The discovery of insulin was among the most exciting and dramatic events in the history of endocrine physiology and therapy. In the United States and Europe, insulin-dependent diabetes mellitus (IDDM), or type 1 diabetes, develops in ~1 in every 600 children. However, the prevalence is only ~1 in 10,000 in eastern Asia. Before 1922, all children with diabetes died within 1 or 2 years of diagnosis. It was an agonizing illness; the children lost weight despite eating well, became progressively weaker and cachectic, were soon plagued by infections, and eventually died of overwhelming acidosis. No effective therapy was available, and few prospects were on the horizon. It was known that the blood sugar level was elevated in this disease, but beyond that, there was little understanding of its pathogenesis.
In 1889, Minkowski and von Mering demonstrated that removing the pancreas from dogs caused hyperglycemia, excess urination, thirst, weight loss, and death—in short, a syndrome closely resembling type 1 diabetes. Following this lead, a group of investigators in the Department of Physiology at the University of Toronto prepared extracts of pancreas and tested the ability of these extracts to lower plasma [glucose] in pancreatectomized dogs. Despite months of failures, these investigators persisted in their belief that such extracts could be beneficial. Finally, by the winter of 1921, Frederick Banting (a surgeon) and Charles Best (at the time, a medical student) were able to demonstrate that an aqueous extract of pancreas could lower blood glucose level and prolong survival in a pancreatectomized dog. N51-2 Within 2 months, a more purified extract was shown to lower blood glucose level in a young man with diabetes. By the end of 1923, insulin (as the islet hormone was named) was being prepared from beef and pork pancreas on an industrial scale, and patients around the world were receiving effective treatment of their diabetes. For the discovery of insulin, Frederick Banting and John Macleod received the 1923 Nobel Prize in Physiology or Medicine. N51-3
Frederick Banting and Charles Best
Contributed by Emile Boulpaep, Walter Boron
In 1923, just 2 years after Frederick Banting (a young faculty member just 5 years out of medical school) and Charles Best (a 22-year-old medical student working in Banting's laboratory) discovered insulin at the University of Toronto, the Nobel Prize in Physiology or Medicine was awarded to Frederick Banting and the head of the research team and chairman of Banting's department, John Macleod. The short delay between the discovery and the award of the prize indicates the enormous significance of the discovery.
It is interesting that Frederick Banting protested the award of the Nobel Prize to John Macleod and gave half of his portion of the monetary award to Charles Best.
Banting FG, Best CH. Pancreatic extracts, 1922. J Lab Clin Med. 1990;115:254–272.
Banting FG, Best CH, Collip JB, et al. Pancreatic extracts in the treatment of diabetes mellitus: Preliminary report, 1922. CMAJ. 1991;145:1281–1286.
Frederick Banting and John Macleod
Contributed by Emile Boulpaep, Walter Boron
For more information about Frederick Banting and John Macleod and the work that led to their Nobel Prize, visit http://nobelprize.org/medicine/laureates/1923/index.html (accessed October 2014).
Since that time, the physiology of the synthesis, secretion, and action of insulin has been studied more extensively than that of any other hormone. Now, nearly a century later, much is known about the metabolic pathways through which insulin regulates carbohydrate, lipid, and protein metabolism in its major targets: the liver, muscle, and adipose tissue. However, the sequence of intracellular signals that triggers insulin secretion by pancreatic β cells, as well as the signal-transduction process triggered when insulin binds to a plasma membrane receptor on target tissues, remain areas of intense study.
Insulin replenishes fuel reserves in muscle, liver, and adipose tissue
What does insulin do? Succinctly put, insulin efficiently integrates body fuel metabolism both during periods of fasting and during feeding (Table 51-2). When an individual is fasting, the β cell secretes less insulin. When insulin levels decrease, lipids are mobilized from adipose tissue and amino acids are mobilized from body protein stores within muscle and other tissues. These lipids and amino acids provide fuel for oxidation and serve as precursors for hepatic ketogenesis and gluconeogenesis, respectively. During feeding, insulin secretion increases promptly, which diminishes the mobilization of endogenous fuel stores and stimulates the uptake of carbohydrates, lipids, and amino acids by insulin-sensitive target tissues. In this manner, insulin directs tissues to replenish the fuel reserves depleted during periods of fasting.
Effects of Nutritional States
AFTER A 24-hr FAST
2 hr AFTER A MIXED MEAL
Plasma [glucose], mg/dL
Plasma [insulin], µU/mL
Plasma [glucagon], pg/mL
Lipids mobilized for fuel
Glucose oxidized or stored as glycogen
As a result of its ability to regulate the mobilization and storage of fuels, insulin maintains plasma [glucose] within narrow limits. Such regulation provides the central nervous system (CNS) with a constant supply of glucose needed to fuel cortical function. In higher organisms, if plasma [glucose] (normally ≅ 5 mM) declines to <2 to 3 mM (hypoglycemia; Box 51-1) for even a brief period, confusion, seizures, and coma may result. Conversely, persistent elevations of plasma [glucose] are characteristic of the diabetic state. Severe hyperglycemia (plasma glucose levels > 15 mM) produces an osmotic diuresis (see Box 35-1) that, when severe, can lead to dehydration, hypotension, and vascular collapse.
Clinical Manifestations of Hypoglycemia and Hyperglycemia
Early symptoms are principally autonomic and include palpitations, tachycardia, diaphoresis, anxiety, hyperventilation, shakiness, weakness, and hunger. More severe hypoglycemia manifests principally as neuroglycopenia, with confusion, aberrant behavior, hallucinations, seizures, hypothermia, focal neurological deficits, and coma.
Early manifestations include weakness, polyuria, polydipsia, altered vision, weight loss, and mild dehydration. For prolonged or severe hyperglycemia (accompanied by metabolic acidosis or diabetic ketoacidosis), manifestations include Kussmaul hyperventilation (deep, rapid breathing; see p. 716), stupor, coma, hypotension, and cardiac arrhythmias.
Contributed by Eugene Barrett
Hypoglycemia, which can be viewed most simply as the opposite of diabetes mellitus, has many causes. Perhaps the most frequent setting is a patient with type 1 diabetes who skips a meal or fails to adjust the insulin dose when exercising. Many diabetic patients who seek to maintain tight control over their blood sugar experience frequent hypoglycemic reactions, which they quickly learn to abort with a carbohydrate snack. Patients with type 2 diabetes who take an excessive dose of sulfonylureas are subject to severe hypoglycemia, which may require continuous treatment for several days because the half-life of some of these drugs is quite long.
We saw in Chapter 50 that epinephrine—acting as a β-adrenergic agonist—is a hyperglycemic agent; that is, it promotes glycogenolysis in liver and muscle (see p. 1033). Thus, β blockers rarely cause hypoglycemia in healthy individuals because these people can appropriately regulate their insulin secretion. However, because β blockers can mask the early adrenergic response to mild hypoglycemia (sweating, tachycardia, tremulousness), diabetic patients taking both insulin and β blockers commonly progress to severe hypoglycemia without warning. Another drug that can induce hypoglycemia is pentamidine, an agent used to treat Pneumocystis jiroveci pneumonia. Pentamidine is a β-cell toxin that leads to an acute, excessive release of insulin, which can be followed by hypoglycemia.
Alcoholic patients are at great risk of hypoglycemia. Ethanol suppresses gluconeogenesis, and hepatic glycogen stores may already be low because of poor nutrition. Other severe illnesses that can produce persistent hypoglycemia include liver disease, renal failure, and some large tumors that produce a hypoglycemia-inducing peptide, usually IGF-2. Rarely, an insulinoma may develop, which is an islet cell tumor (usually benign) that releases high and unregulated concentrations of insulin into the bloodstream.
