Introduction to Glycogen Metabolism
Glycogen is a polymer of glucose residues linked by α-(1,4)- and α-(1,6)-glycosidic bonds. Stores of readily available glucose, to supply the tissues of the body with an oxidizable energy source, are found as glycogen, solely in the liver. Although the liver is the only tissue that can release glucose from glycogen to the blood, other tissues also synthesize glycogen and release glucose from glycogen but this glucose is used for direct cellular energy needs. Two primary tissues that store glucose as glycogen, as a store of energy, are skeletal muscle and the brain, principally astrocytes, however, the kidney, heart, and adipose tissue also store glucose as glycogen. The glucose in muscle and brain glycogen, and other non-hepatic tissues, is not available to other tissues, because of the presence of hexokinase which has a very high affinity for glucose, thereby rapidly phosphorylating any glucose as well as due to the lack of glucose-6-phosphatase.
The major site of daily glucose consumption (75%) is the brain via aerobic pathways. Most of the remainder of is utilized by erythrocytes, skeletal muscle, and heart muscle. The body obtains glucose either directly from the diet or from amino acids and lactate via gluconeogenesis. Glucose obtained from these two primary sources either remains soluble in the body fluids or is stored in a polymeric form, glycogen. Glycogen is considered the principal storage form of glucose and is found mainly in liver, although other tissues such as skeletal muscle, brain, kidney, heart, and adipose tissue are all capable of glycogen synthesis and breakdown. With up to 10% of its weight as glycogen, the liver has the highest specific content of any body tissue. Muscle has a much lower amount of glycogen per unit mass of tissue (1%–2%), but since the total mass of muscle is so much greater than that of liver, total glycogen stored in muscle is about twice that of liver. The amount of glycogen in the brain is on the order of 0.1% of total tissue weight. Stores of glycogen in the liver are considered the main buffer of blood glucose levels.
Glycogen homeostasis involves the concerted regulation of the rate of glycogen synthesis (glycogenesis) and the rate of glycogen breakdown (glycogenolysis). These two processes are reciprocally regulated such that hormones that stimulate glycogenolysis (e.g. glucagon, cortisol, epinephrine, norepinephrine) simultaneously inhibit glycogenesis. Conversely, insulin, which directs the body to store excess carbon for future use, stimulates glycogenesis while simultaneously inhibiting glycogenolysis.
Cytosolic Glycogen Catabolism
Degradation of stored glycogen, termed glycogenolysis, occurs through one of two pathways, cytosolic and lysosomal. The cytosolic pathway involves the actions of glycogen phosphorylase (GP) and the glycogen debranching enzyme (GDE). Within the cytosolic compartment, in particular in skeletal muscle, approximately 40% of glycogen phosphorylase is associated with membranes of the sarcoplasmic reticulum (SR: muscle cell endoplasmic reticulum, ER). As discussed below glycogen phosphorylase activity is regulated by its state of phosphorylation, a reaction catalyzed by phosphorylase kinase (PHK: also called glycogen synthase glycogen phosphorylase kinase). In skeletal muscle cells approximately 55% of PHK activity is associated with the SR. The activity of PHK is itself regulated by phosphorylation and by its Ca2+-binding subunit, calmodulin. The association of glycogen phosphorylase and PHK on SR membranes allows for rapid glycogenolysis in response to muscle cell activation and release of SR stored Ca2+.
Humans express three genes encoding proteins with glycogen phosphorylase activity. One gene (PYGL) expresses the hepatic form of the enzyme, a second (PYGM) expresses the muscle form (referred to as myophosphorylase), and the third (PYGB) expresses the brain form.
The PYGM gene is located on chromosome 11q13.1 and is composed of 20 exons that generate two splice variant mRNAs. The two different PYGM proteins that result are referred to as isoform 1 (842 amino acids) and isoform 2 (754 amino acids).
The PYGL gene is located on chromosome 14q22.1 and, like the PYGM gene, is composed of 20 exons that generate two splice variant mRNAs. The two different PYGL proteins that result are referred to as isoform 1 (872 amino acids) and isoform 2 (813 amino acids).
The PYGB gene is located on chromosome 20p11.21 and is also composed of 20 exons that encode a protein of 843 amino acids. Although preferentially expressed in the brain, low level expression of the PYGB gene is seen in adult liver and skeletal muscle.
The enzymatic functions of the different glycogen phosphorylase gene encoded enzymes are identical but their mechanisms of regulation have tissue specific characteristics in addition to some similarities. In addition, mutations in specific glycogen phosphorylase genes explain the tissue-specific nature of several of the glycogen storage diseases discussed below.
Biologically active glycogen phosphorylase exists as a homodimer. Each subunit binds the vitamin B6-derived cofactor, pyridoxal phosphate, PLP. In addition to the PLP-binding sites and the catalytic sites of homodimeric glycogen phosphorylase, the enzyme contains allosteric regulatory sites and phosphorylation sites as detailed in the section below covering Regulation of Glycogen Catabolism. The majority of glycogen phosphorylase is bound to glycogen granules through a domain referred to as the glycogen storage site. The binding to glycogen allows glycogen phosphorylase to rapidly release stored glucose in response to physiological demands. The catalytic action of phosphorylase is to phosphorolytically remove single glucose residues from α-(1,4)-linkages within the glycogen molecules. The product of this reaction is glucose-1-phosphate and a glycogen molecule with one less glucose residue. The advantage of the reaction proceeding through a phosphorolytic step is that:
1. The glucose is removed from glycogen is an activated state, i.e. phosphorylated and this occurs without ATP hydrolysis.
2. The concentration of Pi in the cell is high enough to drive the equilibrium of the reaction in the favorable direction since the free energy change of the standard state reaction is positive.
