Glucokinase

Glucokinase () is an enzyme that facilitates phosphorylation of glucose to glucose-6-phosphate. Glucokinase occurs in cells in the liver, pancreas, gut, and brain of humans and most other vertebrates. In each of these organs it plays an important role in the regulation of carbohydrate metabolism by acting as a glucose sensor, triggering shifts in metabolism or cell function in response to rising or falling levels of glucose. Glucokinase can only phosphorylate glucose if the concentration of this substrate is high enough; its K_m for glucose is 100 times higher than that of hexokinases I, II, and III.

Glucokinase (GK) is a hexokinase isozyme, related homologously and presumably by evolution to at least three other hexokinases. However, it is coded by a separate gene and its distinctive kinetic properties allow it to serve a very different set of functions. All of the hexokinases can mediate phosphorylation of glucose to glucose-6-phosphate (G6P), which is the first step of both glycogen synthesis and glycolysis. However, the lower affinity for glucose of glucokinase compared to the other hexokinases, and the restricted localization of glucokinase to a few cell types leaves the role of the other three hexokinase forms ascribed to being more important initiators of glycolysis and glycogen synthesis for most tissues and organs.

Substrates and products of GK: glucose is the important substrate

The principal substrate of physiologic importance is glucose, and the most important product is glucose-6-phosphate (G6P). The other necessary substrate, from which the phosphate is derived, is ATP. A simplified version of the glucokinase reaction is:

Glucose + ATP → glucose-6-phosphate + ADP

ATP participates in the reaction in a form complexed to magnesium as a cofactor. Furthermore, under certain conditions, glucokinase, like other hexokinases, can induce phosphorylation of other hexoses (6 carbon sugars) and similar molecules. Therefore the general glucokinase reaction is more accurately described as:

Hexose + MgATP^{2-^{→ hexose-PO_{3_{^{2-^{+ MgADP^{-^{+ H⁺}}}}}}}}

Among the hexose substrates are mannose, fructose, and glucosamine, but the affinity of glucokinase for these requires concentrations not found in cells for significant activity.

Terminology: physiologists and physicians prefer glucokinase

Alternative names for this enzyme are human hexokinase IV, hexokinase D, and ATP:D-hexose 6-phosphotransferase, EC 2.7.1.1 (previously 2.7.1.2). The common name, glucokinase, is derived from its relative specificity for glucose under physiologic conditions. Some biochemists have unsuccessfully argued that the name glucokinase should be abandoned as misleading, as this enzyme can phosphorylate other hexoses in the right conditions, and there are distantly related enzymes in bacteria with more absolute specificity for glucose which better deserve the name and the EC 2.7.1.2. Nevertheless, glucokinase remains the name preferred in the contexts of medicine and mammalian physiology.

Characteristics of the glucokinase molecule: a monomer with an actin fold

Glucokinase is a monomeric protein of 465 amino acids and a molecular weight of about 50 kD. There are at least two clefts, one for the active site, binding glucose and MgATP, and the other for a putative allosteric activator which has not yet been identified.

This is about half the size of the other mammalian hexokinases, which retain a degree of dimeric structure. Several sequences and the three-dimensional structure of the key active sites-- the ATP binding domain, for example,-- are shared with hexokinases, bacterial glucokinases, and other proteins, and the common structure is termed an actin fold.

Kinetic and functional properties of GK: low glucose affinity but uninhibited

Two important kinetic properties distinguish glucokinase from the other hexokinases, allowing it to function in a special role as glucose sensor.

Glucokinase has a lower affinity for glucose than the other hexokinases. Glucokinase changes conformation and/or function in parallel with rising glucose concentrations in the physiologically important range of 4-10 mmol/L (72-180 mg/dl). It is half-saturated at a glucose concentration of about 8 mmol/L (144 mg/dl).
Glucokinase is not inhibited by its product, glucose-6-phosphate. This allows continued signal output (e.g, to trigger insulin release) amid significant amounts of its product.

These two features allow it to regulate a "supply-driven" metabolic pathway: i.e., the flow through the pathway is driven by the glucose supply, not the demand for end-products.

