Start with why this should not work

Glucagon raises blood glucose. That is its job description. It is the hormone your pancreas releases when your sugar drops too low, and it tells your liver to dump stored glucose into your bloodstream.

Now put that in a drug for type 2 diabetes.

If you proposed this in 2005, you would have been walked out of the room. The entire logic of diabetes pharmacology ran the other way. Block glucagon. Inhibit the enzyme. Suppress the pathway. Pharmacology in that era was subtractive, and for good reason: it is easier to prove you turned something off than to prove you turned the right things on in the right ratio.

Somebody ran the experiment anyway. The subtractive version was tested. LY2409021, a glucagon receptor antagonist, was given to patients with type 2 diabetes.

It increased their liver fat.

Hold that result. We come back to it in Part 4, and it is the single most satisfying reversal in this entire body of work.

What glucagon actually does when you stop looking at it sideways

The blood sugar effect is real, and it is one item on a longer list. Look at the rest of it.

Glucagon activates hormone-sensitive lipase, which means it drives lipolysis. It inhibits lipogenesis. It increases ketogenesis. It promotes satiety. It raises energy expenditure. It stimulates insulin secretion.

Read that list without the label attached and you have described something close to an ideal obesity target. Most glucagon receptors sit on hepatocytes, which is why the liver effects are as pronounced as they are, but they also appear in the kidneys, the gastrointestinal tract, the pancreas and the central nervous system.

The problem was never that glucagon is bad. The problem was the one item that made the other six unusable on their own.

So the design answer is not to remove glucagon. It is to bring a chaperone.

The three-receptor arrangement

Retatrutide hits three receptors at once: GLP-1, GIP, and glucagon.

The first two are the incretins. They raise insulin secretion in response to food, delay glucagon release, and slow gastric emptying. GLP-1 receptors are dense in pancreatic beta cells, the gut and the central nervous system, which is where the appetite suppression comes from. GIP receptors sit in beta cells and adipose tissue, and appear to influence how fat cells handle lipid.

People with type 2 diabetes have a blunted incretin response, which is a large part of why their glucose runs high after meals.

Now the arrangement becomes readable. The incretin arms handle glucose. That frees the glucagon arm to do the six things on the list that nobody could previously use.

It is not three drugs stapled together. It is one arm covering the liability of another so the other can be deployed at all.

The proof that all three are actually firing

Claiming triple agonism is easy. The mouse work in this review shows the receipts.

Researchers took mice lacking the GLP-1 receptor entirely and additionally blocked their GIP receptors with a monoclonal antagonist. Two of three targets removed. Then they gave retatrutide.

Plasma glucose still rose about an hour after dosing. Something was still working, and by elimination it had to be the glucagon receptor.

Then they pretreated the same mouse line with a glucagon receptor antibody. The glucose rise disappeared.

That is a clean closed loop. Remove two receptors, effect persists. Remove the third, effect stops. All three are engaged, and the glucagon arm is doing real work.

Separately, glucose tolerance improved in GLP-1 receptor null mice, GIP receptor null mice and normal mice alike, which tells you the glucose benefit is not riding on any single pathway.

The engineering, which is better than the pharmacology

Here is the detail most coverage skips, and it is the most impressive thing in the paper.

Retatrutide is a single peptide, 39 amino acids long. It is built on a GIP backbone, not a GLP-1 backbone, which is already a choice most people would not have made.

Three positions carry non-standard amino acids, and each one has a specific job:

  • Aib at position 2 protects the peptide from being chopped up by DPP-4, the enzyme that destroys natural incretins within minutes.
  • Aib at position 20 tunes GIP activity and the pharmacokinetic profile.
  • MeL at position 13 tunes GIP and glucagon activity together.

Then a 20-carbon fatty acid is attached via a linker at the lysine at position 17. That chain binds albumin, the most abundant protein in your blood, and albumin carries the drug around instead of letting your kidneys flush it. Half-life goes to roughly six days. That is what makes a once weekly injection possible instead of a daily one.

Cryo-electron microscopy has since shown how it docks: a continuous helix driving into the core of each receptor's transmembrane domain, with the tail end interacting with the extracellular portion, adopting a different conformation for each of the three receptors.

One molecule. Three differently shaped locks. A different fit for each.

The detail that reveals the whole philosophy

Retatrutide is 2.9 times less potent than human glucagon at the glucagon receptor.

Read that as a limitation and you have misread it. The goal was never maximum glucagon signalling. Full-strength glucagon agonism would have blown past what the incretin arms could offset, and you would be back to a drug that raises blood sugar in diabetics.

They dialled it down until it fit.

The review is candid that the precise contribution of glucagon receptor stimulation to the weight loss remains unclear, and that this is worth resolving. That is a fair thing to admit while the field is still arguing about the ratio.

The reframe

The last generation of metabolic drugs asked what to block.

This one asked what to combine, and at what strength, so that a hormone everyone had written off could finally be used.

That is a different question, and it produced a different class of answer. In Part 3 we get to the numbers it produced, and to the reason those numbers forced a field to move its own goalposts.

This series covers a peer-reviewed narrative review of published trial data. Retatrutide has not been approved by the FDA. Nothing here is medical advice.