Abstract Body

Considerable experimental and observational evidence suggests that progression of pathology in tauopathy occurs via trans-cellular propagation of pathological assemblies. According to this model, aggregates of unique conformation (strains) that form in one cell escape and enter adjacent or connected cells to serve as templates for their own replication. We have observed that tau aggregates propagate a unique structure indefinitely in cultured HEK293T biosensor cells that express tau repeat domain (RD) containing two disease-associated mutations. These strains reliably seed identical structures in cells, but it has been unclear to what extent this process reflects the structure of a tau seed. We have used an alanine scan to probe amino acids required for incorporation of tau monomer into a seed using an extended RD sequence (aa 246-408) that ecompasses all known filament structures derived by cryo-EM. We have tested this system against recombinant assemblies with structures determined by cryo-EM, including fibril structures associated with Alzheimer disease, and other fibrils of different structure derived from the same fragment. The alanine scan precisely categorized structures, and parsed dozens of brain samples into neuropathologically determined diagnostic categories. Energetic analyses of the role of alanine residues within assemblies reveals a remarkable concordance between experimental findings and computational predictions. Residues critical for replication were often located within the core of the protofilament. This suggests monomer folding is a critical component of fibril assembly. Given a potential role for cofactors in assembly formation, this may explain why stable strains do not precisely replicate within all cells. We hypothesize that cofactors dictate fidelity of strain replication. This would explain faithful replication of strains that evolve in one cell type, and discrepancies when they are propagated to different cells.

Aims

We seek to uncover early disease mechanisms of Frontotemporal lobar degeneration with Tau pathology (FTLD-Tau) and Alzheimer’s Disease (AD) by using human neural organoids (hNOs) with the microtubule associated protein tau (MAPT) R406W gene mutation. We identified a cholesterol transporter, GRAM Domain Containing 1B (GRAMD1B), is significantly increased in excitatory neurons (EX) of mutants compared to isogenic controls. The pathophysiological role of GRAMD1B in the brain is unknown. Our objective is to elucidate the relationship between cholesterol transporter, GRAMD1B, tau pathology, and lipid metabolism in neurodegeneration.

Methods

We applied a comprehensive set of techniques to hNOs including single-cell RNA sequencing, shotgun lipidomics, micro-electrode arrays, RNAScope single molecule fluorescent in situ hybridization (smFISH), western blot assay, and immunostaining. We further characterized GRAMD1B in human post-mortem brain tissue in relationship to pathological tau, free cholesterol, and lipid droplets by immunostaining.

Results

We analyzed the transcriptomic profile of FTLD-Tau patient-specific and isogenic control hNOs using scRNA-seq. The cholesterol transporter, GRAMD1B, is highly expressed in EX of MAPT R406W hNOs compared to isogenic controls. Changes in GRAMD1B gene expression was confirmed with RNAScope smFISH, and protein expression by immunostaining. MAPT R406W hNOs exhibit increased tau phosphorylation, altered functional activity, and altered lipid profiles of conditioned media. Furthermore, increased GRAMD1B correlated well with pathological tau and increased lipid droplets in human brains with FTLD-Tau and AD. The effects of GRAMD1B on tau phosphorylation and spreading via CDK5R1 and LRP1, respectively, is under investigation.

Conclusions

As far as we know, this is the first study to investigate GRAMD1B in FTLD-Tau and AD pathogenesis. Based on our preliminary findings in hNOs and human postmortem brain, we propose GRAMD1B may be a novel player in FTLD-Tau, AD and other Tauopathies.

Aims

We hypothesized that the differences in the clinical duration of the disease found in progressive supranuclear palsy (PSP) patients is a consequence of the molecular diversity of tau

Methods

We performed an extensive biochemical characterisation of the soluble, oligomeric tau species in 20 different brain regions from a cohort of 25 deeply phenotyped patients with PSP. This was complemented by detailed clinical, neuropathological and genetic data together with tau seeding amplification assays, proteomics and transcriptomics.

