Physiology
Insulin and Glucagon: Blood Sugar Control
The antagonistic pancreatic hormones — beta-cell insulin lowers glucose, alpha-cell glucagon raises it — that hold blood sugar near 90 mg/dL
Insulin and glucagon are the two opposing hormones that keep blood glucose pinned inside a narrow fasting band of roughly 70 to 110 mg/dL, a control problem your pancreas solves minute by minute for a lifetime. Both are secreted from the islets of Langerhans: beta cells release insulin when glucose rises after a meal, pulling glucose into muscle and fat through GLUT4 transporters and storing it as glycogen (glycogenesis); alpha cells release glucagon when glucose falls between meals, driving the liver to break down glycogen (glycogenolysis) and manufacture new glucose (gluconeogenesis). Because each hormone opposes the other and each shuts off as glucose returns to its setpoint, the pair form a classic negative-feedback loop. Insulin was isolated by Frederick Banting and Charles Best in Toronto in 1921; the failure of this loop defines diabetes mellitus, which now affects roughly half a billion people worldwide.
- Fasting setpoint~70–110 mg/dL
- Glucose in whole bloodstream~4 g (one sugar cube)
- Insulin cellsBeta cells (~65% of islet)
- Glucagon cellsAlpha cells (~20% of islet)
- Insulin isolatedBanting & Best, 1921–22
- People with diabetes~500 million worldwide
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why blood sugar control matters
- Your brain runs almost entirely on glucose. The adult human brain burns roughly 120 grams of glucose a day — about 60% of the body's resting glucose use — yet cannot store fuel and is largely impermeable to fatty acids. A blood glucose crash below ~50 mg/dL causes confusion, seizures, and coma within minutes. Holding glucose steady is not a convenience; it is what keeps you conscious.
- The total supply is tiny. An average adult's entire bloodstream contains only about 4 grams of dissolved glucose — one sugar cube. A single sugary drink can deliver more than that in one gulp, which is why the insulin response has to be fast and precise: without it, one meal would spike glucose to toxic levels.
- Chronic high glucose damages every vessel. Sustained hyperglycemia glycates proteins (measured clinically as HbA1c, the glucose stuck to hemoglobin over ~3 months) and drives the long-term complications of diabetes: retinopathy, nephropathy, neuropathy, and accelerated atherosclerosis. Diabetes is a leading cause of blindness, kidney failure, and lower-limb amputation.
- The loop must work in both directions. Insulin-only control would be fatal — you would have no defense against fasting hypoglycemia during the night. Glucagon, adrenaline, cortisol, and growth hormone form the counter-regulatory team that pushes glucose back up, with glucagon and adrenaline acting within minutes.
- Insulin is one of medicine's great success stories. Before 1922, type 1 diabetes was a death sentence — children lived months. Injected insulin turned it into a manageable chronic condition overnight, one of the fastest bench-to-bedside translations in medical history.
- It anchors modern metabolic disease. Insulin resistance sits at the center of type 2 diabetes, metabolic syndrome, non-alcoholic fatty liver disease, and polycystic ovary syndrome. Understanding this one feedback loop is the entry point to the most common chronic diseases of the 21st century.
How the insulin–glucagon loop works
The controlled variable is blood glucose itself, and the sensors are the islet cells that sit directly in the bloodstream. Beta cells sense glucose metabolically, not by a classical receptor. Glucose enters the beta cell through GLUT2 (a high-capacity, insulin-independent transporter), is phosphorylated by glucokinase — the rate-limiting "glucose sensor" whose activity rises steeply across the physiological range — and is oxidized to raise the ATP/ADP ratio. Rising ATP closes ATP-sensitive potassium channels (KATP, the target of sulfonylurea drugs), the membrane depolarizes, voltage-gated calcium channels open, and the calcium influx triggers exocytosis of insulin-filled granules. Insulin is released biphasically: a fast first-phase spike of pre-docked granules, then a sustained second phase.
Once in the blood, insulin lowers glucose through the insulin receptor, a receptor tyrosine kinase. Ligand binding triggers autophosphorylation, recruitment of IRS-1/2, and activation of the PI3K–AKT pathway. AKT drives the signature effect: translocation of GLUT4 transporter vesicles to the surface of skeletal muscle and adipocytes, so those tissues suck glucose out of the blood (uptake can rise more than tenfold within minutes). AKT also inactivates GSK-3, releasing the brake on glycogen synthase so glucose is stored as glycogen (glycogenesis), and it suppresses the liver's glucose-making machinery by phosphorylating and excluding FOXO1, cutting transcription of PEPCK and glucose-6-phosphatase. Insulin is, in one word, the storage hormone.
