STRC Research

Phase 8h-lite #6 · Tier-1 RWM permeability + Tanimoto refresh after Phase 7I lead pivot

3 min read

Phase 8h-lite #6 · Tier-1 RWM permeability + Tanimoto refresh

Trigger. Phase 7I closed the v52_top3 batch and pivoted the lead away from adamantyl_CONHOH_-Cl (now mech-anchor) to a 4-compound Tier-1 set headed by CONHOMe and phosphonate classes. Re-running the two ligand-dependent Phase 8h-lite probes (#2 RWM permeability and #5 Tanimoto vs ototoxins) on the new set; #1 druggability is pocket-only and unchanged; #4 pose stability and ADMET re-check were already executed by the parallel Phase 7I work (tier1_pose_stability.csv, tier1_admet_recheck.csv).

Tier-1 RWM permeability

Method: Salt 2001 baseline + Stokes-Einstein MW^(−1/3) + Avdeef logP/TPSA + Henderson-Hasselbalch f_neutral. Polyprotic HH for phosphonates (pKa1 ≈ 2.0, pKa2 ≈ 7.2; cf. methylphosphonic acid).

Tier-1 candidateheadMWlogPTPSAlipo×f_neutralP_eff× TMPAbasal_avg @ 90 mincall
aq3__adamantyl__CONHOMe__-ClCONHOMe3862.1162.717.01.0002.4×10⁻⁷12.9×~100%🟢 best
aq3__adamantyl__CONHOMe__-CNCONHOMe3761.4486.54.01.0005.7×10⁻⁸3.0×~24%⚠ marginal
aq3__1-indanyl__phosphonate__-CF3phosphonate4082.9081.924.3≈ 05.3×10⁻¹³0.00×~0%🔴 blocked
aq3__4-F-biphenyl__phosphonate__-CNphosphonate4192.61105.78.8≈ 01.9×10⁻¹³0.00×~0%🔴 blocked
aq3__adamantyl__CONHOH__-Cl (ref/anchor)CONHOH3721.9473.710.20.9751.5×10⁻⁷7.6×~61%mech-anchor

Key implications

  • CONHOMe-Cl is the best topical-deliverable Tier-1. logP 2.11 + TPSA 62.7 + neutral head → P 12.9× TMPA. The neutral methyl-ester head is the prodrug; cochlear esterases would liberate the active CONHOH at the K1141 pocket (paper-§3 mech-proof reference compound).
  • CONHOMe-CN is marginal. TPSA 86.5 Ų + low logP 1.44 cut the Avdeef factor to 4.0 → P only 3× TMPA → basal_avg ~24%. Acceptable but not first-pick.
  • Phosphonate Tier-1 are NOT topical-deliverable as free acids. Diprotic with pKa2 ~7.2 → at pH 7.4 the population is ~62% mono-anion + ~38% di-anion + virtually 0% neutral. The HH derate is 10⁵-fold; even with their high neutral-form lipo factor (24 / 9), P_eff drops to 10⁻¹³ cm/s — three orders of magnitude below TMPA. Prodrug or formulation strategy required:
    • bis(POM) phosphonate ester (cf. tenofovir disoproxil) hydrolysed by plasma / cochlear esterases
    • cyclic phosphonate prodrug (cf. cyclic-HPMPC)
    • salt with bulky lipophilic counter-ion + nanoemulsion (Mβ-CD / P407 carrier — see DR5 formulation)
  • CONHOH reference (Phase 8h-lite #2) unchanged: 7.6× TMPA, basal_avg ~61%. Held as mech-anchor compound; therapy lead pivots to CONHOMe-Cl.

Tier-1 Tanimoto vs ototoxin panel

Morgan FP r=2 2048 bits, threshold 0.40, 9-compound class panel (aminoglycosides, loop diuretics, cisplatin, salicylate, macrolide, quinine).

Tier-1 candidatemax Tanimotoclosest ototoxinflag
aq3__1-indanyl__phosphonate__-CF30.106aspirin✅ no class motif
aq3__adamantyl__CONHOMe__-Cl0.154aspirin✅ no class motif
aq3__adamantyl__CONHOMe__-CN0.145aspirin✅ no class motif
aq3__4-F-biphenyl__phosphonate__-CN0.116aspirin✅ no class motif
aq3__adamantyl__CONHOH__-Cl (ref)0.127aspirin✅ no class motif

All Tier-1 + reference well below 0.40 — no shared Morgan-FP motif with any major ototoxin class. Aminoglycoside class max across all 5 leads ≤ 0.09. The lead-pivot from CONHOH → CONHOMe / phosphonate does not introduce any new ototoxicity-motif risk.

Forward-tree update (paper-section logic)

  • Paper §6 delivery: now reads “CONHOMe-Cl deliverable as topical RWM ear-drop, P 12.9× TMPA. Phosphonate Tier-1 require prodrug formulation (bis-POM ester or cyclic phosphonate, cf. tenofovir disoproxil precedent).” Phase 8h-lite #2 update is the cleanest cell-edit.
  • Paper §3 mech-proof: keep adamantyl_CONHOH_-Cl as the reference compound (it is what Phase 5k +4.53 kT/e was measured on), pivot CONHOMe-Cl as the lead candidate. Explicit split prevents the mech-vs-lead confusion that a one-compound paper would entail.
  • Phase 5k re-run on phosphonate (heavy planned): di-anion phosphonate carries a stronger Coulomb pull on K1141 than CONHO⁻ — predicted Δ at K1141 NZ approaches +9 kT/e (twice the CONHO⁻ value), formal-anion preference ~−5 kcal/mol. If this matches APBS, mech 4 → mech 5 candidate ON THE PHOSPHONATE COMPOUND. CONHOH stays at mech 4 for clean disambiguation.

Connections

Graph View