Molecular pathology of glucose-6-phosphatase

It was known in the 1950s that hepatic microsomal glucose-6-phosphatase plays an important role in the regulation of blood glucose levels. All attempts since then to purify a single polypeptide with glucose-6-phosphatase activity have failed. Until recently, virtually nothing was known about the molecular basis of glucose-6-phosphatase or its regulation. Recent studies of the type 1 glycogen storage diseases, which are human genetic deficiencies that result in impaired glucose-6-phosphatase activity, have greatly increased our understanding of glucose-6-phosphatase. Glucose-6-phosphatase has been shown to comprise at least five different polypeptides, the catalytic subunit of glucose-6-phosphatase with its active site situated in the lumen of the endoplasmic reticulum; a regulatory Ca2+ binding protein; and three transport proteins, T1, T2, and T3, which respectively allow glucose-6-phosphate, phosphate, and glucose to cross the endoplasmic reticulum membrane. Purified glucose-6-phosphatase proteins, immunospecific antibodies, and improved assay techniques have led to the diagnosis of a variety of new type 1 glycogen storage diseases. Recent studies of the type 1 glycogen storage diseases have led to a much greater understanding of the role and regulation of each of the glucose-6-phosphatase proteins.     

Identification and characterisation of a new human glucose-6-phosphatase isoform

The liver endoplasmic reticulum glucose-6-phosphatase catalytic subunit (G6PC1) catalyses glucose 6-phosphate hydrolysis during gluconeogenesis and glycogenolysis. The highest glucose-6-phosphatase activities are found in the liver and the kidney; there have been many reports of glucose 6-phosphate hydrolysis in other tissues. We cloned a new G6Pase isoform (G6PC3) from human brain encoded by a six-exon gene (chromosome 17q21). G6PC3 protein was able to hydrolyse glucose 6-phosphate in transfected Chinese hamster ovary cells. The optimal pH for glucose 6-phosphate hydrolysis was lower and the K(m) higher relative to G6PC1. G6PC3 preferentially hydrolyzed other substrates including pNPP and 2-deoxy-glucose-6-phosphate compared to the liver enzyme.     

Subcellular localization of glucose-6-phosphatase in animal tissues.

1. Glucose-6-phosphatase (EC 3.1.3.9 D-glucose-6-phosphate phosphohydrolase) was found to be localized mainly in the endoplasmic reticulum (microsomal fraction) of all species of vertebrate liver tissue examined. 2. Hepatopancreas tissue from gastropod molluscs was found to be unique in showing the localization of glucose-6-phosphatase in the cytosol (soluble fraction).     

Use of Ethylenediaminetetraacetic Acid in the Cytologic Demonstration of Amylophosphorylase

Ethylenediaminetetraacetic acid (EDTA) was found to enhance strikingly the staining reaction of rat livers to the Takeuchi and Kuriaki method for phosphorylase (J. Histochem. Cytochera., 3: 153, 1955). Frozen sections of rat livers were prepared as described previously (S. H. Hori, Stain Techn., 39: 275, 1964), incubated in the medium containing: glucose-1-phosphate, 6 mM; adenosine-5-phosphate, 0.3 mM; EDTA, 20 mM; acetate buffer, pH 5.8, 80 mM, and the stain, developed with an aqueous solution of I₂-KI, 1:2:300. The result obtained with this technique was similar to that with the lead technique of the author, but it showed a more intense reaction in centrolobular areas than in periportal areas.     