Many individuals complain of postprandial hypoglycemia, frequently called reactive hypoglycemia. Despite long years of skepticism, investigators now believe that at least some of these patients do indeed experience true symptoms of hypoglycemia within a few hours of eating. There is no absolute glucose level at which symptoms occur; many people can tolerate extremely low levels of glucose without any problems. However, a rather high rate of decline in the plasma glucose level after a meal may cause symptoms. One cause of postprandial hypoglycemia may be a delay in the timing of insulin release after a meal. Thus, the β cells release too much insulin too late after a meal, so the blood glucose level initially rises markedly and then falls rapidly. In some patients, this defect may herald the development of diabetes mellitus.
β cells synthesize and secrete insulin
The Insulin Gene
Circulating insulin comes only from the β cells of the pancreatic islet. It is encoded by a single gene on the short arm of chromosome 11. Exposing islets to glucose stimulates insulin synthesis and secretion. Though the process is not completely understood, this stimulation requires that the glucose be metabolized.
Transcription of the insulin gene product and subsequent processing produces full-length messenger RNA (mRNA) that encodes preproinsulin. Starting from its 5′ end, this mRNA encodes a leader sequence and then peptide domains B, C, and A. Insulin is a secretory protein (see pp. 34–35). As the preprohormone is synthesized, the leader sequence of ~24 amino acids is cleaved from the nascent peptide as it enters the rough endoplasmic reticulum. The result is proinsulin (Fig. 51-2), which consists of domains B, C, and A. As the trans Golgi packages the proinsulin and creates secretory granules, proteases slowly begin to cleave the proinsulin molecule at two spots and thus excise the 31–amino-acid C peptide. The resulting insulin molecule has two peptide chains, designated the A and B chains, that are joined by two disulfide linkages. The mature insulin molecule has a total of 51 amino acids, 21 on the A chain and 30 on the B chain. In the secretory granule, the insulin associates with zinc. The secretory vesicle contains this insulin, as well as proinsulin and C peptide. All three are released into the portal blood when glucose stimulates the β cell.
FIGURE 51-2 Synthesis and processing of the insulin molecule. The mature mRNA of the insulin gene product contains a 5′ untranslated region (UTR); nucleotide sequences that encode a 24–amino-acid leader sequence, as well as B, C, and A peptide domains; and a 3′ UTR. Together, the leader plus the B, C, and A domains constitute preproinsulin. During translation of the mRNA, the leader sequence is cleaved in the lumen of the rough endoplasmic reticulum (ER). What remains is proinsulin, which consists of the B, C, and A domains. Beginning in the trans Golgi, proteases cleave the proinsulin at two sites, releasing the C peptide as well as the mature insulin molecule, which consists of the B and A chains that are connected by two disulfide bonds. The secretory granule contains equimolar amounts of insulin and the C peptide, as well as a small amount of proinsulin. These components all are released into the extracellular space during secretion.
Secretion of Insulin, Proinsulin, and C Peptide
C peptide has no established biological action. Yet because it is secreted in a 1 : 1 molar ratio with insulin, it is a useful marker for insulin secretion. Proinsulin does have modest insulin-like activities; it is ~th as potent as insulin on a molar basis. However, the β cell secretes only ~5% as much proinsulin as insulin. As a result, proinsulin does not play a major role in the regulation of blood glucose.
Most of the insulin (~60%) that is secreted into the portal blood is removed in a first pass through the liver. In contrast, C peptide is not extracted by the liver at all. As a result, whereas measurements of the insulin concentration in systemic blood do not quantitatively mimic the secretion of insulin, measurements of C peptide do. C peptide is eventually excreted in the urine, and the quantity of C peptide excreted in a 24-hour period is a rough measure of the amount of insulin released during that time.
Glucose is the major regulator of insulin secretion
In healthy individuals, the plasma glucose concentration remains within a remarkably narrow range. After an overnight fast, it typically averages between 4 and 5 mM; the plasma [glucose] rises after a meal, but even with a very large meal it does not exceed 10 mM. Modest increases in plasma [glucose] provoke marked increases in the secretion of insulin and C peptide and hence raise plasma [insulin], as illustrated by the results of an oral glucose tolerance test (OGTT) as shown in Figure 51-3A. Conversely, a decline in plasma [glucose] of only 20% markedly lowers plasma [insulin]. The change in the concentration of plasma glucose that occurs in response to feeding or fasting is the main determinant of insulin secretion. In a patient with type 1 diabetes mellitus caused by destruction of pancreatic islets, an oral glucose challenge evokes either no response or a much smaller insulin response, but a much larger increment in plasma [glucose] that lasts for a much longer time (see Fig. 51-3B).
FIGURE 51-3 Glucose tolerance test results. A, When a person ingests a glucose meal (75 g), plasma [glucose] (green curve) rises slowly, reflecting intestinal uptake of glucose. As a result, plasma [insulin] (solid red curve) rises sharply. When a lower glucose dose is given intravenously (IV) over time—in a manner that reproduces the green curve—plasma [insulin] rises only modestly (dashed red curve). The difference between the insulin responses indicated by the solid and dashed red lines is due to the “incretin effect” of oral glucose ingestion. B, In a patient with type 1 diabetes, the same oral glucose load as that in A causes plasma [glucose] to rise to a higher level and to remain high for a longer time. The diagnosis of diabetes is made if the plasma glucose level is above 200 mg/dL at the second hour. C, If a large IV glucose challenge (0.5 g glucose/kg body weight given as a 25% glucose solution) is administered as a bolus, plasma [glucose] rises much more rapidly than it does with an oral glucose load. Sensing a rapid rise in [glucose], the β cells first secrete some of their stores of presynthesized insulin. Following this “acute phase,” the cells secrete both presynthesized and newly manufactured insulin in the “chronic phase.”
A glucose challenge of 0.5 g/kg body weight given as an intravenous bolus raises the plasma glucose concentration more rapidly than glucose given orally. Such a rapid rise in plasma glucose concentration leads to two distinct phases of insulin secretion (see Fig. 51-3C). The acute-phase or first-phase insulin response lasts only 2 to 5 minutes, whereas the second-phase insulin response persists as long as the blood glucose level remains elevated. The insulin released during the acute-phase insulin response to intravenous glucose arises from preformed insulin that had been packaged in secretory vesicles docked at, or residing near, the β-cell plasma membrane. The second-phase insulin response also comes from preformed insulin within the vesicles with some contribution from newly synthesized insulin. One of the earliest detectable metabolic defects that occurs in both type 1 and type 2 diabetes is loss of the first phase of insulin secretion, as determined by an intravenous glucose tolerance test. If a subject consumes glucose or a mixed meal, plasma [glucose] rises much more slowly—as in Figure 51-3A—because the appearance of glucose in plasma depends on gastric emptying and intestinal absorption. Given that plasma [glucose] rises so slowly, the acute-phase insulin response can no longer be distinguished from the chronic response, and only a single phase of insulin secretion is apparent. However, the total insulin response to an oral glucose challenge exceeds the response observed when comparable changes in plasma [glucose] are produced by intravenously administered glucose (see Fig. 51-3A). This difference is referred to as the incretin effect (Box 51-2).
