Search bioRxiv⌕ Search

Biology subjects

Georgescauld, F.

Publications and source records attributed to Georgescauld, F..

2 recordsLinked to original sources

Dynamic conformational ensembles of soluble Tau encode neuronal toxicity prior to aggregation

Tau aggregation is a defining feature of Alzheimers disease and related tauopathies, yet the conformational states of Tau in neurons prior to aggregation remain poorly understood. Existing structural models are derived largely from fibrillar assemblies and provide limited insight into the dynamic, soluble Tau species that initiate pathology. Here, we combine hydrogen-deuterium exchange mass spectrometry with super-resolution imaging and neuronal models to define the conformational ensemble of soluble Tau under physiological and disease-relevant conditions. We show that soluble Tau populates distinct, dynamic conformations characterized by regional stabilization and long-range intramolecular interactions that are invisible to fibril-based structures. Disease-associated perturbations selectively remodel these conformational ensembles, exposing aggregation-prone regions and altering Tau subcellular organization in neurons. Notably, these Tau species inhibit axonal transport, which is essential for neuronal health, linking specific ensemble states to neuronal toxicity. These findings establish soluble Tau conformation as a dynamic, regulatable state that precedes aggregation and encodes disease relevance. By defining the structural logic of Tau before fibril formation, this work provides a framework for understanding early tauopathy mechanisms and for targeting Tau pathology at its earliest stages. SUMMARYTau pathology is a hallmark of Alzheimers disease (AD) and related dementias (ADRDs). Although Tau is often described as intrinsically disordered, it is a dynamic protein with distinct but poorly defined conformations. Here we conduct a systematic time-resolved structure-function analysis of normal and pathologic Tau, including hyperphosphorylated, mutant Tau, and posttranslational-modification-mimetic Tau. To characterize dynamic conformational changes of Tau, we combined state-of-the-art hydrogen deuterium exchange mass spectrometry with structured illumination microscopy, demonstrating a novel Tau-MT binding mode: "dynamic oscillation". To correlate Tau structure with neuronal function, we evaluated axonal transport as a sensitive readout of neuronal health. Many toxic Tau forms share a common signature of increased exposure of the N-terminal phosphate activating domain (PAD) in vitro and in vivo. Aberrant exposure of PAD correlates with Tau pathology and axonal transport defects. Tau phosphorylation at S262 alone is sufficient to alter Tau-microtubule interactions beyond R1-R4 motifs, globally changing Tau conformation, disrupting "dynamic oscillation" on MTs, and inhibiting axonal transport. Frontotemporal dementia-associated P301L-Tau remains associated with microtubules but also inhibits axonal transport. Our results reveal a well-defined conformation of soluble WT Tau in neurons and its highly dynamic interaction with microtubules, altered by AD/ADRD-Tau forms. Our multidisciplinary approach comprising biochemical manipulations, innovative MS tools, advanced microscopy, cellular assays, and mouse and human data pair Tau conformations with distinct neuronal functions and pathologies in health and disease.

neuroscience↗

Lifting the curse from high dimensional data: Automated projection pursuit clustering for the variety of biological data modalities

Unsupervised clustering is a powerful machine-learning technique widely used to analyze high-dimensional biological data. It plays a crucial role in uncovering patterns, structure, and inherent relationships within complex datasets without relying on predefined labels. In the context of biology, high-dimensional data may include transcriptomics, proteomics, and a variety of single-cell omics data. Most existing clustering algorithms operate directly in the high-dimensional space, and their performance may be negatively affected by the phenomenon known as the curse of dimensionality. Here, we show an alternative clustering approach that alleviates the curse by sequentially projecting high-dimensional data into a low-dimensional representation. We validated the effectiveness of our approach, named APP, across various biological data modalities, including flow and mass cytometry data, scRNA-seq, multiplex imaging data, and T-cell receptor repertoire data. APP efficiently recapitulated experimentally validated cell-type definitions and revealed new biologically meaningful patterns.

bioinformatics↗