Progressive Supranuclear Palsy (PSP) is the prototypical 4-repeat (4R) tauopathy, a class of neurodegenerative disorders characterized by the predominant accumulation of 4R tau protein isoforms. PSP is also known as Steele-Richardson-Olszewski syndrome, named after the three neurologists who first described it in 1964[@williams2020]. Unlike Alzheimer's disease (AD), which features equal representation of 3R and 4R tau in neurofibrillary tangles, PSP selectively accumulates 4R tau isoforms, reflecting a distinct molecular pathophysiology[@goedert2018].
This page synthesizes the core mechanistic knowledge of PSP tauopathy, covering tau isoform biology, clinical subtypes, genetic architecture, glial pathology, tau PET imaging, and the therapeutic pipeline. PSP serves as the archetype for understanding how isoform-specific tau dysregulation drives selective neuronal vulnerability and disease progression.
The MAPT gene on chromosome 17q21 encodes tau protein, which exists in six isoforms in the adult human brain. These isoforms arise from alternative mRNA splicing of exons 2, 3, and 10. The critical distinction is the inclusion or exclusion of exon 10:
In the normal adult human brain, the 3R:4R ratio is approximately 1:1, maintained by precise splicing regulation of exon 10[@goedert2018][@bhatt2024].
PSP demonstrates marked predominance of 4R tau in all inclusions — neuronal tangles, glial inclusions, and neuropil threads. Multiple mechanisms drive this imbalance:
1. Exon 10 Splicing Dysregulation
Changes in splicing factor activity shift the 3R:4R balance toward 4R. SF3B1, SRSF2, andhnRNPA1 are implicated in this shift. Both neuronal and glial cells in PSP show altered splicing patterns that favor exon 10 inclusion[@bhatt2024].
2. Enhanced 4R Aggregation Propensity
4R tau has higher aggregation propensity than 3R tau due to:
3. Differential Clearance
The autophagy-lysosome pathway clears 3R and 4R tau with different efficiency. 4R tau is less efficiently degraded by chaperone-mediated autophagy (CMA) due to structural differences in the LAMP-2A recognition motif[@levy2024].
| Feature | PSP (4R Tauopathy) | AD (3R+4R Tauopathy) |
|---|---|---|
| Tau isoform composition | 4R only (~100%) | Equal 3R and 4R (~50:50) |
| Primary filament type | Straight filaments (SF) | Paired helical filaments (PHF) |
| Filament core structure | C-shaped protofilament, excludes R1 | Extended β-sheets across all repeats |
| Key phospho-epitopes | pS356, pS262 | pT181, pS396, pS404 |
| Astroglial pathology | Tufted astrocytes | Neurofibrillary tangles in astrocytes (rare) |
| Oligodendroglial pathology | Coiled bodies | Oligodendroglial tau (less prominent) |
| Amyloid co-pathology | Rare (~15% of cases) | Universal (~100%) |
| Clinical onset | 60-70 years, rapid progression | 65-80 years, slower progression |
| Typical survival | 5-7 years | 8-12 years[@goedert2018][@falcon2019] |
The phosphorylation pattern of tau in PSP differs from AD and other tauopathies:
The original NINDS-SPSP criteria recognized one clinical phenotype[@williams2020]. However, it became clear that PSP encompasses a spectrum of clinical presentations that correlate with underlying regional tau burden.
The Movement Disorder Society (MDS) 2017 criteria formalized multiple clinical subtypes[@stamelou2023][@boxer2017]:
The classic phenotype, accounting for ~50% of PSP cases:
Mimics Parkinson's disease early in the course:
Characterized by primary speech impairment:
Overlaps with corticobasal syndrome:
Predominantly motor phenotype:
Visual processing variant:
Rare variant with cerebellar involvement:
The clinical heterogeneity of PSP subtypes reflects differential regional vulnerability to tau pathology:
| Subtype | Primary Tau Burden Regions | Gait | Speech | Cognition |
|---|---|---|---|---|
| PSP-RS | Brainstem, basal ganglia | Early falls | Dysarthria | Executive |
| PSP-P | Basal ganglia, cortical | Variable | Variable | Executive |
| PSP-PAG | Motor cortex, basal ganglia | Preserved | Apraxia | Language |
| PSP-CBS | Frontoparietal cortex | Late falls | Dysarthria | Visuospatial |
| PSP-PAGF | Frontal-striatal | Freezing | Variable | Relatively spared |
| PSP-PCA | Occipitoparietal cortex | Variable | Variable | Visuospatial |
| PSP-PI | Brainstem, cerebellum | Ataxic gait | Dysarthria | Executive[@stamelou2023] |
The MAPT gene exists in two major haplogroups defined by a ~1.1 Mb inversion polymorphism on chromosome 17q21:
Genome-wide association studies have demonstrated that the H1 haplotype is the major genetic risk factor for PSP:
| MAPT Genotype | PSP Risk | Mechanism |
|---|---|---|
| H1/H1 | ~3-fold increased risk | Increased 4R tau expression, higher exon 10 splicing |
| H1/H2 | Intermediate risk | Partial protection from H2 haplotype |
| H2/H2 | Lowest risk | Reduced MAPT expression, lower 4R:3R ratio |
The H1 haplotype is in nearly complete linkage disequilibrium with the PSP risk allele identified in GWAS studies[@hoglinger2011][@wen2021].
Beyond MAPT, several other genetic loci influence PSP susceptibility[@hoglinger2011][@jabbari2018]:
| Gene | Function | Mechanism |
|---|---|---|
| STX6 (Syntaxin 6) | Vesicle trafficking | Alters endosomal sorting, affects tau clearance |
| EML5 | Microtubule stability | Modifies neuronal cytoskeletal dynamics |
| MOBP (Myelin-Associated Oligodendrocyte Protein) | Myelin integrity | Affects oligodendrocyte vulnerability |
| SLCO1A2 | Organic anion transport | May affect drug delivery to CNS |
| TNIP1 | NF-κB regulation | Modulates neuroinflammation |
| PM20D1 | NPF2/DFNB25 | Lysosomal function, potential role in tau clearance |
While most PSP cases are sporadic, specific MAPT mutations can cause PSP-like phenotypes:
| Mutation | Effect on Tau | Phenotype |
|---|---|---|
| N279K | Increases exon 10 splicing | PSP-like, earlier onset |
| P301L | Increases exon 10 splicing | PSP/FTD spectrum |
| S305S | Increases exon 10 splicing | PSP phenotype |
| +10 intronic | Activates cryptic splice site | PSP with parkinsonism |
| R406W | Reduces microtubule binding | PSP-like, memory prominent |
Flortaucipir is the most extensively studied tau PET ligand, originally developed for AD[@lowe2024].
Binding Characteristics in PSP:
Regional Binding Patterns:
| Region | PSP (Flortaucipir) | CBD (Flortaucipir) | AD (Flortaucipir) |
|---|---|---|---|
| Globus pallidus | ++ (prominent) | ++ | + (variable) |
| Subthalamic nucleus | ++ | + | + (variable) |
| Brainstem | + (moderate) | + | +++ (caudate) |
| Motor cortex | + (variable) | ++ (asymmetric) | + |
| Frontal cortex | + (variable) | ++ | +++ |
| Posterior cortex | + (rare) | + | +++ (temporal) |
Clinical Utility:
A second-generation tau PET ligand with higher affinity for straight filaments characteristic of 4R tauopathies[@andreone2024]:
Next-generation tracers specifically targeting 4R tau structures are under development[@andreone2024]:
| Tau PET Finding | Clinical Correlation | Utility |
|---|---|---|
| Pallidal burden | Disease severity, falls | Prognostic marker |
| Subthalamic signal | PSP-RS diagnosis | Diagnostic support |
| Cortical binding | CBD vs PSP distinction | Differential diagnosis |
| Longitudinal change | Disease progression | Trial endpoint[@chen2024] |
Tufted astrocytes are pathognomonic for PSP, distinguishing it from other 4R tauopathies[@komori2014][@smith2025]:
Morphological Features:
Topographical Distribution:
Pathogenic Significance:
Coiled bodies are tau-positive inclusions in oligodendrocytes, characteristic of PSP[@komori2014]:
Morphological Features:
Regional Distribution:
Functional Significance:
| Feature | PSP (Tufted Astrocytes) | CBD (Astrocytic Plaques) |
|---|---|---|
| Morphology | Thorn-shaped, radiating processes | Annular plaque pattern |
| Location | Striatum, motor cortex, brainstem | Cortex, subcortical |
| Distribution | Perivascular, periventricular | Diffuse cortical |
| Immunostaining | AT8+, pS396+ | AT8+, pS396+ |
| Specificity | Pathognomonic for PSP | Characteristic of CBD |
| Correlation | Disease severity | Cortical dysfunction |
Microglial activation accompanies tau pathology in PSP[@honda2024]:
| Agent | Target | Mechanism | Phase | Key Results |
|---|---|---|---|---|
| BIIB080 (Cinmeremab) | Pan-tau (N-terminal) | Binds extracellular tau, promotes clearance | Phase 1/2 (PSP) | Reduced CSF tau, acceptable safety[@muller2025] |
| 萨罗珠单抗 (Sartumab, ABBV-8E12) | Aggregated tau | Targets misfolded tau conformations | Phase 2 (PSP) | Safety established, efficacy pending[@sato2024] |
| Gosuranemab (BIIB092) | N-terminal tau | Blocks extracellular tau propagation | Phase 2 (PSP) | Did not meet primary endpoint, target engagement shown[@siegler2023] |
| Semorinemab (RO7105705) | Tau oligomers | Targets toxic oligomeric species | Phase 2 (PSP) | Ongoing[@mendonca2024] |
| UCB 0107 | Phospho-tau | Targets specific phospho-epitopes | Phase 1/2 | Recruiting[@siegler2023] |
BIIB080 (Cinmeremab): An antisense approach via antibody — phase 1 results demonstrated dose-dependent reduction in CSF total tau and p-tau181 in PSP patients. The antibody targets the N-terminal region common to all tau isoforms, potentially enhancing clearance of both 3R and 4R species[@muller2025].
萨罗珠单抗 (Sartumab): Phase 2 trials showed acceptable safety and tolerability in PSP patients. Mechanistic studies suggest binding to aggregated tau species with subsequent Fc-mediated microglial clearance[@sato2024].
Antisense oligonucleotides directly targeting MAPT mRNA represent a promising approach[@muller2025]:
| Agent | Mechanism | Phase | Status |
|---|---|---|---|
| LMTM (Hydromethylthionine) | Tau aggregation inhibition, mitochondrial function | Phase 3 | Mixed results, post-hoc benefit in mild PSP[@siegler2023] |
| TPI-287 | Microtubule stabilizer, 4R-specific | Phase 1 | Discontinued due to toxicity |
| Davunetide (AL-108) | Microtubule stabilizer, neuroprotective peptide | Phase 2/3 | Failed primary endpoint[@boxer2017] |
| Target | Approach | Stage | Rationale |
|---|---|---|---|
| GSK-3β | Kinase inhibitors | Preclinical | Reduce tau phosphorylation |
| mTOR | Rapalogs (sirolimus, everolimus) | Preclinical | Enhance autophagy |
| TREM2 | Agonist antibodies | Preclinical | Modulate microglial response |
| Iron chelation | Deferiprone | Phase 2 | Reduce brain iron accumulation |
| Treatment | Mechanism | Response in PSP |
|---|---|---|
| Levodopa/Carbidopa | Dopamine replacement | Minimal (~20% transient benefit) |
| Amantadine | NMDA antagonist | May reduce falls, anecdotal |
| Zolpidem | GABA-A modulator | Some improvement in axial symptoms |
| Botulinum toxin | Neuromuscular blockade | For cervical dystonia |
| Symptom | Treatment Options |
|---|---|
| Dysphagia | Swallowing therapy, dietary modification, PEG tube |
| Depression/apathy | SSRIs, SNRIs, behavioral approaches |
| Cognitive impairment | Cholinesterase inhibitors (limited benefit) |
| Sleep disturbances | Melatonin, clonazepam for REM behavior disorder |
| Urinary dysfunction | Anticholinergics (oxybutynin), catheterization[@mendonca2024] |
Patient stratification:
Outcome measures:
| Feature | PSP | CBD | AD |
|---|---|---|---|
| Core clinical syndrome | Vertical gaze palsy + postural instability | Asymmetric apraxia + parkinsonism | Memory impairment |
| Ocular motor findings | Supranuclear gaze palsy (vertical > horizontal) | Variable, usually absent early | Rare |
| Parkinsonism | Symmetric, axial > limb | Asymmetric, limb > axial | Minimal |
| Cortical signs | Frontal (disinhibition) | Alien limb, apraxia, cortical sensory loss | Memory, visuospatial |
| Response to levodopa | Poor | Poor to moderate | Not applicable |
| Gait/falls | Early, characteristic | Variable | Variable, late |
| Cognitive profile | Executive dysfunction | Visuospatial, language | Memory encoding[@dickson2018] |
| Feature | PSP | CBD | AD |
|---|---|---|---|
| Tau isoform | 4R predominant | 4R predominant | 3R + 4R |
| Primary filament type | Straight filaments | Straight + twisted ribbons | Paired helical filaments |
| Filament core | C-shaped, excludes R1 | C-shaped, variant packing | Extended β-sheets |
| Neuronal inclusions | Globose tangles | Ballooned neurons, tangles | Neurofibrillary tangles |
| Astroglial inclusions | Tufted astrocytes | Astrocytic plaques | Rare, if any |
| Oligodendroglial inclusions | Coiled bodies | Fewer coiled bodies | Oligodendroglial tau |
| Regional distribution | Brainstem > basal ganglia > cortex | Cortex > basal ganglia > brainstem | Cortex (entorhinal > limbic > isocortical) |
| Amyloid co-pathology | ~15% (low) | ~20% (low) | ~100% (defining) |
| Lewy body co-pathology | ~10% | ~5% | ~20%[@kovacs2020][@falcon2019] |