Nonhuman and Mutant Insulin
Cloning of the insulin gene has led to an important therapeutic advance, namely, the use of recombinant human insulin for the treatment of diabetes. Human insulin was the first recombinant protein available for routine clinical use. Before the availability of human insulin, either pork or beef insulin was used to treat diabetes. Pork and beef insulin differ from human insulin by one and three amino acids, respectively. The difference, although small, is sufficient to be recognized by the immune system, and antibodies to the injected insulin develop in most patients treated with beef or pork insulin; occasionally, the reaction is severe enough to cause a frank allergy to the insulin. This problem is largely avoided by using human insulin.
Sequencing of the insulin gene has not led to a major understanding of the genesis of the common forms of human diabetes. However, rare patients with diabetes make a mutant insulin molecule with a single amino-acid substitution in either the A or B chain. In each case that has been described, these changes lead to a less-active insulin molecule (typically only ~1% as potent as insulin on a molar basis). These patients have either glucose intolerance or frank diabetes, but very high concentrations of immunoreactive insulin in their plasma. In these individuals, the immunoreactivity of insulin is not affected to the same extent as the bioactivity.
In addition to revealing these mutant types of insulin, sequencing of the insulin gene has allowed identification of a flanking polymorphic site upstream of the insulin gene that contains one of several common alleles. In some populations, certain polymorphisms are associated with an increased risk of development of type 1 diabetes mellitus.
Metabolism of glucose by the β cell triggers insulin secretion
The pancreatic β cells take up and metabolize glucose, galactose, and mannose, and each can provoke insulin secretion by the islet. Other hexoses that are transported into the β cell but that cannot be metabolized (e.g., 3-O-methylglucose or 2-deoxyglucose) do not stimulate insulin secretion. Although glucose itself is the best secretagogue, some amino acids (especially arginine and leucine) and small keto acids (e.g., α-ketoisocaproate, α-ketoglutarate), as well as ketohexoses (fructose), can also weakly stimulate insulin secretion. The amino acids and keto acids do not share any metabolic pathway with hexoses other than oxidation via the citric acid cycle (see p. 1185). These observations have led to the suggestion that the ATP generated from the metabolism of these varied substances may be involved in insulin secretion. In the laboratory, depolarizing the islet cell membrane by raising extracellular [K+] provokes insulin secretion.
From these data has emerged a relatively unified picture of how various secretagogues trigger insulin secretion. Key to this picture is the presence in the islet of an ATP-sensitive K+ channel and a voltage-gated Ca2+ channel in the plasma membrane (Fig. 51-4). The K+ channel (KATP; see p. 198) is an octamer of four Kir6.2 channels (see p. 196) and four sulfonylurea receptors (SURs; see p. 199; Box 51-3), Glucose triggers insulin release in a seven-step process:
Step 1: Glucose enters the β cell via the GLUT2 glucose transporter by facilitated diffusion (see p. 114). Amino acids enter through a different set of transporters.
Step 2: In the presence of glucokinase (the rate-limiting enzyme in glycolysis), the entering glucose undergoes glycolysis as well as oxidation via the citric acid cycle (see p. 1185), phosphorylating ADP and raising [ATP]i. Some amino acids also enter the citric acid cycle. In both cases, the following ratios increase: [ATP]i/[ADP]i, [NADH]i/[NAD+]i, and [NADPH]i/[NADP+]i (NADH and NAD+ are the reduced and oxidized forms of nicotinamide adenine dinucleotide [NAD], and NADPH and NADP+ are the reduced and oxidized forms of NAD phosphate) N51-5
Step 3: The increase in the ratio [ATP]i/[ADP]i, or [NADH]i/[NAD+]i, or [NADPH]i/[NADP+]i causes KATP channels (see p. 198) to close.
Step 4: Reducing the K+ conductance of the cell membrane causes the β cell to depolarize (i.e., the membrane potential is less negative).
Step 5: This depolarization activates voltage-gated Ca2+ channels (see pp. 190–191).
Step 6: The increased Ca2+ permeability leads to increased Ca2+ influx and increased intracellular free Ca2+. This rise in [Ca2+]i additionally triggers Ca2+-induced Ca2+ release (see pp. 242–243).
Step 7: The increased [Ca2+]i, perhaps by activation of a Ca2+-calmodulin phosphorylation cascade, ultimately leads to insulin release.
FIGURE 51-4 Mechanism of insulin secretion by the pancreatic β cell. Increased levels of extracellular glucose trigger the β cell to secrete insulin in the seven steps outlined in this figure. Metabolizable sugars (e.g., galactose and mannose) and certain amino acids (e.g., arginine and leucine) can also stimulate the fusion of vesicles that contain previously synthesized insulin. In addition to these fuel sources, certain hormones (e.g., glucagon, somatostatin, cholecystokinin [CCK]) can also modulate insulin secretion. ER, endoplasmic reticulum; IP3, inositol 1,4,5-trisphosphate; PLC, phospholipase C.
An entire class of drugs—the sulfonylurea agents—is used in the treatment of patients with type 2 diabetes, or non–insulin-dependent diabetes mellitus (NIDDM). Type 2 diabetes arises from two defects: (1) β cells are still capable of making insulin but do not respond adequately to increased blood [glucose], and (2) insulin target tissues are less sensitive or “resistant” to insulin.
The sulfonylurea agents were discovered accidentally. During the development of sulfonamide antibiotics after the Second World War, investigators noticed that the chemically related sulfonylurea agents produced hypoglycemia in laboratory animals. These drugs turned out to have no value as antibiotics, but they did prove effective in treating the hyperglycemia of type 2 diabetes. The sulfonylureas enhance insulin secretion by binding to the SUR subunits (see p. 199) of KATP channels, thereby decreasing the likelihood that these channels will be open. This action enhances glucose-stimulated insulin secretion (see Fig. 51-4). By increasing insulin secretion, sulfonylureas overcome insulin resistance and decrease blood glucose in these patients.
Unlike insulin, which must be injected, sulfonylureas can be taken orally and are therefore preferred by many patients. However, they have a therapeutic role only in type 2 diabetes; the β cells in patients with type 1 diabetes are nearly all destroyed, and these patients must be treated with insulin replacement therapy.
The Pentose Phosphate Pathway (or Hexose Monophosphate Shunt)
Contributed by Emile Boulpaep, Walter Boron
Figure 58-1 mentions that glucose-6-phosphate can have three major fates. The anabolic series of reactions summarized in this figure convert glucose-6-phosphate to glycogen. The glycolytic pathway summarized in Figure 58-6A is a catabolic pathway that converts glucose-6-phosphate to pyruvate. The third fate—the pentose phosphate pathway—is another catabolic series of reactions that converts glucose-6-phosphate to ribose-5-phosphate.
The pentose phosphate pathway has two major products, NADPH and ribose-5-phosphate. The cell can use the reducing equivalents in NADPH (i.e., energy “currency”) to reduce double bonds in the energy-consuming synthesis of fatty acids and steroids. These reactions are particularly important in such tissues as liver, adipose tissue, mammary gland, and adrenal cortex. Note that the cell cannot use NADH to create NADPH. Thus, the pentose phosphate pathway is critical. The second product of the pathway, ribose-5-phosphate, is important for the synthesis of ribonucleotides, which is particularly important in growing and regenerating tissues. The pentose phosphate pathway involves four reactions, the first and third of which involve the conversion of NADP+ to NADPH and H+.