The glucose-1-phosphate produced by the action of phosphorylase is converted to glucose-6-phosphate by phosphoglucomutase (phosphohexose mutase): this enzyme, like phosphoglycerate mutase of glycolysis, contains a phosphorylated amino acid in the active site (in the case of phosphoglucomutase it is a Ser residue). The enzyme phosphate is transferred to C-6 of glucose-1-phosphate generating glucose-1,6-bisphosphate as an intermediate. The phosphate on C-1 is then transferred to the enzyme regenerating active enzyme and glucose-6-phosphate is the released product.
There are four different phosphoglucomutase genes in humans identified as PGM1, PGM2, PGM3, and PGM5. The protein encoded by the PGM5 gene is called phosphoglucomutase-like protein 5. The PGM1 gene is expressed in most tissues, whereas PGM2 expression predominates in red blood cells.
The PGM1 gene is located on chromosome 1p31.3 and is composed of 13 exons that generate three alternatively spliced mRNAs and three isoforms of this enzyme. Mutations in the PGM1 gene are associated with the congenital disorder of glycosylation, CDG1T (once referred to as glycogen storage disease type 14, GSD14). The PGM2 gene is located on chromosome 4p14 and is composed of 15 exons that encode a protein of 612 amino acids. The PGM3 gene is located on chromosome 6q14.1 and is composed of 19 exons that generate four alternatively spliced mRNAs and four distinct isoforms of this enzyme. The PGM5 gene is located on chromosome 9q21.11 and is composed of 13 exons that encode a protein of 567 amino acids.
As mentioned above, the phosphorylase mediated release of glucose from glycogen yields a charged glucose residue (glucose-1-phosphate) without the need for hydrolysis of ATP. An additional necessity of releasing phosphorylated glucose from glycogen ensures that the glucose residues do not freely diffuse from the cell. In the case of muscle cells this is acutely apparent since the purpose in glycogenolysis in muscle cells is to generate substrate for glycolysis.
The conversion of glucose-6-phosphate to glucose occurs only in the liver, kidney and intestine via the action of glucose-6-phosphatase does not occur in other tissues that synthesize glycogen (e.g. skeletal muscle and brain) as all other cells lack this enzyme. Therefore, any free glucose released from glycogen stores, through the action of the debranching enzyme, in skeletal muscle and brain will be oxidized in the glycolytic pathway. In the liver the action of glucose-6-phosphatase allows glycogenolysis to generate free glucose for maintaining blood glucose levels.
Glycogen phosphorylase cannot remove glucose residues from the branch points (α-1,6 linkages) in glycogen. The activity of phosphorylase ceases approximately four glucose residues from the branch point. The removal of the these branch point glucose residues requires the action of glycogen debranching enzyme (GDE). The official name of GDE is amylo-1,6-glucosidase, 4-α-glucanotransferase (gene symbol: AGL) which contains 2 activities: glucotransferase and glucosidase. The AGL gene is located on chromosome 1p21.2 and is composed of 36 exons that generate five alternative spliced mRNAs that collectively encode two distinct isoforms (1 and 3) of the enzyme. Isoform 1 contains 1532 amino acids and isoform 3 contains 1516 amino acids.
The transferase activity of debranching enzyme removes the terminal three glucose residues of one branch and attaches them to a free C-4 end of a second branch. The glucose in α-(1,6)-linkage at the branch is then removed by the action of glucosidase. This glucose residue is uncharged since the glucosidase-catalyzed reaction is not phosphorylytic. This means that theoretically glycogenolysis occurring in skeletal muscle could generate free glucose which could enter the blood stream. However, the activity of hexokinase in muscle is so high that any free glucose is immediately phosphorylated and enters the glycolytic pathway. Indeed, the precise reason for the temporary appearance of the free glucose from glycogen is the need of the skeletal muscle cell to generate energy from glucose oxidation, thereby, precluding any chance of the glucose entering the blood.
Lysosomal Glycogen Catabolism
When glycogen granules are not recruited for cytosolic degradation they become targets for increased phosphorylation making them less soluble inducing their degrdation via the lysosomal pathway. Lysosomal glycogen degradation is catalyzed by the enzyme lysosomal acid α-glucosidase (also called acid maltase) which is encoded by the GAA gene. The significance of this pathway for glycogen degradation is evidenced from the lethal disorder, Pompe disease, that is the result of mutations in the GAA gene. Lysosomal glycogen is enriched in very large molecular weight granules. The pathway of lysosomal glycogen degradation represents 5% of total muscle glycogen and 10% of total liver glycogen degradation. One significant role for lysosomal glycogen degradation is in the neonate where liver lysosomal glycogen breakdown is the product of glycogen autophagy. This pathway for newborn glycogenolysis is most likely a mechanism for providing extra glucose-derived energy during and after birth.
Lysosomal glycogen breakdown, although catalyzed by α-glucosidase, also requires the action of the dual specificity phosphatase, laforin, and a second enzyme, an E3 ubiquitin ligase, encoded by the NHLRC1 gene (NHL repeat containing E3 ubiquitin protein ligase 1; also known as EPM2B). The NHLRC1 encoded protein is called malin. The precise role for malin in the overall process of glycogen dephosphorylation is unclear but larforin-malin interaction is a required event for the process. The laforin-malin interaction is also involved in the regulation of the removal of glycogen-associated proteins not only via the autophagy–lysosomal pathway but also via the ubiquitin–proteasome pathway. Malin has been shown to ubiquitylate several glycogen-associated proteins including laforin, glycogen synthase, glycogen debranching enzyme, and the regulatory subunit of protein phosphatase 1 which is often referred to as protein targeting glycogen (PTG). PTG within skeletal muscle is encoded by the PPP1R3A (protein phosphatase 1 regulatory subunit 3A) gene. Liver PTG is encoded by the PPP1R3B gene.
