tau biosensor cell line is a reliable cell-based assay that forms amyloid-like inclusions upon the addition of extracellular seed- competent tau species.

Next, we investigated the impact of extracellular seed-competent tau on the neuronal activity of primary cortical cultures derived from wild-type mice. We established an experimental setup that included microfluidic devices and calcium imaging, which allowed us to specifically treat the axons with tau, as well as monitor changes in spontaneous neural activity in a time-course manner. Although we demonstrate that cortical neurons in our microfluidic platforms display typical patterns of neuronal network activity, we do not detect changes after treating them with seed-competent tau. We then investigated how the presence of various extracellular seed-competent tau may affect neural metabolism (i.e., as an indicator of cellular viability) also in primary cortical cultures. Nevertheless, similar to what we observed in the analysis of neuronal activity, within the course of 10 days, no differences between tau-treated and untreated cells are found.

Finally, recent evidence suggests that the cellular prion protein is involved in the pathology of other prion-like proteins, such as amyloid-β and α-synuclein; however, much less is known about its role in tauopathies. We inoculated human Alzheimer’s disease-derived samples into the hippocampus of transgenic mouse models with different expressions of the cellular prion protein. We found that all mice, regardless of their genotype, have similar profiles of tau-related lesions in their brains. Hence, our findings indicate that the cellular prion protein does not have a paramount role in the onset, seeding, or spreading of tau pathology.

Taken together, our work underscores the need for more pathologically relevant models to study certain aspects of sporadic human tauopathies, which could lead to the development of effective therapeutic strategies.