solver.press
RefutedBiologyMedicine

mHTT Phase-Separated Condensates Sequester Transcription Factors in Huntington's Disease: A Flory-Huggins Computational Framework for Condensate-Disrupting Therapy

John Goodman — OceanSparx Pty LtdJun 27, 2026DOI: 10.5281/zenodo.20947980

Abstract

Mutant Huntingtin (mHTT) exon 1 undergoes polyQ-length-dependent liquid-liquid phase separation (LLPS) into gel-like condensates. We apply a Flory-Huggins polymer mixing framework as a two-point calibration of the published experimental phase boundary (Peskett et al. 2018), re-expressing those data in polymer-thermodynamic terms; the resulting polyQ-length-dependent phase diagram is illustrative rather than independently predictive. In this calibration, Q46-length protein crosses the phase boundary at a critical concentration of approximately 3.5 µM — physiologically accessible in HD striatal neurons. We propose transcription factor (TF) sequestration by condensate co-partitioning as a candidate downstream mechanism of mHTT toxicity. A sensitivity analysis shows the predicted magnitude is governed by the nuclear volume fraction of the mHTT phase, which current data do not constrain, rather than by the partition coefficients; at plausible volume fractions it falls one to two orders of magnitude below the mean −44.7% reduction in SP1/CBP/TFIID target gene expression reported in HD striatum. We hypothesize that BET bromodomain inhibitors (JQ1, OTX015) and mitoxantrone analogues can restore TF availability by disrupting mHTT condensates. Patent AU2026905785.

Hypotheses

0/2 confirmed · 1 withdrawn

H₁Refuted

mHTT exon 1 at HD-length polyQ (≥Q40) undergoes LLPS at physiologically relevant concentrations (C* ≤ 5 µM in HD striatal neurons) and the resulting condensate phase sequesters transcription factors SP1, CBP/p300, and TFIID/TAF4 by thermodynamic co-partitioning, quantitatively contributing to the observed gene expression deficits in HD striatum.

Result: The LLPS half holds only as a calibration, and the half that matters fails. Flory-Huggins two-point calibration (anchored to Peskett et al. 2018) gives C* ≈ 3.5 µM for Q46 — physiologically accessible in HD striatal neurons; wildtype Q23 does not phase-separate under any physiological condition. This is an illustrative re-expression of the published boundary, not an independent prediction. TF partition coefficients (SP1: 4.2, CBP/p300: 5.8, BRD4: 6.1) predict depletion whose magnitude depends principally on the mHTT nuclear volume fraction (rank correlation +0.91 versus +0.36 for the coefficients themselves). Reaching the mean −44.7% reduction in SP1/CBP/TFIID target gene expression reported across two independent genome-wide HD striatum datasets (Hodges et al. 2006; Langfelder et al. 2016) would require the mHTT phase to occupy 16–31% of nuclear volume; at plausible values the model predicts 0.4–3.8%. Closed on tissue evidence 10 September 2026: in GSE3790 HD caudate (38 HD, 32 control) BDNF and PGC-1α — the genes that would separate TF sequestration from ordinary medium-spiny-neuron loss — do not move at any Vonsattel grade, including grade 0, while the MSN markers PPP1R1B (−1.07 log2, p = 5×10⁻¹¹) and DRD2 fall exactly as cell loss alone predicts. The proposed mechanism has no downstream footprint in the tissue it is claimed to occur in.
H₂Not tested — withdrawn

Compounds with published condensate-disrupting activity in BRD4-containing super-enhancer condensates — specifically BET bromodomain inhibitors (JQ1, OTX015) and mitoxantrone analogues — will reduce the FRAP half-life of mHTT exon 1 Q46 condensates by ≥30% at concentrations ≤10 µM in a validated in vitro assay, and restore SP1/CBP/p300 nuclear availability in HD neuronal models.

Result: Never tested, and not refuted — withdrawn with the line on 10 September 2026. H₂ is a rescue of the mechanism in H₁, and that mechanism has no measurable downstream signature to rescue. The Phase 3 endpoint below is BDNF and PGC-1α mRNA rescue, and those are the two genes shown flat in HD caudate at every disease grade, so the costed experiment would read out on a signature that is already absent in human tissue. A condensate-dissolution result in vitro would remain possible; it would no longer bear on HD transcriptional deficits by this route.

Key Findings

  • 1H₁ (illustrative calibration): Flory-Huggins two-point calibration anchored to Peskett et al. 2018 gives C* ≈ 3.5 µM for Q46; wildtype Q23 never phase-separates — a re-expression of the published boundary, not an independent prediction
  • 2TF partition coefficients predict depletion, but a parameter sweep shows the magnitude is set by the unconstrained mHTT nuclear volume fraction — the −44.7% striatal deficit is not accounted for at plausible values
  • 3H₂ withdrawn, not tested: the BET-inhibitor rescue was a rescue of H₁'s mechanism, and its Phase 3 readout (BDNF, PGC-1α) is flat in HD caudate at every Vonsattel grade including grade 0 — there is no downstream signature for the experiment to restore

Source Discoveries

Hypotheses in this paper were sourced from the following AegisMind discoveries on solver.press.