Another distinctive property of glucokinase is its moderate cooperativity with glucose, with a Hill coefficient (n_H) of about 1.7. Glucokinase has only a single binding site for glucose and is the only monomeric regulatory enzyme known to display substrate cooperativity. The nature of the cooperativity has been postulated to involve a "slow transition" between two different enzyme states with different rates of activity. If the dominant state depends upon glucose concentration, it would produce an apparent cooperativity similar to that observed.

Because of this cooperativity, the kinetic interaction of glucokinase with glucose does not follow classical Michaelis-Menten kinetics. Rather than a K_m for glucose, it is more accurate to describe a half-saturation level S_0.5, which is the concentration at which the enzyme is 50% saturated and active.

The S_0.5 and nH extrapolate to an "inflection point" of the curve describing enzyme activity as a function of glucose concentration at about 4 mmol/L. In other words, at a glucose concentration of about 72 mg/dl, which is near the low end of the normal range, glucokinase activity is most sensitive to small changes in glucose concentration.

The kinetic relationship with the other substrate, MgATP, can be described by classical Michaelis-Menten kinetics, with an affinity at about 0.3-0.4 mmol/L, well below a typical intracellular concentration of 2.5 mmol/L. The fact that there is nearly always an excess of ATP available implies that ATP concentration rarely infuences glucokinase activity.

The maximum specific activity (K_cat, also known as the turnover rate) of glucokinase when saturated with both substrates is 62/s.

A "minimal mathematical model" has been devised based on the above kinetic information to predict the beta cell glucose phosphorylation rate (BGPR) of normal ("wild type") glucokinase and the known mutations. The BGPR for wild type glucokinase is about 28% at a glucose concentration of 5 mmol/l, indicating that the enzyme is running at 28% of capacity at the usual threshold glucose for triggering insulin release.

Molecular mechanism of catalysis: critically dependent on sulfhydryl groups

The sulfhydryl groups of several cysteines surround the glucose binding site. All except cys 230 are essential for the catalytic process, forming multiple disulfide bridges during interaction with the substrates and regulators. At least in the beta cells, the ratio of active to inactive glucokinase molecules is at least partly determined by the balance of oxidation of sulfhydryl groups or reduction of disulfide bridges.

These sulfhydryl groups are quite sensitive to the oxidation status of the cells, making glucokinase one of the components most vulnerable to oxidative stress, especially in the beta cells.

Genetic aspects of GK: one gene but two promoters

The human glucokinase is coded for by the GCK gene on chromosome 7. This single autosomal gene has 10 exons.

The most distinctive feature of the gene is that it begins with two promoter regions. The first exon from the 5' end contains two tissue-specific promoter regions. Transcription can begin at either promoter (depending on the tissue) so that the same gene can produce a slightly different molecule in liver and in other tissues. The two isoforms of glucokinase differ only by 13-15 amino acids at the N-terminal end of the molecule, which produces only a minimal difference in structure. The two isoforms have the same kinetic and functional characteristics.

The first promoter from the 5' end, referred to as the "upstream" or neuroendocrine promoter, is active in pancreatic islet cells, neural tissue, and enterocytes (small intestine cells) to produce the "neuroendocrine isoform" of glucokinase. The second promoter, the "downstream" or liver promoter, is active in hepatocytes and directs production of the "liver isoform". The two promoters have little or no sequence homology and are separated by a 30 kbp sequence of unknown function.

The two promoters are functionally exclusive and governed by distinct sets of regulatory factors, so that glucokinase expression can be controlled separately in the different tissue types. Some of the factors involved in control of transcription are mentioned below, but the full regulatory system has not been fully elucidated. It is also notable that the two promoters correspond to two broad categories of glucokinase function: in liver glucokinase acts as the gateway for the "bulk processing" of available glucose, while in the neuroendocrine cells, it acts as a sensor, triggering cell responses that affect body-wide carbohydrate metabolism.

Tissue distribution of GK: key locations for carbohydrate metabolism

Glucokinase has been discovered in specific cells in four types of mammalian tissue. All play crucial roles in responding to rising or falling levels of blood glucose.

