Spatial transcriptomics has emerged as a transformative technology for understanding the molecular architecture of Progressive Supranuclear Palsy (PSP), a 4-repeat (4R) tauopathy characterized by selective vulnerability of specific brain regions and neuronal populations. Unlike traditional bulk RNA sequencing, spatial transcriptomics preserves the anatomical context of gene expression, enabling precise mapping of transcriptional changes within affected brain regions and identification of cell-type-specific responses that underlie the characteristic patterns of neurodegeneration in PSP.
This page synthesizes findings from spatial transcriptomic studies in PSP, focusing on region-specific gene expression patterns, comparisons with other 4R tauopathies like corticobasal degeneration (CBD) and Pick's disease, and insights into the molecular mechanisms that drive selective neuronal vulnerability.
Spatial transcriptomics encompasses several technologies that measure gene expression while preserving spatial location within tissue sections:
- Spatial Transcriptomics (10x Visium): Uses slide-bound RNA capture probes with unique spatial barcodes, enabling transcriptome-wide expression mapping at 55 μm resolution
- MERFISH (Multiplexed Error-Robust Fluorescence In Situ Hybridization): Single-cell resolution with hundreds to thousands of genes probed simultaneously
- Stereo-seq: Combines DNA nanoball (DNB)-based arrays with in situ sequencing for high-resolution spatial mapping
- Slide-seq: Uses bead-based arrays for ~10 μm spatial resolution
These technologies have been applied to postmortem brain tissue from PSP patients, revealing spatial patterns of gene expression that correlate with neuropathological hallmarks including neurofibrillary tangles (NFTs), tufted astrocytes, and coiled bodies.
Spatial transcriptomics is particularly valuable for studying tauopathies like PSP because:
- Region-specific pathology: PSP demonstrates characteristic patterns of neurodegeneration in the basal ganglia, brainstem, and frontal cortex
- Cell-type diversity: Multiple cell types (neurons, astrocytes, oligodendrocytes, microglia) contribute to disease pathogenesis
- Temporal progression: Different brain regions show varying stages of pathology, enabling study of disease progression
- Comparative analysis: Spatial approaches allow direct comparison between affected and relatively preserved regions within the same tissue section
The basal ganglia, particularly the globus pallidus and subthalamic nucleus, are among the most affected regions in PSP.
Spatial transcriptomic analysis of the globus pallidus in PSP reveals:
- Mitochondrial dysfunction signatures: Downregulation of nuclear-encoded mitochondrial genes including NDUFA1, COX5A, and ATP5F1B
- Inflammatory response: Upregulation of complement system components (C1Q, C3) and glial fibrillary acidic protein (GFAP) in astrocytes
- Tau-related pathways: Altered expression of microtubule-associated proteins and synaptic function genes
- Energy metabolism: Reduced expression of glycolysis and oxidative phosphorylation enzymes
The globus pallidus externa (GPe) and interna (GPi) show distinct transcriptional programs, with the GPi demonstrating more pronounced mitochondrial impairment consistent with its greater vulnerability to neurodegeneration.