When you stop eating and glucose drifts toward 70 mg/dL, beta cells fall silent and alpha cells release glucagon. Glucagon is a 29-residue peptide that acts almost entirely on the liver via the glucagon receptor, a Gs-coupled GPCR. Receptor binding activates adenylyl cyclase, raises cyclic AMP, and switches on protein kinase A (PKA). PKA phosphorylates phosphorylase kinase, which activates glycogen phosphorylase to chop glucose units off glycogen (glycogenolysis), while simultaneously phosphorylating and inactivating glycogen synthase — reciprocal control ensures storage and breakdown are never running at once. PKA also lowers fructose-2,6-bisphosphate, tipping the liver away from glycolysis and toward gluconeogenesis, building fresh glucose from lactate, glycerol, and glucogenic amino acids. Liver glycogen (about 100 g, roughly a day's supply) is spent first; during prolonged fasting, gluconeogenesis carries the load.
The elegance is in the feedback. Insulin secretion is gated by high glucose and self-limiting — as glucose falls, the ATP signal fades and insulin stops. Glucagon is gated by low glucose and is itself suppressed by insulin and by paracrine somatostatin from neighboring delta cells. The two arms are reciprocal at the level of the ratio: it is the insulin-to-glucagon ratio at the hepatocyte, not either hormone alone, that decides whether the liver stores glucose or releases it. This is why the pancreas is often called a "fuel gauge with two needles that move in opposite directions."
Insulin vs glucagon at a glance
| Feature | Insulin | Glucagon |
|---|---|---|
| Source cell | Beta cells (islet core, ~65%) | Alpha cells (islet mantle, ~20%) |
| Released when glucose is | High (after a meal) | Low (fasting, between meals) |
| Effect on blood glucose | Lowers it | Raises it |
| Structure | 51 aa, two chains (A + B), 3 disulfide bonds | 29 aa, single chain |
| Receptor type | Receptor tyrosine kinase (insulin receptor) | Gs-coupled GPCR → cAMP/PKA |
| Main target tissues | Muscle, adipose, liver | Liver (almost exclusively) |
| Signature transporter | Recruits GLUT4 to cell surface | (no direct transport role) |
| Metabolic programs turned ON | Glycogenesis, lipogenesis, glucose uptake | Glycogenolysis, gluconeogenesis, ketogenesis |
| Metabolic programs turned OFF | Gluconeogenesis, glycogenolysis, lipolysis | Glycogenesis, glycolysis |
| Called the ... hormone | Storage / "fed state" | Mobilization / "fasted state" |
Negative feedback: the glucose thermostat
Blood sugar control is the canonical example used to teach homeostasis, and it maps cleanly onto the parts of a control system. The setpoint is ~90 mg/dL. The sensors are the islet cells reading glucose directly. The effectors are the liver, muscle, and fat. And the error correction runs in both directions, which is what makes it robust.
| Control element | When glucose is too HIGH | When glucose is too LOW |
|---|---|---|
| Sensor fires | Beta cell detects rise → insulin | Alpha cell detects fall → glucagon |
| Liver | Stores glucose as glycogen; stops making glucose | Breaks down glycogen; makes new glucose |
| Muscle & fat | Take up glucose via GLUT4 | Spare glucose; burn fatty acids instead |
| Net glucose flux | Blood → tissues (glucose falls) | Liver → blood (glucose rises) |
| Loop closes when | Glucose drops to setpoint, insulin stops | Glucose rises to setpoint, glucagon stops |
| Backup team | — | Adrenaline, cortisol, growth hormone |
Common misconceptions
- "Insulin's only job is to lower blood sugar." Insulin is the master anabolic signal for the whole fed state: it drives lipogenesis and fat storage, promotes protein synthesis, and inhibits lipolysis and ketogenesis. Lowering glucose is the most visible action, but insulin resistance produces high triglycerides and fatty liver precisely because insulin does far more than move sugar.
- "Glucagon and glycogen are the same thing." They only sound alike. Glucagon is the hormone; glycogen is the branched storage polymer of glucose. Glucagon triggers glycogenolysis — the breakdown of glycogen. Keeping the three straight (glucagon, glycogen, glucose) is the single most common exam trap in this topic.