DIFFERENTIATION OF ENDOPLASMIC RETICULUM IN HEPATOCYTES : I. Glucose-6-Phosphatase Distribution In Situ

The distribution of glucose-6-phosphatase activity in rat hepatocytes during a period of rapid endoplasmic reticulum differentiation (4 days before birth-1 day after birth) was studied by electron microscope cytochemistry. Techniques were devised to insure adequate morphological preservation, retain glucose-6-phosphatase activity, and control some other possible artifacts. At all stages examined the lead phosphate deposited by the cytochemical reaction is localized to the endoplasmic reticulum and the nuclear envelope. At 4 days before birth, when the enzyme specific activity is only a few per cent of the adult level, the lead deposit is present in only a few hepatocytes. In these cells a light deposit is seen throughout the entire rough-surfaced endoplasmic reticulum. At birth, when the specific activity of glucose-6-phosphatase is approximately equal to that of the adult, nearly all cells show a positive reaction for the enzyme and, again, the deposit is evenly distributed throughout the entire endoplasmic reticulum. By 24 hr postparturition all of the rough endoplasmic reticulum, and in addition the newly formed smooth endoplasmic reticulum, contains heavy lead deposits; enzyme activity at this stage is 250% of the adult level. These findings indicate that glucose-6-phosphatase develops simultaneously within all of the rough endoplasmic reticulum membranes of a given cell, although asynchronously in the hepatocyte population as a whole. In addition, the enzyme appears throughout the entire smooth endoplasmic reticulum as the membranes form during the first 24 hr after birth. The results suggest a lack of differentiation within the endoplasmic reticulum with respect to the distribution of glucose-6-phosphatase at the present level of resolution.     

DIFFERENTIATION OF ENDOPLASMIC RETICULUM IN HEPATOCYTES : II. Glucose-6-Phosphatase in Rough Microsomes

Electron microscope cytochemical localization of glucose-6-phosphatase in the developing hepatocytes of fetal and newborn rats indicates that the enzyme appears simultaneously in all the rough endoplasmic reticulum of a cell, although asynchronously within the hepatocyte population as a whole. To confirm that the pattern of cytochemical deposits reflects the actual distribution of enzyme sites, a method to subfractionate rough endoplasmic reticulum was developed. The procedure is based on the retention of the cytochemical reaction product (precipitated lead phosphate) within freshly prepared rough microsomes reacted in vitro with glucose-6-phosphate and lead ions. Lead phosphate increases the density of the microsomes which have glucose-6-phosphatase activity and thereby makes possible their separation from microsomes lacking the enzyme; separation is obtained by isopycnic centrifugation on a two-step density gradient. The procedure was applied to rough microsomes isolated from rats at several stages during hepatocyte differentiation and the results obtained agree with those given by cytochemical studies in situ. Before birth, when only some of the cells react positively for glucose-6-phosphatase, only a commensurate proportion of the rough microsome fraction can be rendered dense by the enzyme reaction. At the time of birth and in the adult, when all cells react positively, practically all microsomes acquire deposit and become dense after reaction. Thus, the results of the microsome subfractionation confirm the cytochemical findings; the enzyme is evenly distributed throughout all the endoplasmic reticulum of a cell and there is no regional differentiation within the rough endoplasmic reticulum with respect to glucose-6-phosphatase. These findings suggest that new components are inserted molecule-by-molecule into a pre-existing structural framework. The membranes are thus mosaics of old and new molecules and do not contain large regions of entirely "new" membrane in which all of the components are newly synthesized or newly assembled.     

The pentose phosphate pathway in the endoplasmic reticulum

Approximately the same levels of six of the seven enzymes catalyzing reactions of the pentose phosphate pathway are in the cisternae of washed microsomes from rat heart, spleen, lung, and brain. Renal and hepatic microsomes also have detectable levels of these enzymes except ribulose-5-phosphate epimerase and ribose-5-phosphate isomerase. Their location in the cisternae is indicated by their latencies, i.e. requirement for disruption of the membrane for activity. In addition, transketolase, transaldolase, and glucose-6-phosphatase, a known cisternal enzyme, are inactivated by chymotrypsin and subtilisin only in disrupted hepatic microsomes under conditions in which NADPH-cytochrome c reductase, an enzyme on the external surface, is inactivated equally in intact and disrupted microsomes. The failure to detect the epimerase and isomerase in hepatic microsomes is due to inhibition of their assays by ketopentose-5-phosphatase. Xylulose 5-phosphate is hydrolyzed faster than ribulose 5-phosphate. A mild heat treatment destroys hepatic xylulose-5-phosphatase and glucose-6-phosphatase without affecting acid phosphatase. These results plus the established wide distribution of glucose dehydrogenase, the microsomal glucose-6-phosphate dehydrogenase, and its localization to the lumen of the endoplasmic reticulum suggest that most mammalian cells have two sets of enzymes of the pentose phosphate pathway: one is cytoplasmic and the other is in the endoplasmic reticulum. The activity of the microsomal pentose phosphate pathway is estimated to be about 1.5% that of the cytoplasmic pathway.     

