Aims

We (and others) have previously shown that pathogenic (mutated and/or hyperphosphorylated) Tau accumulates at synapses causing microglia-independent synapse degeneration. In this study we aim to unveil the molecular mechanisms that are activated in synapses accumulating pathogenic Tau and that lead to synapse degeneration.

Methods

We combined a proteomic and genetic screen to identify mediators of Tau-induced synapse degeneration.

First, we used PS19 mice as a tauopathy model and focus on their hippocampal mossy fiber-CA3 synapses, which strongly accumulate pathogenic Tau. We purified mossy fiber-CA3 synaptosomes from PS19 and wild type mice and performed mass spectrometry to identify changes in the synaptic proteome induced by pathogenic Tau.

Second, we used an in-house generated Drosophila tauopathy model expressing human P301L Tau that shows progressive degeneration of synapses in the lamina resulting in electroretinogram (ERG) defects. We generated human P301L Tau expressing flies with reduced expression of the hits identified in the proteomic screen and recorded ERGs to assess rescue/enhancement of the Tau-induced synaptic phenotype.

Results

We identify several proteins upregulated in synaptosomes sorted from PS19 mice compared to wild type controls. Most of them are proteins physiologically associated to microtubules and neuronal axons. Among them, collapsin response mediator proteins (Crmp) were of particular interest because they are involved in Semaphorin-induced pruning of mossy fiber synapses during development. We verify that the synaptic (but not total) levels of Crmp are increased in PS19 mice as well as in AD patients by western blot. Finally, we find that loss of Crmp rescues pathogenic Tau-induced synaptic defects in fruit flies.

Conclusions

Crmp mediates developmental Semaphorin-induced synaptic pruning and pathogenic Tau-induced synapse degeneration, suggesting an exciting possibility: that both processes share common mechanisms.

Aims

Amyloid and tau proteinopathy and neurodegeneration are accompanied by changes in white matter (WM) microstructure in Alzheimer’s disease (AD). In contrast to the diffusion tensor model, fixel-based analysis¹ more accurately calculates a combined metric of fiber density and cross-section (FDC) from multishell diffusion-weighted MRI (dMRI) in voxels containing crossing fibers. Since tau is thought to spread trans-synaptically and thus disrupts structural connections, we hypothesized lower FDC in WM bundles connecting early tau accumulating brain regions (e.g. temporal cingulum bundle connecting hippocampus and posterior cingulate).

Methods

Twenty-two aMCI (73±8 years, 14F/8M, 21Ab+/1unknown, MMSE=25.1±2.1) and 22 cognitively unimpaired controls (CU) (70±9 years, 10F/12M, 4Ab+/18Ab-, MMSE=29.6±0.8) underwent ¹⁸F‑MK‑6240 PET and dMRI on a GE Signa 3T PET‑MR. PET images were corrected for partial volume effects and standardized uptake value ratios (SUVR) were calculated relative to the cerebellar cortex. dMRI data were preprocessed using MRTrix3 to obtain FDC values. ¹⁸F‑MK‑6240 PET SUVR and FDC were compared between aMCI and CU using a voxel- and fixel-based analysis, respectively.

Results

The voxel-based ¹⁸F‑MK‑6240 PET analysis showed increased binding in aMCI compared to CU with highest effect size in the medial temporal cortex, and spreading towards the lateral temporal cortex, posterior cingulate and precuneus (Figure 1, peak-height p_unc<0.001, cluster-height p_FWE<0.05, k_ext=100 voxels).

The whole-brain fixel-based analysis, corrected for total intracranial volume, showed fixels with significantly decreased FDC in the fornix, splenium of the corpus callosum and in parts of the temporal cingulum bundle (p_FWE<0.05) in aMCI compared to CU (Figure 2).

Conclusions

Our results indicate WM degeneration occurs already in the aMCI stage of AD in WM bundles connecting early tau accumulating brain regions. Further work will be directed towards directly correlating these modalities.

1. Raffelt D.A., et al. Neuroimage2017;144:58–73.

Aims

Protein aggregates in the brain are largely uncharacterised due to their solubility, low concentration, small size, and structural heterogeneity. This study aims to sensitively measure brain-derived aggregate quantity, morphology, and post-translational modifications across Alzheimer’s disease (AD).

Methods

Post-mortem brain samples were obtained from 13 donors, including all AD Braak stages and two brain regions, middle temporal gyrus (MTG) and primary somatosensory cortex (SOM). Our aggregate-specific Simoa® assays measured aggregate quantity. Single-molecule pull-down (SiMPull) uncovered tau aggregate size and shape with super-resolution microscopy, and quantified tau with multiple phosphorylation sites by AT8-T181 fluorescence colocalisation microscopy.

Results

Exponential increases in total and phosphorylated tau (p-tau) aggregate quantities were identified from Braak 3 donors onwards. No notable changes in amyloid-beta aggregate quantities were detected. For SOM, earlier AD-associated tau aggregate detection was achieved with our ultra-sensitive Simoa® assays than immunohistochemistry Braak staging. In late-stage AD, MTG showed ~10-100-fold greater tau aggregate quantity than SOM.

According to AT8/HT7, ~3-fold greater proportions of tau aggregates were phosphorylated in late-stage AD donors compared to early-stage and non-AD donors. Compared with AT8+ or T181+ alone, tau aggregates phosphorylated at both sites rose from ~10% in non-AD to ~50% in late-stage AD donors.

Longer fibril-shaped tau aggregates were detected with AD: significantly greater quantities of these HT7+ aggregates (>826 nm, >0.93 eccentricity) were found in MTG of late-stage compared to mid-stage AD donors, and to a lesser non-significant extent in SOM.