Glycogenin in de novo Glycogen Synthesis
For de novo glycogen synthesis to proceed the first few glucose residues are attached to a protein known as glycogenin. Glycogenin binds to actin filaments therough a domain in the C-terminus of the protein. The interaction of glycogenin with actin filaments is required to initiate the synthesis of glycogen. Glycogenin functions as a homodimer and catalyzes its own glycosylation, attaching C-1 of a UDP-glucose to a tyrosine residue (Y194) on the enzyme. This reaction is carried out by one subunit adding the glucose to the other subunit. Following the addition of the first glucose residue, each glycogenin subunit will then add a further 6-17 glucose residue in an intra-subunit reaction via α(1,4) glycosidic bonds. The attached glucose then serves as the primer required by glycogen synthase (GS) to attach additional glucose molecules via the mechanism described below. Glycogenin binds to conserved amino acid sequences in the N-terminus of glycogen synthase. As the glycogen molecule grows on glycogenin, glycogen synthase is released from glycogenin and binds to the lengthening polymer via a glycogen-binding module in the C-terminus of the enzyme.
There are two glycogenin genes in humans identified as GYG1 and GYG2. The GYG1 gene is located on chromosome 3q24 and is composed of 10 exons that generate three alternatively spliced mRNAs. These three mRNAs produce three glycogenin-1 isoforms identified as isoform 1 (350 amino acids), isoform 2 (333 amino acids), and isoform 3 (279 amino acids). The GYG1 gene is predominantly expressed in muscle but is also expressed in many other tissues as well. Mutations in the GYG1 gene are associated with the recently (2010) characterized glycogen storage disease identified as type 15 (GSD15). The GYG2 gene is located on chromosome Xp22.33 and is composed of 14 exons that generate five alternatively spliced mRNAs leading to the generation of multiple glycogenin-2 isoforms. The GYG2 gene expression in restricted to the liver.
Glycogen Synthase in Glycogen Synthesis
Like the action of glycogenin, glycogen synthase utilizes UDP-glucose as its substrate. Glycogen synthase add glucose residues from UDP-glucose to terminal glucose on glycogenein as well as to the non-reducing end glucose of a molecule of glycogen. The reaction catalyzed by glycogen synthase results in α(1,4) glycosidic linkage (non-branched) of the glucose residues. The activation of glucose to be used for glycogen synthesis is carried out by the enzyme UDP-glucose pyrophosphorylase 2. This enzyme exchanges the phosphate on C-1 of glucose-1-phosphate for UDP. The energy of the phosphoglycosyl bond of UDP-glucose is utilized by glycogen synthase to catalyze the incorporation of glucose into glycogen. UDP is subsequently released from the enzyme. The human UDP-glucose pyrophosphorylase 2 enzyme is encoded by the UGP2 gene that is located on chromosome 2p15 and is composed of 14 exons that generate two alternatively spliced mRNAs. These two mRNAs encode two different isoforms of the enzyme, isoform a (508 amino acids) and isoform b (497 amino acids).
There are two distinct glycogen synthase enzymes in humans. One is more widely expressed and predominates in skeletal muscle, the other predominates in the liver. The broadly expressed enzyme is encoded by the GYS1 gene and the liver, heart, and pancreas enzyme is encoded by the GYS2 gene. The GYS1 gene is located on chromosome 19q13.33 and is composed of 16 exons that produce two alternatively spliced mRNAs encoding two isoforms of the muscle enzyme. Isoform 1 is composed of 737 amino acids and isoform 2 is composed of 673 amino acids. The GYS2 gene is located on chromosome 12p12.1 and is composed of 18 exons that produce a protein of 703 amino acids.
The α-1,6 branches in glycogen are produced by amylo-(1,4 to 1,6)-transglucosidase, also termed the glycogen branching enzyme (gene symbol: GBE1). This enzyme transfers a terminal fragment of 6-7 glucose residues (from a polymer at least 11 glucose residues long) to an internal glucose residue at the C-6 hydroxyl position. The GBE1 gene is located on chromosome 3p12.2 and is composed of 16 exons that encode a protein of 702 amino acids.
Normal glycogen has small amounts of phosphate, as phosphomonoesters, at positions C2, C3, and C6 of the glucose residues in the molecule. Evidence has indicated that the phosphates on C2 and C3 are added by glycogen synthase but the precise mechanism for formation of the C6 phosphomonoesters remains elusive. Removal of the phosphates from glycogen is catalyzed by the dual specificity phosphatase encoded by the EPM2A gene (EPM2A glucan phosphatase). The EPM2A of the gene name refers to Epilepsy, Progressive Myoclonus type 2A. The protein encoded by the EPM2A gene is called laforin (also known as laforin glycogen phosphatase) since loss of the enzyme is associated with the lethal neurodegenerative disease called Lafora disease. The level of phosphate in skeletal muscle glycogen is approximately 1 per 1,500 glucose residues. Laforin dephosphorylates glycogen and this dephosphorylation is required to facilitate the normal branching. The highly branched nature of glycogen maintains it water solubility. The consequences of loss-of-function of laforin is hyperphosphorylation of glycogen which renders it insoluble. The precipitation of hyperphosphorylated glycogen (histologically referred to as Lafora bodies), particularly in neurons, is the underlying cellular cause of the pathology of Lafora disease.
Regulation of Glycogen Catabolism
Consideration of the regulation of glycogen catabolism (glycogenolysis) involves tissue-specific differences in the the reasons for storing glucose as glycogen and the tissues-specific differences in the various forms of glycogen phosphorylase. Despite the tissue of expression, functional glycogen phosphorylase is a homodimeric enzyme that exist in two distinct conformational states: the T state (for tense, less active; referred to as phosphorylase b) and the R state (for relaxed, more active; referred to as phosphorylase a).
Phosphorylase is capable of binding to glycogen when the enzyme is in the R state (phosphorylase a). The skeletal muscle phosphorylase a conformation, but not liver phosphorylase, is enhanced by binding of AMP which serves as an allosteric activator. In liver and skeletal muscle the activity of the phosphorylase a conformation is inhibited by binding of the allosteric inhibitors ATP and glucose-6-phosphate. Within hepatocytes glucose also acts as an allosteric inhibitor, an effect not exerted on skeletal muscle or brain phosphorylase.
