The body is very good at destroying peptides. Native GLP-1, the hormone that the whole incretin class imitates, survives about two minutes in circulation: an enzyme called DPP-4 clips two amino acids off one end, and the kidneys filter out what is left. A hormone that is released after every meal does not need to last longer than that.
A drug does. The engineering problem of the last twenty years has been how to make a GLP-1-like peptide survive long enough to be injected once a week rather than every few hours. The answer that won is a fatty acid.
The albumin trick
Albumin is the most abundant protein in blood plasma, roughly 40 grams per litre. It is large enough that the kidney cannot filter it, and it circulates for about three weeks. Its day job is to carry fatty acids, which it binds reversibly in several pockets on its surface.
If a peptide is given a fatty-acid tail, it will bind to those same pockets. While bound, it is effectively part of a protein too large to filter and too well-folded for DPP-4 to reach. The binding is reversible, so a small fraction of the peptide is always free, and that fraction is what reaches the receptor. The albumin acts as a slow-release reservoir carried in the bloodstream.
Three molecules, one idea
Semaglutide carries a C18 fatty di-acid attached through a short spacer to a lysine at position 26, alongside two amino-acid substitutions that protect it from DPP-4. The result is a half-life of roughly a week1.
Tirzepatide uses a C20 di-acid at position 20 of a 39-residue peptide and reaches a half-life of about five days2.
Retatrutide also carries a C20 chain and sits at about six days3.
Hours; native GLP-1 shown for scale
Lau 2015 [1], Coskun 2018 [2], Coskun 2022 [3]. Chart by TrueLab Care.
The chain length, the spacer, and the position on the peptide are all tuned. Too much albumin affinity and too little free peptide reaches the receptor; too little and the kidney clears it. The design papers for each compound describe the optimisation as a balance rather than a maximisation123.
What a six-day half-life implies
Three consequences follow that are easy to miss.
Accumulation. A drug with a six-day half-life, given weekly, does not reach a stable blood level after one dose. It builds up over four to five half-lives, roughly a month, before the amount cleared each week equals the amount added. Any effect measured in the first fortnight is measured at a concentration still climbing.
Persistence. After the last dose, the compound takes the same month to clear. Side effects do not stop when dosing stops; they decay.
Error propagation. In an experimental system, a mistake in concentration persists across the entire accumulation window rather than washing out between exposures. Reconstitution arithmetic matters more for a weekly peptide than for an hourly one.
Why this matters for a research compound
The fatty-acid chain is also what makes these molecules physically different from a simple peptide. They are more hydrophobic, more prone to sticking to plastic surfaces, and more sensitive to the choice of solvent. A certificate of analysis that confirms the correct mass by spectrometry is confirming the chain is present; a chromatogram that shows a single peak is confirming it has not been partially cleaved. Both are worth checking on any vial labelled with one of these names.