Does tau aggregation specifically cause PS externalization in living cells versus general cellular stress responses?

Novel Hypotheses: PS Exposure Specificity in Tauopathy

Hypothesis 1: Caspase-3 Cleavage of Tau Creates PS-Targeting Specificity

Description: Tau cleavage by activated caspase-3 generates a membrane-binding C-terminal fragment that directly facilitates PS externalization. This creates a tau-specific "find-me" signal distinct from general apoptotic pathways. The caspase-cleaved tau fragment exhibits enhanced binding to inner leaflet phospholipids, disrupting membrane asymmetry preferentially in tau-bearing cells.

Target: Caspase-3, tau cleavage products Confidence: 0.72 Evidence basis: Caspase-3 activation is documented in tauopathy; tau cleavage fragments are found in NFT-bearing neurons; caspase-cleaved proteins can expose PS through downstream effects on scramblase activity.

Hypothesis 2: Hyperphosphorylated Tau Inhibits PMCA Pumps, Causing Sustained PS Exposure

Description: In tauopathies, hyperphosphorylated tau physically interacts with and inhibits plasma membrane calcium ATPase (PMCA) pumps. This calcium dysregulation activates scramblases (TMEM16F, XKR4) specifically, leading to prolonged PS externalization. Normal stressed cells restore calcium homeostasis rapidly, whereas tau-bearing cells exhibit sustained PS exposure enabling selective targeting.

Target: PMCA1/4, TMEM16F scramblase Confidence: 0.65 Evidence basis: Tau-PMCA interaction reported in proteomic studies; calcium dysregulation is established in tauopathy; scramblase activation is calcium-dependent.

Hypothesis 3: PS Exposure is Cell-Type Context-Dependent Rather Than Pathway-Specific

Description: PS exposure functions as a universal "danger-associated molecular pattern" (DAMP) but its downstream interpretation differs by cell type. In neurons, PS triggers anti-inflammatory engulfment (neuroprotective); in microglia, PS indicates activation state. Targeting strategies should therefore focus on neuronal-specific PS-binding receptors (e.g., modified MERTK ligands) rather than PS itself.

Target: PS-binding receptor complexes (MERTK, AXL, TIMD4) Confidence: 0.58 Evidence basis: MERTK/Axl are expressed differentially; neuronal PS exposure in AD does not always trigger phagocytosis; immune cells exhibit "don't eat me" signals concurrently.

Hypothesis 4: Oxidative Stress-Induced PS Exposure is Bnip3-Dependent in Tauopathy

Description: Mitochondrial stress in tauopathy upregulates Bnip3, which localizes to the outer mitochondrial membrane and facilitates mitochondrial-ER contact sites enriched for PS metabolism. This creates a mitochondria-to-plasma membrane PS externalization pathway specific to metabolically compromised tau-bearing neurons, absent in general cellular stress where Bnip3 is not induced.

Target: Bnip3, MFN2 (mitochondrial dynamics) Confidence: 0.54 Evidence basis: Bnip3 elevation in AD brain; mitochondrial dysfunction is early tauopathy feature; inter-organelle contact sites regulate phospholipid distribution.

Hypothesis 5: Regional Vulnerability Reflects Microglial PS-Sensing Heterogeneity

Description: Different brain regions exhibit varying microglial PS-receptor expression profiles (e.g., higher CX3CR1+ microglia in hippocampus with more efficient PS clearance). Tau pathology in regions with lower PS-sensing capacity leads to accumulation of PS-exposed neurons, explaining regional vulnerability patterns. Selectivity would derive from region-specific microglial states, not PS exposure itself.

Target: CX3CR1, microglial PS receptors, regional transcriptomic signatures Confidence: 0.61 Evidence basis: Regional microglia heterogeneity documented; hippocampal vulnerability in AD is established; PS exposure correlates with neuronal loss in specific regions.