- "Type 2 diabetes means you don't make insulin." Early type 2 diabetics often make more insulin than normal — the problem is that muscle, fat, and liver respond poorly (insulin resistance), so the pancreas overcompensates. Beta-cell exhaustion and failure can come later. Type 1, by contrast, is true insulin deficiency from autoimmune beta-cell destruction.
- "The brain needs insulin to take up glucose." Neurons import glucose mainly through insulin-independent GLUT1 and GLUT3, so the brain keeps feeding even when insulin is low — the reason it is so vulnerable to hypoglycemia (no fuel) but not to lack of insulin. GLUT4, the insulin-recruited transporter, is a muscle-and-fat story.
- "Diabetes is purely an insulin problem." Glucagon is dysregulated too. In diabetes, alpha cells release inappropriately high glucagon, telling the liver to dump glucose it should be storing — worsening the hyperglycemia. This "bihormonal" view is why glucagon-suppressing therapies (like GLP-1 agonists) work.
- "Eating sugar is what causes type 2 diabetes directly." The proximate driver is chronic energy surplus, obesity, and inactivity producing insulin resistance, not sugar molecules alone. Genetics load the gun; a persistent caloric surplus and visceral fat pull the trigger. Any excess calories — fat, refined carbohydrate, alcohol — feed the process.
Discovery and famous experiments
- Langerhans describes the islets (1869). As a 22-year-old medical student in Berlin, Paul Langerhans noticed clusters of cells scattered through the pancreas in his doctoral thesis but had no idea what they did. The "islets of Langerhans" were named for him decades later, once their endocrine role was understood.
- Von Mering and Minkowski (1889). Oskar Minkowski and Josef von Mering removed the pancreas from a dog to study digestion; the animal developed severe glucosuria and the classic signs of diabetes. Flies swarming the dog's sugar-laden urine gave the first hard proof that the pancreas controls blood sugar — narrowing the search to an internal secretion.
- Banting and Best isolate insulin (1921). At the University of Toronto, surgeon Frederick Banting and student Charles Best ligated dogs' pancreatic ducts to let the enzyme-secreting tissue wither, then extracted the surviving islets. The extract dramatically lowered a diabetic dog's blood glucose. Biochemist James Collip purified it for human use; on 11 January 1922, 14-year-old Leonard Thompson became the first patient, his ketoacidosis reversing. Banting and J.J.R. Macleod won the 1923 Nobel Prize (Banting sharing his with Best, Macleod with Collip), and the patent was sold to the university for one dollar.
- Kimball and Murlin name glucagon (1923). Working with pancreatic extracts, C.P. Kimball and John Murlin noticed a hyperglycemic contaminant that raised, rather than lowered, blood glucose. They named it "glucagon" (glucose + agon, to drive on), recognizing the second, opposing hormone.
- Sanger sequences insulin (1951–1955). Frederick Sanger determined the complete amino-acid sequence of insulin — the first protein ever fully sequenced — proving that proteins have a defined, unique sequence. The work earned him the 1958 Nobel Prize in Chemistry (his first of two).
- Recombinant human insulin (1982). Genentech and Eli Lilly produced human insulin in engineered E. coli, sold as Humulin — the first recombinant-DNA drug approved for human use, ending reliance on pig and cow pancreas extracts and demonstrating the whole promise of genetic engineering.
Frequently asked questions
How do insulin and glucagon work together to control blood sugar?
They are an antagonistic hormone pair released from the same organ — the pancreatic islets of Langerhans — that clamp blood glucose within a roughly 70 to 110 mg/dL fasting range. After a meal, rising glucose is sensed by beta cells, which secrete insulin. Insulin lowers glucose by recruiting GLUT4 transporters to muscle and fat, activating glycogen synthase for glycogenesis, and shutting off hepatic glucose production. Between meals, when glucose drops toward 70 mg/dL, alpha cells secrete glucagon. Glucagon raises glucose by stimulating glycogenolysis (breaking liver glycogen back to glucose) and gluconeogenesis (making new glucose from amino acids, lactate, and glycerol). Because each hormone opposes the other and each is switched off as glucose returns to setpoint, the system is a textbook negative-feedback loop — the controlled variable, glucose, directly regulates its own controllers.
How does insulin actually lower blood glucose?