Hepatocytes are the predominant cell in the liver parenchyma. The liver takes up glucose and stores it as glycogen after a carbohydrate meal when blood glucose is plentiful. During fasting conditions, the liver releases glucose from glycogen, generates glucose by gluconeogenesis, and exports it into the blood. The separate liver promoter allows glucokinase to be regulated differently in hepatocytes than in the neuroendocrine cells.
Neuroendocrine cells of the pancreas, gut, and brain share some common aspects of glucokinase production, regulation, and function. These tissues are collectively referred to as "neuroendocrine" cells in this context.
- Beta cells and alpha cells of the pancreatic islets
  - Beta cells release insulin in response to rising levels of glucose. Insulin enables many types of cells to import and use glucose, and signals the liver to synthesize glycogen. Alpha cells produce less glucagon in response to rising glucose levels, and more glucagon if blood glucose is low. Glucagon serves as a signal to the liver to break down glycogen and release glucose into the blood.
- Glucose-sensitive neurons of the hypothalamus
  - In response to rising or falling levels of glucose, cells in the hypothalamus polarize or depolarize. Among the neuroendocrine reactions of the central nervous system to hypoglycemia is activation of the adrenergic responses of the autonomic nervous system.
- Enterocytes of the small intestine
  - This is the least well understood of the glucokinase sensor systems. It seems likely that responses to incoming glucose during digestion play a role in the incretin amplification of insulin secretion during a meal, or in the generation of satiety signals from gut to brain.

Function and regulation of GK in the liver: initiator of glycogen synthesis

Most of the glucokinase in a mammal is found in the liver, and glucokinase provides approximately 95% of the hexokinase activity in hepatocytes. Phosphorylation of glucose to glucose-6-phosphate (G6P) by glucokinase is the first step of both glycogen synthesis and glycolysis in the liver. These are two of the anabolic responses to a carbohydrate meal.

When ample glucose is available, glycogen synthesis proceeds at the periphery of the hepatocytes until the cells are replete with glycogen. Excess glucose is then increasingly converted into triglycerides for export and storage in adipose tissue. Glucokinase activity in the cytoplasm rises and falls with available glucose.

G6P, the product of glucokinase, is the principal substrate of glycogen synthesis, and glucokinase has a close functional and regulatory association with glycogen synthesis. When maximally active, GK and glycogen synthase appears to be located in the same peripheral areas of hepatocyte cytoplasm in which glycogen synthesis occurs. The supply of G6P affects the rate of glycogen synthesis not only as the primary substrate, but by direct stimulation of glycogen synthase and inhibition of glycogen phosphorylase.

Glucokinase activity can be rapidly amplified or damped in response to changes in the glucose supply, typically resulting from eating and fasting. Regulation occurs at several levels and speeds, and is influenced by many factors which mainly affect two general mechanisms:

Glucokinase activity can be amplified or reduced in minutes by actions of the glucokinase regulatory protein (GKRP). The actions of this protein are influenced by small molecules such as glucose and fructose.
The amount of glucokinase can be increased by synthesis of new protein. Insulin is the principal signal for increased transcription, operating mainly by way of a transcription factor called sterol regulatory element binding protein-1c (SREBP1c). This occurs within an hour after a rise in insulin levels, as after a carbohydrate meal.

Rapid regulation of GK activity in hepatocytes

Glucokinase can be rapidly activated and inactivated in hepatocytes by a novel regulatory protein (glucokinase regulatory protein), which operates to maintain an inactive reserve of GK which can be made quickly available in response to rising levels of portal vein glucose.

GKRP moves between nucleus and cytoplasm of the hepatocytes and may be tethered to the microfilament cytoskeleton. It forms reversible 1:1 complexes with GK, and can move it from the cytoplasm into the nucleus. It acts as a competitive inhibitor with glucose, such that the enzyme activity is reduced to near zero while bound. GK:GKRP complexes are sequestered in the nucleus while glucose and fructose levels are low. Nuclear sequestration may serve to protect GK from degradation by cytoplasmic proteases. GK can be rapidly released from GKRP in response to rising levels of glucose. Unlike GK in beta cells, GK in hepatocytes is not associated with mitochondria.

Fructose in tiny (micromolar) amounts (after phosphorylation by ketohexokinase to fructose-1-phosphate (F1P)) accelerates release of GK from GKRP. This sensitivity to the presence of small amounts of fructose allows GKRP, GK, and ketohexokinase to act as a "fructose sensing system" which signals that a mixed carbohydrate meal is being digested and accelerates the utilization of glucose. Conversely, fructose-6-phosphate (F6P) potentiates binding of GK by GKRP. F6P decreases phosphorylation of glucose by GK when glycogenolysis or gluconeogenesis are underway. F1P and F6P both bind to the same site on GKRP. It is postulated that they produce 2 different conformations of GKRP, one able to bind GK and the other not.

