How many forms of glycogen storage disease type I?

 Glucose-6-phosphatase is a multicomponent enzymatic system of the endoplasmic reticulum, which catalyses the terminal steps of gluconeogenesis and glycogenolysis by converting glucose-6-phosphate to glucose and inorganic phosphate. Glycogen storage diseases type I (GSD I) are a group of metabolic disorders arising from a defect in a component of this enzymatic system, i.e. the glucose-6-phosphate hydrolase (GSD Ia), the glucose-6-phosphate translocase (GSD Ib) and possibly also the translocases for inorganic phosphate (GSD Ic) or glucose (GSD Id). The genes encoding the glucose-6-phosphate hydrolase and the glucose-6-phosphate translocase have both been cloned and assigned to human chromosomes 17q21 and 11q23, respectively. Investigation of patients with GSD I shows that those with GSD Ia are mutated in the glucose-6-phosphate hydrolase gene, whereas those diagnosed as GSD Ib, GSD Ic or GSD Id are mutated in the glucose-6-phosphate translocase gene, and are therefore GSD Ib patients, in agreement with the fact that they all have neutropenia or neutrophil dysfunction. This suggests that the biochemical assays used to differentiate GSD Ic and GSD Id from GSD Ib are not reliable. Conclusion In practice therefore appears to be only two types of GSD I (Ia and Ib), which can be differentiated by (1) measurement of glucose-6-phosphatase activity in fresh and detergent-treated homogenates and (2) by mutation search in the genes encoding the glucose-6-phosphate hydrolase and the glucose-6-phosphate translocase. Received: 20 July 1999 and in revised form: 1 October 1999 / Accepted: 1 October 1999     

Impaired monocyte function in glycogen storage disease type Ib

Phagocyte (neutrophil and monocyte) function was evaluated in a boy with glycogen storage disease type Ib. Neutrophils were found to be defective in motility and respiratory burst and monocytes showed a defect in respiratory burst but not in motility. These results suggested that the glucose-6-phosphate transport system plays a role in the function of neutrophils and monocytes and that these two phagocytes are different from each other in their energy metabolism for motility.Abbreviations GSD Ia Glycogen storage disease type Ia - GSD Ib glycogen storage disease type Ib - G6P glucose-6-phosphate - G6Pase glucose-6-phosphatase - PHA phytohemagglutinin - CL chemiluminescence - ZAS zymosan-activated serum - FMLP N-formylmethionyl-leucyl-phenylalanine - CSA colony-stimulating activity     

Glycogen storage disease type Ib: familial bleeding tendency

A mild bleeding tendency with characteristics of the von Willebrand disease was documented in family members of a girl with glycogen storage disease type Ib (GSD) Ib). It was assumed that a defective glucose-6-phosphate dependent microsomal glycoprotein synthesis was involved in the bleeding disorder of the patient and the GSD Ib heterozygotes.Abbreviations GSD Ia glycogen storage disease type Ia - GSD Ib glycogen storage disease type Ib - G6P glucose-6-phosphate - G6P-ase glucose-6-phosphatase     

Glucose-6-phosphate: A key compound in glycogenosis I and favism leading to hyper- or hypolipidaemia

The glycogen storage disorders (GSD)-I,-III,-VI and VIII are associated with hypertriglyceridaemia or mixed hyperlipidaemia which poses the question whether these patients have an increased risk for atherosclerosis. The atherogenicity of triglycerides has remained controversial, while increased plasma cholesterol levels are generally accepted as a significant risk factor for coronary heart disease. However, clinical data show that one has to differentiate between metabolic conditions where triglycerides are atherogenic and those which are not significantly related to early onset of atherosclerosis but may cause other disorders such as pancreatitis. Among the disorders of carbohydrate metabolism patients with diabetes mellitus frequently have enhanced plasma triglycerides associated with a higher risk for coronary heart disease, while patients with certain types of glycogen storage disease have high triglyceride levels but do not seem to have an enhanced risk for atherosclerosis. Here we have compared the biochemical abnormalities and the atherogenic risk of three different disorders of glucose metabolism including GSD-I (glucose-6-phosphatase deficiency), favism (glucose-6-phosphate dehydrogenase deficiency), and diabetes mellitus which are related to either hyper- or hypolipidaemia. The available data indicate that glucose-6-phosphate (Glc-6-P) is a central molecule in cellular glucose metabolism which critically influences pentose phosphate cycle activity and, via NADPH₂-generation, regulates glutathione peroxidase activity for radical detoxification and also cholesterol and triglyceride synthesis. Radical detoxification is a major protective factor for cell membrane integrity and together with an appropriate renewal of membrane lipids may protect against the development of atherosclerosis. In GSD-I, the accumulation of Glc-6-P leads to enhanced radical detoxification and thereby increases the generation of reducing-equivalents. In contrast in both glucose-6-phosphate dehydrogenase deficiency and diabetes mellitus there is an impaired activity of the glutathione-peroxidase system and the NADPH₂-generation pentose phosphate cycle which might predispose these patients for membrane damage thus leading to haemolysis, platelet activation and atherosclerosis.     

