CD38 is the dominant NAD+-consuming enzyme in aged tissues; its expression increases with age and inflammatory senescence, driving the NAD+ decline that underlies mitochondrial dysfunction, impaired sirtuin activity, and accelerated aging phenotypes. Prior art: small-molecule inhibitors (78c, isatuximab, daratumumab) validate the target but lack tissue specificity and oral bioavailability as peptide scaffolds. This plan uses the 2I65 (apo) and 2PGJ (substrate-bound) crystal structures to define a hotspot-conditioned BindCraft run. Planned procedures: (1) geometric neighbor profiling of catalytic residues C119, T144, D179, N182, C201, S274 from PDB 2PGJ; (2) BindCraft binder design conditioned on those hotspots, targeting 65-90 aa miniprotein scaffolds; (3) AF2-Multimer forward-fold validation of top-5 designs (independent oracle, not RFdiff); (4) ESM-2 perplexity filter (ppl < 8) on all sequences before AF2; (5) Rosetta FastRelax ddG estimation on AF2-validated complexes; (6) Foldseek search of designed scaffolds against PDB to score novelty. Success criteria: AF2 ipTM > 0.7, pLDDT(binder) > 80, Rosetta ddG < -10 REU, Foldseek TM-score to nearest PDB hit < 0.6.
Details
- disease
- aging
- target_ref
- UniProt:P28907
- primary_endpoint
- At least 1 de novo binder with BindCraft ipTM >= 0.70 against CD38 chain A (PDB:2PGJ), conditioned on hotspot residues C119/T144/D179/N182/C201/S274; AF2-Multimer forward-fold ipTM >= 0.70 as independent validation criterion.
- identification_strategy
- in_silico_KO
Raw fields (3)
- assay_spec
BindCraft single-pass binder design against CD38 catalytic cleft hotspot residues C119/T144/D179/N182/C201/S274 from PDB:2PGJ chain A (pending hotspot geometry confirmation from pdb_hotspot_profile); planned output: 60-80 aa binder length; success criterion: ipTM >= 0.70 on top-1 design; ESM-2 pseudo-perplexity filter (cutoff <= 8.0) applied to top-5 sequences; AF2-Multimer forward-fold validation required (pLDDT >= 80, AF2 ipTM >= 0.70 independent of design model); Foldseek structure search against PDB to confirm novelty (TM-score < 0.5 to nearest non-CD38 hit).
- hypothesis
A de novo peptide binder engineered to occlude the CD38 catalytic cleft — anchoring to hotspot residues C119, T144, D179, N182, C201, and S274 — will competitively inhibit CD38 NADase activity and restore NAD+ levels in aged tissues, with selectivity over structurally unrelated NAD-consuming enzymes (PARP1, SIRT1).
- kill_criteria
Abort or revise if: (1) UniProt P28907 canonical sequence conflicts with 2PGJ chain A residue numbering — re-index hotspots C119/T144/D179/N182/C201/S274 before BindCraft conditioning; (2) BindingDB Ki or IC50 reveals a published peptide already achieving <100 nM against CD38 catalytic cleft with no selectivity gap — reassess novelty; (3) hotspot centroid geometry from 2PGJ confirms residues C119 and C201 are disulfide-bonded and buried — reconsider surface accessibility before proceeding to binder design.