Glycogen phosphorylase is also subject to covalent modification by phosphorylation as a means of regulating its activity. The major site for this regulatory phosphorylation is Ser 14 on both subunits of the homodimeric enzyme. The basal activity of the unmodified phosphorylase enzyme (phosphorylase b) is sufficient to generate enough glucose-1-phosphate, for entry into glycolysis, for the production of sufficient ATP to maintain the normal resting activity of the cell. This is true in both liver, brain, and skeletal muscle cells.
In response to lowered blood glucose the α cells of the pancreas secrete glucagon which binds to cell surface receptors that are predominantly found on hepatocytes. Glucagon receptors are also present on white adipocytes and cardiomyocytes but at significantly lower levels than those seen on hepatocytes. Because of this distribution of receptors, it is easy to understand why liver cells are the primary target for the action of glucagon.
The glucagon receptor is a Gs-coupled GPCR. The response of cells to the binding of glucagon to its cell surface receptor is, therefore, the activation of the enzyme adenylate cyclase. Activation of adenylate cyclase leads to a large increase in the formation of cAMP which then binds to, and activates the enzyme, cAMP-dependent protein kinase (PKA: see Figure below). The original model of cAMP-mediated activation of PPKA involved the binding of cAMP to the regulatory subunits of PKA leading to the release of the regulatory subunits and subsequent activation of the catalytic subunits. Recently it has been demonstrated that release of the regulatory subunits is not required for cAMP-mediated activation of PKA. When activated the catalytic subunits phosphorylate a number of target proteins on serine and threonine residues.
Of significance to this discussion is the PKA-mediated phosphorylation of phosphorylase kinase (PHK) as shown in the Figure above. There are three isoforms of phosphorylase kinase, one is primarily expressed in skeletal muscle, one primarily expressed in the liver, and one expressed primarily in the brain. All three isoforms of phosphorylase kinase are multi-subunit (hexadecameric) enzymes composed of four copies of each of the unique subunits: α, β, γ, and δ. The difference between the skeletal muscle (as well as heart), liver, and brain isoforms is the result of two distinct α proteins and two distinct γ proteins, each of which are encoded by different genes. The α and β subunits are the regulatory subunits that are phosphorylated. The γ subunit is the catalytic subunit. The δ subunit is calmodulin (described below) which can be encoded for by one of three calmodulin genes; CALM1, CALM2, CALM3.
The two different α subunits of phosphorylase kinase are encoded by the PHKA1 (although not exclusive it is referred to as the muscle form) and PHKA2 (although not exclusive it is referred to as the liver form) genes. The PHKA1 gene is located on chromosome Xq13.1 and is composed of 32 exons that generate three alternatively spliced mRNAs that collectively encode three distince protein isoforms. One of the PHKA1 isoforms (identified as α’) lacks an internal 59 amino acids, compared to the longest isoform, and predominates in cardiac myocytes and slow-twitch skeletal muscle cells. The PHKA2 gene is located on chromosome Xp22.13 and is composed of 34 exons that encode a protein of 1235 amino acids. The two different γ subunits are encoded by the PHKG1 (although not exclusive it is referred to as the muscle form) and PHKG2 (although not exclusive it is referred to as the liver form) genes. The PHKG1 gene is located on chromosome 7p11.2 and is composed of 13 exons that generate three alternatively spliced mRNAs that collectively encode three distinct protein isoforms. The PHKG2 gene is located on chromosome 16p11.2 and is composed of 10 exons that generate two alternatively spliced mRNAs encoding isoform 1 (406 amino acids) and isoform 2 (374 amino acids). The common β subunit is encoded by the PHKB gene which is located on chromosome 16q12.1 and is composed of 35 exons that undergo alternative splicing at two sites resulting in proteins with different internal segments of 28 amino acids and different N-termini. These differences are seen in the skeletal muscle β subunit which is distinct from the form expressed in the brain and several other tissues.
Mutations in the PHKA2 gene result in the X-linked liver glycogen storage diseases identified as types 9a1 and 9a2 (GSD9A1 and GSD9A2). Mutations in the PHKB gene result in the autosomal recessive liver and muscle glycogen storage disease identified as type 9B (GSD9B). Mutations in the PHKG2 gene cause the hepatic glycogen storage disease identified as type 9C (GSD9C). Mutations in the PHKA1 gene cause the X-linked muscle glycogen storage disease identified as type 9D (GSD9D). The various Glycogen Storage diseases are listed in the Table at the end of this page.
This identical cascade of events, responsible for the regulation of glycogen phosphorylase activity in hepatocytes, occurs in skeletal muscle cells and brain astrocytes as well. However, in skeletal muscle cells the induction of the cascade is the result of epinephrine binding to adrenergic receptors on the surface of muscle cells or as a result of acetylcholine binding nicotinic acetylcholine receptors at a neuromuscular junction. Epinephrine is released from the adrenal glands in response to sympathetic nervous system outflow from the brain indicating an immediate need for enhanced glucose utilization in muscle, the so called fight-or-flight response. Muscle cells lack glucagon receptors and therefore, do not respond in any way to pancreatic effects of low blood glucose. The presence of glucagon receptors on muscle cells would be futile anyway since the role of glucagon release is to increase blood glucose concentrations and muscle glycogen stores cannot contribute to blood glucose levels.
Regulation of phosphorylase kinase activity is also effected by two distinct mechanisms involving Ca2+ ions. The ability of Ca2+ ions to regulate phosphorylase kinase is through the function of one of the subunits of this enzyme, specifically the γ subunit. The γ subunit of phosphorylase kinase is the ubiquitous protein, calmodulin. As indicated above, there are three different calmodulin genes, any one of which can encode the γ subunit of phosphorylase kinase. Calmodulin is a calcium binding protein and the binding of Ca2+ induces a conformational change in calmodulin which in turn enhances the catalytic activity of phosphorylase kinase towards its substrate, phosphorylase b. This activity is crucial to the enhancement of glycogenolysis in muscle cells where muscle contraction is induced via acetylcholine stimulation at the neuromuscular junction. The effect of acetylcholine release from nerve terminals at a neuromuscular junction is to depolarize the muscle cell leading to increased release of sarcoplasmic reticulum stored Ca2+, thereby activating phosphorylase kinase. Thus, not only does the increased intracellular calcium increase the rate of muscle contraction it increases glycogenolysis which provides the muscle cell with the glucose it needs to oxidize to satisfy the increased ATP it needs for contraction.