Hypothesis 6: PS Externalization Requires Tau Aggregation-Phase Transition

Description: Monomeric soluble tau does not induce PS exposure; PS externalization specifically occurs when tau undergoes liquid-to-solid phase transition (condensate formation or fibrillization). This aggregate-state specifically alters membrane curvature and local lipid organization, triggering scramblase recruitment. This explains why early tau pathology may not show PS exposure.

Target: Tau liquid-liquid phase separation intermediates, membrane curvature proteins (BAR family) Confidence: 0.47 Evidence basis: Tau forms biomolecular condensates; membrane curvature affects PS distribution; aggregation correlates with neuronal loss, not early tau accumulation.

Hypothesis 7: Dual-Signal Model—PS + Specific Tau Epitope Enables Selective Targeting

Description: PS exposure alone is insufficient for selective tau targeting; rather, a dual-signal mechanism exists where externalized PS recruits microglia while exposed tau N-termini (truncated by caspases) serve as the "specific eat-me signal." Therapeutic strategies should develop bispecific constructs recognizing both PS and specific tau neoepitopes simultaneously, avoiding off-target effects in stressed-but-non-tau tissues.

Target: Tau N-terminal fragments (e.g., aa 1-150), PS with engineered bridging proteins Confidence: 0.69 Evidence basis: Truncated tau fragments in AD CSF; PS externalization correlates with neuronal loss; bispecific antibodies show enhanced selectivity in oncology models.

Summary Table

| # | Title | Target | Confidence |
|---|-------|--------|------------|
| 1 | Caspase-3 cleavage model | Caspase-3, tau fragments | 0.72 |
| 2 | PMCA inhibition pathway | PMCA, TMEM16F | 0.65 |
| 3 | Cell-type context model | MERTK/Axl receptors | 0.58 |
| 4 | Bnip3 mitochondrial pathway | Bnip3, MFN2 | 0.54 |
| 5 | Regional vulnerability model | CX3CR1, microglia | 0.61 |
| 6 | Phase transition model | Tau condensates, BAR proteins | 0.47 |
| 7 | Dual-signal targeting | Tau N-fragments + PS | 0.69 |

Primary recommendation: Hypothesis 7 (dual-signal model) and Hypothesis 1 (caspase-dependent model) offer the most actionable pathways for selective targeting while minimizing off-target risks in inflamed/stressed tissues.

Critical Evaluation of PS Exposure Specificity Hypotheses in Tauopathy

Preliminary Methodological Concerns

Before addressing individual hypotheses, several overarching issues merit attention:

The fundamental question remains unresolved: These hypotheses address how PS exposure creates targeting selectivity, but none adequately explain why PS externalization would be selectively elevated in tau-bearing neurons. The explanatory arrow appears inverted—each mechanism proposes a downstream effect of tau pathology that then leads to PS exposure, but no mechanism explains why tauopathy neurons would differ from other stressed/apoptotic cells where PS exposure is universal.

Mechanistic specificity problem: PS externalization is an evolutionarily conserved apoptotic signal. Claims that tauopathy creates "selective" PS exposure mechanisms must explain why these mechanisms wouldn't also operate in the myriad other conditions causing cellular stress and caspase activation. The hypotheses largely treat PS exposure as an active, regulated process, when in non-apoptotic contexts PS externalization is primarily a consequence of membrane integrity loss.

Terminology concerns: "PS exposure" in these hypotheses conflates distinct phenomena: (1) the canonical apoptotic phosphatidylserine flip associated with cell death, (2) non-apoptotic PS externalization reported in stressed-but-viable cells, and (3) the specific subacute PS exposure that might enable therapeutic targeting windows. These are mechanistically distinct phenomena with different implications.

Hypothesis 1: Caspase-3 Cleavage of Tau Creates PS-Targeting Specificity

Weaknesses and Challenges

Ubiquity problem: Caspase-3 activation is a universal feature of apoptosis across all cell types and all insults. If caspase-3-mediated tau cleavage generates PS-targeting specificity, then any condition activating caspase-3 should produce equivalent effects—but general apoptosis does not create the proposed selectivity. The hypothesis does not articulate why caspase-3 cleavage of tau would differ functionally from caspase-3 cleavage of the thousands of other substrates activated during apoptosis.