If the cell does not use the ribose-5-phosphate to generate ribonucleotides, the cell can use a complex series of reactions to convert the ribose-5-phosphate to fructose-6-phosphate. This sequence of reactions (i.e., from glucose-6-phosphate to ribose-5-phosphate to fructose-6-phosphate) bypasses or “shunts” the conversion of glucose-6-phosphate to fructose-6-phosphate, which would otherwise be catalyzed by phosphoglucose isomerase (see Fig. 58-6A). For this reason, the pentose phosphate pathway is also called the hexose monophosphate shunt.* However, the reader should be reassured that the shunt does not permit the cell to generate two NADPH molecules for free. Three glucose-6-phosphate molecules (3 × 6 = 18 carbons) must traverse the hexose monophosphate shunt to generate six NADPH molecules (3 × 2 = 6) plus two fructose-6-phosphate (2 × 6 = 12 carbons) molecules, a single glyceraldehyde-3-phosphate (1 × 3 = 3 carbons), and three CO2 molecules that arise from a decarboxylation reaction in the pentose phosphate pathway (3 × 1 = 3 carbons). If those three glucose molecules had gone through the classical glycolytic pathway, they would have generated 3 × 2 = 6 net ATPs and 6 NADHs (see Table 58-3). However, if those same three glucose molecules all go through the pentose phosphate pathway, the net result is only five ATPs, only five NADHs, but six NADPHs. Thus, the cell gives up only one ATP and one NADH for the sake of generating six NADPHs—not a bad deal for the cell!
Nelson DL, Cox MM. Lehninger Principles of Biochemistry. 3rd ed. Worth Publishers: New York; 2000.
Voet D, Voet JG. Biochemistry. 2nd ed. Wiley: New York; 1995.
*Note that the term shunt is a bit of a misnomer, inasmuch as the “shunt” is not a shortcut from glucose-6-phosphate to fructose-6-phosphate (normally catalyzed in one step by phosphoglucose isomerase), but rather a lengthy detour!
Other secretagogues can also modulate insulin secretion via the phospholipase C pathway (see p. 58) or via the adenylyl cyclase pathway (see p. 53) in addition to the pathway just outlined. For example, glucagon, which stimulates insulin release, may bypass part or all of the glucose/[Ca2+]i pathway by stimulating adenylyl cyclase, thus raising cAMP levels and activating protein kinase A (PKA). Conversely, somatostatin, which inhibits insulin release, may act by inhibiting adenylyl cyclase.
Neural and humoral factors modulate insulin secretion
The islet is richly innervated by both the sympathetic and the parasympathetic divisions of the ANS. Neural signals appear to play an important role in the β-cell response in several settings. β-adrenergic stimulation augments islet insulin secretion, whereas α-adrenergic stimulation inhibits it (see Fig. 51-4). Isoproterenol, a synthetic catecholamine that is a specific agonist for the β-adrenergic receptor, potently stimulates insulin release. In contrast, norepinephrine and synthetic α-adrenergic agonists suppress insulin release both basally and in response to hyperglycemia. Because the postsynaptic sympathetic neurons of the pancreas release norepinephrine, which stimulates α more than β adrenoceptors, sympathetic stimulation via the celiac nerves inhibits insulin secretion. In contrast to α-adrenergic stimulation, parasympathetic stimulation via the vagus nerve, which releases acetylcholine, causes an increase in insulin release.
The effect of sympathetic regulation on insulin secretion may be particularly important during exercise, when adrenergic stimulation of the islet increases. The major role for α-adrenergic inhibition of insulin secretion during exercise is to prevent hypoglycemia. Exercising muscle tissue uses glucose even when plasma [insulin] is low. If insulin levels were to rise, glucose use by the muscle would increase even further and promote hypoglycemia. Furthermore, an increase in [insulin] would inhibit lipolysis and fatty-acid release from adipocytes and would thus diminish the availability of fatty acids, which the muscle can use as an alternative fuel to glucose (see p. 1211). Finally, a rise in [insulin] would decrease glucose production by the liver. Suppression of insulin secretion during exercise may thus serve to prevent excessive glucose uptake by muscle, which, if it were to exceed the ability of the liver to produce glucose, would lead to severe hypoglycemia, compromise the brain, and abruptly end any exercise!
Another important setting in which neural and humoral factors regulate insulin secretion is during feeding. Food ingestion triggers a complex series of neural, endocrine, and nutritional signals to many body tissues. The cephalic phase (see pp. 871 and 890) of eating, which occurs before food is ingested, results in stimulation of gastric acid secretion and a small rise in plasma insulin level. This response appears to be mediated by the vagus nerve in both cases. If no food is forthcoming, blood [glucose] declines slightly and insulin secretion is again suppressed. If food ingestion does occur, the acetylcholine released by postganglionic vagal fibers in the islet augments the insulin response of the β cell to glucose.
As already discussed, after a subject drinks a glucose solution, the total amount of insulin secreted is greater than when the same amount of glucose is administered intravenously (see Fig. 51-3A). This observation has led to a search for enteric factors or incretins that augment the islet β-cell response to an oral glucose stimulus. Currently, we know of three peptides released by intestinal cells in response to feeding that enhance insulin secretion: cholecystokinin from I cells, glucagon-like peptide 1 (GLP-1) from L cells, and gastric inhibitory polypeptide (GIP, also called glucose-dependent insulinotropic peptide) from K cells. GLP-1 (see p. 1051), perhaps the most important incretin yet discovered, has a very short half-life in plasma (<2 min), which precludes its therapeutic use. A peptide analog of GLP-1, called exendin-4, is found in the saliva of the Gila monster. A long-lasting synthetic derivative of exendin-4 (exenatide) is approved for use in treating diabetic patients to enhance insulin secretion and to ameliorate hyperglycemia. Because it is a peptide, exenatide cannot be given orally but is effective by subcutaneous injection. Orally available agents have been developed that inhibit the enzyme dipeptidyl-peptidase 4, which breaks down native GLP-1; these gliptins are now available to treat type 2 diabetes.
In the laboratory, incretins stimulate insulin secretion by isolated islets of Langerhans, and glucose increases this secretion even further. The presence of these incretins in the gut mucosa gives the islets advance notice that nutrients are being absorbed and “primes” the β cells to amplify their response to glucose. In addition, vagal stimulation of the β cells primes the islets for an amplified response.
The insulin receptor is a receptor tyrosine kinase
Once insulin is secreted into the portal blood, it first travels to the liver, where more than half is bound and removed from the circulation. The insulin that escapes the liver is available to stimulate insulin-sensitive processes in other tissues. At each target tissue, the first action of insulin is to bind to a specific receptor tyrosine kinase (see pp. 68–70) on the plasma membrane (Fig. 51-5).
FIGURE 51-5 Insulin receptor. The insulin receptor is a heterotetramer that consists of two extracellular α chains and two membrane-spanning β chains. Insulin binding takes place on the cysteine-rich region of the α chains.
The insulin receptor—as well as the closely related insulin-like growth factor 1 (IGF-1) receptor—is a heterotetramer, with two identical α chains and two identical β chains. N51-6 The α and β chains are synthesized as a single polypeptide that is subsequently cleaved. The two chains are joined by disulfide linkage (reminiscent of the synthesis of the A and B chains of insulin itself) in the sequence β-α-α-β. The insulin receptor shares considerable structural similarity with the IGF-1 receptor (see p. 996). The overall sequence homology is ~50%, and this figure rises to >80% in the tyrosine kinase region. This similarity is sufficient that very high concentrations of insulin can stimulate the IGF-1 receptor and, conversely, high levels of IGF-1 can stimulate the insulin receptor.