Effect of diabetes on the rat hepatic glucose-6-phosphatase system in endoplasmic reticulum subfractions

Diabetes-induced alterations in the activities of the components of the glucose-6-phosphatase system (i.e., the enzyme, the glucose-6-P translocase (T(1)), and the phosphate translocase (T(2)) were examined in smooth and rough subfractions of hepatic endoplasmic reticulum from streptozotocin-injected rats. A significant effect of diabetes on the maximal velocity of glucose-6-P hydrolysis by the enzyme was present in both endoplasmic reticulum subfractions (3.1-fold increase in rough endoplasmic reticulum; 3.8-fold increase in smooth endoplasmic reticulum). Based on latency values, diabetes did not result in a proportional increase in capacity of T(1) or T(2). In contrast to the control condition, the relationship between transport capacity and hydrolytic capacity was not significantly different in the two subfractions from diabetic animals. Elucidation of the effects of diabetes on the components of the glucose-6-phosphatase system associated with smooth and rough endoplasmic reticulum membranes enhances our understanding of the hepatic contribution to diabetic hyperglycemia.     

The antihyperglycemic effect of estrone sulfate in genetically obese-diabetic (ob/ob) mice is associated with reduced hepatic glucose-6-phosphatase.

Excessive glucose production by the liver contributes significantly to diabetic hyperglycemia. The enzyme system glucose-6-phosphatase plays a key role in regulating hepatic glucose production and therefore its inhibition is a potential therapeutic target for the correction of hyperglycemia. It has previously been shown that sulfated steroids, such as estrone sulfate and dehydroepiandrosterone sulfate, inhibit the glucose-6-phosphatase system in vitro, principally through inhibition of endoplasmic reticulum glucose-6-phosphate transport. We report here that in the obese/diabetic ob/ob mouse model, orally administered estrone sulfate reduces the abnormally elevated hepatic glucose-6-phosphatase enzyme activity and enzyme protein levels that are characteristic in the ob/ob mouse, and that this reduction is associated with normalization of blood glucose levels. Other sulfated and non-sulfated steroids also reduced, to a lesser extent, glucose-6-phosphatase enzyme activity - with the exception of dehydroepiandrosterone sulfate, which had no apparent effect on this system in ob/ob mice. Estrone sulfate is therefore an effective antihyperglycemic agent in ob/ob mice, and the glucose-6-phosphatase system can be successfully targeted for the therapeutic management of hyperglycemia in this animal model of non-insulin-dependent diabetes mellitus.     

Loss of endoplasmic reticulum membrane integrity: an image analysis of the glucose-6-phosphatase system in human hepatocyte.

Histochemical and cytochemical methods induce a loss of endoplasmic reticulum (ER) membrane integrity in hepatocytes. In order to evaluate the degree of ER membrane integrity, glucose-6-phosphatase (G6P-A) was localized in light and electron microscopy using glucose-6-phosphate (G6P) and mannose-6-phosphate (M6P) as substrates. In case of ER membrane alteration, M6P diffuses inside the ER and is hydrolysed by a non-specific phosphohydrolase. G6P and M6P hydrolysis was quantified with image analysis methods. In light microscopy, the ratio of reaction of M6P hydrolysis/G6P hydrolysis gave 75% of non specific reaction. In electron microscopic study this ratio was about 30%. These results showed that enzyme localization methods in electron microscopy produced less ER membrane alteration than light microscopic methods.     

