Human iPSC-derived 2D and 3D disease models to study tau-driven vulnerabilities in Frontotemporal Dementia and identify opportunities for drug repurposing

Tauopathies are a heterogeneous group of neurodegenerative disorders characterized by the abnormal accumulation of tau protein aggregates. The overlapping clinical and pathological features among these conditions complicate both early diagnosis and accurate classification, underscoring the complexity of tau’s role in neurodegeneration and the critical need for targeted therapeutic approaches. Tau, encoded by the MAPT gene, regulates microtubule stability and dynamics within neurons. MAPT pre-mRNA undergoes developmentally regulated alternative splicing, generating six isoforms in the adult human brain that differ by the presence of zero, one, or two N-terminal inserts (0N, 1N, 2N) and three or four C-terminal microtubule-binding repeats (3R or 4R). A balanced 3R/4R ratio is essential for neuronal homeostasis, and perturbations in this equilibrium are a hallmark of several tauopathies. Among these, frontotemporal dementia (FTD) represents one of the most severe clinical presentations, characterized by progressive degeneration of the frontal and temporal lobes and resulting in significant behavioural, cognitive, and language impairments. The intronic MAPT IVS10+16 mutation, in particular, has been associated with familial FTD, since it disrupts the ratio of tau isoforms in favour of the 4R variants, which exhibit increased aggregation propensity and compromising microtubule binding, ultimately compromising axonal transport and synaptic connectivity. Despite significant advances in understanding tau-related mechanisms and numerous clinical trials, no interventions have yet proven capable of slowing or altering disease progression. Since neurodegeneration begins many years before the onset of clinical symptoms, the early detection of pathological changes becomes essential. This scenario highlights the need for highly sensitive biomarkers that can identify the disease at an initial stage, when therapeutic approaches may still exert meaningful efficacy. In this context, retina has gained increasing attention as an accessible site for neurodegeneration research. As an extension of the CNS, the retina mirrors cerebral pathology, and retinal imaging biomarkers such as tau accumulation, layer thinning, and glial reactivity are emerging as diagnostic and prognostic tools in FTD and other tauopathies. The development of suitable preclinical models remains a major challenge, as mouse models often fail to replicate human tau biology due to species-specific differences in MAPT structure and splicing. Human induced pluripotent stem cell (iPSC) technology has therefore emerged as a promising strategy for modeling neurodegenerative disease mechanisms in a physiologically relevant context. In this thesis I investigated the impact of MAPT IVS10+16 mutation in retinal and cortical development. I generated both monolayer and 3D retinal and cortical cultures, as well as neural rosettes derived from control and isogenic MAPT IVS10+16 iPSC lines and examined the tauopathy-related neurodegenerative processes. Retinal and cortical neurons exhibited shared disrupted molecular pathways, providing a powerful and complementary platform for disease modeling and therapeutic screening. These findings demonstrate that MAPT IVS10+16 leads to delayed retinal and cortical development and impaired neuronal maturation. The mutation induces an excess of 4R tau and accumulation of pathological tau species, altering cytoskeletal dynamics and synapse formation. Furthermore, the results support a growing concept that tauopathy may have neurodevelopmental roots, with early mitochondrial defects creating vulnerabilities that emerge clinically decades later under aging-related stress. Thus, a therapeutic strategy was investigated targeting these metabolic impairments with Bezafibrate (BZ), a PPAR agonist and enhancer of mitochondrial biogenesis. Pharmacological treatment restored mitochondrial morphology, reduced tau hyperphosphorylation and rescued synaptic connectivity in MAPT-mutant cultures. These findings position mitochondrial targeted modulation as a promising avenue for early intervention in FTD and related tauopathies. The dissertation is structured into four main sections. The opening chapter provides a general introduction together with the state of the art on tau physiology, pathogenic roles, and therapeutic strategies. Chapter 2 presents the experimental results, structured into three parts that progressively explore (I) tau aggregates toxicity and propagation mechanisms in iPSC-derived retinal neurons, (II) the impact of the MAPT IVS10+16 mutation on retinal development, and (III) the mechanistic interplay between mitochondrial dysfunction, neuroepithelial patterning, and tau-driven neurodegeneration in cortical models. Chapter 3 summarized the methods and protocols employed through the experiments. The final chapter discusses the broader implications of the findings and outlines future perspectives for advancing iPSC-based disease modeling.