Mechanistic gap: The claim that a cytosolic tau fragment "directly facilitates PS externalization" lacks biophysical plausibility without an articulated mechanism. Tau fragments lack known membrane-anchoring domains or lipid modification signals. The proposal that these fragments "disrupt membrane asymmetry preferentially in tau-bearing cells" implies either (a) a dominant-negative effect on scramblase regulation or (b) physical disruption of membrane leaflet organization—both requiring substantially more molecular detail to assess.

Temporal inconsistency: Caspase-3 activation is a late-stage apoptotic event occurring after mitochondrial outer membrane permeabilization and cytochrome c release. If caspase-3-cleaved tau fragments are generating PS exposure as the primary targeting signal, this occurs in cells already committed to death, reducing therapeutic utility to post-mortem debris clearance rather than selective intervention in dying neurons.

Fragment heterogeneity: The hypothesis assumes a specific caspase-cleaved fragment with membrane-binding properties, but caspase-3 cleaves tau at multiple sites (notably D13, D25, D391, and others). Different cleavage patterns generate fragments with distinct biochemical properties. Which fragment is responsible? The literature documents multiple cleavage products with different apparent molecular weights that vary across studies and AD samples.

Counter-Evidence

• Shrinivasan et al. (Cell, 2022) demonstrated that caspase-cleaved tau fragments are released in exosome-free fractions and can be detected in CSF—these are secreted products of dying cells, not active signaling intermediates
• Apoptotic neurons in stroke, traumatic brain injury, and other neurodegenerative conditions show robust PS exposure without requiring tau as a specificity determinant
• Direct measurement of caspase-3-cleaved tau membrane binding has not, to my knowledge, been demonstrated in primary neuronal membranes

Falsification Experiments

Primary test: Generate caspase-3-resistant tau mutants (non-cleavable at major sites) and test whether PS externalization in response to various stressors is reduced specifically in tau-overexpressing cells compared to wild-type controls. If caspase-3 cleavage is the specificity mechanism, non-cleavable tau should reduce PS exposure.

Temporal dissection: Use live-cell imaging with caspase-3 reporters and annexin V binding in primary neurons. Does PS exposure in tauopathy models precede or follow caspase-3 activation? If PS exposure follows caspase activation, the hypothesis fails the temporal requirement.

Membrane binding assay: Purify major caspase-cleaved tau fragments (e.g., △D25, △D391) and perform surface plasmon resonance or liposome co-floatation assays. Direct measurement of membrane association will falsify or support the membrane-binding claim.

Cell-type specificity test: Express caspase-cleaved tau fragments in non-neuronal cell lines. Do these fragments induce PS exposure in cells without the neuronal membrane composition or scramblase context?

Revised Confidence: 0.48 (down from 0.72)

The mechanism does not resolve the caspase-3 ubiquity problem and lacks direct membrane-binding evidence. The high original confidence appears to reflect plausibility rather than evidence strength.

Hypothesis 2: Hyperphosphorylated Tau Inhibits PMCA Pumps, Causing Sustained PS Exposure

Weaknesses and Challenges

Non-specific calcium dysregulation: PMCA inhibition would disrupt calcium homeostasis in all cells with hyperphosphorylated tau, not just those destined for PS exposure. This hypothesis does not explain selectivity—any cell type accumulating hyperphosphorylated tau should show equivalent effects.

PMCA compensation: Neurons express multiple calcium extrusion mechanisms (NCX, mitochondrial uptake, SERCA, PMCA isoforms). PMCA inhibition alone, even if specific to tau-bearing cells, would not cause the sustained calcium dysregulation claimed without addressing these compensating systems.

Scramblase specificity assumptions: The calcium-dependent scramblases TMEM16F and XKR4 are primarily expressed in hematopoietic cells (for TMEM16F) or have unclear neuronal expression patterns. Neuronal scramblases mediating PS exposure are not well-characterized. The hypothesis invokes these scramblases without establishing their neuronal relevance.