The second Ca2+ ion-mediated pathway to phosphorylase kinase activation is through activation of α1-adrenergic receptors by epinephrine or norepinephrine as outlined in the following Figure.
Unlike β-adrenergic receptors which are coupled to activation of adenylate cyclase, α1-adrenergic receptors are coupled through Gq-type G proteins that activate phospholipase-Cβ (PLCβ). However, it is important to note that α2-adrenergic receptors couple to a Gi-type G-protein and, as a result of ligand binding to this class of receptor, inhibit the activation of adenylate cyclase. Activation of PLCβ leads to increased hydrolysis of membrane phosphatidylinositol-4,5-bisphosphate (PIP2), the products of which are inositol-1,4,5-trisphosphate (IP3) and diacylglycerol (DAG). DAG binds to and activates protein kinase C (PKC), an enzyme that phosphorylates numerous substrates, one of which is glycogen synthase (see below). IP3 binds to one of several receptors on the surface of the endoplasmic reticulum leading to release of stored Ca2+ ions. The Ca2+ ions then interact with the calmodulin subunits of phosphorylase kinase resulting in its activation. Additionally, the Ca2+ ions activate PKC in conjunction with DAG.
Humans express three distinct IP3 receptors encoded by the ITPR1, ITPR2, and ITPR3 genes. The ITPR1 gene which is located on chromosome 3p26.1 and is composed of 63 exons that generate three alternatively spliced mRNAs encoding three distinct isoforms of the receptor. ITPR1 isoform 1 is a 2710 amino acid protein, isoform 2 is a 2695 amino acid protein, and isoform 3 is a 2743 amino acid protein. The ITPR2 gene is located on chromosome 12p11.23 and is composed of 62 exons that encode a 2701 amino acid protein. The ITPR3 gene is located on chromosome 6p21.31 and is composed of 60 exons that encode a 2671 amino acid protein. Each of the IP3 receptors possesses a cytoplasmic N-terminal ligand-binding domain and is comprised of six membrane-spanning helices that forms the core of the ion pore.
In order to terminate the activity of the enzymes of the glycogen phosphorylase activation cascade, once the needs of the body are met, the modified enzymes need to be unmodified. In the case of Ca2+ induced activation, the level of Ca2+ ion release from muscle stores will terminate when the incoming nerve impulses cease. The removal of the phosphates on phosphorylase kinase and phosphorylase a is carried out by a family of enzy,es identified as phosphoprotein phosphatase-1 (PP1). Each functional PP1 is a heterodimeric enzyme composed of a catalytic subunit and a regulatory subunit. Humans express three distinct PP1 catalytic subunit genes identified as PPP1CA, PPP1CB, and PPP1CC. There are at least 29 PP1 regulatory subunit genes expressed in the human genome. Several of the regulatory subunits are also involved in targeting of PP1 to glycogen. These regulatory subunits are also commonly referred to as protein targeting to glycogen, PTG (see Figure below). The PTG regulator of the muscle isoform of PP1 is encoded by the protein phosphatase 1, regulatory subunit 3A gene located on chromosome 7q31.1 (gene symbol: PPP1R3A). The muscle PTG protein is also commonly referred to as PP1G. The PTG regulator of the liver form of PP1 is encoded by the PPP1R3B gene located on chromosome 8p23.1. The PPP1R3B encoded protein was the originally identified PTG activity.
In order that the phosphate residues placed on various enzymes by PKA and phosphorylase kinase (PHK) are not immediately removed, the activity of PP1 must also be regulated. Within skeletal muscle this is accomplished through the interaction of PP1 catalytic subunits with the regulatory subunits encoded by the PPP1R3A gene (the PTG protein). Since the protein encoded by this gene inhibits the activity of PP1 it was once called phosphoprotein phosphatase inhibitor 1 (PPI-1) or PP1 inhibitor. The PPP1R3A encoded protein (PTG) is phosphorylated by PKA on Ser65 which causes PP1 to dissociate from the PTG complex, thereby preventing glycogen synthase from being dephosphorylated keeping it in the less active state. Conversely, insulin mediated signaling results in phosphorylation of the PPP1R3A protein at Ser46 which results in increased activity of PP1, removal of the inhibitory phosphate from glycogen synthase and a subsequent increase in glycogen synthesis.
Regulation of Glycogen Synthesis
Glycogen synthase is a tetrameric enzyme consisting of four identical subunits. The liver and muscle glycogen synthase proteins are derived from different genes and share only 46% amino acid identity. As indicated above, the liver glycogen synthase is encoded by the GYS2 gene while the muscle form (as well as that expressed in several other tissues) is encoded by the GYS1 gene. Regardless of tissue of expression, the activity of glycogen synthase is regulated by both allosteric effectors and by phosphorylation of serine residues in the subunit proteins. Allosteric regulation of glycogen synthase activity is effected by glucose-6-phosphate and ATP with glucose-6-phosphate being a positive effector and ATP being and inhibitory effector. Phosphorylation of glycogen synthase has been shown to occur on at least nine different serine residues and these phosphorylations are carried out by numerous different kinases. Phosphorylation of glycogen synthase reduces its activity towards UDP-glucose. When in the phosphorylated state, glycogen synthase inhibition can be overcome by the presence of the allosteric activator, glucose-6-phosphate. The two forms of glycogen synthase are identified by the same nomenclature as used for glycogen phosphorylase. The unphosphorylated and most active form is glycogen synthase a and the phosphorylated, less active form is glycogen synthase b.
