Defining the role of Tau in oxidative stress

Supervisors: John Todd, Katherine Bull

The development of hyperphosphorylated, insoluble and filamentous Tau is a hallmark of human neurodegenerative diseases collectively called tauopathies, including Alzheimer’s disease (AD), frontotemporal dementia, corticobasal degeneration and progressive supranuclear palsy. Tauopathy severity correlates with upregulation of oxidative stress markers in the brain and cerebrospinal fluid, and lipid peroxidation and reduced antioxidant function are early features of murine models of AD. Brain atrophy, impaired cerebral glucose metabolism, CNS insulin resistance and oxidative stress are also features of diabetes and aging, suggesting shared metabolic pathways. Thus a link between Tau and antioxidant response could have implications for understanding of both neurodegenerative and systemic metabolic and senescence related pathology, and reveal pathways relevant to multiorgan disease. However a direct causal relationship between Tau and oxidative stress in human cells has not previously been demonstrated.

In unpublished work, we have established a flow cytometry-based quantitative assay to induce and quantify oxidative stress in cells. Applying this assay in human neuronal SH-SY5Y cells, we demonstrate a novel role for MAPT (Tau) in protection from oxidative injury; deletion of MAPT increases susceptibility to oxidative stress, rescued by MAPT overexpression, with proteomic studies highlighting aberrant RNA binding with loss of Tau.

Oxidative stress is a central feature of diabetic nephropathy (DKD), the leading global cause of end stage kidney disease. By single cell RNA-Sequencing, and at the protein level, we and others have identified Tau expression in the kidney, particularly in proximal tubules, and we find Tau expression altered in human renal podocyte cells after induction of oxidative injury. Together these findings lead us to hypothesise a role for Tau in the response to diabetic kidney injury.

In this DPhil project we will extend these findings and test our hypothesis with three key experimental approaches:

• Define transcriptomic pathways linked to Tau antioxidant function in a multicellular context. We have generated neuronal organoids from iPSCs with Tau deletion (Todd, Bull with Phalguni Rath CHG iPSC research lab), and will use scRNA and spatial |(Xenium, Paolo Piazza CHG technology platforms) to explore pathways disrupted by loss of Tau, and under induction of oxidative stress. The Bull lab have experience with generation and computational analysis of tissue snRNA-seq and with high resolution spatial imaging including Xenium (Benjamin et al Nature 2024). Pathways and genes will be compared to the existing in-house neuronal proteomics with and without oxidative stress, in the presence or absence of Tau.

• Identify potential therapeutic candidates for both neuronal and renal oxidative stress, including stress mediated by loss of Tau function. We will use our scalable flow based oxidative stress assay to screen candidate compounds and drugs for ability to rescue neuronal and renal cells from oxidative stress. We will also test whether these compounds can protect the MAPT knockout cells, in which resistance to oxidative stress is reduced. In addition to stress we will measure cell viability, and for screen positive compounds, cell type specific assays, with support from the CHG Cellular Imaging facility and collaborations for additional functional cellular assays: Steph Fowler, IMCM Fellow, Tau imaging and filaments, Ira Milosovic, CHG PI, mesangial and neuronal endocytosis.

• Identify shared human in vivo pathways in neuronal and renal diabetic disease. We will compare pathways highlighted in a) and b) with human renal snRNA and spatial transcriptomics (Xenium or Visium HD with Fadi Issa) in DKD and healthy kidneys obtained from human kidney biopsy tissue. Fresh and FFPE tissue is available via existing ethics agreements with the Oxford Kidney Pathology Atlas (Bull), the Oxford Centre for Histopathology Research and the Oxford Transplant Biobank (Rutger Ploeg). Pilot snRNA seq data has already been generated and samples identified for a planned CosMx spatial experiment.

Spatial transcriptomics work will build on expertise in the Bull lab, including new mathematical tools to deconvolute cells (Benjamin et al Nature 2024).

The goals of the project are to

1. Identify candidate targets and therapeutics for oxidative stress driven disease, with potential applications for common neurological and renal pathologies.
2. Define mechanisms by which Tau modulates the antioxidant response and validate in vitro observations in organoids and human tissue.
3. Profile human diabetic tissue pathology, which can be linked to the in vitro findings, but will also establish a dataset that can be compared to clinical outcomes in DKD, towards new approaches to stratify disease and predict response to treatment.

The clinical DPhil student will benefit from a highly cross disciplinary project linking human tissue multi-omic data to mechanistic validation in organoid and cellular models, relevant to common neurological and renal diseases with significant burden and limited treatment options. They will be supported to develop both molecular ‘wet’ laboratory techniques, building on prior collaborative work in the Todd and Bull groups, and computational biology skills applied to cutting edge sub cellular resolution spatial platforms. There is significant experience and expertise within the proposed groups and the student will be encouraged to take advantage of the OBDS training programme at an early stage, alongside training opportunities within the MSDTC.