Insulin binds the insulin receptor, a receptor tyrosine kinase, which autophosphorylates and recruits IRS-1/2, then PI3K and AKT. AKT triggers the translocation of GLUT4-containing vesicles to the plasma membrane of skeletal muscle and adipocytes — glucose uptake into those tissues can rise more than tenfold within minutes. In the liver and muscle, AKT inactivates GSK-3, which de-represses glycogen synthase, so glucose is polymerized into glycogen (glycogenesis). Insulin simultaneously suppresses the two glucose-raising liver programs: it dephosphorylates and inhibits gluconeogenic gene expression (PEPCK, glucose-6-phosphatase via FOXO1) and blocks glycogenolysis. The net effect is that glucose leaves the blood into storage tissues and the liver stops adding new glucose.
How does glucagon raise blood glucose?
Glucagon is a 29-amino-acid peptide that acts almost exclusively on the liver. It binds the glucagon receptor, a Gs-coupled GPCR, activating adenylyl cyclase, raising cyclic AMP, and activating protein kinase A (PKA). PKA phosphorylates phosphorylase kinase, which activates glycogen phosphorylase — the enzyme that cleaves glucose units off glycogen (glycogenolysis) — while simultaneously phosphorylating and inactivating glycogen synthase, so storage and breakdown are never on at once. PKA and the fall in fructose-2,6-bisphosphate also switch the liver toward gluconeogenesis, building new glucose from lactate, glycerol, and glucogenic amino acids. Liver glycogen (about 100 g, roughly a day's fasting supply) is depleted first; gluconeogenesis then sustains blood glucose during prolonged fasting.
What is the normal blood sugar setpoint?
Fasting plasma glucose in a healthy adult sits around 70 to 110 mg/dL (about 3.9 to 6.1 mmol/L), and even after a large meal it rarely exceeds 140 mg/dL before insulin brings it back. This is remarkably tight regulation: the entire bloodstream of an average adult holds only about 4 grams of dissolved glucose — roughly one sugar cube — yet the brain alone consumes about 120 grams per day and depends almost entirely on a steady glucose supply. The American Diabetes Association defines fasting glucose of 100 to 125 mg/dL as prediabetes and 126 mg/dL or higher (on two occasions) as diabetes. Hypoglycemia below about 70 mg/dL triggers glucagon and adrenaline; symptomatic neuroglycopenia sets in below roughly 50 mg/dL.
Where are insulin and glucagon made?
Both come from the islets of Langerhans — about one to two million micro-organs scattered through the pancreas that together make up only 1 to 2 percent of its mass. Each islet contains several endocrine cell types: beta cells (roughly 60 to 70 percent of islet cells) make insulin, alpha cells (about 20 percent) make glucagon, delta cells make somatostatin, PP cells make pancreatic polypeptide, and epsilon cells make ghrelin. The cells are arranged so their secretions influence each other — insulin and somatostatin both suppress glucagon release, and the islet is richly vascularized so hormones enter the portal blood and reach the liver first. The islets were described by Paul Langerhans in his 1869 medical thesis, decades before their endocrine function was known.
What is the difference between type 1 and type 2 diabetes?
Both are failures of the insulin-glucagon loop, but at different points. Type 1 diabetes is an autoimmune destruction of beta cells — T cells attack the insulin-producing cells, so insulin falls to near zero. It usually appears in childhood or young adulthood, accounts for about 5 to 10 percent of cases, and is fatal without injected insulin. Type 2 diabetes is insulin resistance: the beta cells still make insulin (often extra), but muscle, fat, and liver respond poorly, so glucose stays high; over years the overworked beta cells can also fail. Type 2 accounts for about 90 to 95 percent of the roughly 500 million people with diabetes worldwide and is tied to obesity, inactivity, and genetics. In both types glucagon is also dysregulated — paradoxically high glucagon worsens the hyperglycemia, which is why the disease is now viewed as a two-hormone (bihormonal) disorder, not insulin deficiency alone.
How was insulin discovered?
In the summer of 1921, at the University of Toronto, surgeon Frederick Banting and medical student Charles Best ligated the pancreatic ducts of dogs to let the digestive tissue degenerate, then extracted the surviving islet tissue. The extract lowered blood glucose in a diabetic dog. Biochemist James Collip purified it enough for human use, and on 11 January 1922 a 14-year-old boy named Leonard Thompson, dying of type 1 diabetes, became the first person treated — his blood glucose fell and his ketoacidosis reversed. Banting and lab director J.J.R. Macleod shared the 1923 Nobel Prize in Physiology or Medicine; Banting split his share with Best and Macleod with Collip. They sold the patent to the University of Toronto for one dollar so the drug could be made widely. Frederick Sanger later sequenced insulin (published 1951–1955), the first protein ever sequenced, winning the 1958 Nobel Prize in Chemistry.