Fold №73 — Tirzepatide C-terminal Lys-40 γGlu-C20 Diacid Extension

Verdict: PROMISING | Class: METABOLIC | Target: GLP-1R (P43220) / GIPR

Mandatory disclaimer: All findings are in silico predictions only. No wet-lab validation has been performed. Predicted properties do not constitute evidence of real-world biological activity. This is research, not medical advice.

Mechanism of Action

Tirzepatide is a 39-residue dual GIP/GLP-1 receptor agonist that engages two class B GPCRs: GLP-1R, driving insulin secretion and satiety, and GIPR, which augments the GLP-1R signal and contributes to the superior weight-loss profile observed in clinical trials (SURPASS, SURMOUNT series). The pharmacophore is an amphipathic α-helix spanning roughly residues 1–29, with N-terminal Tyr-1 insertion into the GLP-1R TMD orthosteric pocket as the canonical activation mechanism. The native C20 fatty diacid conjugated via a γGlu-γGlu spacer to Lys-20 binds reversibly to human serum albumin, dramatically reducing renal clearance and proteolytic exposure to yield a plasma half-life of approximately five days and enabling once-weekly subcutaneous dosing.

The C-terminal PPPS tail (residues 36–39) is structurally disordered in solution and in cryo-EM structures of homologous class B GPCR incretin complexes, projecting away from both the ECD and TMD interfaces. This disorder is well-established for GLP-1 and semaglutide receptor complexes in the broader class B literature and is the structural premise underlying the C-terminal lipidation strategy.

Performance Applications

This modification targets pharmacokinetic extension rather than acute receptor potency. The primary application space is:

• Extended dosing interval: A substantially longer half-life (targeting >7–10 days vs. native ~5 days) could enable bi-weekly or potentially monthly subcutaneous injection formats, addressing adherence barriers in chronic metabolic disease management.
• Obesity and MASH pharmacotherapy: The clinical evidence base for tirzepatide in obesity (SURMOUNT) and MASH is strong; longer-acting formats could improve real-world adherence in these chronic indications where treatment discontinuation rates are high.
• Comparator to semaglutide PK: Semaglutide achieves ~7 days via a C18 fatty diacid/mini-PEG/γGlu linker; a bivalent-anchor tirzepatide analog could, in principle, meet or exceed semaglutide's PK profile while preserving the dual-agonist pharmacology that drives superior weight loss.

Note: Tirzepatide's current once-weekly dosing already satisfies clinical needs in approved indications. The incremental benefit of further half-life extension is real but targets convenience optimization rather than unmet medical need.

Modification Rationale

The modification appends a single Lys-40 residue C-terminally to the native PPPS tail, conjugating a γGlu spacer and C20 fatty diacid (eicosanedioic acid) to its ε-amine. Key design decisions:

• C20 chain length: Matches the native Lys-20 lipidation chemistry (also C20 diacid), minimizing synthetic novelty and maintaining physicochemical consistency. This contrasts with Retatrutide Fold №45, which used a C18 chain on the same conceptual scaffold — the C20 choice here is deliberate for Tirzepatide-specific synthetic integration.
• γGlu (single) spacer at Lys-40: The native Lys-20 uses a γGlu-γGlu (double) spacer; using a single γGlu at Lys-40 provides spatial differentiation between the two anchors, potentially reducing the risk of intramolecular steric conflict between the two lipid chains while still providing sufficient linker flexibility for albumin engagement.
• C-terminal positioning: The PPPS tail is the most chemically permissive region of the peptide. Fold №63 (Tirzepatide, i,i+4 lactam in the central helix) was DISCARDED (pLDDT 0.69) due to tool-limit failure on helix-perturbing chemistry — the C-terminal strategy deliberately avoids that failure mode.
• Retatrutide precedent: Fold №45 applied the same conceptual framework to Retatrutide (C-terminal Lys-40, γGlu-C18 diacid) and returned PROMISING (pLDDT 0.70), validating that the PPPS tail of incretin peptides tolerates lipid conjugation without disrupting the helical pharmacophore in structural prediction.

Predicted Properties — Where the Signal Is Moderate

Where the PROMISING signal is genuine:

• pLDDT of 0.716 is consistent with the helical pharmacophore being well-folded in the model, matching the predictions for Retatrutide Fold №45 (0.70) and Tirzepatide Folds №23 and №31 (both 0.71).
• Low aggregation propensity (0.222) is encouraging given the dual-lipidation burden — two C20 fatty acid chains on the same 40-residue peptide create real self-association risk that the heuristic does not flag.
• The heuristic long half-life estimate is consistent with — though not quantitatively predictive of — the bivalent albumin hypothesis.

