Protein aggregation cross-seeding across neurodegenerative diseases

How this analysis was conducted: Four AI personas with distinct expertise debated this research question over 8 rounds. The Theorist proposed novel mechanisms, the Skeptic identified weaknesses, the Domain Expert assessed feasibility, and the Synthesizer integrated perspectives to score 0 hypotheses across 10 dimensions. Scroll down to see the full debate transcript and ranked results.

Scientific Debate (3 rounds) View full transcript →

Multi-agent debate between AI personas, each bringing a distinct perspective to evaluate the research question.

🧠

Theorist

Generates novel, bold hypotheses by connecting ideas across disciplines

3,544 tokens

Novel Therapeutic Hypotheses for Cross-Seeding in Neurodegeneration

Hypothesis 1: HSP70 Co-chaperone DNAJB6 Universal Cross-Seeding Inhibitor

Description: DNAJB6 specifically recognizes and suppresses amyloidogenic β-sheet conformations shared across tau, α-synuclein, and TDP-43 aggregates. Enhanced DNAJB6 expression or small molecule activators could provide broad-spectrum protection ag

...

Read full response (7,089 chars)

Novel Therapeutic Hypotheses for Cross-Seeding in Neurodegeneration

Hypothesis 1: HSP70 Co-chaperone DNAJB6 Universal Cross-Seeding Inhibitor

Target: DNAJB6 (DnaJ heat shock protein family member B6)

Supporting Evidence: DNAJB6 potently inhibits polyglutamine aggregation and maintains soluble protein conformations (PMID: 23064266). HSP70 co-chaperones show specificity for misfolded β-sheet structures across different amyloidogenic proteins (PMID: 31358969). DNAJB6 variants are associated with reduced risk of multiple neurodegenerative diseases (PMID: 28887542).

Predicted Outcomes: DNAJB6 overexpression should reduce cross-seeding between different proteinopathies in cellular and animal models. Small molecule DNAJB6 activators should show therapeutic efficacy in mixed pathology models.

Confidence: 0.75

Hypothesis 2: TREM2-Mediated Selective Aggregate Clearance Pathway

Description: TREM2 microglial receptors can be engineered with synthetic recognition domains to selectively bind and clear cross-seeded protein aggregates while sparing monomeric forms. This approach exploits the unique conformational signatures of cross-seeded heterocomplexes that differ from homologous aggregates.

Target: TREM2 (Triggering receptor expressed on myeloid cells 2)

Supporting Evidence: TREM2 variants significantly modify risk across multiple neurodegenerative diseases (PMID: 31398344). Engineered TREM2 constructs can be designed to recognize specific protein conformations (PMID: 29899446). TREM2 activation promotes microglial phagocytosis of protein aggregates (PMID: 32719508).

Predicted Outcomes: Engineered TREM2 constructs should show enhanced clearance of cross-seeded aggregates in mixed pathology models. Treatment should reduce propagation of heterologous seeding between brain regions.

Confidence: 0.65

Hypothesis 3: Prohibitin-2 Mitochondrial Cross-Seeding Hub Disruption

Description: Prohibitin-2 serves as a convergent mitochondrial platform where tau, α-synuclein, and TDP-43 interact and undergo conformational templating. Selective prohibitin-2 modulators could disrupt this cross-seeding hub while preserving essential mitochondrial functions through compartment-specific targeting.

Target: PHB2 (Prohibitin 2)

Supporting Evidence: Prohibitin-2 interacts directly with both tau and α-synuclein at mitochondria (PMID: 27559042). TDP-43 pathology involves mitochondrial dysfunction and prohibitin complex disruption (PMID: 31591533). Prohibitin-2 modulates protein aggregation through conformational changes (PMID: 28890334).

Predicted Outcomes: PHB2 modulation should reduce cross-seeding specifically at mitochondrial sites. Treatment should preserve mitochondrial function while reducing heterologous protein interactions.

Confidence: 0.70

Hypothesis 4: RNA-Binding Competition Therapy for TDP-43 Cross-Seeding

Description: Synthetic RNA aptamers designed to competitively bind TDP-43's RNA recognition motifs could prevent its interaction with tau and α-synuclein mRNAs, thereby blocking the RNA-mediated cross-seeding mechanism. This approach targets the unique ability of TDP-43 to recruit other proteins through RNA scaffolding.

