Rapid ultraperformance liquid chromatography-tandem mass spectrometry assay for a characteristic glycogen-derived tetrasaccharide in Pompe disease and other glycogen storage diseases.

BACKGROUND: Urinary excretion of the tetrasaccharide 6-α-D-glucopyranosyl-maltotriose (Glc4) is increased in various clinical conditions associated with increased turnover or storage of glycogen, making Glc4 a potential biomarker for glycogen storage diseases (GSD). We developed an ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) assay to detect Glc4 in urine without interference of the Glc4 isomer maltotetraose (M4). METHODS: Urine samples, diluted in 0.1% ammonium hydroxide containing the internal standard acarbose, were filtered, and the filtrate was analyzed by UPLC-MS/MS. RESULTS: We separated and quantified acarbose, M4, and Glc4 using the ion pairs m/z 644/161, 665/161, and 665/179, respectively. Response of Glc4 was linear up to 1500 µmol/L and the limit of quantification was 2.8 µmol/L. Intra- and interassay CVs were 18.0% and 18.4% (10 µmol/L Glc4), and 10.5% and 16.2% (200 µmol/L Glc4). Glc4 in control individuals (n = 116) decreased with increasing age from a mean value of 8.9 mmol/mol to 1.0 mmol/mol creatinine. M4 was present in 5% of urine samples. Mean Glc4 concentrations per age group in untreated patients with Pompe disease (GSD type II) (n = 66) were significantly higher, ranging from 39.4 to 10.3 mmol/mol creatinine (P < 0.001-0.005). The diagnostic sensitivity of Glc4 for GSD-II was 98.5% and the diagnostic specificity 92%. Urine Glc4 was also increased in GSD-III (8 of 9), GSD-IV (2 of 3) and GSD-IX (6 of 10) patients. CONCLUSIONS: The UPLC-MS/MS assay of Glc4 in urine was discriminative between Glc4 and M4 and confirmed the diagnosis in >98% of GSD-II cases.

Bone mineral density and markers of bone turnover in patients with glycogen storage disease types I, III and IX.

Patients with glycogen storage disease (GSD) types I, III and IX show reduced bone mineral content, but there is scarce data on new serum and urine markers of bone turnover or their relationship to bone densitometry. Six GSD I, four GSD III and four GSD IX patients underwent bone mineral density (BMD) measurement by dual-energy X-ray absorptiometry. Free pyridinoline (fPYD):creatinine and free deoxypyridinoline (fDPD):creatinine ratios were analysed on random urines. Procollagen type I C-terminal propeptide, procollagen type I N-terminal propeptide (PINP), carboxyterminal telopeptide of type I collagen and bone-specific alkaline phosphatase were analysed in serum. Some GSD I and GSD III patients had low or very low BMD. There was no difference in total body BMD z-score between the GSD types after adjusting for height (p=0.110). Bone marker analysis showed no consistent pattern. Urine fPYD:creatinine ratio was raised in four GSD I and two GSD III patients, while serum PINP was inappropriately low in some of these patients. There was no clear correlation between any markers of bone destruction and total body z-score, but the patient with the lowest total body z-score showed the highest concentrations of both urinary fPYD:creatinine and fDPD:creatinine ratios. We conclude that some GSD I and GSD III patients have very low bone mineral density. There is no correlation between mineral density and bone markers in GSD patients. The inappropriately low concentration of PINP in association with the raised urinary fPYD:creatinine and fDPD:creatinine ratios seen in two GSD I patients reflect uncoupling of bone turnover. All these findings taken together suggest that some GSD I and GSD III patients may be at an increased risk of osteoporosis.

Studies on alpha-ketoglutaric aciduria in type I glycogenosis.

Urinary excretion of the organic acids in patients with type I and III glycogenosis was investigated. In all patients with type I glycogenosis, urinary alpha-ketoglutarate concentration ws about 10 times the normal value. alpha-Ketoglutaric aciduria was not improved by the acute or prolonged administration of a large dose of factors for pyruvate- and alpha-ketoglutarate dehydrogenase complex. On the other hand, the level of alpha-ketoglutarate in the urine from type I patients decreased in conjunction with the decrease of plasma lactate and pyruvate concentration after repeated oral glucose loading. Oral citrate loading brought an increased excretion of alpha-ketoglutarate in type I glycogenosis. It is possible that alpha-ketoglutarate dehydrogenase in the rate-limiting step in tricarboxylic acid cycle and in patients with glycogenosis type I, the excessive excretion of alpha-ketoglutarate may be caused by the limited activity of alpha-ketoglutarate dehydrogenase with excessive substrate.

Glucose-containing oligosaccharides in the urine of patients with glycogen storage disease type II and type III.

Patients with glycogen storage disease type II and type III were recently found to excrete increased amounts of a glucose-containing tetrasaccharide DGlcp(alpha1 leads to 6)DGlcp(alpha1 leads to 4)DGlcp(alpha1 leads to 4)DGlc [Lennartson, G., Lundblad, A., Sjöblad, S., Svensson, S. and Ockerman, P.A. (1976) Biomed. Mass Spectrom. 3, 51--54]. In addition to this tetrasaccharide, urine from these patients also contains larger oligosaccharides containing only glucose. From urine of patients with glycogen storage disease type II and type III, three and four oligosaccharides respectively have been isolated. Structural studies including sugar analyses, methylation analyses, partial acid hydrolysis and optical rotation revealed that three compounds were present in the urine of both patients. Their proposed structures or partial structures are as follows: DGlcp(alpha1--6)DGlcp(alpha1--6)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlc, DGlcp(alpha1--4)DGlcp(alpha1--6)DGlcp(alpha1--6)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlc, and DGlcp(alpha1--6)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlcp(alpha1--6)DGlcp(alpha1--4)DGlcp(alpha1--4)DGlc. A fourth compound has been partially characterized as a branched heptasaccharide with four (1 leads to 4) linkages and two (1 lead to 6) linkages. Glycogen is possibly the origin of these compounds. However, the number of (1 leads to 6) linkages is higher than expected and may indicate a shorter distance between branches in glycogen than has been generally assumed.