Increased de novo lipogenesis and delayed conversion of large VLDL into intermediate density lipoprotein particles contribute to hyperlipidemia in glycogen storage disease type 1a

Glycogen storage disease type 1a (GSD-1a) is a metabolic disorder characterized by fasting-induced hypoglycemia, hepatic steatosis, and hyperlipidemia. The mechanisms underlying the lipid abnormalities are largely unknown. To investigate these mechanisms seven GSD-1a patients and four healthy control subjects received an infusion of [1-(13)C]acetate to quantify cholesterogenesis and lipogenesis. In a subset of patients, [1-(13)C]valine was given to assess lipoprotein metabolism and [2-(13)C]glycerol to determine whole body lipolysis. Cholesterogenesis was 274 +/- 112 mg/d in controls and 641 +/- 201 mg/d in GSD-1a patients (p < 0.01). Plasma triglyceride-palmitate derived from de novo lipogenesis was 7.1 +/- 9.4 and 86.3 +/- 42.5 micromol/h in controls and patients, respectively (p < 0.01). Production of VLDL did not show a consistent difference between the groups, but conversion of VLDL into intermediate density lipoproteins was relatively retarded in all patients (0.6 +/- 0.5 pools/d) compared with controls (4.3 +/- 1.8 pools/d). Fractional catabolic rate of intermediate density lipoproteins was lower in patients (0.8 +/- 0.6 pools/d) compared with controls (3.1 +/- 1.5 pools/d). Whole body lipolysis was similar, i.e., 4.5 +/- 1.9 micromol/kg/min in patients and 3.8 +/- 1.9 micromol/kg/min in controls. Hyperlipidemia in GSD-1a is associated with strongly increased lipid production and a slower relative conversion of VLDL to LDL.     

Acute myelogenous leukemia and glycogen storage disease 1b

Glycogen storage disease 1b (GSD 1b) is caused by a deficiency of glucose-6-phosphate translocase and the intracellular accumulation of glycogen. The disease presents with failure to thrive, hepatomegaly, hypoglycemia, lactic acidosis, as well as neutropenia causing increased susceptibility to pyogenic infections. We present a case of a young woman with GSD 1b who developed acute myelogenous leukemia while on long-term granulocyte colony-stimulating factor therapy. The presence of two rare diseases in a single patient raises suspicion that GSD 1b and acute myelogenous leukemia are linked. Surveillance for acute myelogenous leukemia should become part of the long-term follow-up for GSD 1b.     

Glucose-6-phosphatase mutation G188R confers an atypical glycogen storage disease type 1b phenotype

Glycogen storage disease type 1a (GSD 1a) is caused by a deficiency in microsomal glucose-6-phosphatase (G6Pase). A variant (GSD 1b) is caused by a defect in the transport of glucose-6-phosphate (G6P) into the microsome and is associated with chronic neutropenia and neutrophil dysfunction. Mutually exclusive mutations in the G6Pase gene and the G6P transport gene establish GSD la and GSD 1b as independent molecular processes and are consistent with a multicomponent translocase catalytic model. A modified translocase/catalytic unit model based on biochemical data in a G6Pase knockout mouse has also been proposed for G6Pase catalysis. This model suggests coupling of G6Pase activity and G6P transport. A 5-mo-old girl with hypoglycemia, hepatomegaly, and lactic acidemia was diagnosed with GSD 1a. She also developed neutropenia, neutrophil dysfunction, and recurrent infections characteristic of GSD 1b. Homozygous G188R mutations of the G6Pase gene were identified, but no mutations in the G6P translocase gene were found. We have subsequently identified a sibling and two unrelated patients with similar genotypic/phenotypic characteristics. The unusual association of neutrophil abnormalities in patients with homozygous G188R mutations in the G6Pase gene supports a modified translocase/catalytic unit model.     

