What a GLP-1 Actually Does Inside Your Body

Every conversation I have about these medications starts in the same place. Someone sits down, looks at the floor, and says some version of: I feel like I’m cheating.

So let me start where I always start. You are not taking a foreign chemical that overrides your body. You are taking a long-lasting copy of a hormone your own intestine has been making after every meal of your life. The interesting question is not whether that is cheating. The interesting question is why a copy works so much better than the original.

The hormone you already make

Tucked into the lining of your lower small intestine and colon are cells called L-cells. When food arrives — particularly fat and carbohydrate — they release glucagon-like peptide-1 into your bloodstream. It is one of two major incretin hormones, the gut’s way of telling the rest of the body that nutrients are on the way.

Native GLP-1 does four things worth knowing:

• It tells the pancreas to release insulin, but only when blood glucose is actually elevated.
• It tells the pancreas to stop releasing glucagon, the hormone that pushes stored sugar back into circulation.
• It slows how fast the stomach empties into the intestine.
• It acts on the brain — the hypothalamus and the brainstem — to reduce hunger and bring on fullness.

That fourth one is the one everybody cares about, and it is also the one that gets described most badly in the press.

Here is the catch with the natural version: it lasts about two minutes. An enzyme called DPP-4 clips it apart almost immediately. Your body designed GLP-1 as a text message, not a broadcast.

What the drug changes

Semaglutide is that same peptide with two structural edits. One is a single amino-acid substitution at the site DPP-4 attacks, so the enzyme can no longer clip it. The other is a fatty acid chain bolted onto the molecule, which lets it bind reversibly to albumin — the taxi protein floating through your bloodstream — and hitch a ride instead of being filtered out by your kidneys.

Two edits. Two minutes becomes roughly one week.

That is the whole trick. Not a new mechanism, not a stimulant, not a thyroid drug. A hormone you already produce, held at a steady level instead of a two-minute flicker.

The drug did not give me a new biology. It gave my existing biology a longer sentence to finish.

Why hunger goes quiet

GLP-1 receptors sit in several places in the brain that regulate eating. The arcuate nucleus of the hypothalamus, which integrates hunger and satiety signals. The area postrema and nucleus tractus solitarius in the brainstem, which sit outside the tight blood-brain barrier and can sample the bloodstream directly. There is also an indirect route: GLP-1 receptors on the vagus nerve, sending the message upward from the gut itself.

The subjective result is what my patients describe, almost word for word, as the food noise stopped. It is not that food becomes disgusting. It is that the background process — the one that was running a low-level negotiation about the pantry while you were trying to answer email — goes quiet. Studies of food preference and reward on these drugs consistently show reduced cravings and a shift away from high-fat, highly palatable foods, not just smaller portions.

Layered on top of that is the slower stomach emptying, which means a given meal keeps you physically full for longer. Early in treatment this effect is strong; it attenuates somewhat over months.

Both of those roads lead to the same destination: you eat less, without spending the day fighting yourself. In the phase 3 trials, this added up to a mean loss of about 15% of body weight over 68 weeks on semaglutide 2.4 mg, against roughly 2.4% on placebo — with both arms receiving lifestyle counselling.

The glucose part, and why hypoglycemia is rare

This is the piece I wish more people understood, because it explains a lot of the safety profile.

GLP-1’s effect on insulin is glucose-dependent. The signal to the beta cell is conditional: release insulin, if glucose is high. When glucose is normal or low, the signal quiets down. That is fundamentally different from injected insulin or from a sulfonylurea, both of which push glucose down whether or not it needs pushing.

It is why a GLP-1 taken on its own almost never drops someone into hypoglycemia. It is also why, if you are on insulin or a sulfonylurea alongside it, those doses usually need to come down — the hypoglycemia risk in that situation comes from the other drug, unmasked by eating less.

Where tirzepatide differs

There is a second incretin hormone: glucose-dependent insulinotropic polypeptide, or GIP. For years GIP was the neglected sibling — its role in obesity was genuinely confusing, and the field argued about whether agonizing or blocking it would help.

Tirzepatide is a single molecule that activates both the GIP and GLP-1 receptors. Whatever GIP is contributing, it is contributing something: in its phase 3 obesity trial, tirzepatide produced mean weight reductions of roughly 15% to 21% at 72 weeks depending on dose, against about 3% on placebo.

I want to be careful here, because this is where marketing gets ahead of science. We do not fully understand why dual agonism outperforms. Proposed explanations include additional central effects on appetite, better tolerability at higher effective doses, and direct effects of GIP on fat tissue. The honest summary is: the outcome is well established, the mechanism is still being argued about in good faith.

What this mechanism does not do

Understanding how a drug works also tells you what it cannot do, and this is where I spend most of my counselling time.

It does not choose what you lose. GLP-1 signalling reduces energy intake. Your body decides how to make up that deficit, and left alone it will take some of it from muscle. That is a nutrition and training problem, not a drug problem, and it is the single thing I am most militant about.

It does not fix the environment you eat in. The drug lowers the volume of the signal. It does not restock your kitchen, renegotiate your work hours, or teach you what a protein-forward breakfast looks like.

It does not persist after you stop. Remove the exogenous hormone and you return to the two-minute version. Appetite comes back, and so, for most people, does weight. That is a design feature of the mechanism, not a personal failure — but it deserves a plan before you ever start.

Where the mechanism shows up in daily life

Three consequences of delayed gastric emptying are worth connecting to the pharmacology above, because each is usually treated as a separate mystery: oral medication is absorbed differently, which matters for a short list of drugs covered in GLP-1s and your other medications; your stomach may not be empty after a standard pre-operative fast, which is the subject of GLP-1s, surgery and procedures; and alcohol reaches the small intestine with far less food in the way, which is why a smaller amount goes further.

I wrote about first encountering all of this in why I started.

The reframe I’d like you to take with you

There is a version of this conversation where the drug is a shortcut and the patient is a cheater, and it is both cruel and biologically illiterate.

Here is a better one. Appetite is not a character trait. It is a hormonal signal, generated by an organ, acting on a receptor, degraded by an enzyme. Some people’s version of that signal is louder, or shorter-lived, or blunted in ways we can now measure. We have built a molecule that adjusts it.

That is what every effective drug in internal medicine does. We are not shocked that a beta blocker lowers heart rate rather than asking the patient to want a slower pulse. This is the same category of intervention. It just happens to sit on top of the one bodily process our culture has decided is a referendum on your worth.

It isn’t. It’s an incretin.