STRC Research

STRC h01 Phase 5p v53 Acyl Sulfonamide Library 2026-04-26

4 min read

STRC h01 Phase 5p — v5.3 acyl-sulfonamide library design + triage 2026-04-26

Background fallback library design: if Phase 5m τRAMD also fails to deliver kinetic selectivity (target STRC τ : TRPM4 τ ≥ 5×) on the v5.2 shortlist, head-group must change. Per STRC h01 Phase 8g-v3-lite + 8h-lite 9 Pocket Max Phi + Dose Sweep 2026-04-26 §next_step, weakly-anionic head-group exploration is the P1 lever. This script lays the chemistry and triages it before the τRAMD verdict lands, so that Boltz-2 + Vina compute can launch immediately if needed.

Design rationale

Phase 8g-v3-lite proved the off-target APBS FAIL is residue-driven — STRC pocket top-10% +21.90 kT/e ≤ TRPM4 +24.39. CONHOMe head-swap from CONHOH narrowed the pose-anion margin by +3.24 kT/e but does not flip verdict. Logical next move: dilute the head-group anion charge density across more atoms so that highly-cationic off-target pocket interiors (TRPM4 K-rich gating ring, TMEM16A R-rich anion-conducting funnel) pull the head-group less than they pull a CONHO⁻.

Acyl sulfonamide is the canonical hydroxamate bioisostere per Hydroxamic Acid Divalent Cation Liability §bioisosteric-fallbacks and per DR4 (STRC h01 DR4 hydroxamic-acid IP landscape). pKa of the acyl-sulfonamide N-H is 4–5 (Bedaquiline-style pyridine-3-carboxamide bioisostere; GSK RIPK1 series). At pH 7.4 the form is anionic, but charge is delocalised across the C=O, S=O, S=O, and N — four heavy atoms — vs CONHO⁻ where charge sits on a single oxygen. The Coulomb pull on a +24 kT/e off-target pocket is therefore reduced by the ratio of effective-charge-radii, roughly 2-3×.

Library

scripts/phase5p_v53_weakanion_library.py constructed via prefix-replacement on the v5.2 lead SMILES backbone (matches [[STRC h01 Phase 8 v5 Library Coulomb-Aware Design 2026-04-24]] convention).

  • Heads (2):
    • acylsulfonamide_SO2Me — CS(=O)(=O)NC(=O)- (canonical Bedaquiline-class head)
    • acylsulfonamide_SO2CF3 — FC(F)(F)S(=O)(=O)NC(=O)- (more electron-withdrawing → lower pKa → stronger anion at pH 7.4)
  • Bodies (2): adamantyl, 1-indanyl (matches v5.2 mut-prefer top hits)
  • Tails (3): -Cl, -CN, -CF3 (matches v5.2 Boltz-2 mut-prefer correlates)

Library size = 2 × 2 × 3 = 12 SMILES. Ring-bioisostere heads (tetrazole, oxadiazol-2(3H)-one, hydroxytriazole) require ring-numbering-aware SMILES assembly that the simple prefix-replacement approach cannot do reliably; deferred to Phase 5p-v2 with hand-written per-combo SMILES if Phase 5m verdict elevates the priority.

Triage gates (cumulative, RDKit)

  1. RDKit parse / sanitize OK
  2. Lipinski rule-of-5 (MW ≤ 500, logP ≤ 5, HBD ≤ 5, HBA ≤ 10)
  3. TPSA ≤ 100 Ų (RWM permeability proxy from STRC h01 Phase 8h-lite Light Computational Evidence Package 2026-04-26 §2)
  4. Formal anion count ≤ 1 at pH 7.4 (avoids the phosphonate di-anion failure mode noted in hub)

