Fold №70 — Ipamorelin Lys-5 → Arg: Probing GHSR-1a Affinity via Guanidinium Salt-Bridge Engagement

Verdict: REFINED | pLDDT 0.79 | ipTM 0.73 | Single-run Boltz-2 prediction

Mechanism of Action

Ipamorelin is a synthetic pentapeptide GHSR-1a agonist that mimics the GH-releasing action of endogenous ghrelin without stimulating ACTH or cortisol secretion. Its minimal scaffold (Aib-His-DBNal-DPhe-Lys-NH2) encodes a precise binding mode at the GHSR-1a orthosteric pocket: the N-terminal Aib nucleates a type-II β-turn; His-2 and DBNal-3 form a hydrophobic-aromatic cluster; DPhe-4 engages the aromatic subpocket; and Lys-5 contributes a C-terminal basic charge that interacts with conserved acidic residues (notably Glu124 in TM3) on the intracellular face of the extracellular-facing pocket. GHSR-1a activation stimulates somatotroph GH release, downstream IGF-1 elevation, and is associated with improvements in body composition, recovery, and metabolic signalling relevant to performance contexts.

Arginine's guanidinium group (pKa ~12.5) is planar, resonance-delocalized, and geometrically capable of bidentate H-bond and salt-bridge contacts that a primary lysine ε-amine (pKa ~10.5) cannot form. Replacing Lys-5 with Arg is hypothesized to tighten the C-terminal anchor at TM3/ECL2 acidic residues, translating to a more constrained and energetically favourable docked pose without perturbing the established DBNal-DPhe hydrophobic pharmacophore.

Performance Applications

As a GHSR-1a agonist, the Arg-5 variant of Ipamorelin shares the parent compound's performance-relevant activity profile: pulsatile GH secretion stimulation, IGF-1 elevation, and downstream anabolic and lipolytic signalling. If the predicted affinity improvement is validated experimentally, Arg-5 Ipamorelin could offer:

• Enhanced receptor engagement at lower molar doses, potentially reducing the peptide burden required for a GH pulse equivalent to native Ipamorelin.
• A tighter, more selective binding mode that leverages the same selectivity profile (no ACTH/cortisol co-stimulation) but with improved potency at GHSR-1a.
• A base scaffold for combinatorial optimization — pairing improved affinity with the pharmacokinetic gains explored in Folds №48 (palmitoylation) and №33 (lactam cyclization) in future design iterations.

These are predicted properties of an in silico model. No in vivo or clinical claims are made.

Modification Rationale

The Lys-5 → Arg substitution was selected as the first Ipamorelin fold to directly interrogate affinity-driving electrostatics at the C-terminus — a dimension the prior Ipamorelin program has not explored. Previous folds addressed:

• Fold №4: N-Me-Aib at position 1 → backbone methylation for DPP-IV/aminopeptidase resistance (REFINED, pLDDT 0.80)
• Fold №33: Lys5–Asp6 side-chain lactam → C-terminal conformational lock (REFINED, pLDDT 0.73)
• Fold №48: γGlu-C16 palmitoyl on Lys-5 ε-amine → albumin-binding half-life extension (REFINED, pLDDT 0.78)

In each prior fold, Lys-5 was treated as a structural or pharmacokinetic handle — its intrinsic pharmacophoric contribution to GHSR-1a binding was never independently tested. Fold №70 isolates that question: does swapping the ε-amine for a guanidinium strengthen receptor engagement? The rationale is supported by SAR precedent in related GHS scaffolds (macimorelin carries an Arg-derived guanidinium motif; ghrelin analogs with Arg at equivalent positions maintain or improve potency) and by the known preference of GHSR-1a acidic pocket residues for geometrically rich bidentate contacts.

Predicted Properties

⚠️ Heuristic biophysical estimates are sequence-derived approximations, not experimental measurements. No quantitative ΔΔG value was returned by Boltz-2; the REFINED verdict reflects structural confidence, not a measured affinity delta.

The predicted docked pose preserves the DBNal-DPhe hydrophobic engagement seen consistently across all REFINED Ipamorelin folds, confirming the core pharmacophore is not disrupted by the C-terminal substitution.

Suggested Next Steps

Computational (in silico):

1. Ensembled Boltz-2 run + Chai-1 cross-validation — Chai-1 agreement data were unavailable for this fold; running both predictors in ensemble mode would strengthen confidence and flag any pose divergence.
2. FEP/MM-GBSA rescoring — Apply free-energy perturbation or molecular mechanics generalized Born surface area calculations on the Boltz-2 pose to obtain a quantitative ΔΔG estimate for Lys → Arg at position 5.
3. Combinatorial fold: Arg-5 + γGlu-Palm (Fold №48 logic) — Test whether a palmitoylated Arg-5 variant (replacing the ε-amine lipidation handle with a backbone-compatible linker at a different site, or N-terminal acylation) can combine improved affinity with half-life extension.
4. Combinatorial fold: Arg-5 + N-Me-Aib-1 (Fold №4 logic) — Pair the predicted affinity gain at position 5 with the established DPP-IV resistance at position 1 in a single variant.

Wet-lab validation:

1. Solid-phase peptide synthesis of Aib-His-DBNal-DPhe-Arg-NH2 using standard Fmoc chemistry; Arg is straightforwardly incorporated.
2. Competitive radioligand binding assay (³H-ghrelin or ¹²⁵I-ghrelin displacement at GHSR-1a-expressing HEK293 cells) to obtain Ki and compare directly with native Ipamorelin.
3. GH stimulation assay (pituitary cell culture or in vivo rat model) to confirm functional agonism and potency ratio vs. parent.
4. Plasma stability assay — compare half-life of Arg-5 variant vs. Lys-5 parent under standard plasma incubation conditions to confirm the substitution does not introduce unexpected proteolytic vulnerability.