Transcriptional regulation of GK in hepatocytes

Insulin acting via the sterol regulatory element binding protein-1c (SREBP1c) is thought to be the most important direct activator of glucokinase gene transcription in hepatocytes. SREBP1c is a basic helix-loop-helix zipper (bHLHZ) transactivator. This class of transactivators bind to the "E box" sequence of genes for a number of regulatory enzymes. The liver promoter in the first exon of the glucokinase gene includes such an E box which appears to be the principal insulin-response element of the gene in hepatocytes. SREBP1c must be present for transcription of glucokinase in hepatocytes. SREBP1c increases in response to a high carbohydrate diet, probably as a direct effect of frequent insulin elevation. Increased transcription can be detected in less than an hour after hepatocytes are exposed to rising insulin levels.

Fructose-2,6-bisphosphate (F2,6P₂) also stimulates GCK transcription, apparently by way of Akt2 rather than SREBP1c. It is not known whether this effect is one of the downstream effects of activation of insulin receptors or independent of insulin action. Levels of F2,6P₂ play other amplifying roles in glycolysis in hepatocytes.

Other transacting factors suspected of playing a role in liver cell transcription regulation include:

Hepatic nuclear factor-4-alpha (HNF4α) is an orphan nuclear receptor important in the transcription of many genes for enzymes of carbohydrate and lipid metabolism. It activates GCK transcription.
Upstream stimulatory factor 1 (USF1) is another basic helix-loop-helix zipper (bHLHZ) transactivator.
Hepatic nuclear factor 6 (HNF6) is a homeodomain transcriptional regulator of the "one-cut class." HNF6 is also involved in regulation of transcription of gluconeogenic enzymes such as glucose-6-phosphatase and phosphoenolpyruvate carboxykinase.

Hormonal and dietary regulation of glucokinase in hepatocytes

Insulin is by far the most important of the hormones that have direct or indirect effects on glucokinase expression and activity in the liver. Insulin appears to affect both glucokinase transcription and activity through multiple direct and indirect pathways. While rising portal vein glucose levels increase glucokinase activity, the concomitant rise of insulin amplifies this effect by induction of glucokinase synthesis. Glucokinase transcription begins to rise within an hour of rising insulin levels. Glucokinase transcription becomes nearly undetectable in prolonged starvation, severe carbohydrate deprivation, or untreated insulin-deficient diabetes.

The mechanisms by which insulin induces glucokinase may involve both of the major intracellular pathways of insulin action, the extracellular signal-regulated kinase (ERK 1/2) cascade, and the phosphoinositide 3-kinase (PI3-K) cascade. The latter may operate via the FOXO1 transactivator.

Conversely, as would be expected given its antagonistic effect on glycogen synthesis, glucagon and its intracellular second messenger cAMP suppresses glucokinase transcription and activity, even in the presence of insulin.

Other hormones such as triiodothyronine (T₃) and glucocorticoids provide permissive or stimulatory effects on glucokinase in certain circumstances. Biotin and retinoic acid increase GCK mRNA transcription as well as GK activity. Fatty acids in significant amounts amplify GK activity in the liver, while long chain acyl CoA inhibits it.

Function and regulation in pancreas and other neuroendocrine tissue

Although most of the glucokinase in the body is in the liver, smaller amounts in the beta and alpha cells of the pancreas, certain hypothalamic neurons, and specific cells (enterocytes) of the gut play an increasingly appreciated role in regulation of carbohydrate metabolism. In the context of glucokinase function, these cell types are collectively referred to as neuroendocrine tissues and they share some aspects of glucokinase regulation and function, especially the common neuroendocrine promoter. Of the neuroendocrine cells, the beta cells of the pancreatic islets are the most studied and best understood. It is likely that many of the regulatory relationships discovered in the beta cells will also exist in the other neuroendocrine tissues with glucokinase.