Studies on vinblastine-induced autophagocytosis in mouse liver. IV. Origin of membranes

The origin of the limiting membranes of autophagic vacuoles (AV) in mouse hepatocytes was studied by cytochemical techniques. Autophagocytosis was induced by an intraperitoneal injection of vinblastine (50 mg/kg). The marker enzymes used were adenosine triphosphatase for the plasma membrane, glucose-6-phosphatase for the endoplasmic reticulum and thiamine pyrophosphatase for the Golgi apparatus and the endoplasmic reticulum. All the three enzymes showed a characteristic localization in both control and vinblastine-treated hepatocytes. The space between the limiting membranes of a few apparently newly formed AV's showed weak glucose-6-phosphatase activity. Neither adenosine triphosphatase nor thiamine pyrophosphatase activities were observed on or between the AV membranes. It was suggested that endoplasmic reticulum membranes may be used as a source of AV membranes in hepatocytes. The lack of glucose-6-phosphatase activity in the limiting membranes even of most of the newly formed AV's suggests a transformation process of the membranes destined to form AV, during which the enzyme activity characteristic for endoplasmic reticulum may disappear from them.     

Studies on vinblastine-induced autophagocytosis in mouse liver

Summary The origin of the limiting membranes of autophagic vacuoles (AV) in mouse hepatocytes was studied by cytochemical techniques. Autophagocytosis was induced by an intraperitoneal injection of vinblastine (50 mg/kg). The marker enzymes used were adenosine triphosphatase for the plasma membrane, glucose-6-phosphatase for the endoplasmic reticulum and thiamine pyrophosphatase for the Golgi apparatus and the endoplasmic reticulum. All the three enzymes showed a characteristic localization in both control and vinblastine-treated hepatocytes. The space between the limiting membranes of a few apparently newly formed AV's showed weak glucose-6-phosphatase activity. Neither adenosine triphosphatase nor thiamine pyrophosphatase activities were observed on or between the AV membranes. It was suggested that endoplasmic reticulum membranes may be used as a source of AV membranes in hepatocytes. The lack of glucose-6-phosphatase activity in the limiting membranes even of most of the newly formed AV's suggests a transformation process of the membranes destined to form AV, during which the enzyme activity characteristic for endoplasmic reticulum may disappear from them.     

The glucose-6-phosphatase enzyme in developing human trachea and oesophagus

Summary Glucose-6-phosphatase is an endoplasmic reticulum system which is found primarily in liver and kidney. Recently, it has become clear that it is also present in lower amounts in a variety of other tissues. Previous histochemical studies of glucose-6-phosphate hydrolysis in trachea have given equivocal results and only one study on adult oesophagus has shown glucose-6-phosphatase, enzymatic activity but without cellular localization. We have now shown, using microassay techniques, that microsomes isolated from human foetal trachea and oesophagus both contain low levels of specific glucose-6-phosphatase activity (mean= 0.9 and 1.5 nmol min⁻¹ mg⁻¹ microsomal protein, respectively) which are less than 10% of the levels in microsomes of human foetal liver of similar age. In the developing trachea, glucose-6-phosphatase immunoreactivity has been found, using a monospecific antibody to the catalytic subunit of the glucose-6-phosphatase enzyme, to be first present at 10–11 weeks' gestation, and thereafter in foetal life, predominantly present in ciliated cells, with smaller amounts in non-ciliated secretory cells, duct lining cells, and occasional basal cells. The foetal oesophageal epithelium is transiently ciliated from 10 to 11 weeks' gestation, but ciliated cells are gradually replaced by squamous cells from 14 to 16 weeks onwards. Glucose-6-phosphatase immunoreactivity in human foetal oesophagus is predominantly confined to ciliated cells, but non-ciliated luminal cells are also reactive, as are occasional basal cells. Mucus secretory cells in foetal trachea and oesophagus are immunonegative, as is the entire epithelium of both organs in the embryo (up to 56 postovulatory days).     