Conclusions

New detailed quantitation of brain region-specific tau protein changes with AD are reported. The data support associations between AD progression, tau phosphorylation, and tau aggregation, with increased aggregate quantity and elongation. Future work will investigate the influence of these measures upon AD symptomology.

Aims

Map the spatial distribution of fibrillar polymorphs of Aβ and tau across the brain during progression of Alzheimer's disease.

Methods

Scanning x-ray microdiffraction is used in coordination with immunohistochemistry to identify specific fibrillar polymorphs in histological sections of human brain tissue. Tissue is scanned with a 5 μ diameter x-ray beam and a full diffraction pattern is collected on a 5 μ grid spanning hundreds of microns, generating tens of thousands of diffraction patterns used to map the spatial localization of molecular constituents. Each pattern is examined for the presence of fibrillar aggregates of tau or Aβ based on the presence of the 4.7Å peak diagnostic for fibrillar structure. Where detected, the shape of this peak is analyzed to identify the predominant polymorph present in the scattering volume.

Results

Proof-of-principle experiments demonstrated the mapping of both Aβ and tau fibrillar structures across thin sections cut from different regions of the brain. Coordinated examination of serial sections by immunohistochemistry was used to aid in the interpretation of scattering patterns and to put the observations in a broader anatomical context. Scattering from lesions in tissue shown to be rich in Aβ fibrils by immunohistochemistry exhibited scattering patterns with a prototypical 4.7 Å cross-β peak and intensity distribution that compared well with that predicted from high resolution structures. Scattering from lesions in tissue with extensive tau pathology exhibited a 4.7 Å cross-β peak with intensity distributions that were distinct from those seen in Aβ-rich regions.

Conclusions

Scanning x-ray microdiffraction can supplement the information obtained by immunohistochemistry through the identification of the predominant fibrillar polymorph present in each lesion, potentially providing clues as to the role of specific polymorphs in disease progression.

Aims

Pathological tau aggregates are key features of Alzheimer's disease and other tauopathies. An important focus of research aims to understand how these aggregates spread in AD patient brains. Despite significant research, the mechanisms behind tau aggregation and spreading remain unclear. Our recently published study used mass spectrometry to identify proteins that interact with tau seeds and found that the presynaptic protein Bassoon plays a critical role in tau spreading and pathology. Considering our finding enhancing the relevance of tau-seed interactor on disease pathogenesis, we further characterize the interactome of the tau-seed isolated from AD human brains using a similar approach.

Methods

In this study, we used Size Exclusion Chromatography (SEC) to separate TBS-soluble AD and healthy brain extracts into fractions and measured tau seeding activity. Then, the fractions with higher seeding activity were subjected to immunoprecipitation to isolate the tau seed and performed quantitative mass spectrometry to identify proteins interacting with this tau seed.

Results

We found that the strongest seeding activity in AD brains was found in fraction 9, which contained high molecular weight (HMW) proteins larger than 2,000 kDa. Interestingly, the age-matched controls had similar levels of tau in fraction 9 but lacked seeding activity. Our bioinformatic analysis revealed that a total of 506 proteins interact with HWM tau without seeding activity isolated from healthy controls and 1201 proteins interact with HWM tau-seed isolated from AD cases. Interestingly, Weighted Gene-Co-Expression Network Analysis (WGCNA) from AD HMW tau-seed, revealed enrichment in proteins related to both synaptic and mitochondrial pathways.

Conclusions

Our research findings significantly contribute to our understanding of the mechanisms involved in tau propagation and offer new insights into the potential for targeting tau-seed interactions as a therapeutic strategy for neurodegenerative tauopathies.

Aims

Abnormal intracellular accumulation of Tau aggregates is a hallmark of Alzheimer's disease (AD) and other Tauopathies, such as Frontotemporal dementia (FTD). Pathological Tau can undergo a range of post-translational modifications (PTMs) that might trigger or modulate disease pathology. Recent studies indicate that modification of Tau by Small ubiquitin-like modifier SUMO isoform 1 (SUMO1) promotes Tau hyperphosphorylation and aggregation in transfected cells. Tau is also modified by SUMO2/3; however, the consequences of this modification have not been investigated. The goal of our study was to evaluate whether SUMO2/3 conjugation modifies the neurodegenerative disease pathology associated with the aggregation-prone mutant Tau P301L, P301S, and R406W variants.

Methods

Using viral approaches on primary hippocampal neurons, transgenic mice expressing mutant Tau and SUMO2, and iPSC-derived neurons from FTD patients, we assessed Tau aggregation and Tau-iduced mitochondrial and synaptic dysfunctions. We also performed single cell RNA sequencing analyses to identify pathways primarily affected by Tau, and rescued by SUMO2.

Results

We found that SUMO2 has quite the opposite effect of SUMO1 on Tau, since it reduces both phosphorylation and aggregation of mutant forms of Tau. Importantly, expression of mutant Tau is accompanied by a significant reduction of SUMO2/3 conjugation levels, and restoring levels of SUMO2 is neuroprotective. Mechanistically, we found that increase of SUMO2 restores mitochondrial-synaptic axis dysfunction in neurons expressing mutant Tau. Furthermore, NucSeq analyses indicate that SUMO2 modulates critical pathways involved in stress response.

Conclusions

Our results uncover an endogenous neuroprotective mechanism in FTD patients-derived cells and mouse models, whereby SUMO2 conjugation reduces Tau neuropathology, and protects against mithocondrial and synaptic dysfunction. These findings bring to light the potential therapeutic implication of manipulating SUMO conjugation to detoxify Tau through selective PTM-based approaches.