Treatment of glycogen storage diseases

This overview shows the present state of the art in the treatment of glycogen storage diseases (GSD) illustrated by some characteristic courses of glucose-6-phosphatase deficiency (GSD type I) and of phosphorylase b-kinase deficiency (GSD type VIa). In the majority of our patients suffering from GSD type I the combination of nocturnal gastric drip feeding (GDF) using oligosaccharides with frequent daytime meals using high amounts of glucose, it's polymers and low amounts of uncooked starch is better accepted and more effective than a round the clock diet using high amounts of uncooked starch without the use of GDF. In one of three patients suffering from GSD type VIa dextro-thyroxine has been shown to be very effective concerning linear growth velocity, liver size, hyperlipidaemia and hypertransaminasaemia. Finally, the need and availability of prenatal diagnosis is discussed in view of the rather limited therapeutical efficacy in most of the GSD.     

Historical highlights and unsolved problems in glycogen storage disease type 1

Thirty-three years after Von Gierke described the first patient with glycogen storage disease type 1 (GSD1) in 1929, the Coris detected glucose-6-phosphatase (G6Pase) deficiency. The first mutation of this enzyme was identified 41 years later and subsequently the gene was mapped to chromosome 17q21, its enzyme topology defined, a nine transmembrane helical model suggested, an enzyme deficient knockout mouse created and by infusing an adenoviral vector associated murine G6Pase gene, correction of the clinical and laboratory abnormalities was observed. A similar successful gene transfer has been performed in enzyme deficient canine puppies. To explain the function of the G6Pase complex, a multicomponent translocase catalytic model has been proposed in which different transporters shuttle glucose-6-phosphate (G6P), inorganic phosphate (Pi) and glucose across the microsomal membrane. It was suggested that GSD1b patients suffered from a G6P transporter (G6PT) defect and the first mutation in the G6PT gene subsequently recognised. The gene mapped to chromosome 11q23 and its structural organisation was defined which showed a close functional linkage between G6PT and hydrolysis. Nordlie identified the first patient with Pi transport deficiency (GSD1c). However putative GSD1c and 1d patients based on kinetic studies were found to harbour mutations in the G6PT gene so that GSD1 patients are presently divided into 1a and non-1a. G6PT deficient patients suffer from numerical and functional leucocyte defects. A mRNA leucocyte G6PT deficiency has been suggested to account for the glucose phosphorylation and subsequent calcium sequestration defects observed in theses cells. Inflammatory bowel disease which occurs frequently in GSD non-1a patients has been related to their leucocyte abnormalities. Dietary management of GSD1 patients, designed to maintain a normal blood glucose level can be achieved during the night by nocturnal gastric infusions of glucose-containing solution or by the administration of uncooked cornstarch around the clock or by a combination of both. Both therapeutic modalities, if conducted in a meticulous manner, have a major impact on the quality of life, prevention of complications and subsequent prognosis. Open questions relate to the source of endogenous glucose production in GSD1 patients which increases as a function of age from 50% to 100% of normal, concomitant with an improvement in the patients fasting tolerance. Several complications, the nature of which is incompletely understood, tend to occur after the first decade: Liver adenomata with a small risk of transforming into hepatoma, progressive renal disease, which may be related to the hyperlipidaemia observed in this disease, often leading to end stage renal failure, osteopenia apparently based on high bone turnover, growth retardation and delayed puberty.Conclusion: this review highlights the present knowledge of glycogen storage disease type 1 and subtypes, discussing unsolved questions, which reflect the limitation of our knowledge in the understanding of this intriguing group of diseases. Published online: 23 August 2002     