Studies on the effect of paraquat on glycogen mobilization in liver of common carp (Cyprinus carpio L.)

1. A herbicide, paraquat (1,1'dimethyl-4,4'-bipyridilium-dichloride) was administered to carp in 0.5-10.0 ppm concentrations, respectively, and blood sugar level, glucose-6-phosphatase and glycogen phosphorylase activities of liver were determined. 2. Paraquat treatment caused an increase of blood sugar level and enhanced phosphorylase and glucose-6-phosphatase activities. 3. Paraquat can induce alterations in endoplasmic reticulum that might contribute to the changes in glucose-6-phosphatase activity, resulting in an increase of blood glucose level and/or all the effects can be attributed to a high level of circulating epinephrine produced by paraquat toxicosis.     

Permeabilization of rat hepatocytes with Staphylococcus aureus alpha- toxin

Pathogenic staphylococci secrete a number of exotoxins, including alpha-toxin. alpha-Toxin induces lysis of erythrocytes and liposomes when its 3S protein monomers associate with the lipid bilayer and form a hexomeric transmembrane channel 3 nm in diameter. We have used alpha-toxin to render rat hepatocytes 93-100% permeable to trypan blue with a lactate dehydrogenase leakage less than or equal to 22%. Treatment conditions included incubation for 5-10 min at 37 degrees C and pH 7.0 with an alpha-toxin concentration of 4-35 human hemolytic U/ml and a cell concentration of 13-21 mg dry wt/ml. Scanning electron microscopy revealed signs of swelling in the treated hepatocytes, but there were no large lesions or gross damage to the cell surface. Transmission electron microscopy indicated that the nucleus, mitochondria, and cytoplasm were similar in control and treated cells and both had large regions of well-defined lamellar rough endoplasmic reticulum. Comparisons of the mannose-6-phosphatase and glucose-6-phosphatase activities demonstrated that 5-10 U/ml alpha-toxin rendered cells freely permeable to glucose-6-phosphate, while substantially preserving the selective permeability of the membranes of the endoplasmic reticulum and the functionality of the glucose-6-phosphatase system. Thus, alpha-toxin appears to have significant potential as a means to induce selective permeability to small ions. It should make possible the study of a variety of cellular functions in situ.     

Rat liver glucose-6-phosphatase system: light scattering and chemical characterization

Glucose-6-phosphatase is a multicomponent system located in the endoplasmic reticulum, involving both a catalytic subunit (G6PC) and several substrate and product carriers. The glucose-6-phosphate carrier is called G6PT1. Using light scattering, we determined K(D) values for phosphate and glucose transport in rat liver microsomes (45 and 33mM, respectively), G6PT1 K(D) being too low to be estimated by this technique. We provide evidence that phosphate transport may be carried out by an allosteric multisubunit translocase or by two distinct proteins. Using chemical modifications by sulfhydryl reagents with different solubility properties, we conclude that in G6PT1, one thiol group important for activity is facing the cytosol and could be Cys(121) or Cys(362). Moreover, a different glucose-6-phosphate translocase, representing 20% of total glucose-6-phosphate transport and insensitive to N-ethylmaleimide modification, could coexist with liver G6PT1. In the G6PC protein, an accessible thiol group is facing the cytosol and, according to structural predictions, could be Cys(284).     