GK in beta cells: a signal for insulin

In islet beta cells, glucokinase activity serves as a principal control for the secretion of insulin in response to rising levels of blood glucose. As G6P is consumed, increasing amounts of ATP initiate a series of processes that result in release of insulin. One of the immediate consequences of increased cellular respiration is a rise in the NADH and NADPH concentrations (collectively referred to as NAD(P)H). This shift in the redox status of the beta cells results in rising intracellular calcium levels, closing of the K_ATP channels, depolarization of the cell membrane, merging of the insulin secretory granules with the membrane, and release of insulin into the blood.

It is as a signal for insulin release that glucokinase exerts the largest effect on blood sugar levels and overall direction of carbohydrate metabolism. Glucose, in turn, influences both the immediate activity and the amount of glucokinase produced in the beta cells.

Rapid regulation of glucokinase activity in beta cells

Glucose immediately amplifies glucokinase activity by the cooperativity effect.

A second important rapid regulator of glucokinase activity in beta cells occurs by direct protein-protein interaction between glucokinase and the "bifunctional enzyme" (phosphofructokinase-2/fructose-2,6-bisphosphatase) which also plays a role in the regulation of glycolysis. This physical association stabilizes glucokinase in a catalytically favorable conformation (somewhat opposite the effect of GKRP binding) that enhances its activity.

Association of glucokinase with insulin secretory granules

Much, but not all, of the glucokinase found in the cytoplasm of beta cells is associated with insulin secretory granules and with mitochondria. The proportion thus "bound" falls rapidly in response to rising glucose and insulin secretion. It has been suggested that binding serves a purpose similar to the hepatic glucokinase regulatory protein-- protecting glucokinase from degradation so that it is rapidly available as the glucose rises. The effect is to amplify the glucokinase response to glucose more rapidly than transcription could do so.

Transcriptional regulation in pancreatic beta cells

In as little as 15 minutes, glucose can stimulate GCK transcription and glucokinase synthesis by way of insulin. Insulin is produced by the beta cells, but some of it acts on beta cell B-type insulin receptors, providing an autocrine positive-feedback amplification of glucokinase activity. Further amplification occurs by insulin action (via A-type receptors) to stimulate its own transcription.

Transcription of the GCK gene is initiated through the "upstream," or neuroendocrine, promoter. This promoter, in contrast to the liver promoter, has elements homologous to other insulin-induced gene promoters. Among the probable transacting factors are Pdx-1 and PPARγ. Pdx-1 is a homeodomain transcription factor involved in the differentiation of the pancreas. PPARγ is a nuclear receptor which responds to glitazone drugs by enhancing insulin sensitivity.

GK in alpha cells: suppression of glucagon?

It has also been proposed that glucokinase plays a role in the glucose sensing of the pancreatic alpha cells, but the evidence is less consistent and some researchers have found no evidence of glucokinase activity in these cells. Alpha cells occur in pancreatic islets, mixed with beta and other cells. While beta cells respond to rising glucose levels by secreting insulin, alpha cells respond by reducing glucagon secretion. When blood glucose concentration falls to hypoglycemic levels, alpha cells release glucagon. Glucagon is a protein hormone which blocks the effect of insulin on hepatocytes, inducing glycogenolysis, gluconeogenesis, and reduced glucokinase activity in hepatocytes. The degree to which glucose suppression of glucagon is a direct effect of glucose via glucokinase in alpha cells, or an indirect effect mediated by insulin or other signals from beta cells, is still uncertain.

GK in the brain: glucose sensing in the hypothalamus

While all neurons use glucose for fuel, certain glucose-sensing neurons alter their firing rates in response to rising or falling levels of glucose. These glucose-sensing neurons are concentrated primarily in the ventromedial nucleus and arcuate nucleus of the hypothalamus, which regulate many aspects of glucose homeostasis (especially the response to hypoglycemia), fuel utilization, satiety and appetite, and weight maintenance. These neurons are most sensitive to glucose changes in the 0.5-3.5 mmol/L glucose range.

Glucokinase has been found in the brain in largely the same areas which contain glucose-sensing neurons, including both of the hypothalamic nuclei. Inhibition of glucokinase abolishes the ventromedial nucleus response to a meal. However, brain glucose levels are lower than plasma levels, typically 0.5-3.5 mmol/L. Although this range is matches the sensitivity of the glucose sensing neurons, it is below the optimal inflection sensitivity for glucokinase. The presumption, based on indirect evidence and speculation, is that neuronal glucokinase is somehow exposed to plasma glucose levels even in the neurons.

GK in enterocytes: subserving incretin?

While glucokinase has been shown to occur in certain cells (enterocytes) of the small intestine and stomach, its function and regulation have not been worked out. It has been suggested that here also, glucokinase serves as a glucose sensor, allowing these cells to provide one of the earliest metabolic responses to incoming carbohydrates. It is suspected that these cells are involved in incretin functions.

Effects of disease on glucokinase activity

Because insulin is one of, if not the most important, regulators of glucokinase synthesis, diabetes of all types diminishes glucokinase synthesis and activity by a variety of mechanisms. Glucokinase activity is sensitive to oxidative stress of cells, especially the beta cells.

Diseases caused by glucokinase mutations

Around 200 mutations of the human glucokinase gene GCK have been discovered that can change the efficiency of glucose binding and phosphorylation, increasing or decreasing the sensitivity of beta cell insulin secretion in response to glucose, and producing clinically significant hyperglycemia or hypoglycemia.

Loss of function mutations cause diabetes

Over 190 of these mutations reduce the functional efficiency of the glucokinase molecule. Heterozygosity for alleles with reduced enzyme activity results in a higher threshold for insulin release and persistent, mild hyperglycemia. This condition is referred to as maturity onset diabetes of the young, type 2 (MODY2).

Homozygosity for GCK alleles with reduced function can cause severe congenital insulin deficiency resulting in persistent neonatal diabetes.

Gain of function mutations cause hyperinsulinemic hypoglycemia

As of 2004, 5 mutations have been found to enhance insulin secretion. Heterozygosity for gain of function mutations reduces the threshold glucose that triggers insulin release. This creates hypoglycemia of varying patterns, including transient or persistent congenital hyperinsulinism, or fasting or reactive hypoglycemia appearing at an older age.

Homozygosity for gain of function mutations has not been found.

Glucokinase as a drug target

Several laboratories sponsored by pharmaceutical companies are researching molecules that activate glucokinase in hope that it will be useful in the treatment of type 2 diabetes.

Comparative biology

Liver glucokinase occurs widely but not universally throughout vertebrate species. Its sequence is highly conserved among most mammals (e.g., rat and human glucokinase is more than 80% homologous). However there are some unusual exceptions: for example, it has not been discovered in cats and bats, though some reptiles, birds, and amphibians have it. Whether glucokinase occurs similarly in the pancreas and other organs has not yet been determined. It has been postulated that the presence of glucokinase in liver reflects the ease with which carbohydrates can be included in the animals' diets.

References

Arden C, Harbottle A, Baltrusch S, Tiedge M and Agius L. Glucokinase is an integral component of the insulin granules in glucose-responsive insulin secretory cells and does not translocate during glucose stimulation. Diabetes 53: 2346-2352, 2004.
Cardenas ML. Glucokinase: its regulation and role in liver metabolism. Austin:R.G. Landes Co., 1995. ISBN 1-57059-207-1; International ISBN 3-540-59285-7. This is the most detailed treatment of liver glucokinase.
Kamata K, Mitsuya M, Nishimura T, Eiki J, Nagata Y. Structural basis for allosteric regulation of the monomeric allosteric enzyme human glucokinase. Structure (Camb). 2004 ;12:429-38. Beautiful structural pictures illustrating the conformational changes and potential regulatory mechanisms.
Mahalingam B, Cuesta-Munoz A, Davis EA, Matschinsky FM, Harrison RW, Weber IT. Structural model of human glucokinase in complex with glucose and ATP: implications for the mutants that cause hypo- and hyperglycemia. Diabetes 1999 Sep;48(9):1698-705. This is the best single article review of the molecular pathophysiology of the human diseases resulting from glucokinase abnormalities.
Matschinsky FM, Magnuson MA, eds. Glucokinase and Glycemic Disease: From Basics to Novel Therapeutics. Basel:Karger, 2004. ISBN 3-8055-7744-3. Much of the information in this article comes from this excellent compilation.

EC 2.7.1

Gourt Home::Gourt :: Glucokinase