Where the signal is weak:

• The ipTM of 0.096 means that GLP-1R TMD engagement, including the canonical Tyr-1 insertion, cannot be confirmed from this prediction. This is a known limitation of Boltz-2 for class B GPCR peptide-receptor complexes and reflects tool resolution rather than a predicted binding failure — but it leaves the receptor pharmacology hypothesis unvalidated.
• Lipid conjugates are not explicitly modeled; both Lys-20 and Lys-40 sidechain positions appear solvent-exposed, which is consistent with the hypothesis, but the albumin-binding geometry of the dual-anchor configuration cannot be assessed in silico.
• No Chai-1 ensemble data are available; single-model predictions carry higher uncertainty.

What Would Strengthen This Signal

Additional computational work:

• Ensemble prediction (Chai-1 multi-seed): Run 5+ seeds to assess structural convergence of the helical core and variability of the C-terminal tail disposition. Retatrutide Fold №45 would benefit from the same treatment as a direct comparator.
• Explicit albumin co-fold: Model the dual-lipidated Tirzepatide in complex with human serum albumin (HSA, PDB 1AO6 fatty acid binding sites) using AlphaFold-Multimer or Boltz-2 multimer mode to assess whether both anchors simultaneously engage albumin or compete stereoelectronically.
• Free energy perturbation (FEP) or MM-GBSA: Estimate relative albumin-binding affinity of single-anchor (native) vs. dual-anchor analog — this is the most direct computational test of the bivalent PK hypothesis.
• Molecular dynamics simulation: Assess conformational stability of the PPPS-K(γGlu-C20) tail and its impact on receptor-proximal residues over nanosecond timescales.

Wet-lab validation experiments:

• SPR or ITC against human serum albumin: Measure Kd and off-rate for native Tirzepatide vs. Fold №73 analog — direct test of whether dual lipidation increases albumin affinity.
• GLP-1R and GIPR cAMP assay (HEK293 or CHO overexpression): Quantify EC50 for both receptors; critical to confirm that the Lys-40 C-terminal lipid does not perturb GIPR C-terminal engagement.
• In vitro DPP-4 stability assay: Confirm that the C-terminal modification does not indirectly alter proteolytic susceptibility at the N-terminal scissile bond.
• Pharmacokinetic study in rodents or NHP: The definitive test — plasma half-life measurement for dual-anchor vs. native Tirzepatide is the only way to confirm the bivalent albumin hypothesis experimentally.
• SEC or DLS aggregation assay: Assess self-association of the dual-lipidated analog, particularly at concentrations relevant to subcutaneous injection formulation.

Key comparator experiment: Synthesize the Retatrutide Fold №45 analog (γGlu-C18 diacid at Lys-40) and the Tirzepatide Fold №73 analog (γGlu-C20 diacid at Lys-40) in parallel and run head-to-head albumin-binding and receptor-activation assays — this would validate the cross-peptide PPPS-tail lipidation strategy as a generalizable platform.

Lab Context & Cross-Fold Connections

This fold is part of a developing Tirzepatide medicinal chemistry series at Alembic Labs:

• Fold №45 (Retatrutide) — the direct structural precedent: C-terminal Lys-40 γGlu-C18 diacid on the homologous PPPS tail, PROMISING (pLDDT 0.70). Fold №73 advances this to Tirzepatide with a C20 chain for chemistry consistency, returning an equivalent PROMISING verdict with slightly higher pLDDT (0.716).
• Fold №63 (Tirzepatide, DISCARDED) — the cautionary precedent: i,i+4 lactam staple in the central helix was DISCARDED (pLDDT 0.69) due to AlphaFold-blind crosslink chemistry. Fold №73 explicitly avoids helix-perturbing modifications.
• Folds №23 and №31 (Tirzepatide, both PROMISING, pLDDT 0.71) — non-canonical substitutions (αMe-Cys-24, Aib-2) in the helical core returned consistent moderate-confidence predictions, establishing the baseline pLDDT range for this peptide scaffold. Fold №73 falls within this established range, suggesting the C-terminal modification does not degrade the global structural prediction.
• Fold №54 (Retatrutide, REFINED) — Lys-17/Asp-21 lactam in the central helix returned REFINED (pLDDT 0.71), confirming that the incretin helix is a productive target for conformational pre-organization; however, this strategy targets potency rather than PK, and Fold №73 occupies the complementary pharmacokinetic space.

The emerging Alembic Labs narrative for Tirzepatide modifications is: the helical core is structurally robust to non-canonical substitutions at positions 2 and 24 (Folds №23, 31), the helix is not tolerant of crosslinking chemistry visible to AlphaFold (Fold №63), and the PPPS C-terminus is a permissive site for lipid conjugation (Fold №73, consistent with Fold №45 for Retatrutide). The next logical step in this series is a triple-combination analog combining Aib-2 (DPP-4 resistance, Fold №31) with C-terminal lipidation (Fold №73) to address both enzymatic stability and PK extension simultaneously.