Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Recurso digital |
| Lingua: | inglese |
| Pubblicazione: |
Zenodo
2026
|
| Soggetti: | |
| Accesso online: | https://doi.org/10.5281/zenodo.18123207 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
Sommario:
- <h2>Abstract</h2> <p>Tauopathies including Alzheimer’s disease (AD) and Progressive Supranuclear Palsy (PSP) lack effective disease-modifying treatments. This report advances and tests a stage-dependent “Secretory Shunt” hypothesis: when autophagic flux becomes impaired, cells may shift toward RAB27A-associated vesicular/exosomal export of p62/SQSTM1–tau cargo, providing intracellular relief while potentially increasing extracellular propagation. Cross-model results suggest this response evolves over time rather than behaving uniformly across disease phase and model system.</p> <p>In chronic human AD postmortem brain (GSE84422; 2,004 Affymetrix microarray samples from 125 subjects across 19 regions), SQSTM1–RAB27A is strongly positive (r = +0.539, p < 10⁻¹⁵⁰), while RAB27A–RAB27B is negative (r = −0.516, p < 10⁻¹³⁶). In acute 4R tauopathy models, RAB27A–RAB27B is positive in both models (r = +0.80–0.86; p-values reported below).</p> <p>We interpret this sign flip as consistent with a stage-evolution model: acute stress mobilizes both secretory paralogs (“emergency response”), whereas chronic disease is consistent with a relative downshift in RAB27B-associated capacity alongside persistence of RAB27A (“shunt takeover”).</p> <h2>Key findings (hypothesis-generating, cross-dataset)</h2> <ul> <li> <p><strong>Core axis conserved:</strong> SQSTM1–RAB27A correlation is positive across all datasets.</p> </li> <li> <p><strong>Stage discordance resolved:</strong> RAB27A–RAB27B flips from positive (acute models) to negative (chronic human brain).</p> </li> <li> <p><strong>Nrf2 coupling signal:</strong> NFE2L2→SQSTM1 shows positive evidence in cell models, motivating a staging-dependent caution hypothesis that requires perturbational validation before any clinical interpretation.</p> </li> <li> <p><strong>Motor independence signal:</strong> MYO5A co-varies with RAB27B but not RAB27A in these analyses, consistent with a distinct shunt-associated pathway.</p> </li> <li> <p><strong>Mechanistic hypothesis tracking:</strong> A structured set of mechanistic hypotheses is evaluated with validated/refined/contradicted status reported in accompanying tables.</p> </li> </ul> <h2>Datasets included (pre-processed for replication)</h2> <ul> <li> <p><strong>GSE84422 (MSBB / Mount Sinai):</strong> chronic human AD postmortem bulk-tissue Affymetrix microarray, n = 2,004 (125 subjects; 19 regions)</p> </li> <li> <p><strong>GSE255902:</strong> acute iPSC-derived 4R tau neurons, RNA-seq, n = 19</p> </li> <li> <p><strong>GSE163150:</strong> acute SH-SY5Y 4R tau cells, RNA-seq, n = 24</p> </li> </ul> <p>Total: <strong>2,047 samples across 3 independent datasets.</strong></p> <h2>What’s in this record</h2> <ul> <li> <p><strong>Manuscript:</strong> <code>secretory_shunt_v4.pdf</code> (plus Markdown and LaTeX sources)</p> </li> <li> <p><strong>Scripts:</strong> <code>reproduce_v4_correlations.py</code> (recomputes reported statistics), <code>generate_figures.py</code> (regenerates figures)</p> </li> <li> <p><strong>Tables/data products:</strong> correlation summaries, hypothesis status tables, and a stage-evolution model JSON</p> </li> <li> <p><strong>Figures:</strong> 6 main + 2 supplementary PDFs</p> </li> </ul> <h2>Reproducibility (quickstart)</h2> <p>To reproduce reported statistics:</p> <pre><code>pip install pandas numpy scipy python reproduce_v4_correlations.py </code></pre> <p>Optional (only if re-downloading from GEO; preprocessed files included in this record):</p> <pre><code>pip install GEOparse requests </code></pre> <p>To regenerate figures:</p> <pre><code>pip install matplotlib numpy python generate_figures.py </code></pre> <h2>Limitations / scope</h2> <ul> <li> <p>GSE84422 is bulk tissue; results can be confounded by cell-type composition and regional effects.</p> </li> <li> <p>GSE84422 includes repeated measures; the packaged processed artifact does not include donor IDs for subject-level modeling, so p-values may be inflated if samples are treated as independent.</p> </li> <li> <p>All results are correlational; mechanistic directionality requires perturbational validation.</p> </li> <li> <p>This report is for research purposes and does not provide medical guidance.</p> </li> </ul> <h2>Version notes (v4.0)</h2> <ul> <li> <p>Corrected GSE84422 modality labeling (microarray, not RNA-seq)</p> </li> <li> <p>Refined therapeutic framework to strictly focus on preclinical hypothesis generation and experimental sequencing</p> </li> <li> <p>Expanded limitations and reproducibility materials</p> </li> </ul> <h2>License / contact</h2> <p>License: <strong>CC-BY 4.0</strong></p> <p>Correspondence: <a href="mailto:aaron@lumenais.com">aaron@lumenais.com</a></p>