Chapter 7 — Key Takeaways

The Incretin System

The core claims

The incretin effect: a given blood glucose level produces substantially more insulin when the glucose arrived orally than when it was infused. The gut informs the pancreas ahead of the glucose. The system is anticipatory, not reactive.

GLP-1 comes from intestinal L cells, cut from proglucagon (the same precursor that yields glucagon in the pancreas), released in response to nutrients in the gut — not in response to circulating GLP-1 levels. That is why agonists don't suppress endogenous production.

GLP-1 does four things at once:

  ① PANCREAS beta cell   → glucose-dependent insulin secretion
  ② PANCREAS alpha cell  → glucagon suppression
  ③ STOMACH              → slowed gastric emptying   ← this is where the nausea comes from
  ④ BRAIN                → reduced appetite

Glucose-dependence is the safety property. GLP-1 amplifies insulin release that glucose is already driving; it does not force it. That is why these drugs alone rarely cause hypoglycemia — and why they can contribute to it alongside insulin or a sulfonylurea.

GIP is the more complicated story. Produced by K cells, probably the larger contributor in healthy people, blunted in type 2 diabetes, apparently promotes fat storage — and adding GIP receptor agonism to GLP-1 agonism helps anyway. Four explanations are on the table, none established, and both GIP agonists and antagonists are being pursued, which is an honest admission of confusion.

DPP-4 made this undruggable for two decades, clipping two residues from GLP-1's activating N-terminus. Half-life: 1–2 minutes. Two strategies followed — inhibit the enzyme (modest effects, since you can only preserve physiological levels) or build an agonist it cannot cut (the drugs in Chapter 8).

The gut-brain routes may differ between physiology and drug. Endogenous GLP-1 likely acts largely locally, via vagal afferents, because so little survives to circulate. An injected agonist floods the system and reaches brain regions with a leaky blood-brain barrier directly. These may not be the same signal — a candidate explanation for why the drug effect so exceeds anything physiology produces.

The rule to carry forward

A peptide with four actions has four mechanistic stories, and which one matters depends entirely on which claim you are evaluating.

For the weight claim: actions ③ and ④. For the glucose claim: ① and ②. For the cardiovascular claim: not established which, if any. A source that lists all four and then asserts a benefit has not told you which action is supposed to produce it.

Evidence ratings issued in this chapter

Claim Rating Why What would change it
Native GLP-1, administered as such, is a viable treatment for type 2 diabetes or obesity ❌ On delivery grounds, not efficacy grounds. IV infusion in humans demonstrably works; a 1–2 minute half-life cannot be delivered outside a hospital A delivery system achieving sustained therapeutic levels of the native peptide — which analog engineering solved by not using the native peptide

This rating is in the book to demonstrate that ❌ can attach for reasons having nothing to do with whether the biology is real. "The mechanism doesn't work" and "this cannot be delivered" are different failures with different remedies.

What this system does NOT explain

  • Why obesity happens. A drug acting on a system is not evidence the system caused the disease.
  • The magnitude of the drug effect. Endogenous GLP-1 does not produce 15% weight loss.
  • Why weight returns on discontinuation. Predicted in principle by override-vs-replace (Ch 2), not explained in detail.
  • The non-metabolic benefits. Cardiovascular, renal, hepatic — mechanism unresolved (Ch 10).
  • Who should take these drugs. A clinical judgment, not a physiological question.

Numbers worth remembering

Quantity Value
Native GLP-1 30 amino acids, half-life 1–2 minutes
GIP 42 amino acids
Share of meal insulin response attributable to incretins roughly half or more, varying by load
Incretin effect in type 2 diabetes substantially reduced

Key terms

incretin effect · incretin · enteroendocrine cell · L cell · K cell · proglucagon · GIP · DPP-4 · glucose-dependence · gastric emptying · gut-brain axis · vagus nerve · area postrema · nucleus tractus solitarius · arcuate nucleus · circumventricular organ · ileal brake · food noise

What you can now evaluate

  • ✅ why these drugs cause nausea, and why it is not a random side effect
  • ✅ why they don't cause the hypoglycemia insulin does — and when they can
  • ✅ why they don't suppress your own GLP-1
  • ✅ why DPP-4 inhibitors are weaker than receptor agonists
  • ✅ what "restores natural satiety signaling" gets wrong
  • ✅ which of a multi-action peptide's actions is load-bearing for a given claim
  • ❌ not yet: what the trials actually showed. Chapter 8.

The one-sentence version

Your gut has been telling your pancreas and your brain what you ate for as long as you have been eating — and the drug that exploits this is not simply more of that signal, which may be exactly why it works so much better.

Next: Chapter 8 — semaglutide, from three molecular modifications through four trial programs to the finding that reframed the entire drug class.