STRC h01 Phase 5n-lite — N-glycosylation map + calnexin retention proxy on STRC 2026-04-26

Tests the “extended calnexin retention” candidate for the rate-limiting step of E1659A QC clearance, surfaced as the most likely candidate by STRC E1659A QC Rate-Limiting Step Lit Gap-Map (DR1 follow-on).

Method

Calnexin/calreticulin (CNX/CRT) bind GlcMan9GlcNAc2 monoglucosylated N-glycans on nascent glycoproteins. Engagement is glycan-driven; the protein-side determinant is the density and spatial clustering of N-linked glycosylation sequons (NX[S/T], X ≠ P) on the surface of the ER-luminal substrate.

Local script scripts/phase5n_lite_nglyc_calnexin_map.py:

  1. Map all NX[S/T] sequons on STRC (UniProt Q7RTU9, 1775 aa).
  2. Compute their distances to (a) E1659A and (b) K1141 in 3D from Phase 5d snap_010 (full-length E1659A MD-relaxed parent).
  3. Compare WT vs E1659A: does the mutation disrupt or create a sequon?
  4. Calnexin-retention proxy = number of sequons within 25 Å (calnexin engagement radius for tetra-glucan plus protein contact) of the mutation site.

Result

QuantityValue
STRC length1775 aa
Total WT sequons14
E1659A sequons created0
E1659A sequons destroyed0
Nearest sequon to E1659N1179 (NLT motif), 29.8 Å
Nearest sequon to K1141N1179, 29.7 Å
Sequons within 25 Å of E16590
Sequons within 50 Å of E16593
Sequons within 25 Å of K11410
Sequons within 50 Å of K11413

Domain distribution

DomainRangeN-glyc sequons
N-terminal1–7999
Central FN3-like800–12004
Linker1201–15001
C-terminal ZP-like1501–17710

Critical observation: the C-terminal ZP-like domain — where E1659A lives — has zero N-glycosylation sequons. Calnexin/CRT cannot engage E1659A directly via the canonical lectin axis.

Interpretation

Three load-bearing findings:

  1. No sequon within 25 Å of E1659. Even allowing 50 Å (twice the canonical engagement radius), only 3 sequons are within range, and all of them are in the central FN3-like or N-terminal domain — not on the same fold-unit as the mutation.
  2. E1659A does not perturb the N-glyc landscape. Zero sequons created or destroyed by the E to A substitution. Effect on QC therefore cannot be via primary N-glyc-sequon disruption.
  3. The C-terminal ZP-like domain is N-glyc-naïve in the canonical sense. Of the 14 sequons in the full protein, 0/14 are in the 1501–1771 ZP-like fold.

Combined: calnexin extended-retention is unlikely to be the rate-limiting step for E1659A QC clearance. The canonical mannose-trimming-clock mechanism that gates calnexin → EDEM → HRD1 progression cannot apply to a domain that calnexin doesn’t see.

Implications for h01 design

  1. Re-ranking of QC rate-limiting-step candidates from STRC E1659A QC Rate-Limiting Step Lit Gap-Map:
    • ↓ calnexin extended-retention (was top candidate)
    • ↑ HRD1/SEL1L core ERAD (HRD1 ubiquitin ligase recognises misfolded glycoproteins via SEL1L; less glyc-dependent than calnexin)
    • ↑ ER-phagy (FAM134B / RTN3 selective autophagy of ER membrane segments; size-driven, not glyc-driven)
    • ↑ BiP/Grp78 retention (sequon-independent; recognises exposed hydrophobic patches on misfolded ZP-like fold)
  2. τ-floor for Phase 5m τRAMD design holds at 60 min (2× F508del CFTR ERAD t½ from Ward & Kopito 1994 just retrieved by sci-hub). But the physical mechanism gating that floor is more likely BiP cycling or HRD1/SEL1L recognition than calnexin lectin-cycle.
  3. Pharmacochaperone binding at K1141 is even safer than Migalastat Dosing-Cycle Principle flagged. K1141 is in the central FN3-like domain (4 nearby sequons) — not on the ZP-like fold that’s actually disordered by E1659A. Allosteric binding at K1141 cannot interfere with whatever ER-residence-time mechanism is operating on the ZP-like domain.
  4. Pre-Phase-5m τRAMD interpretation reframe. Slow STRC k_off (target ≥ 60 min) is still the design constraint, but the comparator is BiP-cycle / HRD1-recognition timing, not calnexin-lectin-cycle timing. The lit basis is now Hebert et al 1996 + Christianson 2008 (SEL1L) + Schmidt & Lederkremer 2024 BiP review (need to retrieve), not Molinari 2007.

Ranking delta

  • tier A → A (held)
  • mech 4 → 4 (held; E1659A is fold-defect, not glyc-defect — confirms thermodynamic-pocket framing per Phase 5k matched-ensemble +5.99 kT/e mut-WT advantage)
  • deliv 3 → 3 (held)
  • misha_fit 4 → 4 (held)
  • next_step refresh: drop “Phase 5n-lite calnexin/CRT consensus MSA scan” from P1-light queue (this proof closes it as negative); add “BiP-cycle / HRD1-SEL1L timing-axis literature retrieval (sci-hub Christianson 2008, Schmidt 2024 BiP review)” to P1-light

Connections