Results

Our study has exposed a striking patient-to-patient heterogeneity in the soluble oligomeric tau species across the 20 brain regions examined. Tau seeding in motor cortex from all subjects revealed a strong correlation between the amount of soluble oligomeric tau species and the tau seeding capacity, strongly implicating these species in driving the seeding activity of tau in PSP.

To identify factors that contribute to differences in seeding ctivity, we quantified the amount of phosphorylated tau and evaluated the physical properties of tau in these diverse brain extracts using a protease-sensitivity digestion assay that differentiates protein conformations and correlated these measures with their tau seeding capacity. Finally, we have performed the first proteomic-wide profiling of the motor cortex in PSP coupled to spatial transcriptomic analysis and uncovered several mechanistic pathways that differentiate patients with high seeding activity and low seeding activity.

Conclusions

Our observations strongly suggest that that distinct molecular populations of tau contribute to the observed phenotypic differences in PSP. These data support the concept that individuals with PSP may have multiple molecular drivers of an otherwise common phenotype and emphasize the need for a novel molecular classification of PSP that will underpin the future development of personalized therapeutic approaches to curtail symptom progression.

Aims

Several studies reported the detrimental effects of extracellular tau oligomers (ex-oTau) on glutamatergic synaptic transmission and plasticity. ex-oTau accumulation in astrocytes alters neuro/gliotransmitter handling thereby affecting synaptic function. Here we aim at determining the specific contribution of glypican-4 (GPC4), a membrane receptor belonging to the family of heparan sulfate proteoglycans, in ex-oTau uploading in astrocytes that leads to synaptic failure.

Methods

We used in vitro and ex-vivo models obtained from wild-type C57BL/6 and transgenic mice in which either APP was knocked-out or it contained the non-phosphorylatable amino acid alanine replacing threonine 688 (APP^TA mice). In these models we performed immunofluorescence analyses, Western blot and real-time PCR experiments, confocal Ca²⁺ imaging, high-performance liquid chromatography, FM1-43 imaging, electrophysiological recordings and Chromatin immunoprecipitation assays.

Results

Treatment of cultured astrocytes with a specific anti-GPC4 antibody significantly reduced oTau internalization in astrocytes and prevented oTau-induced alterations of Ca²⁺-dependent gliotransmitter release. Anti-GPC4 treatment also spared neurons from the astrocyte-mediated ex-oTau synaptotoxicity. Indeed, no significant decreases in synaptic vesicular release and synaptic protein expression were observed in anti-GPC4-treated neurons co-cultured with astrocytes following exposure to ex-oTau. The latter also failed to inhibit LTP at CA3-CA1 synapses. We also found that GPC4 expression depends on APP and, in particular, on its C-terminal domain, AICD, generated by γ-secretase-dependent cleavage of APP, that can bind Gpc4 promoter. Accordingly, GPC4 expression was significantly reduced in both APP-KO ad APP^TA mice, and ex-oTau did not exert any significant detrimental action on synaptic function in APP-KO mice.

Conclusions

Our study provides novel evidence on the crosstalk between APP cleavage and GPC4 expression playing a critical role in oTau accumulation in astrocytes. These phenomena underlie the astrocyte-dependent dysregulation of synaptic function significantly contributing to AD pathophysiology.

Aims

Identification of mechanisms by which tau promotes cellular dysfunction remains as a fundamental gap in the field. Our published data show that tau normally associates with ribosomes, but in pathological conditions, pathological tau shifts the translatome yielding a maladaptive response. However, the molecular mechanisms driving this effect remain unknown. Our aim was to define when and where tau associates with ribosomes in cognitively normal and Alzheimer’s disease brains, and how these interactions affect translation.

Methods

Human Alzheimer’s and age-matched non-demented control brains were subjected to enhanced cross-linking immunoprecipitation (eCLIP) sequencing using different antibodies to total and pathological tau. In vitro experiments were used to validate eCLIP results. Ribosome profiling was performed from the same human brain samples. Computational analyses to identify translation dynamics was performed.