Insulin and IGF-1 Receptors
Contributed by Emile Boulpaep, Walter Boron
Activation of the insulin and IGF-1 receptors (see Fig. 51-5) occurs by somewhat different mechanisms, as we discuss on pages 1041–1042 for the insulin receptor and on page 996 for IGF-1 receptor. In brief, these receptors are tetrameric; they are composed of two α and two β subunits. The α subunit contains a cysteine-rich region and functions in ligand binding. The β subunit is a single-pass transmembrane protein with a cytoplasmic tyrosine kinase domain. The α and β subunits are held together by disulfide bonds (as are the two α subunits), forming a heterotetramer. Ligand binding produces conformational changes that appear to cause allosteric interactions between the two α and β pairs, as opposed to the dimerization characteristic of the first class of receptor tyrosine kinases (see Fig. 3-12C). Thus, insulin binding results in the autophosphorylation of tyrosine residues in the catalytic domains of the β subunits. The activated insulin receptor also phosphorylates cytoplasmic substrates such as IRS-1 (see Fig. 51-6), which, once phosphorylated, serves as a docking site for additional signaling proteins.
The insulin receptor's extracellular α chains have multiple N-glycosylation sites. The β chains have an extracellular, a membrane-spanning, and an intracellular portion. The β subunit of the receptor is glycosylated on its extracellular domains; receptor glycosylation is required for insulin binding and action. The intracellular domain of the β chain possesses tyrosine kinase activity, which increases markedly when insulin binds to sites on the α chains of the receptor. The insulin receptor can phosphorylate both itself and other intracellular substrates at tyrosine residues (see pp. 68–70). The targets of tyrosine phosphorylation (beyond the receptor itself) include a family of cytosolic proteins known as insulin-receptor substrates (IRS-1, IRS-2, IRS-3, and IRS-4) as well as Src homology C terminus (SHC), as illustrated in Figure 51-6. This phosphorylation mechanism appears to be the major one by which insulin transmits its signal across the plasma membrane of insulin target tissues.
FIGURE 51-6 Insulin signal-transduction system. When insulin binds to its receptor—which is a receptor tyrosine kinase (RTK)—tyrosine kinase domains on the intracellular portion of the β chains become active. The activated receptor transduces its signals to downstream effectors by phosphorylating at tyrosine residues on the receptor itself, the IRS family (IRS-1, IRS-2, IRS-3, IRS-4), and other cytosolic proteins (e.g., SHC). SH2-containing proteins dock onto certain phosphorylated tyrosine groups on the IRSs and thus become activated. Not all of the signaling pathways are active in all of insulin's target cells. For example, the liver cell does not rely on the GLUT4 transporter to move glucose in and out of the cell. Likewise, the liver is a very important target for regulation of the gluconeogenic enzymes by insulin, whereas muscle and adipose tissue are not. GS, glycogen synthase; GSK-3, glycogen synthase kinase 3; IF, initiation factor; PDK, phosphatidylinositol-dependent kinase.
The IRS proteins are docking proteins to which various downstream effector proteins bind and thus become activated. IRS-1 has at least eight tyrosines within specific motifs that generally bind proteins containing SH2 (Src homology domain 2) domains (see p. 58), so that a single IRS molecule simultaneously activates multiple pathways. The IGF-1 receptor, which is closely related to the insulin receptor, also acts through IRS proteins.
Figure 51-6 illustrates three major signaling pathways triggered by the aforementioned tyrosine phosphorylations. N51-7 The first begins when phosphatidylinositol 3-kinase (PI3K) binds to phosphorylated IRS and becomes activated. PI3K phosphorylates a membrane lipid phosphatidylinositol 4,5-bisphosphate (PIP2) to form phosphatidylinositol 3,4,5-trisphosphate (PIP3), and it leads to major changes in glucose and protein metabolism.
Insulin Signal Transduction
Contributed by Eugene Barrett, Emile Boulpaep, Walter Boron
Figure 51-6 shows three major pathways. In the first pathway, activation of phosphatidylinositol 3-kinase (PI3K) phosphorylates phosphatidylinositol 4,5-bisphosphate (PIP2) to form phosphatidylinositol 3,4,5-trisphosphate (PIP3; see p. 58), which in turn activates phosphatidylinositol-dependent kinase (PDK). This serine/threonine kinase then activates protein kinase B (PKB), which leads to the insertion of GLUT4 glucose transporters into the plasma membrane. PDK also phosphorylates and thus inactivates glycogen synthase kinase 3 (GSK-3); the net effect is reduced inactivation of glycogen synthase (GS) and enhanced glycogen synthesis. Finally PDK activates mTOR (target of rapamycin), a serine/threonine kinase that phosphorylates the binding protein PHAS-1 and thus releases an active initiation factor (IF), promoting translation of mRNA into protein. mTOR also phosphorylates p70-S6 kinase, which phosphorylates the ribosomal S6 protein.
In the second pathway, the insulin receptor phosphorylates SHC (which stands for Src homology, C terminus) at tyrosine residues, stimulating SOS. In addition, activation of GRB2 also stimulates SOS. The stimulated SOS activates the Ras pathway, as described in Figure 3-13. The activated Raf-1, which is itself a MEK kinase, activates not only MEK but also other MEK kinases, which in turn activate JNK (a kinase) and p38 kinase. MAPK activates both a transcription factor and p90-S6 kinase. The activated p90-S6 kinase phosphorylates a variety of nuclear proteins as well as phosphoprotein phosphatase 1 (PP1); the latter leads to activation of glycogen synthase.
In the third pathway, SH2-containing proteins (shown in blue in Fig. 51-6)—other than PI3K and GRB2, already discussed—bind to specific phosphotyrosine groups on either the insulin receptor or IRS proteins. These SH2-containing proteins have a variety of effects, for example, on enzymes involved in lipid metabolism.
The second signaling pathway begins in one of two ways: (1) the insulin receptor phosphorylates SHC or (2) growth factor receptor–bound protein 2 (GRB2; see p. 69) binds to an IRS and becomes activated. As illustrated in Figure 51-6, both phosphorylated SHC and activated GRB2 trigger the Ras signaling pathway, leading through mitogen-activated protein kinase kinase (MEK) and mitogen-activated protein kinase (MAPK; see pp. 68–69) to increased gene expression and growth. Gene-deletion studies in mice show that IRS-1 deletion does not cause diabetes but results in small mice. In contrast, IRS-2 deletion does cause diabetes, in part because of impaired insulin secretion by the pancreatic β cell!
The third signaling pathway begins with the binding of SH2-containing proteins—other than PI3K and GRB2, already discussed—to specific phosphotyrosine groups on either the insulin receptor or IRS proteins. This binding activates the SH2-containing protein (Box 51-4).
The Insulin Receptor and Rare Forms of Diabetes
The ability of insulin to act on a target cell depends on three things: the number of receptors present on the target cell, the receptor's affinity for insulin, and the receptor's ability to transduce the insulin signal.
Several disorders have been described in which a mutation of the insulin receptor blunts or prevents insulin's actions. One such mutation markedly affects growth in utero, as well as after birth. This rare disorder is called leprechaunism, and it is generally lethal within the first year of life. Other mutations of the receptor have less devastating consequences.
Some individuals make antibodies to their own insulin receptors. Insulin, produced either endogenously or administered to these patients, does not work well because it must compete with these antibodies for sites on the receptor; as a result, the patient is hyperglycemic. Interestingly, other antibodies can be “insulin mimetic”; that is, not only do the antibodies bind to the receptor, but they also actually mimic insulin's action. This mimicry causes severe hypoglycemia in affected individuals.