"Sustained" vs. "transient" PS exposure distinction: The hypothesis claims that tau-bearing cells show "prolonged PS exposure" unlike normal stressed cells. This is an unverified empirical claim. Annexin V binding assays cannot reliably distinguish transient from sustained exposure without real-time kinetic measurements in live cells, which are technically challenging in primary neurons.

Mechanistic connection: The hypothesis states that tau "physically interacts with" PMCA pumps causing inhibition, but does not explain how hyperphosphorylation specifically mediates this interaction, why this would occur in tauopathy but not other proteopathies, or how PMCA inhibition preferentially affects PS externalization pathways.

Counter-Evidence

• Primary neurons from PMCA1/4 knockout mice show developmental abnormalities but PS exposure dynamics in mature neurons under tauopathic stress are uncharacterized
• Hyperphosphorylated tau accumulates in many conditions (aging, other dementias, prodromal AD) where "sustained PS exposure" is not the prominent feature
• If PMCA inhibition is the mechanism, pharmacological PMCA inhibitors should reproduce the specific PS exposure pattern—this has not been demonstrated in the literature

Falsification Experiments

PMCA functionality direct test: Use fluorescent calcium indicators (fura-2, GCaMP) to measure calcium clearance rates in primary neurons from tauopathy models (P301S, rTg4510) vs. controls. Is recovery kinetics genuinely slower? This directly tests the "sustained exposure" claim.

PMCA knockdown rescue: If tau-PMCA interaction is inhibitory, knockdown of PMCA1/4 should phenocopy tauopathy effects. Test whether PMCA knockdown in non-tau neurons increases PS exposure to tauopathy levels.

Scramblase identification: Perform RNA-seq and proteomics on neurons with vs. without hyperphosphorylated tau to identify which scramblases are actually expressed and regulated. Current assumptions about TMEM16F/XKR4 in neurons are not well-supported.

Calcium clamping test: Use BAPTA-AM or similar calcium chelators to prevent calcium dysregulation. Does this prevent PS exposure in tauopathy models even if tau-PMCA interaction occurs?

Revised Confidence: 0.45 (down from 0.65)

The hypothesis has logical structure but lacks empirical support for the specific molecular interactions proposed. Neuronal scramblase biology is underspecified.

Hypothesis 3: PS Exposure is Cell-Type Context-Dependent Rather Than Pathway-Specific

Weaknesses and Challenges

Does not address the selectivity question: This hypothesis addresses downstream interpretation of PS exposure, not why PS is selectively externalized on tau-bearing neurons. The original framing concerns selective targeting of tau pathology; this hypothesis shifts focus to microglial responses, which is therapeutically relevant but mechanistically distinct from explaining PS exposure specificity.

MERTK/Axl expression confusion: The hypothesis states these are "neuronal-specific PS-binding receptors" but the literature indicates the opposite—MERTK and AXL are predominantly expressed in myeloid cells (macrophages, microglia) with limited neuronal expression. TIMD4 is a T-cell immunoglobulin domain protein, not a neuronal receptor. This conflation of cell-type expression patterns is a significant error.

"Don't eat me" co-expression: The hypothesis states immune cells "exhibit 'don't eat me' signals concurrently" with PS exposure. This is biologically correct but mechanistically trivial—simultaneous "eat me" and "don't eat me" signals create signal ambiguity rather than specificity. The hypothesis does not explain how this ambiguity resolves.

DAMP framing issue: The DAMP framework applies to intracellular molecules released from dying cells (HMGB1, ATP, mitochondrial DNA). PS is a membrane phospholipid exposed on apoptotic cells—a fundamentally different category. PS is not typically classified as a DAMP in the immunological literature.