Numerous kinases have been shown to phosphorylate and regulate both hepatic and muscle forms of glycogen synthase. Most detailed analyses of glycogen synthase phosphorylation have been carried out using enzyme isolated from skeletal muscle but related findings with liver glycogen synthase have also been demonstrated. At least nine sites of phosphorylation have been identified in glycogen synthase and these nine sites are clustered into four phosphorylation domains present in the N-terminal and C-terminal ends of the enzyme. Phosphorylation of glycogen synthase occurs through the activities of at least ten distinct kinases. The phosphorylation sites in glycogen synthase are identified as site 1a, 1b, 2, 2a, 3a, 3b, 3c, 4, and 5 with sites 2, 2a, 3a, and 3b being the most significant with respect to the regulation of enzyme activity.
Regulation of glycogen synthase by phosphorylation occurs via both primary and secondary phosphorylation events. The ten kinases that regulate glycogen synthase activity are PKA, PKC, glycogen synthase kinase-3β (encoded by the GSK3B gene), phosphorylase kinase (PHK), a calcium/calmodulin-dependent protein kinase 2 family member kinase (CAMK2 or CAMKII), a member of the casein kinase 1 (CK1) family, casein kinase 2 (CK2), AMPK, PAS domain containing serine/threonine kinase (PASK), dual specificity tyrosine phosphorylation regulated kinase 2 (DYRK2), and p38MAPK (encoded by the MAPK14 gene). Humans express 11 genes in the Ca2+/calmodulin-dependent protein kinase family with the functional CAMKII enzyme being composed of four different subunits, α, β, δ, and γ that are encoded by the CAMK2A, CAMK2B, CAMK2D, and CAMK2G genes, respectively. The CK1 family of kinases is composed of seven monomeric enzymes.
Recent studies demonstrate that the association of AMPK and PHK with the liver form of glycogen synthase is not significant or does not occur. Primary phosphorylation events regulating muscle glycogen phosphorylase activity are initiated by PKA, PKC, CaMPK2, PHK, and CK2. Secondary phosphorylation events are the result of GSK3 and CK1. When glucagon binds its receptor on hepatocytes the resultant rise in activity of PKA leads to increased phosphorylation of glycogen synthase directly by PKA. The regulatory phosphorylation sitesin glycogen synthase targeted by PKA are 1a, 1b, and 2. Within liver glycogen phosphorylase phosphorylation of site 2 has been shown to be the most significant relative to its regulation. Activated PKA also leads to phosphorylaiton and activation of phosphorylase kinase which also phosphorylates glycogen synthase on site 2. In addition, glucagon signaling effects an increase in the activity of CK2. Thus, the net effect of glucagon action on hepatocytes is activation of three distinct kinases that phosphorylate and inhibit glycogen synthase.
Since insulin and glucagon are counter-regulatory hormones it should be clear that they will exert opposing effects on the rate and level of glycogen synthase phosphorylation. As described above, when glucagon binds its receptor on hepatocytes there is a resultant rise in cAMP and a concomitant increase in the activity of PKA. PKA has been shown to phosphorylate glycogen synthase on at least three different sites (sites 1a, 1b, and 2). In addition, PKA activity results in an increase in the activity of phosphorylase kinase which in turn phosphorylates glycogen synthase at one of the same sites as PKA (site 2). The action of insulin, at the level of PKA, is to increase the activity of phosphodiesterase which hydrolyzes cAMP to AMP thereby reducing the level of active PKA. Insulin also exerts a negative effect on the activity of GSK3 such that there is a reduced level of phosphorylation of glycogen synthase by this kinase. Within skeletal muscle GSK3 phosphorylates sites 3a, 3b, 3c, and 4 in glycogen synthase. To see more on the action of insulin at the level of GSK3 visit the Insulin Action page.
Hormones and neurotransmitters that result in release of stored intracellular Ca2+ also lead to negative regulation of glycogen synthase activity. As described above, Ca2+ ions bind to the calmodulin subunit of phosphorylase kinase (PHK: calmodulin is a component of many enzymes that are responsive to Ca2+) and result in its activation leading to increased phosphorylation and inhibition of glycogen synthase. When α1-adrenergic receptors are stimulated there is an increase in the activity of PLCβ with a resultant increase in PIP2 hydrolysis. The products of PIP2 hydrolysis are DAG and IP3. As described above for glycogen phoshorylase, DAG, along with the Ca2+ ions released by IP3, activate PKC which phosphorylates and inactivates glycogen synthase. Phosphorylation of glycogen synthase by PKC occurs in the same domain of the enzyme that is one of the target sites for PKA phosphorylation, namely site 2. Additional responses to calcium are the activation of members of the calmodulin-dependent protein kinase family identified as CaMK2 (Ca2+/calmodulin-dependent kinase 2; also designated CAMKII) which also phosphorylates glycogen synthase. Humans express four CaMK2 genes identified as CAMK2A (CaMK2 alpha), CAMK2B (CaMK2 beta), CAMK2D (CaMK2 delta), and CAMK2G (CaMK2 gamma) each of which generate multiple different protein isoforms as a result of alternative mRNA splicing. Phosphorylation of glycogen synthase by CaMK2 occurs at sites 1b and 2.
The net effects of the various phosphorylations of glycogen synthase result in:
- decreased affinity of the enzyme for UDP-glucose.
- decreased affinity of the enzyme for glucose-6-phosphate.
- increased affinity of the enzyme for ATP and Pi.
Reconversion of glycogen synthase b to glycogen synthase a requires dephosphorylation. This is carried out predominately by the serine/threonine phosphatase described earlier, PP1. This, of course is the same phosphatase involved in the dephosphorylation of glycogen phosphorylase described in detail above. Although another serine/threonine phosphatase, namely protein phosphatase-2A (PP2A), has been shown to dephosphorylate glycogen synthase in vitro, its role in vivo is significantly less than that of PP1.