The subthalamic nucleus is a hallmark affected region in PSP, showing:
- Neuronal loss: Spatial transcriptomics confirms selective loss of excitatory neurons
- Synaptic dysfunction: Downregulation of synaptic vesicle proteins (SYN1, SYP) and glutamate receptors
- Calcium dysregulation: Altered expression of calcium-binding proteins including calbindin and parvalbumin
- Tau pathology genes: Upregulation of genes associated with tau phosphorylation and aggregation
The substantia nigra shows characteristic dopaminergic neuron loss in PSP:
- Dopaminergic markers: Reduced TH (tyrosine hydroxylase) and DAT (dopamine transporter) expression
- Lewy body-related genes: Interestingly, α-synuclein (SNCA) expression patterns differ from Parkinson's disease
- Neuroinflammation: Prominent microglial activation signatures with increased CD68, CD86, and complement genes
- Mitochondrial genes: Downregulation consistent with the known mitochondrial dysfunction in PSP
Spatial transcriptomics of the pontine nuclei reveals:
- Tau pathology genes: Upregulation of MAPT (tau) and associated kinases
- Myelin genes: Oligodendrocyte-specific expression changes affecting myelin maintenance
- Synaptic plasticity: Altered expression of genes involved in long-term potentiation
The prefrontal cortex shows relatively less severe pathology than subcortical regions but demonstrates important transcriptional changes:
- Layer-specific patterns: Spatial mapping reveals layer-specific vulnerability, with layer 5 showing more pronounced changes
- Tau-related genes: Increased expression of tau kinases (GSK3B, CDK5) and decreased phosphatases (PP2A)
- Synaptic genes: Downregulation of postsynaptic density proteins (PSD95, HOMER)
- Astrocytic responses: Reactive astrocyte signatures with increased GFAP and S100B
Spatial transcriptomics enables precise characterization of which neuronal populations are most vulnerable in PSP:
- Corticospinal tract neurons: Large pyramidal neurons in layer 5 of the motor cortex
- Basal ganglia projection neurons: Striatal medium spiny neurons and pallidal neurons
- Brainstem cholinergic neurons: Pedunculopontine nucleus and laterodorsal tegmental nucleus
- Substantia nigra dopaminergic neurons: Particularly the ventrolateral tier**
These neurons share common transcriptional vulnerabilities including:
- High energy demands requiring robust mitochondrial function
- Long axons with extensive myelination
- Calcium dysregulation susceptibility
- Tau expression patterns favoring 4R isoform inclusion
Spatial transcriptomics also identifies neurons that are relatively spared in PSP:
- Cerebellar Purkinje cells: Show minimal transcriptional changes despite being affected in some PSP variants
- Cortical interneurons: Certain interneuron subtypes show relative preservation
- Sensory relay neurons: Maintain stable transcriptional profiles
Spatial transcriptomics reveals distinct microglial activation patterns in PSP:
- Disease-associated microglia (DAM): Upregulation of TREM2, APOE, and TYROBP
- Spatial distribution: Microglial clusters near tau pathology hotspots
- Regional variation: More pronounced activation in basal ganglia than cortex
- Comparison with AD: PSP microglia show distinct transcriptional profiles from Alzheimer's disease microglia
Key microglial genes upregulated in PSP:
- TREM2 (triggering receptor on myeloid cells 2)
- APOE (apolipoprotein E)
- CD33 (sialic acid-binding Ig-like lectin 3)
- CX3CR1 (C-X3-C motif chemokine receptor 1)
- IL1B, IL6 (pro-inflammatory cytokines)
Reactive astrocytes in PSP show region-specific responses:
- Tufted astrocyte signature: GFAP, S100B, and vimentin upregulation
- Metabolic dysfunction: Altered glucose metabolism genes
- Potassium buffering: Dysregulation of Kir channels and glutamate transporters
- Spatial correlation: Astrocyte reactivity correlates with NFT distribution
Oligodendroglial involvement is a hallmark of PSP pathology (coiled bodies):
- Myelin genes: Downregulation of MBP, PLP1, and myelin-related genes
- Spatial patterns: Coiled body-containing regions show pronounced oligodendrocyte dysfunction
- Tau expression: Oligodendrocytes express 4R tau isoforms prominently
- Comparison with CBD: Similar oligodendrocyte involvement but distinct spatial patterns
Spatial transcriptomic comparison between PSP and CBD reveals:
| Feature |
PSP |
CBD |
| Primary pathology location |
Subthalamic nucleus, globus pallidus |
Motor cortex, basal ganglia |
| Astrocytic lesions |
Tufted astrocytes |
Astrocytic plaques |
| Oligodendrial lesions |
Coiled bodies |
Coiled bodies (similar) |
| Microglial activation |
Prominent in basal ganglia |
Prominent in cortex |
| Neuronal vulnerability |
Subcortical > cortical |
Cortical > subcortical |
Shared pathways:
- 4R tau expression and aggregation
- Mitochondrial dysfunction
- Neuroinflammation
- Oligodendrocyte pathology
Distinct pathways:
- PSP: Brainstem involvement, oculomotor dysfunction genes
- CBD: Motor cortex vulnerability, apraxia-related genes
Comparison with Pick's disease (3R tauopathy):
- Tau isoform: 3R vs 4R distinguishes the diseases spatially
- Neuronal vulnerability: Different cortical layer patterns
- Astrocytic pathology: Pick bodies vs tufted astrocytes
- Regional emphasis: Frontal cortex > basal ganglia in Pick's
Comparison with Alzheimer's disease (AD):
- Tau pathology: 3R+4R NFTs vs 4R in PSP
- Amyloid relationship: No direct amyloid involvement in PSP
- Microglial patterns: Different activation states
- Spatial progression: AD follows specific cortical hierarchy vs PSP's subcortical pattern
Spatial transcriptomics has identified altered tau-related gene expression:
-
Tau kinases (upregulated):
- GSK3B (glycogen synthase kinase 3 beta)
- CDK5 (cyclin-dependent kinase 5)
- MAPK8 (JNK1)
- DYK
-
Tau phosphatases (downregulated):
- PP2A (protein phosphatase 2A)
- PPP2R2A (regulatory subunit)
-
Tau post-translational modification:
- Altered expression of SUMOylation and ubiquitination genes
A consistent finding across spatial transcriptomic studies:
- Complex I-V subunit downregulation
- mtDNA-encoded gene alterations
- Mitophagy pathway changes (PINK1, PARK2)
- Energy metabolism reprogramming
Spatially resolved inflammation mapping:
- Complement activation: C1Q, C3, C4 upregulation in affected regions
- Cytokine profiles: Regional variation in IL1B, IL6, TNF expression
- Microglial states: Distinct activation states correlating with local tau pathology
Spatial transcriptomics reveals:
- Presynaptic protein downregulation (SYN, SYP)
- Postsynaptic density alterations (PSD95, HOMER)
- Excitotoxicity-related genes
- Calcium homeostasis disruption
- Resolution limits: Visium (55 μm) includes multiple cell types per spot
- Postmortem tissue: RNA quality varies with agonal state and postmortem interval
- Disease staging: Different regions represent different disease stages
- Cell type deconvolution: Computational methods needed to separate signals
Best practices for spatial transcriptomics in PSP:
- Integrate with neuropathological staging
- Combine with single-cell data for cell-type resolution
- Correlate with PET imaging data (tau PET, FDG-PET)
- Link with genetic data (MAPT haplotypes, risk genes)
Spatial transcriptomics identifies:
- Fluid biomarker candidates: Genes with brain-region-specific expression
- Imaging targets: Spatially mapped gene expression informs PET ligand development
- Disease progression markers: Genes correlating with regional pathology severity
Key pathways for therapeutic intervention:
- Tau aggregation inhibitors: Based on spatial patterns of tau-related gene expression
- Microglial modulators: TREM2, CD33-targeted therapies
- Mitochondrial protectors: Complex I-V targeting
- Synaptic protectors: Based on synaptic gene expression patterns
Spatial transcriptomics enables:
- Regional vulnerability profiling: Predict which brain regions will be affected
- Treatment response prediction: Based on individual spatial expression patterns
- Disease subtype classification: Spatial patterns may distinguish PSP subtypes
- Single-cell spatial transcriptomics: Resolving cell-type-specific patterns at single-cell resolution
- Multimodal integration: Combining spatial transcriptomics with spatial proteomics and metabolomics
- Longitudinal studies: Animal models enabling temporal mapping of disease progression
- In vivo spatial approaches: Development of spatial transcriptomic probes for live imaging
Current limitations in understanding PSP through spatial transcriptomics:
- Limited studies in early/prodromal PSP
- Insufficient diversity in tissue banks
- Need for standardized analysis pipelines
- Integration with clinical phenotyping