Neither receptor mutations nor antireceptor antibodies appear to be responsible for any of the common forms of diabetes seen clinically. However, abnormal insulin-receptor signaling may be involved in many patients with type 2 diabetes. Indeed, activation of inflammatory pathways involving the p38 subset of MAPKs (see p. 69) and nuclear factor κB (see pp. 86–87) can lead to phosphorylation of the insulin receptor (and of IRS proteins), principally at serine residues. This serine phosphorylation occurs commonly in animal models of insulin resistance and type 2 diabetes as well as in human diabetes, and can interfere with the normal metabolic actions of insulin.
High levels of insulin lead to downregulation of insulin receptors
The number of insulin receptors expressed on the cell surface is far greater than that needed for the maximal biological response to insulin. For example, in the adipocyte, the glucose response to insulin is maximal when only ~5% of the receptors are occupied; that is, the target cells have many “spare” receptors for insulin.
The number of insulin receptors present on the membrane of a target cell is determined by the balance among three factors: (1) receptor synthesis, (2) endocytosis of receptors followed by recycling of receptors back to the cell surface, and (3) endocytosis followed by degradation of receptors. Cells chronically exposed to high concentrations of insulin have fewer surface receptors than do those exposed to lower concentrations. This dynamic ability of cells to decrease the number of specific receptors on their surface is called downregulation. Insulin downregulates insulin receptors by decreasing receptor synthesis and increasing degradation. Such downregulation is one mechanism by which target tissues modulate their response to hormones. Downregulation of insulin receptors results in a decrease in the sensitivity of the target tissue to insulin without diminishing insulin's maximal effect.
One example of how downregulation can affect insulin's action is shown in Figure 51-7, which illustrates the effect of increases in insulin concentration on glucose uptake in adipocytes from normal individuals and individuals with type 2 diabetes. Adipocytes from patients with type 2 diabetes (Box 51-5) have fewer insulin receptors per unit of surface area than do adipocytes from normal individuals. The markedly lower glucose transport across the entire physiological range of insulin concentrations in diabetic adipocytes is characteristic of insulin resistance. In healthy control adipocytes, glucose transport is maximal when only a few (~5%) of the receptors are occupied. In diabetic adipocytes, a much higher concentration of insulin is required, and a larger fraction of the insulin receptors is occupied. However, the major effects in type 2 diabetes apparently are not the result of a decrease in receptor number, but rather are caused by impairment in signaling downstream from the receptor. This impairment includes diminished activity of the insulin receptor tyrosine kinase, PI3K activity, and perhaps other steps along the pathway to GLUT4 recruitment to the plasma membrane (see Fig. 51-6). It is the summation of these multiple defects, only some of which have been identified, that leads to insulin resistance.
FIGURE 51-7 Response to insulin of normal and downregulated adipocytes.
Diabetes is the most common serious metabolic disease in humans. The hallmark of diabetes is an elevated blood glucose concentration, but this abnormality is just one of many biochemical and physiological alterations. Diabetes is not one disorder, but can arise as a result of numerous defects in regulation of the synthesis, secretion, and action of insulin. The type of diabetes that most commonly affects children is called type 1 IDDM. The diabetes that generally begins in adulthood and is particularly common in obese individuals is called type 2 or NIDDM.
Type 1 Diabetes N51-4
Type 1 diabetes is caused by an immune-mediated selective destruction of the β cells of the pancreas. The other cell types present in the islet are spared. The consequence of the loss of insulin, with the preservation of glucagon, can be viewed as an accelerated form of fasting or starvation. A healthy person who is fasting for several days continues to secrete insulin at a low rate that is sufficient to balance the action of glucagon in modulating the production of glucose and ketones by the liver. However, in type 1 diabetes, insulin deficiency is severe, and glucose and ketone production by the liver occur at a rate that greatly exceeds the rate at which they are being used. As a result, the concentration of these substances in blood begins to rise. Even when glucose concentrations reach levels 5 to 10 times normal, no insulin is secreted because β cells are absent. The increased glucose and ketones provide an immense solute load to the kidney that causes osmotic diuresis. In addition, the keto acids that are produced are moderately strong organic acids (pK < 4.0), and their increased production causes severe metabolic acidosis (see p. 635). If these patients are not treated with insulin, the acidosis and dehydration lead to death from diabetic ketoacidosis.
With appropriate diagnosis and the availability of insulin as an effective treatment, persons with type 1 diabetes can lead full, productive lives. Indeed, some patients have been taking insulin successfully for treatment of type 1 diabetes for >75 years. As technology has improved, patients have been able to monitor their blood glucose levels themselves and adjust their insulin dosages accordingly, using specifically designed insulin analogs that have either short or very long half-lives, or insulin pumps that continuously deliver insulin via a subcutaneous catheter. Thus, individuals with type 1 diabetes can avoid not only severe, life-threatening episodes of ketoacidosis but also the long-term consequences of diabetes—namely, blood vessel injury that can lead to blindness, kidney failure, and accelerated atherosclerosis.
Type 2 Diabetes
In type 2 diabetes, the cause of hyperglycemia is more complex. These individuals continue to make insulin. β cells not only are present but also are frequently hyperplastic (at least early in the course of the disease). For reasons still being defined, the β cells do not respond normally to increases in plasma glucose level by increasing insulin secretion. However, altered insulin secretion is only part of the problem. If we administered identical doses of insulin to the liver, muscle, and adipose tissue of a person with type 2 diabetes and a healthy control, we would find that the patient with type 2 diabetes is resistant to the action of insulin. Thus, both the secretion of insulin and the metabolism of glucose in response to insulin are abnormal in type 2 diabetes. Which problem—decreased insulin release or insulin resistance—is more important in provoking development of the diabetic state likely varies among individuals. Usually, these patients make enough insulin—and it is sufficiently active—that the severe ketoacidosis described above in patients with type 1 diabetes does not develop.
The insulin resistance seen in individuals with type 2 diabetes appears to bring with it an increase in the prevalence of hypertension, obesity, and a specific dyslipidemia characterized by elevated TAGs and depressed high-density lipoproteins (see Fig. 46-15). Insulin resistance (along with one or more of these other metabolic abnormalities) is frequently found in individuals before the development of type 2 diabetes and is referred to as metabolic syndrome. This constellation of abnormalities is estimated to affect >45 million individuals in the United States alone. Because each component of this syndrome has adverse effects on blood vessels, these individuals are at particularly increased risk of early atherosclerosis.
Tight control of glucose concentrations in both type 1 and type 2 diabetes, together with careful management of blood pressure and plasma lipids, can retard the development of many of the long-term complications of diabetes.
In liver, insulin promotes conversion of glucose to glycogen stores or to triacylglycerols
Insulin's actions on cellular targets frequently involve numerous tissue-specific enzymatic and structural processes. As we will see in this and the next two sections, the three principal targets for insulin action are liver, muscle, and adipose tissue.
Because the pancreatic veins drain into the portal venous system, all hormones secreted by the pancreas must traverse the liver before entering the systemic circulation. For insulin, the liver is both a target tissue for hormone action and a major site of degradation.
The concentration of insulin in portal venous blood before extraction by the liver is three to four times greater than its concentration in the systemic circulation. The hepatocyte is therefore bathed in a relatively high concentration of insulin and is thus well positioned to respond acutely to changes in plasma [insulin].
After feeding, the plasma [insulin] rises, triggered by glucose and by neural and incretin stimulation of β cells. In the liver, this insulin rise acts on four main processes involved in fuel metabolism. These divergent effects of insulin entail the use of multiple enzymatic control mechanisms, indicated by numbered boxes in Figure 51-8.
