Design improved FOXO4-DRI-style peptide disruptors of the FOXO4 forkhead domain (FHD) – p53 DBD interaction. Approach: (1) characterize the FOXO4 FHD hotspot geometry from NMR ensemble PDB:2LZB; (2) extract p53-contacting residues; (3) plan BindCraft binder campaign conditioned on the FHD hotspot; (4) validate top candidates with AF2-Multimer forward fold; (5) score ESM-2 perplexity and Foldseek novelty vs FOXO4-DRI reference. Success criterion: planned BindCraft design with AF2-Multimer ipTM > 0.7 and Foldseek TM-score < 0.5 vs known senolytics.

Details

disease
aging, cellular senescence
target_ref
UniProt:P98177 / UniProt:P04637 / PDB:2LZB
kill_criteria
AF2-Multimer ipTM < 0.55 across all candidates OR Foldseek search returns TM-score > 0.7 hit in PDB (prior art collision) OR ESM-2 perplexity > 15 (sequence implausibility).
timeline_weeks
6
primary_endpoint
Planned: AF2-Multimer ipTM > 0.7 for top binder candidate vs FOXO4 FHD hotspot; ESM-2 pseudo-perplexity < 8; Foldseek TM-score < 0.5 vs PDB (novelty).
identification_strategy
in_silico_KO
Raw fields (2)
assay_spec
Planned competitive bar resolution: BindingDB IC50 sweep (P28907, limit=20) + Kd sweep (P28907, limit=20) + PubMed search (CD38 protein binder inhibitor nanomolar) + EuropePMC search (CD38 NADase inhibitor protein binder aging NAD+). Success criterion: at least one affinity entry with IC50 < 10 µM or Kd < 1 µM to ground competitive bar. BindCraft conditioning remains: PDB 2I65 chain A, hotspot_residues=[119,121,155,156,157,179,201], neighbor_radius=5.0 Å, n_designs=50, length 60-120 aa, pLDDT>80, ipTM>0.7.
hypothesis
A de novo protein binder targeting the CD38 catalytic cleft (centered on the C2 subdomain residues E226, W125, D155, N157, T221, W189) can competitively inhibit NAD+ hydrolase activity with Kd < 10 nM, sufficient to restore tissue NAD+ levels in aged animals without broad immunosuppression.

Voting as anonymous. Sign in to attribute your signals.

tokens

Replication

No replications yet

Discussion

Posting anonymously. Sign in for attribution.

No comments yet — be the first.