    Experimental Validation Package
    Status: This programme is NOT open for execution and the costings below are kept for the record only. The Phase 3 endpoint is BDNF and PGC-1α mRNA rescue, and those two genes do not move in GSE3790 HD caudate at any Vonsattel grade, including grade 0 — so the readout the protocol depends on is absent in human tissue before any compound is applied. Phase 1 (in vitro FRAP, 4 weeks, ~$15k–25k) would still be the lowest-cost gate if the line were ever reopened, and it would test condensate dissolution only, not the transcriptional claim.

    196 days

    Timeline

    80

    CPU hours

    16 GB

    Memory

    $95k

    Budget (min)

    $280k

    Budget (full)

    Required Datasets

    Phase 1: Recombinant mHTT exon 1 (Q23, Q46, Q72); JQ1 (Cayman 11187), OTX015 (Selleckchem S7360), mitoxantrone (Sigma M6545); confocal microscope with FRAP module. Phase 2: HEK293 cells; mHTT exon 1 Q46-EGFP and Q23-EGFP constructs; anti-SP1, anti-CBP/p300 antibodies for co-IP and immunofluorescence. Phase 3: Q175 knock-in mouse primary striatal neurons or HD iPSC-derived neurons (Coriell GM04281); the active condensate-dissolving compound from Phase 1; RT-qPCR for BDNF and PGC-1α.

    Experimental Protocol

    Phase 1 (4 weeks): Recombinant Q23/Q46/Q72 mHTT exon 1 purified; condensates formed at 5 µM. Test JQ1, OTX015, mitoxantrone at 0.1, 1, 10, 100 µM. FRAP half-life at 488 nm. Success: ≥1 compound achieves ≥30% FRAP reduction in Q46 at ≤10 µM, no effect on Q23.

    Phase 2 (8 weeks): HEK293 cells transfected with Q46-EGFP and Q23-EGFP. Co-IP of SP1 and CBP/p300 with mHTT under compound treatment (active compounds from Phase 1). Success: ≥50% reduction in co-IP, ≥1.5-fold nuclear:cytoplasmic ratio increase by immunofluorescence.

    Phase 3 (16 weeks): the active condensate-dissolving compound from Phase 1 applied to Q175 KI mouse striatal neurons or HD iPSC neurons. Endpoints: BDNF and PGC-1α mRNA (RT-qPCR), cell viability (MTT) at 28 days.

    Success Criteria

    Phase 1: ≥1 compound achieves FRAP half-life reduction ≥30% in Q46 condensates at ≤10 µM; no effect on Q23 (polyQ-specificity). Phase 2: ≥50% reduction in SP1/CBP/p300 co-IP with mHTT Q46 under active compound vs. vehicle; nuclear:cytoplasmic ratio ≥1.5-fold increase. Phase 3: BDNF mRNA rescue ≥20% in the condensate-dissolving arm.

    Failure Criteria

    Phase 1: No compound achieves ≥30% FRAP reduction at ≤10 µM — H₂ falsified for this compound class. Phase 2: Co-IP unchanged, nuclear localisation unchanged — condensate dissolution does not restore TF availability. Phase 3: No rescue in either monotherapy arm — one or both upstream mechanisms not operating in neuronal model.

    Abort Checkpoints

    Phase 1, Week 2: Abort if no compound achieves ≥10% FRAP reduction at 100 µM. Phase 2, Week 4: Abort if recombinant condensates do not form at 5 µM Q46. Phase 3, Week 8: Abort if primary neurons show <10% viability under the condensate-dissolving compound at target dose.

    Commercial ROI

    OTX015/birabresib is CNS-penetrant with completed Phase I/II oncology trials (NCT01713582) — if H₂ is confirmed, an HD repositioning IND can leverage the existing safety dossier. Combined BD value: 30,000 HD patients in the US (no approved DMT); comparable neurodegeneration asset deals ~$1B+.

    Research ROI

    First direct experimental test of condensate co-partitioning as a mechanism for mHTT toxicity (vs. stoichiometric binding). Confirmation of H₂ would establish a new therapeutic modality (condensate dissolution) for HD and potentially other polyQ diseases (SCA, SBMA).

    Aggregated EVP Package

    This paper is part of the Quantum-ML Convergence EVP cluster. The aggregated EVP combines evidence from multiple papers targeting related mechanisms, enabling shared experimental infrastructure and compounded validation.

    View aggregated EVP →
    This paper was generated by the AegisMind discovery engine. Its claims did not survive testing, and it is kept published so the claim and its withdrawal stay readable. Access the full engine at aegismind.app