FIGURE 51-8 Effect of insulin on hepatocytes. Insulin has four major effects on liver cells. First, insulin promotes glycogen synthesis from glucose by enhancing the transcription of glucokinase (1) and by activating glycogen synthase (2). Additionally, insulin together with glucose inhibits glycogen breakdown to glucose by diminishing the activity of G6Pase (4). Glucose also inhibits glycogen phosphorylase (3). Second, insulin promotes glycolysis and carbohydrate oxidation by increasing the activity of glucokinase (1), phosphofructokinase (5), and pyruvate kinase (6). Insulin also promotes glucose metabolism via the hexose monophosphate shunt (7). Finally, insulin promotes the oxidation of pyruvate by stimulating pyruvate dehydrogenase (8). Insulin also inhibits gluconeogenesis by inhibiting the activity of PEPCK (9), fructose-1,6-bisphosphatase (10), and G6Pase (4). Third, insulin promotes the synthesis and storage of fats by increasing the activity of acetyl CoA carboxylase (11) and fatty-acid synthase (12) as well as the synthesis of several apoproteins packaged with VLDL. Insulin also indirectly inhibits fat oxidation because the increased levels of malonyl CoA inhibit CAT I (13). The inhibition of fat oxidation helps shunt fatty acids to esterification as TAGs and storage as VLDL or lipid droplets. Fourth, by mechanisms that are not well understood, insulin promotes protein synthesis (14) and inhibits protein breakdown (15).
Glycogen Synthesis and Glycogenolysis
Physiological increases in plasma [insulin] decrease the breakdown and utilization of glycogen and—conversely—promote the formation of glycogen from plasma glucose. Although moderately increased levels of insulin allow gluconeogenesis to persist, the hepatocytes store the gluconeogenic product—glucose-6-phosphate—as glycogen rather than releasing it as glucose into the bloodstream. At high concentrations, insulin can inhibit the gluconeogenic conversion of lactate/pyruvate and amino acids to glucose-6-phosphate.
Glucose enters the hepatocyte from the blood via GLUT2, which mediates the facilitated diffusion of glucose. GLUT2 is present in abundance in the liver plasma membrane, even in the absence of insulin, and its activity is not influenced by insulin. Insulin stimulates glycogen synthesis from glucose by activating glucokinase (numbered box 1 in Fig. 51-8) and glycogen synthase (box 2). The latter enzyme contains multiple serine phosphorylation sites. Insulin causes a net dephosphorylation of the protein, thus increasing the enzyme's activity. At the same time that glycogen synthase is being activated, increases in both insulin and glucose diminish the activity of glycogen phosphorylase (box 3). This enzyme is rate limiting for the breakdown of glycogen. The same enzyme that dephosphorylates (and thus activates) glycogen synthase also dephosphorylates (and thus inhibits) phosphorylase. Thus, insulin has opposite effects on the opposing enzymes, with the net effect that it promotes glycogen formation. Insulin also inhibits glucose-6-phosphatase (G6Pase; box 4), which otherwise converts glucose-6-phosphate (derived either from glycogenolysis or gluconeogenesis) to glucose. Glycogen is an important storage form of carbohydrate in both liver and muscle. The glycogen stored during the postprandial period is then available for use many hours later as a source of glucose.
Glycolysis and Gluconeogenesis
Insulin promotes the conversion of some of the glucose taken up by the liver into pyruvate and—conversely—diminishes the use of pyruvate and other three-carbon compounds for gluconeogenesis. Insulin induces transcription of the glucokinase gene (numbered box 1 in Fig. 51-8) and thus results in increased synthesis of this enzyme, which is responsible for phosphorylating glucose to glucose-6-phosphate and initiating the metabolism of glucose. In acting to promote glycolysis and diminish gluconeogenesis, insulin induces the synthesis of a glucose metabolite, fructose-2,6-bisphosphate. This compound is a potent allosteric activator of phosphofructokinase (box 5), a key regulatory enzyme in glycolysis. Insulin also stimulates pyruvate kinase (box 6), which forms pyruvate, and stimulates pyruvate dehydrogenase (box 8), which catalyzes the first step in pyruvate oxidation. Finally, insulin promotes glucose metabolism by the hexose monophosphate shunt (box 7). N51-5
In addition, insulin also inhibits gluconeogenesis at several steps. Insulin diminishes transcription of the gene encoding phosphoenolpyruvate carboxykinase (PEPCK; numbered box 9 in Fig. 51-8), thus reducing the synthesis of a key regulatory enzyme required to form phosphoenolpyruvate from oxaloacetate early in the gluconeogenic pathway. The increased levels of fructose-2,6-bisphosphate also inhibit the activity of fructose-1,6-bisphosphatase (box 10), which is also part of the gluconeogenic pathway.
Insulin promotes the storage of fats and inhibits the oxidation of fatty acids (see Fig. 58-10) through allosteric and covalent modification of key regulatory enzymes, as well as by transcription of new enzymes (numbered boxes in Fig. 51-8). The pyruvate that is now available from glycolysis can be used to synthesize fatty acids. Insulin promotes dephosphorylation of acetyl coenzyme A (CoA) carboxylase 2 (ACC2; box 11), the first committed step in fatty-acid synthesis in the liver. This dephosphorylation leads to increased synthesis of malonyl CoA, which allosterically inhibits carnitine acyltransferase I (CAT I; box 13). This enzyme converts acyl CoA and carnitine to acylcarnitine, a reaction necessary for long-chain fatty acids to cross the inner mitochondrial membrane, where they can be oxidized. Thus, malonyl CoA inhibits fatty-acid transport and fat oxidation. At the same time, insulin stimulates fatty-acid synthase (box 12), which generates fatty acids. Thus, because insulin promotes the formation of malonyl CoA and fatty acids but inhibits fatty-acid oxidation, this hormone favors esterification of the fatty acids with glycerol within the liver to form triacylglycerols (TAGs). The liver can either store these TAGs in lipid droplets or export them as very-low-density lipoprotein (VLDL) particles (see p. 968). Insulin also induces the synthesis of several of the apoproteins that are packaged with the VLDL particle. The hepatocyte then releases these VLDLs, which leave the liver via the hepatic vein. Muscle and adipose tissue subsequently take up the lipids in these VLDL particles and either store them or oxidize them for fuel. Thus, by regulation of transcription, by allosteric activation, and by regulation of protein phosphorylation, insulin acts to promote the synthesis and storage of fat and diminish its oxidation in liver. N51-8
Nonalcoholic Fatty Liver Disease
Contributed by Fred Suchy
There are differences in the hepatic metabolism of glucose and fructose of importance to human health. Hepatic glucose metabolism is tightly regulated by phosphofructokinase, which is inhibited by ATP and citrate. Thus, when energy status is sufficient, hepatic uptake of dietary glucose is inhibited and much of the consumed glucose will bypass the liver and reach the systemic circulation. In contrast, dietary fructose is metabolized to fructose-1-phosphate by fructokinase, which is not regulated by hepatic energy status and the inhibitory effects of high ATP and citrate levels. Thus, fructose uptake and metabolism by the liver is unregulated, and relatively little of ingested fructose reaches the systemic circulation. In the liver, the large fructose load can result in increased de novo lipogenesis and inhibition of fatty-acid oxidation. This process contributes to the development of hepatic insulin resistance and nonalcoholic fatty liver disease (NAFLD). Owing to the current epidemic of obesity, NAFLD is now the most common liver disorder in adults and children.
Insulin stimulates the synthesis of protein and simultaneously reduces the degradation of protein within the liver (numbered boxes in Fig. 51-8). The general mechanisms by which insulin stimulates protein synthesis (box 14) and restrains proteolysis (box 15) by the liver are complex and are less well understood than the mechanisms regulating carbohydrate and lipid metabolism.