Glycogen storage disease type I: diagnosis and phenotype/genotype correlation

Glycogen storage disease type Ia (GSD Ia) is caused by mutations in theG6PC gene encoding the phosphatase of the microsomal glucose-6-phosphatase system. GSD Ia is characterized by hepatomegaly, hypoglycemia, lactic acidemia, hyperuricemia, hyperlipidemia and short stature. Other forms of GSD I (GSD I non-a) are characterized by the additional symptom of frequent infections caused by neutropenia and neutrophil dysfunction. GSD I non-a is caused by mutations in a gene encoding glucose-6-phosphatase translocase (G6PT1). We report on the molecular genetic analyses of G6PC and G6PT 1 in 130 GSD Ia patients and 15 GSD I non-a patients, respectively, and provide an overview of the current literature pertaining to the molecular genetics of GSD I. Among the GSD Ia patients, 34 different mutations were identified, two of which have not been described before (A65P; F117C). Seventeen different mutations were detected in the GSD I non-a patients. True common mutations were identified neither in GSD Ia nor in GSD I non-a patients,Conclusion: Glycogen storage disease type Ia and and type I non-a are genetically heterogenous disorders. For the diagnosis of the various forms of glycogen storage disease type I, molecular genetic analyses are reliable and convenient alternatives to the enzyme assays in liver biopsy specimens. Some genotype-phenotype correlations exist, for example, homozygosity for oneG6PC mutation, G188R, seems to be associated with a glycogen storage disease type I non-a phenotype and homozygosity for the 727G>T mutation may be associated with a milder phenotype but an increased risk for hepatocellular carcinoma. Published online: 27 July 2002     

Hypercalcemia in glycogen storage disease type I patients of Turkish origin

Glycogen storage disease type I (GSD I) is an autosomal recessive disorder caused by defects in the glucose-6-phosphatase complex. Deficient activity in the glucose-6-phosphatase-a (G6Pase) catalytic unit characterizes GSD IA and defects in the glucose-6-phosphate transporter protein (G6PC) characterize GSD IB. The main clinical characteristics involve fasting hypoglycemia, hyperuricemia, hyperlactatemia, and hyperlipidemia. Hypercalcemia arose as an unknown problem in GSD I patients, especially in those with insufficient metabolic control. The aim of the present study was to obtain the prevalence of hypercalcemia and to draw attention to the metabolic complications of GSD I patients, including hypercalcemia in poor metabolic control. Hypercalcemia frequency and the affecting factors were studied cross-sectionally in 23 GSD I pediatric subjects. Clinical diagnosis of GSD I was confirmed in all patients either through documentation of deficient G6Pase enzyme activity levels on liver biopsy samples or through G6PC gene sequencing of DNA. Hypercalcemia was detected in 78.3% of patients with GSD I. Different from the previous report about hypercalcemia in a GSD IA patient who had R83H and 341delG mutations, we could not identify any genotype-phenotype correlation in our GSD I patients. Hyperlactatemia and hypertriglyceridemia correlated significantly with hypercalcemia. Furthermore, no differences in serum calcium concentrations could be demonstrated between patients with optimal metabolic control. We observed hypercalcemia in our series of GSD I patients during acute metabolic decompensation. Therefore, we speculate that hypercalcemia should be considered as one of the problems of GSD I patients during acute attacks. It may be related with prolonged lactic acidosis or may be a pseudohypercalcemia due to hyperlipidemia that can be seen in GSD I patients with poor metabolic control.     

Improved neutrophil function in a glycogen storage disease type 1b patient after liver transplantation

Patients with glycogen storage disease type 1b (GSD1b) not only show hepatomegaly, hypoglycaemia and lactic acidosis, but also neutropenia and neutrophil dysfunction. Here, we report improvement of neutropenia and neutrophil function in a 22-year-old male GSD1b patient who had undergone living-related partial liver transplantation (LT) at 18 years of age. After LT, the patients infectious episodes decreased, gastrointestinal symptoms ameliorated, neutrophil counts increased, and neutrophil function tests normalised. Conclusion:although it is not known whether this improvement was causally related to liver transplantation, this may be the first recorded case of restoration of neutrophil dysfunction in a glycogen storage disease type 1b patient.Abbreviations G6PT glucose-6-phosphate translocase - GSD glycogen storage disease - GSD1b glycogen storage disease type 1b - LT liver transplantation - PMA phorbol-12-myristate-13-acetate - rhGCSF recombinant human granulocyte colony stimulating factor     

Impaired metabolic function of polymorphonuclear leukocytes in glycogen storage disease Ib

To elucidate the basis for the recurrent infections in patients with glycogen storage disease (GSD) Ib we tested polymorphonuclear leukocyte (PMN) function in one patient. Bactericidal capacity and phagocytosis-induced O₂ consumption were reduced. Also, phorbol myristate acetate-stimulated superoxide production and glucose oxidation through the hexose monophosphate shunt were diminished compared to control subjects. Therefore it could be speculated that in PMN of patients with GSD Ib, glucose-6-phosphate has no access to the enzymes of the hexose monophosphate shunt due to a transport-related defect as shown for glucogenesis in hepatocytes.This work was supported by the Deutsche Forschungsgemeinschaft (Ga 148/6-1)     

Uric acid metabolism in therapy of glycogen storage disease type I.