The activity of PP1 is also affected by insulin as described such that the effects of insulin exert an opposing effect to that of glucagon and epinephrine on overall glycogen homeostasis. This should appear obvious since the role of insulin is to increase the uptake of glucose from the blood and to store it as glycogen.
Nuclear Glycogen Metabolism and Control of Gene Expression
Although the cytosol is the principal site of glycogen accumulation, the molecule is also found in the nucleus, in the mitochondria, and associated with the endoplasmic reticulum, ER (or the sarcoplasmic reticulum in muscle). The presence of glycogen in the nucleus suggests that there is likely to be a nuclear specific glycogen metabolic pathway(s). There is no known mechanism for transporting glycogen across cell membranes and given that glycogen synthase has been found in the nucleus, it has been proposed that glycogen synthesis is likely to occur de novo in the nucleus. In addition to glycogen synthase (encoded by the GYS1 and GYS2 genes), UDP-glucose pyrophosphorylase 2 (encoded by the UGP2 gene), and the glucose-6-phosphate transporter, G6PT1 (encoded by the SLC37A4 gene) have been found in the nucleus. Evidence has shown that glucose-6-phosphate (G6P) is the substrate for incorporation of glucose into nuclear glycogen. In addition to enzymes of glycogen metabolism, all the glycolytic enzymes (excluding hexokinase) and the pyruvate dehydrogenase complex (PDHc) are also found in the nucleus. Indeed, nuclear PDHc regulates histone acetylation.
One of the major functions of nuclear glycogen is to serve as a carbon pool supplying the metabolic substrates for histone modification that is independent of cytoplasmic metabolites of glycogen and glucose. Nuclei have been shown to metabolize glucose-6-phosphate to the glycolytic intermediates, including fructose-6-phosphate (F6P), 3-phosphoglycerate (3PG), and pyruvate. However, nuclei do not convert free gluocse to glycolytic intermediates. Metabolism of nuclear glycogen is enhanced by the presence of the E3 ubiquitin ligase, malin [encoded by the NHLRC1 (NHL repeat containing E3 ubiquitin protein ligase 1) gene], and requires the presence of glycogen phosphorylase. One important function for nuclear glycogen metabolism is most likely to be the generation of nuclear localized pyruvate which can then serve as a source of the nuclear acetate required for histone acetylation. In in vitro experiments, where malin is overexpressed, it was found that there was increased protein acetylation in the nucleus and that this required the presence of glycogen phosphorylase. Malin overexpressing cells have increased acetylation of numerous histones including H1.4, H2A1, H2A3, H3, and H4. Histone acetylation is directly associated with changes in transcription. Indeed, in malin overexpressing cells increased expression of genes involved in pathways of smooth muscle contraction, DNA methylation, DNA demethylation, and DNA alkylation has been documented while decreased expression of genes involved in chromosome segregation, cell proliferation, and cellular catabolic processes were also found.
Nuclear glycogenolysis contributes to the pyruvate pool and subsequently to histone acetylation. This process is significant not only for normal cells but is also contributory to the genesis of many types of cancer. Cancers that have reduced expression of the NHLRC1 gene show downregulation of glycogenolysis due to the role of malin in nuclear localized glycogen phosphorylase. The reduction in nuclear glycogen metabolism leads to a lack of substrate for histone acetylation and contributes to the altered epigenetic landscape seen in many cancers.
Clinical Consequences of Defects in Glycogen Homeostasis
Glycogen Storage Diseases
Since glycogen molecules can become enormously large, an inability to degrade glycogen can cause cells to become pathologically engorged; it can also lead to the functional loss of glycogen as a source of cell energy and as a blood glucose buffer. Although glycogen storage diseases are quite rare, their effects can be most dramatic. The debilitating effect of many glycogen storage diseases depends on the severity of the mutation causing the deficiency. In addition, although the glycogen storage diseases are attributed to specific enzyme deficiencies, other events can cause the same characteristic symptoms. For example, Type 1 glycogen storage disease (von Gierke disease) is attributed to lack of glucose-6-phosphatase. However, the catalytic activity of this enzyme is localized to a domain of the membrane-localized enzyme present in the lumen of the endoplasmic reticulum (ER); in order to gain access to the phosphatase, glucose-6-phosphate must pass through a specific translocase in the ER membrane (see Figure below). Mutation of either the phosphatase or the translocase makes transfer of liver glycogen to the blood a very limited process. Thus, mutation of either gene leads to symptoms associated with von Gierke disease, which occurs at a rate of about 1 in 200,000 people.
The metabolic consequences of the hepatic glucose-6-phosphate deficiency of von Gierke disease extend well beyond just the obvious hypoglycemia that results from the deficiency in liver being able to deliver free glucose to the blood. The inability to release the phosphate from glucose-6-phosphate results in diversion into glycolysis and production of pyruvate as well as increased diversion onto the pentose phosphate pathway. The production of excess pyruvate, at levels above of the capacity of the TCA cycle to completely oxidize it, results in its reduction to lactate resulting in lactic acidemia. In addition, some of the pyruvate is transaminated to alanine leading to hyperalaninemia. Some of the pyruvate will be oxidized to acetyl-CoA which can’t be fully oxidized in the TCA cycle and so the acetyl-CoA will end up in the cytosol where it will serve as a substrate for triglyceride and cholesterol synthesis resulting in hyperlipidemia. The oxidation of glucose-6-phosphate via the pentose phosphate pathway leads to increased production of ribose-5-phosphate which then activates the de novo synthesis of the purine nucleotides. In excess of the need, these purine nucleotides will ultimately be catabolized to uric acid resulting in hyperuricemia and consequent symptoms of gout. The interrelationships of these metabolic pathways is diagrammed in the Figure below.
The glycogen storage diseases are divided into two primary categories: those that result principally from defects in liver glycogen homeostasis and those that represent defects in muscle glycogen homeostasis. The liver glycogen storage diseases result in hepatomegaly and hypoglycemia or cirrhosis, whereas the muscle glycogen storage diseases result in skeletal and cardiac myopathies and/or energy impairment. The most notable muscle glycogen storage disease is Pompe disease (type II GSD) due to it being featured in the recent movie “Extraordinary Measures”.