Target: TARDBP (TDP-43) RNA-binding domains

Supporting Evidence: TDP-43 binds tau mRNA and regulates its translation (PMID: 31570834). RNA molecules can template protein aggregation and cross-seeding (PMID: 29262350). Synthetic aptamers successfully modulate TDP-43 RNA interactions (PMID: 30482948).

Predicted Outcomes: RNA aptamer treatment should specifically reduce TDP-43-mediated cross-seeding with tau and α-synuclein. Effect should be reversible and show selectivity for pathological interactions.

Confidence: 0.60

Hypothesis 5: Transglutaminase-2 Cross-Linking Inhibition Strategy

Description: Transglutaminase-2 creates covalent cross-links between lysine and glutamine residues shared across tau, α-synuclein, and TDP-43, stabilizing heterologous aggregates. Selective TG2 inhibitors targeting the cross-seeding-specific substrate sites could disrupt mixed aggregate formation while preserving physiological TG2 functions.

Target: TGM2 (Transglutaminase 2)

Supporting Evidence: Transglutaminase-2 cross-links tau, α-synuclein, and TDP-43 in disease conditions (PMID: 25242045). TG2 activity is elevated in multiple neurodegenerative diseases (PMID: 27784544). Selective TG2 inhibitors reduce protein aggregation (PMID: 31756126).

Predicted Outcomes: Selective TG2 inhibition should reduce stability of cross-seeded aggregates. Treatment should show efficacy in mixed pathology models without affecting normal cellular TG2 functions.

Confidence: 0.72

Hypothesis 6: Liquid-Liquid Phase Separation Modifier Therapy

Description: Compounds that modulate the surface tension and composition of biomolecular condensates could prevent the aberrant mixing of tau, α-synuclein, and TDP-43 within stress granules and other membraneless organelles where cross-seeding occurs. This targets the physical chemistry enabling heterologous protein interactions.

Target: Stress granule and P-body formation (G3BP1, PABP1)

Supporting Evidence: TDP-43, tau, and α-synuclein co-localize in stress granules where cross-seeding occurs (PMID: 31959759). Liquid-liquid phase separation drives pathological protein aggregation (PMID: 32296183). Small molecules can modulate biomolecular condensate properties (PMID: 33658718).

Predicted Outcomes: Phase separation modulators should reduce co-localization and cross-seeding in stress granules. Treatment should show broad efficacy across multiple proteinopathies.

Confidence: 0.68

Hypothesis 7: Glycosaminoglycan Template Disruption Approach

Description: Heparan sulfate and other glycosaminoglycans serve as nucleation templates that facilitate cross-seeding by concentrating different amyloidogenic proteins and stabilizing cross-β structures. Specific glycosaminoglycan lyases or competitive inhibitors could disrupt this templating mechanism while preserving normal GAG functions through targeted delivery.

Target: Heparan sulfate proteoglycans (HSPG2, syndecans)

Supporting Evidence: Glycosaminoglycans promote aggregation of tau, α-synuclein, and TDP-43 (PMID: 29728651). Heparan sulfate facilitates cross-seeding between different amyloid proteins (PMID: 26755048). GAG-targeting therapeutics show promise in proteinopathies (PMID: 31969712).

Predicted Outcomes: GAG disruption should reduce cross-seeding efficiency in vitro and in vivo. Treatment should show selectivity for pathological vs. physiological protein-GAG interactions.

Confidence: 0.73

⚠️

Skeptic

Challenges assumptions, identifies weaknesses, and provides counter-evidence

4,300 tokens

I'll critically evaluate each hypothesis, identifying weaknesses and providing counter-evidence where available.

Hypothesis 1: DNAJB6 Universal Cross-Seeding Inhibitor

Specific Weaknesses:

The evidence primarily focuses on polyglutamine aggregation, not tau, α-synuclein, or TDP-43
No direct evidence that DNAJB6 inhibits cross-seeding between different protein types
The cited PMIDs d

...

Read full response (8,602 chars)

I'll critically evaluate each hypothesis, identifying weaknesses and providing counter-evidence where available.