Factors which may explain lower serum uric in a new therapy of patients with glycogen storage disease (GSD) type I have been studied. [1-14C]Glycine incorporation into urine uric acid was 0.68% of the injected dose during a 6-day period of frequent high carbohydrate feedings, 0.40% with the same diet and nocturnal nasogastric feeding by Vivonex, and 0.18% in a control patient with GSD type III. Fractional renal uric acid excretion in the patient with GSD type I increased from 11.3% to 26.3% after beginning nocturnal nasogastric feeding of Vivonex. Red cell phosphoribosylpyrophosphate leve,ls were not changed by the therapy. Addition of Vivonex nocturnal feedings to frequent high carbohydrate feedings (1) decreased the accelerated de novo purine synthesis to a level still higher than control and (2) increased fractional renal uric acid excretion.     

Glycogen storage disease type Ib without neutropenia

We report 2 patients with atypical glycogen storage disease type Ib without neutropenia or infectious complications. Neither patient was deficient in hepatic glucose-6-phosphatase activities in microsome-disrupted homogenates; both had mutations in the glucose-6-phosphate transporter gene, suggesting an allelic variant of glycogen storage disease type Ib.     

Cholesterol synthesis and de novo lipogenesis in premature infants determined by mass isotopomer distribution analysis

Premature infants change from placental supply of mainly carbohydrates to an enteral supply of mainly lipids earlier in their development than term infants. The metabolic consequences hereof are not known but might have long-lasting health effects. In fact, knowledge of lipid metabolism in premature infants is very limited. We have quantified de novo lipogenesis and cholesterogenesis on d 3 of life in seven premature infants (birth weight, 1319 +/- 417 g; gestational age, 30 +/- 2 wk). For comparison, five healthy adult subjects were also studied. All subjects received a 12-h [1-(13)C] acetate infusion, followed by mass isotopomer distribution analysis (MIDA) on lipoprotein-palmitate and plasma unesterified cholesterol. The fraction of lipoprotein-palmitate synthesized at the end of the infusion period was 5.4 +/- 3.9% in infants, which was in the same range as found in adult subjects on a normal diet, suggesting that hepatic de novo lipogenesis is not a major contributor to fat accumulation in these premature neonates. The fractional contribution of newly synthesized cholesterol to plasma unesterified cholesterol was 7.4 +/- 1.3% after a 12-h infusion. The calculated rate of endogenous cholesterol synthesis was 31 +/- 7 mg/kg/d, a value approximately three times higher than that found in adult subjects (10 +/- 6 mg/kg/d). These results indicate that the cholesterol-synthesizing machinery is well developed in premature infants.     

Biochemical diagnosis of Type 1b glycogen storage disease

A child with the classical signs and symptoms of Type 1 glycogen storage disease is presented who on investigation was shown to have a recently described variant of this disease known as Type 1b glycogen storage disease. A reliable and simple procedure for the diagnosis and differentiation of Types 1 and 1b glycogen storage disease is described, as the conventional diagnostic approach of assaying glucose-6-phosphate phosphohydrolase in frozen tissue will not diagnose Type 1b glycogen storage disease.
A portion of biopsy tissue should be maintained at a temperature near 0°C (but not frozen) and the remainder frozen. Glucose-6-phosphate phosphohydrolase assays are carried out on the tissue homogenates of both portions. In Type 1 glycogen storage disease, glucose-6-phosphate phosphohydrolase activity will be tow or absent in both frozen and unfrozen tissues. In Type 1b glycogen storage disease the frozen tissue homogenate will exhibit normal glucose-6-phosphate phosphohydrolase activity due to the disruption of the microsomes by tee crystals, white in the unfrozen tissue low levels of glucose-6-phosphate phosphohydrolase activity will be detected.     