| Molecular Feature | PSP | CBD | AD |
|---|---|---|---|
| Key phospho-epitopes | pS356, pS262 | pS262, pS396 | pT181, pS396, pS404 |
| Key acetylation sites | K274, K281 | K280, K281 | K280, K369 |
| MAPT haplotype risk | H1/H1 (strong) | H1/H1 (moderate) | H1/H1 (weak) |
| MAPT mutations | Rare | Rare | Rare |
| Tau seeding potency | Moderate | Moderate-high | High |
| Tau strain characteristics | PSP-specific (straight filament) | CBD-specific (twisted ribbons) | AD-specific (PHF)[@koga2023] |
| Therapeutic Approach | PSP | CBD | AD |
|---|---|---|---|
| Anti-tau antibodies (pan-tau) | Applicable | Applicable | Applicable |
| Anti-tau antibodies (4R-specific) | More relevant | More relevant | Less relevant |
| Aggregation inhibitors | Relevant | Relevant | Relevant |
| ASOs targeting MAPT | High relevance (reduces 4R) | High relevance (reduces 4R) | Moderate (reduces all tau) |
| Amyloid-targeting therapies | Not applicable | Not applicable | Applicable |
| Anti-inflammatory (TREM2) | Applicable | Applicable | Applicable[@mendonca2024] |
Tau pathology in PSP spreads through neural networks in a prion-like manner[@chen2025][@brendza2023]:
1. Tau Release
2. Cellular Uptake
3. Template-Directed Misfolding
4. Propagation Through Networks
Early (Brainstem)
Intermediate (Subcortical)
Late (Cortical)
Different PSP phenotypes may reflect distinct tau strain properties[@koga2023]:
Neuronal Vulnerability:
Glial Vulnerability:
Oligomeric tau represents the most toxic species in tauopathies, including PSP[@patel2025]:
1. Early Oligomers (dimers, trimers)
2. Intermediate Oligomers (4-12 mers)
3. Protofibrils
| Property | PSP Oligomers | AD Oligomers | CBD Oligomers |
|---|---|---|---|
| Size | 6-10mers (compact) | 12-20mers (larger) | 8-14mers |
| Morphology | Short, round | Elongated, fibrillar | Variable |
| pS356 content | High (specific) | Low | Low-moderate |
| Membrane binding | High | Moderate | Moderate |
| Synaptic localization | Prominent | Variable | Moderate |
| Neurotoxicity potency | High | Very high | High[@patel2025] |
Synaptic Dysfunction:
Mitochondrial Dysfunction:
ER Stress and UPR Activation:
Membrane Permeabilization:
PSP represents the archetypal 4R tauopathy, with a distinct mechanistic profile from other tauopathies like CBD and AD. The selective accumulation of 4R tau isoforms, driven by MAPT H1 haplotype risk and exon 10 splicing dysregulation, produces a unique filament architecture (straight filaments with C-shaped protofilaments) that underlies the disease phenotype.
The clinical heterogeneity of PSP subtypes reflects differential regional vulnerability to tau pathology, from the classic Richardson syndrome (brainstem-predominant) to PSP-PAG (speech-predominant) and PSP-CBS (cortical-predominant). This variation offers opportunities for precision medicine approaches.
The therapeutic pipeline for PSP is maturing, with anti-tau antibodies (BIIB080, sartumab, gosuranemab), MAPT-targeting ASOs (BIIB080), and aggregation inhibitors in various stages of clinical development. Tau PET imaging enables patient selection and target engagement assessment, while fluid biomarkers (p-tau181, p-tau217, pS356) provide complementary diagnostic and monitoring tools.
Glial tau pathology — tufted astrocytes and coiled bodies — represents a distinguishing feature of PSP and an important therapeutic target. Understanding the interplay between neuronal and glial tau pathology will be critical for developing effective disease-modifying treatments.