Result — 6/12 PASS

nameMWlogPHBAHBDTPSAanionPASS
v5.3__aq3__adamantyl__acylsulfonamide_SO2Me__-Cl434.01.515287.60 (Lipinski)✅
v5.3__aq3__adamantyl__acylsulfonamide_SO2Me__-CN424.50.8462111.4—❌ TPSA
v5.3__aq3__adamantyl__acylsulfonamide_SO2Me__-CF3467.51.885287.6—✅
v5.3__aq3__1-indanyl__acylsulfonamide_SO2Me__-Cl415.91.485287.6—✅
v5.3__aq3__1-indanyl__acylsulfonamide_SO2Me__-CN406.50.8162111.4—❌ TPSA
v5.3__aq3__1-indanyl__acylsulfonamide_SO2Me__-CF3449.51.845287.6—✅
v5.3__aq3__adamantyl__acylsulfonamide_SO2CF3__-Cl487.92.405287.6—✅
v5.3__aq3__adamantyl__acylsulfonamide_SO2CF3__-CN478.51.7362111.4—❌ TPSA
v5.3__aq3__adamantyl__acylsulfonamide_SO2CF3__-CF3521.52.775287.6—❌ Lipinski (MW > 500)
v5.3__aq3__1-indanyl__acylsulfonamide_SO2CF3__-Cl469.92.375287.6—✅
v5.3__aq3__1-indanyl__acylsulfonamide_SO2CF3__-CN460.41.7062111.4—❌ TPSA
v5.3__aq3__1-indanyl__acylsulfonamide_SO2CF3__-CF3503.42.735287.6—❌ Lipinski (MW > 500)

Pass rate 50 %. All -CN tails fail TPSA (additional polar nitrile pushes TPSA from 87.6 → 111.4); two -CF3 + SO2CF3 combos fail Lipinski MW.

Top-3 next-gen lead candidates

Ranked by lipophilicity-balanced TPSA (RWM permeability-favoured):

  1. v5.3__aq3__1-indanyl__acylsulfonamide_SO2Me__-Cl — MW 415.9, logP 1.48, TPSA 87.6. Best RWM permeability profile of the PASS set.
  2. v5.3__aq3__adamantyl__acylsulfonamide_SO2Me__-Cl — MW 434.0, logP 1.51, TPSA 87.6. Direct head-swap of the v5.2 lead (CONHOMe → acylsulfonamide), holds adamantyl body for ADMET cleanliness inheritance.
  3. v5.3__aq3__adamantyl__acylsulfonamide_SO2Me__-CF3 — MW 467.5, logP 1.88, TPSA 87.6. Higher Boltz-2 mut-prefer probability if the -CF3 tail correlation from v5.2 holds.

Path forward (only if Phase 5m τRAMD verdict requires head pivot)

  1. Parameterise PASS-list SMILES via openff Molecule + am1bcc (now installed, AmberTools/sqm path-fixed in scripts).
  2. Vina dock all 6 PASS candidates against STRC E1659A K1141 pocket.
  3. Boltz-2 ipTM scoring on top-3 by Vina against WT + E1659A + TRPM4 (3-target run).
  4. APBS pocket-top10 resample on top-3 with anion charge spread across S=O₂N(H)C(=O) (vs CONHO⁻). Predicted improvement: STRC anion-O ⟨φ⟩ similar; off-target anion-O ⟨φ⟩ reduced 30–50 % due to charge delocalisation (reduces TRPM4 anti-selectivity from −2.49 kT/e to predicted −0.5 kT/e — would put us within WEAK gate).
  5. If APBS gate still fails, escalate to Phase 5p-v2 with hand-written ring-bioisostere SMILES (tetrazole, oxadiazolone, hydroxytriazole).

Ranking delta

  • tier A → A (held)
  • mech / deliv / misha_fit unchanged (this is design-prep not a proof)
  • next_step adds: “v5.3 acyl-sulfonamide PASS-list ready (6 SMILES) — execute Boltz-2 + Vina + APBS pipeline only if Phase 5m τRAMD STRC:TRPM4 ratio < 5× on v5.2 shortlist; otherwise stay on CONHOMe-Cl lead”

Connections

Graph View

Backlinks