Albumin Binding and Half-Life Extension

The problem being solved

A peptide of incretin size is small enough to pass the glomerulus and be excreted, and exposed enough to be cleaved by circulating proteases. Blocking DPP-4 with a substitution at position 8 handles one enzyme; it does nothing about renal clearance. Something has to make the molecule effectively larger without changing what the receptor sees.

Why albumin is the obvious carrier

Serum albumin is the most abundant plasma protein, has a half-life of about 19 days, and evolved to bind fatty acids and transport them through the circulation. A peptide carrying a fatty acid chain therefore has a ready-made, high-capacity, long-lived carrier already present in the blood. Nothing has to be co-administered.

Why the binding must be reversible

This is the part that makes the design work. A peptide permanently bound to a 66 kDa protein could not engage its receptor; it would be protected and useless. Because binding is an equilibrium, a small free fraction exists at any moment and is available to signal, while the bound pool acts as a slow-release depot that continuously replenishes it. The effective half-life reflects the whole pool, not the free fraction.

How chain length tunes the result

Longer and more lipophilic chains bind albumin more tightly, shifting the equilibrium toward the bound state and extending half-life. Liraglutide's C16 palmitic acid gives roughly 13 hours, enough for daily dosing. Semaglutide's C18 diacid, attached through a γGlu-2xOEG linker, gives about a week. Retatrutide uses a C20 diacid at Lys17 through the same linker chemistry. Same strategy, three intensities.

What the linker does

The γGlu-2xOEG spacer between peptide and fatty acid is not inert padding. It sets the distance and flexibility between the albumin-binding element and the receptor-binding peptide, so that albumin binding does not sterically block the peptide from reaching its receptor. Getting the linker wrong can compromise either binding event.

Why this recurs across compounds targeting different receptors

Semaglutide targets GLP-1R. Tirzepatide targets GIPR and GLP-1R. Retatrutide adds the glucagon receptor. Their receptor pharmacology differs completely, yet all three carry a fatty diacid on a lysine through a similar linker, because they share a clearance problem rather than a target. Recognising that explains the structural family resemblance across molecules that otherwise have little in common.

Quick reference