Counter-Evidence

• Single-cell RNA-seq datasets (Allen Brain Cell Atlas, human AD microglia) consistently show MERTK and AXL expression in myeloid cells, not neurons
• The claim that PS triggers "anti-inflammatory engulfment" in neurons conflicts with the established function of PS as an apoptotic "eat me" signal that promotes phagocytosis regardless of cell type
• PS receptors identified on neurons (e.g., for developmentalaxon guidance) are distinct from the phagocytic receptors invoked here (MERT

Practical Feasibility Assessment: PS Targeting in Tauopathy

Executive Summary

Based on the revised confidence scores and mechanistic critique, I will assess the top three hypotheses with therapeutic potential: Hypothesis 7 (Dual-Signal), Hypothesis 5 (Regional Vulnerability), and Hypothesis 1 (Caspase-3). Hypotheses 2, 3, and 6 are deprioritized due to either low revised confidence or fundamental methodological errors identified in the critique.

Hypothesis 7: Dual-Signal Model (Tau N-fragments + PS)

1. Druggability and Therapeutic Potential

Therapeutic Rationale: This is the strongest candidate for translation. Bispecific constructs are an established modality (Blinatumomab approved, multiple candidates in oncology trials), and the dual-targeting approach addresses the core problem: off-target toxicity in stressed-but-non-tau tissues.

Druggability Assessment:

• Target accessibility: Externalized PS is on the cell surface (annexin V crystallography shows binding pocket druggable); tau N-terminal fragments (aa 1-150) are theoretically targetable if conformational epitopes are accessible extracellularly
• Molecular format: Bispecific antibody (single-domain vs. full IgG) vs. engineered fusion protein vs. small molecule dual-ligand
• Key uncertainty: Whether caspase-cleaved tau N-terminal fragments are actually exposed on the extracellular surface of dying neurons in vivo—current evidence is from CSF measures, not surface staining

Potential: HIGH—if the dual-signal is real, this enables true selectivity unprecedented in neurodegeneration.

2. Existing Compounds and Clinical Trials

| Asset | Stage | Holder | Notes |
|-------|-------|--------|-------|
| N/A for this specific mechanism | — | — | No direct competitors |

Adjacent programs:

• Lu cedimab (Piramal) — Annexin V-based imaging agent (Phase II complete in amyotrophic lateral sclerosis); demonstrates PS targeting is safe in humans
• Multiple bispecific oncology antibodies — Established safety/manufacturing infrastructure transferable
• Anti-tau antibodies (Semorinemab, Gosuranemab) — Target different epitopes (mid-domain, N-terminus), but manufacturing pathways applicable

Development pathway: Requires demonstration that tau N-terminal fragments are exposed on the neuronal surface in vivo—a significant gap. If confirmed, bispecific constructs can be developed using standard antibody engineering.

3. Development Cost and Timeline

| Phase | Estimated Cost | Timeline |
|-------|---------------|----------|
| Target validation (surface exposure of tau N-frags) | $2-4M | 18-24 months |
| Lead bispecific engineering + in vitro potency | $8-12M | 24-36 months |
| IND-enabling studies (GLP tox, manufacturing) | $15-20M | 18-24 months |
| Total to Phase I | $25-36M | 5-7 years |

Key cost drivers: Bispecific antibody manufacturing is 3-5x more expensive than monoclonal antibodies; however, regulatory pathways are well-established following Blinatumomab precedent.

De-risking experiments (should precede investment):

Tissue staining of AD/PSP brain sections with conformation-specific antibodies for surface tau N-fragments

Correlation of surface tau fragment expression with annexin V positivity in postmortem tissue

Development of assay to detect surface tau fragments on live neurons from tauopathy iPSC lines

4. Safety Concerns

| Concern | Severity | Mitigation Strategy |
|---------|----------|---------------------|
| Off-target phagocytosis of stressed-but-healthy neurons | MEDIUM | Dual-targeting reduces but doesn't eliminate risk; requires careful tissue cross-reactivity screening |
| Immune complex deposition (bispecific + Tau + PS) | MEDIUM | Fc-silenced formats; monitoring for infusion reactions |
| Binding to soluble tau fragments (circulating) | HIGH | Engineering for preferential membrane-bound antigen recognition; affinity tuning |
| Microglial over-activation | MEDIUM | Tissue distribution studies; monitoring for neuroinflammation |
| Species cross-reactivity (mouse vs. human tau) | HIGH | Required for preclinical tox; use hTau/P301S models |

Verdict: VIABLE — This hypothesis has the highest translational potential but requires target validation work before committing to bispecific development.

