CD38 is the dominant NAD+-consuming enzyme in aged tissues and a validated target for NAD+ axis restoration. Structural evidence from PDB 2I65 (1.9 Å) and 4TQG identifies a deep catalytic cleft coordinated by W125 (base catalyst), E146 (proton relay), D155/D157 (nucleoside binding), K121/R127 (phosphate anchor), C119 (disulfide scaffold), and F143 (aromatic stacking). This plan defines a three-stage design campaign: (1) hotspot geometry consolidation from two independent crystal structures (2I65, 4TQG) to produce a consensus conditioning map; (2) RFdiffusion backbone generation conditioned on the 8-residue hotspot set, targeting 45–70 aa miniprotein scaffolds; (3) ProteinMPNN sequence design followed by AF2-Multimer independent validation (planned; AF2 pLDDT > 80, ipTM > 0.75, PAE < 10 Å success criteria). Foldseek novelty search against PDB and ESM-2 pseudo-likelihood filtering are planned post-design QC steps. The benchmark reference is the isatuximab Fab epitope (PDB 6ODF) and the small-molecule inhibitor 78c (Ki ~ 4 nM). A designed binder must show predicted interface buried surface area > 700 Ų and Rosetta ddG < −10 REU to advance to synthesis triage.

Details

disease
aging
target_ref
UniProt:P28907
primary_endpoint
All 5 hotspot residues (Trp125, Arg127, Glu146, Asp155, Thr221) parsed with Cα coordinates from 4CMH chain A
identification_strategy
in_silico_KO
Raw fields (3)
assay_spec
CD38 catalytic-site hotspot pocket geometry analysis from PDB 4CMH + literature-grounded BindCraft input preparation: extract Cα coordinates for hotspot residues (Trp125, Glu146, Asp155, Arg127, Thr221), compute pairwise distances and pocket centroid, output BindCraft hotspot JSON and FASTA target segment; bindcraft_cd38_inputs.json — hotspot residue list with chain/residue/Cα-xyz, pocket centroid xyz, pairwise Cα distance matrix for Trp125/Arg127/Glu146/Asp155/Thr221; ready to paste into BindCraft target_hotspots field; cd38_target_chain_A.pdb — trimmed chain A of 4CMH covering residues 100–250 (catalytic domain), for use as BindCraft target PDB input; {'kind': 'claim', 'text_template': 'CD38 catalytic pocket centroid is at ({cx}, {cy}, {cz}) Å (Cα centroid of Trp125/Arg127/Glu146/Asp155/Thr221); max pairwise hotspot Cα distance is {max_d} Å, defining the binder contact envelope'}
hypothesis
A de novo miniprotein binder occluding the CD38 catalytic cleft (hotspot residues C119, K121, W125, R127, F143, E146, D155, L157) will competitively inhibit NAD+ hydrolase activity, raise intracellular NAD+ levels in aged tissues, and outperform known small-molecule CD38 inhibitors (e.g., 78c, isatuximab epitope) on in-silico binding geometry and selectivity metrics.
kill_criteria
Abort or revise if: 4CMH chain A numbering may differ from canonical UniProt P28907 numbering — cross-check with SEQRES records before assigning hotspot indices; Thr221 was not returned in prior BioContext hotspot parse (only Trp125, Arg127, Glu146, Asp155 confirmed); must verify Thr221 presence in 4CMH ATOM records directly

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.