Increased levels of hemostatic proteins are independent of inflammation in glycogen storage disease type Ia

OBJECTIVES: Glycogen storage disease type Ia (GSD-Ia), a congenital deficiency of hepatic glucose-6-phosphatase activity, is often associated with hyperproteinemia. To document the mechanism of hyperproteinemia, the proteins of the hemostatic system were analyzed according to their site of synthesis: hepatocyte, endothelial cell, or both. The role of inflammation was investigated by the measurement of tumor necrosis factor alpha (TNF-alpha) and interleukin-6 (IL-6) levels in plasma. METHODS: Twenty-seven patients with GSD-Ia were evaluated, as were 14 patients with other types of GSD and 30 healthy control subjects. Of the 41 patients with GSD, 15 also had hepatic adenoma (14 patients with GSD-Ia and 1 with GSD type III). RESULTS: In patients with GSD-Ia, there was a two-fold increase in all hepatocyte-synthesized proteins (i.e., factor VII, protein C, C4b binding protein) compared with control subjects and patients with other types of GSD. The proteins with mixed endothelial and hepatocyte origin (i.e., antithrombin and protein S) also were significantly increased but to a lesser extent. In contrast, the mean concentration of von Willebrand factor, which is exclusively synthesized in endothelial cells, was normal, as was the concentration of TNF-alpha and IL-6. CONCLUSIONS: These results suggest that the hyperproteinemia of GSD-Ia (including hemostatic proteins) is attributable to hepatocyte dysfunction and not related to an inflammatory process.     

Radical trapping in glycogen storage disease 1a

Oxidative mechanisms involving lipid peroxidation in the subendothelium of the arterial vessel wall play a key role in atherogenesis. Despite severe hyperlipidaemia, patients with glycogen storage disease type la (GSD1a) do not develop premature atherosclerosis. Therefore, we analysed parameters of antioxidative defence and oxidative stress in plasma and serum of patients with GSD1a (n=17) and compared them with those of patients with type 1 diabetes mellitus (n=17), familial hypercholesterolaemia (n=18) and healthy controls (n=20). We measured the total radical trapping ability parameter (TRAP), single plasma antioxidants (sulfhydryl-groups, uric acid, vitamin C, alpha-tocopherol, coenzyme-Q10), markers of lipid peroxidation, lipoprotein (a) and homocysteine. Patients with GSD1a showed an elevated TRAP (P<0.01) compared to the three other groups. This can mainly be attributed to elevated uric acid levels (P<0.05 versus control). Lipoprotein (a) was significantly lower in the GSD1a group compared to the three other groups (P<0.05).Conclusion: Patients with glycogen storage disease type 1a show an increased antioxidative defence in plasma which may protect them against lipid peroxidation and thus against premature atherosclerosis. Our finding of low lipoprotein (a) levels in this small group of patients warrants further investigation in a greater number of patients before assessing its role in atherogenesis in glycogen storage disease type 1a. Published online: 19 September 2002     

Is glycogen storage disease 1a associated with atherosclerosis?

Deficiency of microsomal glucose-6-phosphatase in liver and kidney leads to glycogen storage disease type la (GSD 1a). Notwithstanding intensive dietary therapy, moderate to severe dyslipidaemia and microal-buminuria, both known atherosclerotic risk factors, remain present. Although more patients reach adult age, no information is still available about accelerated athero-sclerosis. the aim of our study was to investigate whether GSD 1a was associated with premature atherosclerosis. In nine adolescent patients (mean age 22.7±3.4 years) and nine matched healthy control subjects, lipid profile, blood pressure, ankle-brachial indices, aortic distensibility and intima-media thickness (IMT) of the carotid and femoral arteries were determined. As expected, lipid profiles were significantly unfavourable in the patient group compared with the control group. No differences were found in blood pressure, ankle-brachial indices and aortic distensibility between both groups. IMT segments were comparable in both groups, with even thinner segments in the patient group. In different multivariate models, GSD la remained an independent predictor for a thinner IMT (R²=0.90; \=−0.69;P=0.018).Conclusion: glycogen storage disease type 1a is not associated with premature atherosclerosis, despite the existence of longstanding dyslipidaemia and microalbuminuria. Published online: 2 July 2002