Several glycogenoses are the result of deficiencies in enzymes of glycolysis whose symptoms and signs are similar to those seen in McArdle disease (type V GSD). These include deficiencies in muscle phosphoglycerate kinase and muscle pyruvate kinase as well as deficiencies in fructose 1,6-bisphosphatase, lactate dehydrogenase and phosphoglycerate mutase.
Table of Glycogen Storage Diseases
|Type: Name||Enzyme Affected||Gene||Primary Organ||Manifestations|
|GSD0A||liver isozyme of glycogen synthase||GYS2||liver||hypoglycemia, early death, hyperketonia, low blood lactate and alanine|
|glucose-6-phosphatase||G6PC||liver||hepatomegaly, severe fasting hypoglycemia, hyperlipidemia, hyperuricemia, kidney failure (Fanconi syndrome), thrombocyte dysfunction|
|GSD1b||microsomal glucose-6-phosphate transporter (G6PT1): this protein is a member of the solute carrier protein family and the gene is identified as SLC37A4||SLC37A4||liver||like 1a, but also associated with neutropenia and increased susceptibility to bacterial infections|
|lysosomal acid α-glucosidase|
also called acid maltase
|GAA||skeletal and cardiac muscle||infantile form = death by 2|
juvenile form = myopathy
adult form = muscular dystrophy-like
Cori or Forbes
|glycogen debranching enzyme||AGL||liver, skeletal and cardiac muscle||infant hepatomegaly, myopathy|
|glycogen branching enzyme||GBE1||liver, muscle||infantile hypotonia, hepatosplenomegaly, cirrhosis|
|muscle phosphorylase||PYGM||skeletal muscle||excercise-induced cramps and pain, myoglobinuria|
|GSD6: Hers||liver phosphorylase||PYGL||liver||hepatomegaly, mild fasting hypoglycemia, hyperlipidemia and ketosis, improvement with age|
|GSD7: Tarui||muscle-specific subunit of PFK-1||PKFM||muscle, RBC||like V, also hemolytic anemia|
|GSD9A1/A2||α subunit of hepatic phosphorylase kinase||PHKA2||liver||mildest form of GSD, hepatomegaly, growth retardation, elevated plasma AST and ALT, hypercholesterolemia, hypertriglyceridemia, fasting hyperketosis|
|GSD9B||common β subunit of phosphorylase kinase||PHKB||liver and muscle||marked hepatomegaly in early childhood, fasting hypoglycemia|
|GSD9C||γ subunit hepatic phosphorylase kinase||PHKG2||liver||increased glycogen in muscle as well as liver, hepatosplenomegaly, short stature, hypoglycemia, muscle weakness|
|GSD9D||α subunit muscle phosphorylase kinase||PHKA1||muscle||nighttime muscle cramping in childhood, late-onset exercise-induced muscle fatigue and cramping|
|GSD10||phosphoglycerate mutase||PGAM2||muscle||exercise-induced cramps, occasional myoglobinuria, exercise intolerance|
|GSD11||muscle-specific subunit of lactate dehydrogenase||LDHA||muscle||exercise-induced myoglobinuria, easily fatigued|
|Fanconi-Bickel (hepatorenal glycogenosis with renal Fanconi syndrome); was referred to as GSD11 but term no longer valid for this disease||glucose transporter-2 (GLUT-2)||SLC2A2||liver||is a GSD secondarily related to nonfunctional glucose transport; failure to thrive, hepatomegaly, rickets, proximal renal tubular dysfunction; also associated with a form of permanent neonatal diabetes mellitus|
|GSD12||aldolase A||ALDOA||liver, RBC||hepatosplenomegaly, non-spherocytic hemolytic anemia|
|GSD13||muscle predominant form of enolase: β-enolase||ENO3||muscle||myalgia, exercise intolerance|
|CDG1T (once called GSD14)||predominant form of phosphoglucomutase||PGM1||multiple affected tissues||this disease is a type 1 congenital disorder of glycosylation; associated with cleft lip, bifid uvula, short stature, hepatomegaly, hypoglycemia, exercise intolerance|
|GSD15||muscle predominant form of glycogenin||GYG1||muscle||muscle weakness, glycogen accumulation in heart, cardiac arrhythmias|
Lafora disease is a fatal autosomal recessive disease that is the result of a defect in glycogen metabolism. The disease gets its name from the Spanish neuropathologist, Gonzalo Rodriguez Lafora, who initially characterized the disorder. Lafora disease is characterized by the presence of inclusion bodies (the composition of which is glycogen), called Lafora bodies, in the cytoplasm of the cells of numerous tissues including neurons, heart, skeletal muscle, and liver.
Lafora disease is caused by loss-of-function mutations in either of two genes. One gene (EPM2A) encodes the EPM2A glucan (glycogen) phosphatase, which is commonly called laforin. The other gene (NHLRC1) encodes the NHL repeat containing E3 ubiquitin protein ligase 1, which is commonly called malin. The NHLRC1 gene was also known as EPM2B.
Laforin is a dual specificity phosphatase that removes phosphate from glycogen. The lack of laforin activity in Lafora disease leads to hyperphosphorylation of glycogen. Hyperphosphorylated glycogen disrupts the processes of branching and debranching leading to longer glucose chains than in a normal glycogen molecule. These abnormal glycogens (called polyglucans) are insoluble and ultimately lead to the formation of Lafora bodies. Normal laforin functions to maintain the low level of phosphomonoesters in glycogen which prevents the molecule from precipitating. As its gene name implies, malin is a ubiquitin ligase whose precise function in the process of glycogen dephosphorylation is incompletely understood However, evidence indicates that malin ubiquitylates several enzymes in glycogen homeostasis in a laforin-dependent manner. Together laforin and malin interact to regulate glycogen phosphorylation and chain length pattern, the latter of which is crucial to the solubility of glycogen in the cell.