Hypothesis 1: Caspase-3 Cleavage of Tau (Revised Confidence: 0.48)

1. Druggability and Therapeutic Potential

Therapeutic Rationale: Targeting caspase-3 cleavage of tau is conceptually cleaner than a dual-signal approach because it addresses the mechanism upstream. However, the critique identified the "ubiquity problem"—caspase-3 is activated in all apoptosis.

Druggability Assessment:

• Direct approach: Caspase-3 inhibitors exist but are notoriously non-selective and CNS-penetrance is poor
• Alternative approach: Prevent tau cleavage by developing caspase-3 cleavage-resistant tau, or block the downstream fragment's membrane interactions
• Key uncertainty: The mechanism requires that a tau fragment actively drives PS exposure, not merely correlates with apoptosis. No direct evidence exists for membrane-binding by caspase-cleaved tau fragments.

Potential: MODERATE — The mechanism is plausible but requires substantial validation before therapeutic investment.

2. Existing Compounds and Clinical Trials

| Asset | Stage | Company | Notes |
|-------|-------|---------|-------|
| Emricasan (IDN-7313) | Phase II complete | Conatus/Novartis | Pan-caspase inhibitor; failed in NASH; demonstrates CNS exposure concerns |
| Selonsertib (GS-4997) | Discontinued | Gilead | ASK1 inhibitor (not caspase) |

Key finding: Broad caspase inhibitors have been tested systemically and failed due to toxicity (liver, infections). Neuronal-specific caspase-3 inhibition would require substantially different approach.

Alternative strategy: Develop agents that specifically prevent caspase-3 cleavage of tau (e.g., stapled peptides blocking the D391 cleavage site). No current programs.

3. Development Cost and Timeline

Problematic: The falsification experiments identified are technically challenging and expensive before therapeutic investment is warranted.

| Phase | Estimated Cost | Timeline |
|-------|---------------|----------|
| Falsification experiments (membrane binding assays, non-cleavable tau mutant generation) | $1-2M | 12-18 months |
| If validated: Lead optimization, caspase-resistant tau strategies | $15-25M | 3-4 years |
| IND-enabling | $15-20M | 18-24 months |
| Total to Phase I (if validated) | $31-47M | 6-8 years |

Critical path: Must first demonstrate that caspase-cleaved tau fragments have membrane-binding activity—currently unproven. This is the gating factor.

4. Safety Concerns

| Concern | Severity | Mitigation Strategy |
|---------|----------|---------------------|
| Systemic caspase inhibition toxicities (infection, liver) | HIGH | Must be neuronal-specific; oral inhibitors not viable |
| Disruption of normal synaptic pruning (caspase-3 dependent) | HIGH | Neuronal subtype-specific targeting required |
| "Ubiquity problem": won't selectivity | HIGH | Need to address why this creates selectivity over general apoptosis |

Verdict: CONDITIONAL — Viability depends entirely on falsification experiments. If caspase-cleaved tau fragments do not directly bind membranes, this hypothesis is dead. If they do, this represents a novel mechanism worth pursuing but with significant safety challenges.

Hypothesis 5: Regional Vulnerability (Microglial PS-Sensing Heterogeneity)

1. Druggability and Therapeutic Potential

Therapeutic Rationale: This hypothesis shifts the therapeutic target from neurons (PS exposure) to microglia (PS sensing). While mechanistically interesting, the therapeutic goal becomes enhancing microglial PS clearance in vulnerable regions rather than selective PS targeting—a substantially different therapeutic approach.

