The Incretin Effect, Explained

Give someone 50 grams of glucose to drink. On another day, infuse glucose into a vein and adjust the drip until their blood glucose traces the same curve, point for point. Identical glucose exposure, but the oral route releases far more insulin. That surplus is the incretin effect: the share of your insulin response that exists only because the glucose passed through your gut. Two laboratories, in Denver and London, demonstrated it independently in 1964, and it is the physiology the GLP-1 drug class was built to exploit.

How big is the gap?

Large. In the study that defined the field (Nauck and colleagues, Diabetologia, 1986), 8 healthy controls and 14 people with type 2 diabetes each received a 50 g oral glucose load and, on a separate day, an “isoglycaemic” intravenous infusion matched to the oral glucose curve. In the healthy controls, incretin factors accounted for 72.8 ± 6.9% of the total insulin response; measured by C-peptide, which avoids the confounding effect of hepatic insulin extraction, the figure was 58.4 ± 7.6%.

The effect also scales with how much you eat. A 2013 PLoS One study by Mari and colleagues ran the same isoglycaemic protocol at three glucose doses in 8 people with normal glucose tolerance and 8 with type 2 diabetes. In the healthy group, oral glucose produced 1.3-, 1.7- and 2.2-fold the insulin secretion of the matched intravenous infusion at 25 g, 75 g and 125 g respectively. Bigger load, louder gut signal.

Most of the insulin your pancreas releases after a carbohydrate meal is not a response to blood glucose. It is a response to hormones your intestine released when the food arrived.

The two hormones doing the work

The signals come from two gut hormones: GIP (glucose-dependent insulinotropic polypeptide), secreted by K cells in the upper small intestine, and GLP-1 (glucagon-like peptide-1), secreted by L cells further down. In a 2020 review in Biology, Boer and Holst summarise one analysis of a 50 g oral load that attributed 44% of the insulin response to GIP, 22% to GLP-1, and 33% to glucose alone; at 75 g, endogenous GIP and GLP-1 together account for almost 70% of the response, with GIP the larger contributor.

That inverts the popular picture: in healthy physiology, GIP, not GLP-1, does more of the insulin-releasing work. GLP-1 earned its therapeutic prominence for other reasons, since it also suppresses glucagon, slows gastric emptying, and acts in the brain to reduce appetite. Both hormones act in a glucose-dependent way, amplifying insulin release when glucose is elevated and mostly standing down when it is not; Boer and Holst note this is why incretin-based treatments carry a very low intrinsic risk of hypoglycaemia on their own.

Why the natural hormone makes a terrible drug

• DPP-4, an enzyme on the surface of endothelial cells, cleaves the first two amino acids off GLP-1’s N-terminus, producing the inactive fragment GLP-1(9–36)NH₂.
• Degradation begins before the hormone has left the gut wall; less than 10% of secreted GLP-1 reaches its target organs intact.
• Hence the two drug strategies: block the enzyme (DPP-4 inhibitors), or redesign the peptide so the enzyme cannot cut it, covered in why GLP-1 drugs last a week.

What changes in type 2 diabetes

The amplification is substantially weaker. In the same 1986 Nauck experiment, the incretin contribution to the insulin response was 36.0 ± 8.8% in the diabetic group (roughly half that of controls), and the C-peptide figure collapsed to 7.6 ± 14.5%.

Measure	Normal glucose tolerance	Type 2 diabetes
Incretin share of insulin response, 50 g oral glucose (Nauck 1986)	72.8 ± 6.9%	36.0 ± 8.8%
Incretin share of C-peptide response (Nauck 1986)	58.4 ± 7.6%	7.6 ± 14.5%
Insulin secretion vs. matched IV infusion at 25 / 75 / 125 g (Mari 2013)	1.3× / 1.7× / 2.2×	1.0× / 1.1× / 1.3×

Two limits on interpretation. These experiments show the incretin effect is reduced in type 2 diabetes, not that the reduction causes it; Boer and Holst treat it as an important pathogenic factor, but the design cannot separate cause from consequence. And the samples are small (22 people in 1986, 16 in 2013), normal for an invasive two-day metabolic protocol and still a reason to read the decimal places loosely.

The takeaway

The incretin effect is one of the better-established findings in metabolic physiology: measured since 1964, quantified with a clean within-person control, and reproduced across glucose loads. In healthy adults, gut hormones account for roughly 60–70% of the insulin response to oral glucose; in type 2 diabetes, that contribution is roughly halved. The drug class followed directly: if the natural signal is degraded within minutes and blunted in disease, build a version that resists degradation. For how the individual hormones differ, see how GLP-1, GIP and glucagon differ.

Sources

• Reduced incretin effect in type 2 (non-insulin-dependent) diabetes — Nauck et al., Diabetologia 1986
• Incretin Hormones and Type 2 Diabetes — Mechanistic Insights and Therapeutic Approaches — Boer & Holst, Biology 2020
• Mechanisms of the Incretin Effect in Subjects with Normal Glucose Tolerance and Patients with Type 2 Diabetes — Mari et al., PLoS One 2013
• The Origin and Understanding of the Incretin Concept — Rehfeld, Frontiers in Endocrinology 2018