Results

Tau associated robustly with RNA, many of which have not been previously reported. Early and late-stage AD tau-RNA complexes shifted from coding to non-coding sequences. The consensus sequences that associated with tau contain numerous splice sites and start codons. RNA-stabilizing species associated with tau in AD brains. Tau overexpression impaired RNA processing, transport, and translation in vitro. We also identified changes to ribosome dynamics using ribosome profiling. For example, the types of proteins translated in disease brains participate in different gene ontologies compared to controls. Ribosome profiles revealed that tau interferes with ribosome maturation, suggesting that this enriches maladaptive ribosomes.

Conclusions

Our data suggest that production and assembly of the translational machinery is inadequate, leading to a maladaptive response to tau pathology. Overcoming these deficits may confer functional benefits. Moreover, tau exerts devastating effects on RNA availability for translation. Therefore, tau-mediated maladaptive translation may be a pathogenic event, and efforts to rescue these effects are currently underway.

Aims

TMEM106B is a risk modifier of multiple neurological conditions, where a single coding variant and multiple non-coding SNPs influence the balance between susceptibility and resilience. Two key questions that emerge from past work are whether the lone T185S coding variant contributes to protection, and if the presence of TMEM106B is helpful or harmful in the context of disease. Here, we address both questions while expanding the scope of TMEM106B study from TDP models to tauopathy.

Methods

We generated two mouse models to explore the impact of TMEM106B within the context of tau pathology. One model featured a constitutive null mutation, while the other incorporated a homozygous T186S knock-in mutation (equivalent to the human T185S variant). These lines were bred with a transgenic P301S tau model (PS19) to yield eight genotypes for study: WT, KO, Tau, and KO:Tau; plus WT, Tau, KI, and KI:Tau. Animals underwent a battery of cognitive assays at 6 or 9 months of age before brain tissue was harvested to examine the effect of TMEM manipulation on neuropathology and transcription.

Results

We found that TMEM106B deletion accelerated cognitive decline, hindlimb paralysis, tau pathology, and neurodegeneration. TMEM106B deletion also increased transcriptional correlation with human AD and the functional pathways enriched in KO:tau mice also aligned with those of AD. In contrast, the coding variant protected against tau-associated cognitive decline, synaptic impairment, neurodegeneration, and paralysis without affecting tau pathology.

Conclusions

Our findings reveal that TMEM106B is a critical safeguard against tau aggregation, and that loss of this protein has a profound effect on the sequelae of tauopathy. Our study further demonstrates that the coding variant is biologically active and contributes to neuroprotection downstream of tau pathology to preserve cognitive function.

Aims

Previous studies of our group showed that the abnormal tau protein impairs mitochondrial function and dynamics. Mitochondria are coupled to the endoplasmic reticulum (ER) via mitochondria-associated ER membranes (MAMs), which are known to be altered in Alzheimer’s disease (AD). An important MAMs function is the regulation of cholesterol synthesis and transfer to mitochondria. Therefore, we aimed to elucidate the impact of disease-associated tau on the MAMs with a specific focus on cholesterol homeostasis.

Methods

We used SH-SY5Y cells expressing mutant tau (P301Ltau) and the corresponding vector-expressing cells to study the impact of tau on the mitochondrial morphology, the ER-mitochondria contacts, and intracellular cholesterol homeostasis. Key findings were recapitulated in pR5 (P301Ltau mutant) mice, as well as in patient-derived induced pluripotent stem cells (iPSCs) bearing the P301Ltau mutation.

Results

Our key results were that:

- Contact points between the ER and mitochondria are decreased in the presence of P301L mutant tau protein.

- P301Ltau disturbs cholesterol metabolism and its conversion to pregnenolone within mitochondria.

- The P301Ltau-induced disruption of ER-mitochondria contacts and cholesterol homeostasis is restored by GSK3β inhibition.

Conclusions

This study provides the first description of the impact of disease-associated tau on the ER-mitochondria interactions and the consequences on mitochondrial cholesterol metabolism. Our results suggest that correcting damaged ER–mitochondria associations may also correct other tau-inducted mitochondrial and cellular dysfunctions.

This work was supported by grants from the Synapsis Foundation (Dementia Research Switzerland), the Universität Basel Forschungsfonds and the Novartis Foundation for Medical-Biological Research.