The Protective Effects of Eugenol on Carbon Tetrachloride induced Hepatotoxicity in Rats

Our earlier studies in vitro have shown that eugenol inhibits liver microsomal monooxygenase activities and carbon tetrachloride (CCl₄)-induced lipid peroxidation (Free Rad. Res. 20,253-266,1994). The objective of the present investigation was to study the in vivo protective effect of eugenol against CCI₄ toxicity. Eugenol (5 or 25 mg/kg body wt) given orally for 3 consecutive days did not alter the levels of serum glutamic oxalacetic transaminase (SGOTJ, microsomal enzymes such as cytochrome P450 reductase, glucose-6-phosphatase (G-6-Pase) xenobiotic-metabolizing enzymes (aminopyrine-N-demethylase, N-nitrosodimethylamine-demethylase and ethoxyresorufin-O-deethylase) and liver histology. Doses of eugenol (5 or 25 mg/kg) administered intragastrically to each rat on three consecutive days i.e. 48 hr, 24 hr and 30 min before a single oral dose of CCU (2.5 ml/kg body wt) prevented the rise in SGOT level without appreciable improvement in morphological changes in liver. Eugenol pretreatment also did not influence the decrease in microsomal cytochrome P₄₅₀ content, G-6-Pase and xenobiotic-metabolizing enzymes brought about by CCI₄. Since eugenol is metabolized and cleared rapidly from the body, the dose schedule was modified in another experiment. Eugenol (0.2,1.0,5.0 or 25 mg/kg) when given thrice orally i.e. prior to (-1 hr) along with (0 hr) and after (+ 3 hr) the i.p. administration of CCI₄ (0.4 ml/kg) prevented significantly the rise in SGOT activity as well as liver necrosis. The protective effect was more evident at 1 mg and 5 mg eugenol doses. However, the decrease in microsomal G-6-Pase activity by CCI₄ treatment was not prevented by eugenol suggesting that the damage to endoplasmic reticulum is not protected. The protective effect of eugenol against CC1₄ induced hepatotoxicity is more evident when it is given concurrently or soon after rather than much before CCU treatment.     

Multiple roles of glucose-6-phosphatases in pathophysiology: State of the art and future trends

Background

The endoplasmic reticulum enzyme glucose-6-phosphatase catalyzes the hydrolysis of glucose-6-phosphate to glucose and inorganic phosphate. The enzyme is a part of a multicomponent system that includes several integral membrane proteins; the catalytic subunit (G6PC) and transporters for glucose-6-phosphate, inorganic phosphate and glucose. The G6PC gene family presently includes three members, termed as G6PC, G6PC2, and G6PC3. Although the three isoforms show a moderate amino acid sequence homology, their membrane topology and catalytic site are very similar. The isoforms are expressed differently in various tissues. Mutations in all three genes have been reported to be associated with human diseases.

Scope of review

The present review outlines the biochemical features of the G6PC gene family products, the regulation of their expression, their role in the human pathology and the possibilities for pharmacological interventions.

Major conclusions

G6PCs emerge as integrators of extra- and intracellular glucose homeostasis. Beside the well known key role in blood glucose homeostasis, the members of the G6PC family seem to play a role as sensors of intracellular glucose and of intraluminal glucose/glucose-6-phosphate in the endoplasmic reticulum.

General significance

Since mutations in the three G6PC genes can be linked to human pathophysiological conditions, the better understanding of their functioning in connection with genetic alterations, altered expression and tissue distribution has an eminent importance.     

The molecular basis of the type 1 glycogen storage diseases.

Microsomal glucose-6-phosphatase catalyses the last step in liver glucose production. Glucose-6-phosphatase deficiency, now termed type 1 glycogen storage disease, was first described almost 40 years ago but until recently very little was known about the molecular basis of the various type 1 glycogen storage diseases. Recently we have shown that at least six different proteins are needed for normal glucose-6-phosphatase activity in liver. Four of the proteins have been purified and three cloned. Study of the type 1 glycogen storage diseases has stimulated investigations of the mechanisms of small molecule transport across the endoplasmic reticulum membrane and demonstrated the existence of novel endoplasmic reticulum transport proteins for glucose and phosphate.