In summary, insulin modulates the activity of multiple regulatory enzymes, which are responsible for the hepatic metabolism of carbohydrates, fat, and protein. Insulin causes the liver to take up glucose from the blood and either store the glucose as glycogen or break it down into pyruvate. The pyruvate provides the building blocks for storage of the glucose carbon atoms as fat. Insulin also diminishes the oxidation of fat, which normally supplies much of the ATP used by the liver. As a result, insulin causes the liver, as well as other body tissues, to burn carbohydrates preferentially.
In muscle, insulin promotes the uptake of glucose and its storage as glycogen
Muscle is a major insulin-sensitive tissue and the principal site of insulin-mediated glucose disposal. Insulin has four major effects on muscle.
First, in muscle, unlike in the liver, glucose crosses the plasma membrane principally via GLUT4, an insulin-sensitive glucose transporter. GLUT4, which is found virtually exclusively in striated muscle and adipose tissue, belongs to a family of proteins that mediate the facilitated diffusion of glucose (see p. 114). Insulin markedly stimulates GLUT4 in both muscle (Fig. 51-9) and fat (see below) by a process involving recruitment of preformed transporters from a membranous compartment in the cell cytosol out to the plasma membrane. Recruitment places additional glucose transporters in the plasma membrane, thereby increasing the Vmax of glucose transport into muscle and increasing the flow of glucose from the interstitial fluid to the cytosol. As discussed above, a different glucose transporter, GLUT2, mediates glucose transport into hepatocytes (see Fig. 51-8) and β cells, and insulin does not increase the activity of that transporter.
FIGURE 51-9 Effect of insulin on muscle. Insulin has four major effects on muscle cells. First, insulin promotes glucose uptake by recruiting GLUT4 transporters to the plasma membrane. Second, insulin promotes glycogen synthesis from glucose by enhancing the transcription of hexokinase (1) and by activating glycogen synthase (2). Third, insulin promotes glycolysis and carbohydrate oxidation by increasing the activity of hexokinase (1), phosphofructokinase (3), and pyruvate dehydrogenase (4). These actions are similar to those in liver; note that there is little or no gluconeogenesis in muscle. Fourth, insulin promotes protein synthesis (5) and inhibits protein breakdown (6).
The enzymatic steps regulated by insulin are indicated by numbered boxes in Figure 51-9.
The second effect of insulin on muscle is to enhance the conversion of glucose to glycogen by activating hexokinase (numbered box 1 in Fig. 51-9)—different from the glucokinase in liver—and glycogen synthase (box 2).
Third, insulin increases glycolysis and oxidation by increasing the activity of phosphofructokinase (box 3) and pyruvate dehydrogenase (box 4).
Fourth, insulin also stimulates the synthesis of protein in skeletal muscle (box 5) and slows the degradation of existing proteins (box 6). The result is preservation of muscle protein mass, which has obvious beneficial effects in preserving strength and locomotion. The insulin-induced increase in glucose utilization permits the muscle to diminish fat utilization and allows it to store as TAGs some of the fatty acid that it removes from the circulation. The stored TAGs and glycogen are a major sources of energy that muscle can use later when called on to exercise or generate heat.
Exercise and insulin have some interesting parallel effects on skeletal muscle. Both increase the recruitment of GLUT4 transporters to the sarcolemma and both increase glucose oxidation; therefore, both increase glucose uptake by muscle. Additionally, exercise and insulin appear to have synergistic effects on the above processes. Clinically, this synergism is manifest as a marked increase in insulin sensitivity induced by exercise and is exploited as part of the treatment of patients with diabetes mellitus.
In muscle, as in the liver, insulin directs the overall pattern of cellular fuel metabolism by acting at multiple sites. In both tissues, insulin increases the oxidation of carbohydrate, thus preserving body protein and fat stores. Carbohydrate ingested in excess of that used immediately as an oxidative fuel is either stored as glycogen in liver and muscle or is converted to lipid in the liver and exported to adipose tissue and muscle.
In adipocytes, insulin promotes glucose uptake and conversion to TAGs for storage
Adipose tissue is the third major insulin-sensitive tissue involved in the regulation of body fuel. Again, insulin has several sites of action in adipocytes. All begin with the same receptor-mediated action of insulin to stimulate several cellular effector pathways. Insulin has four major actions on adipocytes.
First, like muscle, adipose tissue contains the insulin-sensitive GLUT4 glucose transporter. In insulin-stimulated cells, preformed transporters are recruited from an intracellular compartment to the cell membrane, which markedly accelerates the entry of glucose into the cell.
Second, insulin promotes the breakdown of glucose to metabolites that will eventually be used to synthesize TAGs. Unlike in muscle or liver, little of the glucose taken up is stored as glycogen. Instead, the adipocyte glycolytically metabolizes much of the glucose to α-glycerol phosphate, which it uses to esterify long-chain fatty acids into TAGs. The glucose not used for esterification goes on to form acetyl CoA and then malonyl CoA and fatty acids. Insulin enhances this flow of glucose to fatty acids by stimulating pyruvate dehydrogenase (numbered box 1 in Fig. 51-10) and acetyl CoA carboxylase (box 2).
FIGURE 51-10 Effect of insulin on adipocytes. Insulin has four major effects on adipocytes. First, insulin promotes glucose uptake by recruiting GLUT4 transporters to the plasma membrane. Second, insulin promotes glycolysis, which leads to the formation of α-glycerol phosphate. Insulin also promotes the conversion of pyruvate to fatty acids by stimulating pyruvate dehydrogenase (1) and acetyl CoA carboxylase (2). Third, insulin promotes the esterification of α-glycerol phosphate with fatty acids to form TAGs, which the adipocyte stores in fat droplets. Conversely, insulin inhibits HSL (3), which would otherwise break the TAGs down into glycerol and fatty acids. Fourth, insulin promotes the synthesis of LPL in the adipocyte. The adipocyte then exports this enzyme to the endothelial cell, where it breaks down the TAGs contained in chylomicrons and VLDL, yielding fatty acids. These fatty acids then enter the adipocyte for esterification and storage in fat droplets as TAGs.
Third, insulin promotes the formation of TAGs by simple mass action; the increased levels of α-glycerol phosphate increase its esterification with fatty acids (principally C-16 and C-18) to yield TAGs. Some of the fatty acids are a result of the glucose metabolism noted above. Most of the fatty acids, however, enter the adipocyte from chylomicrons and VLDLs (see Table 46-4) in the blood. The cell sequesters these TAGs in lipid droplets, which form most of the mass of the adipose cell. Conversely, insulin restrains the activity of adipose triacylglycerol lipase (ATGL; see p. 1182), which converts TAGs to diacylglycerols (DAGs), and hormone-sensitive lipase (HSL), which converts DAGs to monoacylglycerols (MAGs). In fat, these enzymes (numbered box 3 in Fig. 51-10) mediate the conversion of stored TAGs to fatty acids and glycerol for export to other tissues.
Fourth, insulin induces the synthesis of a different enzyme—lipoprotein lipase (LPL). This lipase does not act on the lipid stored within the adipose cell. Rather, the adipocyte exports the LPL to the endothelial cell, where it resides on the extracellular surface of the endothelial cell, facing the blood and anchored to the plasma membrane. In this location, the LPL acts on TAGs in chylomicrons and VLDLs and cleaves them into glycerol and fatty acids. These fatty acids are then available for uptake by nearby adipocytes, which esterify them with glycerol phosphate to form TAGs. This mechanism provides an efficient means by which insulin can promote the storage of lipid in adipose tissue.