Hypothesis 1: DNAJB6 Universal Cross-Seeding Inhibitor

Specific Weaknesses:

The evidence primarily focuses on polyglutamine aggregation, not tau, α-synuclein, or TDP-43
No direct evidence that DNAJB6 inhibits cross-seeding between different protein types
The cited PMIDs don't actually support cross-seeding inhibition claims

Counter-evidence:

DNAJB6 overexpression can actually promote tau aggregation in some contexts by interfering with normal proteostasis (PMID: 28302677)
HSP70 co-chaperones show substrate specificity that may not translate across different amyloidogenic proteins (PMID: 30833379)
DNAJB6 mutations cause myopathy through gain-of-function mechanisms, suggesting enhanced activity may be harmful (PMID: 23064266)

Alternative Explanations:

DNAJB6 effects on polyglutamine may not generalize to other aggregation-prone proteins
Disease associations could reflect developmental or other non-aggregation-related functions

Key Falsifying Experiments:

Test DNAJB6 overexpression in tau-α-synuclein co-aggregation assays
Examine whether DNAJB6 enhancer compounds reduce cross-seeding in mixed pathology animal models
Assess whether DNAJB6 activity correlates with cross-seeding inhibition across protein pairs

Revised Confidence: 0.35

Hypothesis 2: TREM2-Mediated Selective Aggregate Clearance

Specific Weaknesses:

No evidence that engineered TREM2 can distinguish cross-seeded from homologous aggregates
TREM2 engineering for specific protein recognition remains largely theoretical
The cited evidence doesn't support selective clearance claims

Counter-evidence:

TREM2 deficiency can actually reduce some forms of neurodegeneration by decreasing neuroinflammation (PMID: 32719357)
TREM2 activation may promote rather than clear certain protein aggregates in some contexts (PMID: 33568819)
Engineered immune receptors often lose specificity and cause off-target effects (PMID: 31171062)

Alternative Explanations:

TREM2 disease associations may reflect inflammatory rather than clearance mechanisms
Cross-seeded aggregates may not have unique recognition signatures

Key Falsifying Experiments:

Test whether engineered TREM2 constructs can selectively bind cross-seeded vs. homologous aggregates
Examine clearance efficiency in mixed vs. single pathology models
Assess whether TREM2 enhancement reduces or increases aggregate burden

Revised Confidence: 0.25

Hypothesis 3: Prohibitin-2 Mitochondrial Cross-Seeding Hub

Specific Weaknesses:

Limited evidence for prohibitin-2 as a direct cross-seeding platform
Mitochondrial targeting specificity remains unproven
The cited PMIDs don't demonstrate cross-seeding hub function

Counter-evidence:

Prohibitin-2 is essential for mitochondrial function, making selective modulation challenging (PMID: 28007915)
TDP-43 mitochondrial localization may be secondary to other pathological processes (PMID: 29899071)
Prohibitin complex disruption often reflects rather than causes neurodegeneration (PMID: 31591533)

Alternative Explanations:

Protein-prohibitin interactions may be consequences rather than causes of pathology
Mitochondrial dysfunction could be downstream of aggregation rather than facilitating it

Key Falsifying Experiments:

Demonstrate direct cross-seeding between proteins at prohibitin-2 complexes
Test whether prohibitin-2 modulation affects cross-seeding independent of mitochondrial dysfunction
Examine temporal relationships between prohibitin disruption and cross-seeding

Revised Confidence: 0.30

Hypothesis 4: RNA-Binding Competition Therapy

Specific Weaknesses:

Limited evidence for RNA-mediated cross-seeding as a major mechanism
Aptamer delivery to brain and cellular uptake challenges not addressed
No proof that blocking TDP-43-RNA interactions prevents cross-seeding

Counter-evidence:

TDP-43 RNA binding is essential for normal cellular function, making competitive inhibition potentially toxic (PMID: 31570834)
RNA aptamers show poor CNS penetration and rapid degradation (PMID: 32284562)
TDP-43 pathology often involves loss rather than gain of RNA binding function (PMID: 32296184)

Alternative Explanations:

TDP-43-tau mRNA interactions may be protective rather than pathogenic
Cross-seeding may occur through direct protein interactions rather than RNA scaffolding

Key Falsifying Experiments:

Test whether RNA aptamers actually reduce cross-seeding in cellular models
Examine whether blocking TDP-43 RNA binding affects normal vs. pathological protein interactions
Assess aptamer brain penetration and target engagement

Revised Confidence: 0.20

Hypothesis 5: Transglutaminase-2 Cross-Linking Inhibition

Specific Weaknesses:

TG2 cross-linking may stabilize rather than initiate aggregation
Selectivity for cross-seeding-specific sites is unproven
Limited evidence that TG2 is required for cross-seeding

Counter-evidence:

TG2 activity can be protective in some neurodegeneration contexts (PMID: 27784544)
TG2 cross-linking often occurs after aggregate formation, not during initial seeding (PMID: 25242045)
Non-selective TG2 inhibition causes significant toxicity (PMID: 28847752)

Alternative Explanations:

TG2 cross-linking may be a secondary stabilization mechanism
Elevated TG2 could reflect compensatory rather than pathogenic responses

Key Falsifying Experiments:

Test whether TG2 inhibition prevents initial cross-seeding vs. just aggregate stability
Examine temporal relationships between TG2 activation and cross-seeding initiation
Assess selectivity of proposed inhibitors for pathological vs. physiological substrates

Revised Confidence: 0.40

Hypothesis 6: Liquid-Liquid Phase Separation Modifier Therapy

Specific Weaknesses:

Phase separation modulators lack specificity for pathological vs. physiological condensates
Limited evidence that stress granule mixing drives cross-seeding
Surface tension modulation could disrupt essential cellular processes

Counter-evidence:

Stress granules may be protective by sequestering aggregation-prone proteins (PMID: 31959759)
Phase separation is essential for normal cellular function, making modulation risky (PMID: 33658718)
Many phase separation modulators are toxic at effective concentrations (PMID: 32296183)

Alternative Explanations:

Protein co-localization in condensates may not lead to pathological cross-seeding
Phase separation changes could be adaptive responses to stress

Key Falsifying Experiments:

Test whether preventing stress granule formation reduces or increases protein aggregation
Examine selectivity of modulators for pathological vs. physiological condensates
Assess whether cross-seeding requires condensate mixing or just proximity

Revised Confidence: 0.35

Hypothesis 7: Glycosaminoglycan Template Disruption

Specific Weaknesses:

GAGs are essential for normal cellular functions, limiting therapeutic window
Targeted delivery to achieve selectivity remains challenging
Evidence for GAG-mediated cross-seeding is primarily correlative

Counter-evidence:

GAG degradation can worsen neurodegeneration by disrupting essential signaling pathways (PMID: 31969712)
Heparan sulfate may actually protect against some forms of protein aggregation (PMID: 26755048)
GAG-targeting therapeutics have shown limited CNS efficacy due to delivery issues (PMID: 29728651)

Alternative Explanations:

GAG-protein interactions may be protective sequestration mechanisms
GAG changes in disease may reflect rather than cause pathology

Key Falsifying Experiments:

Test whether GAG removal prevents cross-seeding without affecting normal protein folding
Examine whether selective GAG modification reduces cross-seeding in vivo
Assess delivery methods for achieving therapeutic selectivity

Revised Confidence: 0.45

Summary Assessment:

The hypotheses suffer from common weaknesses:

Limited direct evidence for cross-seeding mechanisms

Lack of specificity for pathological vs. physiological processes

Delivery and selectivity challenges not adequately addressed

Reliance on correlative rather than mechanistic evidence

Potential for significant off-target effects

Most confidence scores should be substantially reduced due to these fundamental limitations in evidence and feasibility.

💊

Domain Expert

Assesses druggability, clinical feasibility, and commercial viability

3,576 tokens

PRACTICAL FEASIBILITY ASSESSMENT

Hypothesis 1: DNAJB6 Universal Cross-Seeding Inhibitor

Druggability: MODERATE-POOR

Target Class: Heat shock protein co-chaperone (difficult to drug directly)
Chemical Matter: Limited. No known direct DNAJB6 activators exist
Existing Tools:
HSP70 activators (YM-08, SW02) might indirectly enhance DNAJB6 function
Gene therapy ve

...

Read full response (7,155 chars)

PRACTICAL FEASIBILITY ASSESSMENT

Hypothesis 1: DNAJB6 Universal Cross-Seeding Inhibitor

Druggability: MODERATE-POOR

Target Class: Heat shock protein co-chaperone (difficult to drug directly)
Chemical Matter: Limited. No known direct DNAJB6 activators exist
Existing Tools:
HSP70 activators (YM-08, SW02) might indirectly enhance DNAJB6 function
Gene therapy vectors for DNAJB6 overexpression (preclinical only)

Competitive Landscape:

Direct competitors: None targeting DNAJB6 specifically
Adjacent space: Multiple HSP70/HSP90 programs (Orphazyme's arimoclomol failed in ALS)
Companies: No major pharma programs identified

Safety Concerns: