| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Vanderbilt University School of Medicine (R.L.P., D.K.G.), Nashville, Tennessee 37232, and Department of Veterans Affairs (D.K.G.), Nashville, Tennessee 37212
Address all correspondence and requests for reprints to: Daryl K. Granner, 707 Light Hall, Vanderbilt University School of Medicine, Nashville, Tennessee 37232-0615. E-mail: daryl.granner{at}vanderbilt.edu.
Diabetes mellitus affects more than 170 million individuals worldwide and is increasing at an alarming rate (1). Some 90% of these individuals have type 2 diabetes mellitus (T2DM). The duration and extent of hyperglycemia associated with T2DM is responsible for many of the complications of the disease. This hyperglycemia results from several metabolic defects, including defective insulin secretion, increased hepatic glucose production, and reduced glucose utilization (1, 2). Therapeutic agents have been targeted against each of these abnormal processes, and are often used in combination in an effort to restore euglycemia (1, 3). A single agent that could correct several of these abnormal processes would be a helpful addition to the therapeutic armamentarium.
The normal plasma glucose set point of approximately 5 mM is dependent on the activity of glucokinase (GK), which senses glucose and controls metabolic flux in certain key cell types (4, 5, 6). The limited tissue expression and unique kinetic properties of GK allow it to play a critical role in pancreatic ß-cell insulin secretion, hepatic glucose utilization, and neural/neuroendocrine cellular responses (6, 7). A drug discovery process aimed at increasing the activity of GK is a logical therapeutic objective, and several compounds that activate this enzyme, so-called GK activators (GKAs), have been discovered (6, 8, 9, 10, 11, 12). In this issue, Efanov et al. (13) continue the discovery process with a report that describes the characterization of a novel GKA, LY2121260.
LY2121260 increases the activity of GK throughout the physiological range of glucose concentrations by increasing the maximum velocity of the enzyme reaction and increasing the affinity the enzyme has for glucose. Efanov et al. (13) determined the crystal structure of LY2121260 bound to GK in the presence of glucose in an attempt to explain these findings. They conclude that the compound must activate GK by binding to an allosteric regulatory site. In addition to having an acute effect on GK activity, LY2121260 results in an increased amount of GK in insulinoma cells, presumably by enhancing the ability of glucose to stabilize the enzyme against degradation. The consequences of these actions are an increase of glucose utilization in primary hepatocytes and an increase of glucose-stimulated insulin secretion from pancreatic ß-cells. The net effect is a decrease of fasting plasma glucose and improved glucose tolerance in experimental animals. Results such as this provide the basis for thinking that a new class of therapeutic agents, such as the GKAs, might be useful in the treatment of diabetes. The mechanism of action of new potential therapeutic agents is always of interest, and there is a sound experimental foundation for understanding how agents such as LY2121260 attack such a complicated system as the regulation of GK.
What is the rational basis for attempting to manipulate the glucose sensor for therapeutic gain? Important clues come from studies of humans with GK mutations that have an impact on glucose homeostasis (6, 14). First, some 150 different mutations of the glucokinase gene result in impaired glucose-mediated insulin secretion and a mild, autosomal dominant form of diabetes known as maturity onset diabetes of the young, type 2 (6, 14, 15). Many of these mutations also result in decreased accumulation of hepatic glycogen and in a failure to decrease postprandial hepatic glucose production. Some of these mutations decrease GK activity by reducing the stability of the enzyme or by decreasing the maximum velocity of the enzyme (Vmax). In most instances, they decrease the affinity of the enzyme for its substrates, glucose and/or ATP, which is measured as an increase of the S0.5 (substrate concentration at half-maximum velocity). A more severe disease, permanent neonatal diabetes mellitus-GK, occurs when both alleles of the GK gene are mutated (6, 14, 16, 17).
Second, gain-of-activity GK mutations are linked to a rare subtype of persistent hyperinsulinemic hypoglycemia of infancy (PHHI-GK), which is characterized by a shift of the glucose-stimulated insulin secretion curve to the left. This results in the inappropriate secretion of insulin in the face of hypoglycemia (6, 14, 18, 19, 20). The mutations associated with PHHI-GK reduce the S0.5 for glucose and/or increase Vmax of the enzyme. The glucose vs. velocity curve of wild-type GK shows a sigmoidal response, with a Hill coefficient of approximately 1.5. This indication of cooperativity is reduced or lost in some of the GK mutations that cause PHHI-GK. Interestingly, these mutant enzymes are more like the 100 kDa hexokinases (HKs), which do not show cooperativity for glucose (Hill coefficients of 1.0 and a hyperbolic glucose vs. velocity curve) and bind glucose with higher affinity (a lower S0.5). Some individuals with PHHI-GK have GK mutations that are not located in the known catalytic site of the enzyme and appear to cluster around a separate domain where a natural, or pharmaceutical, activator of enzyme activity could bind and exert an effect on GK. Although a natural ligand (activator) has been sought, none has been found.
Studies of maturity onset diabetes of the young type 2, permanent neonatal diabetes mellitus-GK, and PHHI-GK firmly establish GK as an important glucose sensor in ß-cells and liver (6, 14, 15, 16, 18, 19, 20). In addition, mouse models that directly test the effects of GK mutations on cellular processes are being established and studied (6, 21). This accumulated evidence suggests that adjusting GK activity could be used to regulate the plasma glucose concentration. Several companies have reported initial success in this endeavor (8, 9, 12), and LY2121260 can now be added to the list (13).
How do GKAs tweak GK? Although the exact mechanism of GK activation by GKAs is not known, some inferences may be drawn from previous reports of the crystal structures of GK (10, 12), and as described by Efanov et al. (13). These studies show that the 50-kDa GK is folded into two domains of unequal size, the so-called "large" and "small" domains. These domains are separated by a large cleft, which forms the active site for phosphorylation of glucose at the 6-position. Hexokinase I (HKI), a 100-kDa enzyme composed of two 50-kDa halves, shares this general large and small domain structure in its catalytically active C-terminal half. However, GK has a domain that is not found in HKI and, by inference, in the other HKs. This flexible domain is distinct from the catalytic site, is exposed to solvent when glucose is bound to the enzyme, and forms a hydrophobic pocket located in the hinge region between the large and small domains. LY2121260 binds to this site (13), as do at least two GKAs described previously (10, 12). The corresponding region of HKI (and presumably the other mammalian HKs) is more rigid and has no space for a GKA to bind. This provides an important structural basis for the specificity of GKA action in the HK family.
How do GKAs increase flux through GK? Their most important activity may be to reduce the S0.5 of GK for glucose. GK normally cycles between three conformations, a "super-open" (inactive) conformation at glucose concentrations below the S0.5, an "open" (ready for catalysis) conformation at glucose concentrations above the S0.5, and a "closed" conformation (active catalysis) that forms after glucose and ATP bind (10). The time required to cycle between the super-open and open conformations is longer than that required to move from the open and closed conformations. Some GKAs, such as LY2121260, appear to change the glucose vs. GK activity curve from the characteristic sigmoidal shape (Hill coefficient, >1) to a hyperbolic response that is typically seen with other HKs. A working hypothesis for the mechanism of action of this type of GKA can be derived from the published crystal structure data (10). The previously mentioned small domain is involved. The GKAs bind to an allosteric site and maintain a structure of this small domain that favors high affinity glucose binding, i.e. the open conformation. This open conformation is a prerequisite for catalysis and the GKA prevents the enzyme from cycling back to the low-affinity, inactive, super-open conformation, which is characteristic of GK in the absence of glucose. Therefore, by inference, GK with bound GKA has a shorter catalytic cycle, as only transitions between the open and closed structures occur. Under these circumstances, GK has an increased affinity for glucose and an increased turnover rate of catalytic events. Collectively, these actions of GKAs shift the glucose vs. enzyme activity curve to the left, and, in pancreatic ß-cells, this results in a corresponding leftward shift of glucose-stimulated insulin secretion, as seen by Efanov et al. (13) and others (12).
GKAs can increase GK activity in a cell in other ways. For example, some GKAs, including LY2121260 (13), increase the Vmax of GK, which provides additional activity at a given glucose concentration. An increased amount of GK protein will also increase GK activity in a cell. GK is not a particularly stable enzyme; it is quite easily inactivated by thermal denaturation or oxidation and is subject to degradation (6, 22). GK, with glucose bound, is resistant to these processes. GKAs, by increasing the affinity of GK for glucose, may thus prevent this inactivation and protein degradation. In this case, the concentration of GK would increase in the cell. Efanov et al. (13) observed this end result in islets and islet-derived cell lines treated with LY2121260.
GK has considerable control strength over glucose utilization in the liver, especially in the direction of glycogen synthesis (23). As in ß-cells, hepatic GK activity is regulated by glucose concentration, but an additional mechanism, the interaction of GK with the glucokinase regulatory protein (GKRP), comes into play in the liver (6, 24). GKRP binds to GK and allosterically inhibits the enzyme by decreasing its apparent affinity for glucose. This inhibition of GK is released by fructose 1-phosphate (F1P). Fructose 6-phosphate, when bound to GKRP, blocks F1P from binding, and thereby stabilizes the GKRP-GK interaction. GK, when bound to GKRP, is located primarily in the nucleus in an inactive pool but is available for hepatic glucose metabolism when fructose and other precursors of F1P stimulate GK release from GKRP and stimulate GK translocation from its inactive nuclear pool to the cytoplasm. Increases of plasma glucose and/or insulin also cause a rapid translocation of GK from the nucleus (25). Kinetic studies of GKRP inhibition of GK indicate that GKRP is a competitive inhibitor with respect to glucose (6, 24). Although not explored by Efanov et al. (13), GKAs stabilize the binding of glucose to GK and therefore shift the GK reaction mechanism toward glucose binding and catalytic activity. Thus, GKAs could increase glucose flux in the liver, by counteracting the inhibition of GK by GKRP.
Can this basic science be translated into improved control of hyperglycemia in T2DM? GKAs promote insulin secretion in rodents, as noted above. The increased plasma insulin should reduce hepatic glucose production and increase peripheral glucose utilization. These effects contribute to the reduction of fasting plasma glucose and improvement in glucose tolerance noted in normal animals and various animal models of diabetes, as described by Efanov et al. (13) and others (8, 9, 12). In one study, rats were fasted for 18 h and were then subjected to a pancreatic clamp in which plasma insulin and glucose levels were kept constant (12). In this condition, GKA treatment promoted net hepatic glucose clearance independent of increased plasma insulin. Thus, the GKAs have a definite glucose-lowering action in experimental animals. It will be interesting to see how effective they are in humans.
The GKAs represent a very interesting class of potential therapeutic agents. The general inferences made here about their mechanism of action should be tempered by the possibility that, because several different chemical entities are involved and GK regulation is quite complex, different actions (metabolic effects) may be found for one or more of these compounds. Indeed, even a single agent, such as LY2121260, appears to have more than one action (see above). This is important because persons are treated for T2DM for decades and, therefore, the therapeutic benefit to safety ratio must be high. The acute concern with all treatments for T2DM is hypoglycemia. GKAs, by reducing the S0.5, essentially make GK more sensitive to glucose concentrations that normally would not be as effective. Indeed, a reduction of fasting plasma glucose, and hypoglycemia, have been reported as a consequence of GKA treatment in some of the animal studies (9, 12). The possible long-term complications of increased hepatic glycogen stores and lipid deposition in liver and muscle, perhaps as a consequence of prolonged hyperinsulinemia, represent concerns that may not become apparent until these agents have been used long-term in man.
These caveats not withstanding, the GKA drug discovery process represents an excellent example of translational research. Solid basic and clinical science, with data collected over several decades, forms the basis of a logical effort to find a new, safe therapy for T2DM.
 |
Footnotes |
|---|
Abbreviations: F1P, Fructose 1-phosphate; GK, glucokinase; GKA, GK activator; GKRP, glucokinase regulatory protein; HK, hexokinase; PHHI-GK, persistent hyperinsulinemic hypoglycemia of infancy-GK; S0.5, substrate concentration at half maximum velocity; T2DM, type 2 diabetes mellitus; Vmax, maximum velocity.
Received June 8, 2005.
Accepted for publication June 22, 2005.
|
References |
|---|
 Top References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. F. Pilch, R. P. Souto, L. Liu, M. P. Jedrychowski, E. A. Berg, C. E. Costello, and S. P. Gygi Cellular spelunking: exploring adipocyte caveolae J. Lipid Res., October 1, 2007; 48(10): 2103 - 2111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Okamoto, W. Ogawa, A. Nishizawa, H. Inoue, K. Teshigawara, S. Kinoshita, Y. Matsuki, E. Watanabe, R. Hiramatsu, H. Sakaue, et al. Restoration of Glucokinase Expression in the Liver Normalizes Postprandial Glucose Disposal in Mice With Hepatic Deficiency of PDK1 Diabetes, April 1, 2007; 56(4): 1000 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Baltrusch, S. Langer, L. Massa, M. Tiedge, and S. Lenzen Improved Metabolic Stimulus for Glucose-Induced Insulin Secretion through GK and PFK-2/FBPase-2 Coexpression in Insulin-Producing RINm5F Cells Endocrinology, December 1, 2006; 147(12): 5768 - 5776. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