Druggability Assessment:

• CX3CR1 as target: CX3CR1 antagonists exist (e.g., monoclonal antibodies in oncology); CX3CR1 agonists not well-established
• Microglial receptor enhancement: Upregulating MERTK/Axl on microglia in vulnerable regions could enhance clearance—ligands exist (Gas6, Protein S)
• Key uncertainty: Whether regional vulnerability is truly determined by microglial receptor heterogeneity vs. other factors (neuronal subtype, metabolic profile, blood-brain barrier properties)

Potential: LOW-MODERATE — This is a mechanistic hypothesis about disease progression rather than a direct therapeutic target. Enhancing microglial PS sensing could be beneficial, but the link to tau-selective targeting is indirect.

2. Existing Compounds and Clinical Trials

| Asset | Stage | Holder | Notes |
|-------|-------|--------|-------|
| No direct PS-sensing modulators in neurodegeneration | — | — | — |
| Anti-CX3CR1 antibodies | Phase I (oncology) | Biocad, others | Demonstrates target safety |
| MERTK agonists | Preclinical | Various | No clinical-stage programs for neurodegeneration |

Adjacent programs: Microglial modulation is an active field (TREM2 agonists, CSF1R inhibitors), but PS-sensing specifically is not being targeted.

3. Development Cost and Timeline

Problematic: This hypothesis is primarily explanatory (why some regions are vulnerable) rather than actionable for drug development. Direct translation requires identifying a specific targetable mechanism linking regional microglial heterogeneity to tau vulnerability.

| Phase | Estimated Cost | Timeline |
|-------|---------------|----------|
| Mechanistic validation (regional microglial profiling, causality tests in models) | $3-5M | 18-24 months |
| If validated: Target engagement strategy (likely need new chemical entities) | $20-30M | 3-4 years |
| Total to Phase I (highly speculative) | $23-35M | 5-6 years |

Verdict: LOW PRIORITY — This hypothesis explains a phenomenon but does not directly enable targeting. Recommend deprioritization unless specifically investigating disease progression mechanisms rather than therapeutic targeting.

Hypothesis Prioritization Summary

| Rank | Hypothesis | Revised Confidence | Viability | Development Cost | Recommendation |
|------|------------|-------------------|-----------|-----------------|-----------------|
| 1 | Hypothesis 7: Dual-Signal | 0.69 | HIGH | $25-36M | PRIORITIZE — Proceed with target validation |
| 2 | Hypothesis 1: Caspase-3 | 0.48 | MODERATE | $31-47M | CONDITIONAL — Falsification required first |
| 3 | Hypothesis 5: Regional | 0.61 | LOW-MODERATE | $23-35M | DEPRIORITIZE — Explanatory, not actionable |
| 4 | Hypothesis 2: PMCA | 0.45 | LOW | — | ABANDON — Insufficient mechanism |
| 5 | Hypothesis 6: Phase transition | 0.47 | LOW | — | DEPRIORITIZE — Premature |
| 6 | Hypothesis 3: Cell-type | — | N/A | — | ABANDON — Fundamental errors |
| 7 | Hypothesis 4: Bnip3 | 0.54 | LOW | — | DEPRIORITIZE — Mitochondrial complexity |

Critical Path for Hypothesis 7

Before committing $25-36M to bispecific development:

Critical Experiment: Immunohistochemistry of postmortem AD/PSP brain tissue with conformation-specific antibodies for tau N-terminal fragments, co-stained with annexin V and neuronal markers (NeuN, MAP2). Target validation is the gating step.

Decision point: If surface tau fragment staining correlates with annexin V positivity in neurons (not just extracellular debris), proceed to bispecific engineering. If surface exposure is not detectable, the dual-signal hypothesis fails.

Risk mitigation: Consider developing an annexin V-tau antibody fusion (simpler than bispecific) as a fallback if dual-epitope confirmation is equivocal.

Bottom line: The dual-signal hypothesis (H7) is the only candidate with both mechanistic plausibility and established therapeutic modality precedent. Investment is warranted conditional on target validation.