Enzyme Assay for Diagnosis of Guanidinoacetate Methyltransferase Deficiency

Creatine in the human body is derived from two sources: the diet and endogenous synthesis. Biosynthesis involves two enzymes: l-arginine:glycine amidinotransferase (EC 2.1.4.1) and guanidinoacetate N-methyltransferase (GAMT; EC 2.1.1.2). Defects in both of these enzymes lead to inborn errors of metabolism (1)(2). Deficiency of l-arginine:glycine amidinotransferase has been reported very recently in two patients by Item et al. (3) and Bianchi et al. (4). GAMT deficiency was first described in 1994 (1)(5). The clinical presentation of the disease varies widely (2)(6). Some patients present with intractable epilepsy, extrapyramidal movement disorder, and severe retardation, whereas other patients have much milder symptoms and show only mild retardation and autistic behavior. The aspecific clinical picture of GAMT deficiency and the relatively good treatment options makes a good diagnostic strategy important. Magnetic resonance spectroscopy of the brains of affected patients shows absence of the creatine/creatine phosphate peak. However, magnetic resonance spectroscopy may not be performed on all patients with mild symptoms. Screening for metabolic disorders may also aid in the diagnosis because generalized increases in amino acids, organic acids, and other metabolites can be found when their concentrations are calculated relative to the creatinine concentration (2). The relatively low creatinine excretion in GAMT patients, which is attributable to the creatine deficiency, gives falsely increased values for a range of metabolites when they are calculated relative to creatinine. When GAMT deficiency is suspected, analysis of guanidinoacetate in plasma should be performed (7)(8). Confirmation of the defect can be obtained by enzymatic and DNA analysis (9)(10). Ilas et al. (11) described an assay for the determination of GAMT in fibroblasts and lymphoblasts. This method uses 14C-labeled guanidinoacetate as substrate. Product formation ([14C]creatine) is determined by HPLC with detection of the radioisotope. Because in healthy individuals GAMT activity is low in lymphoblasts and even lower in fibroblasts, large numbers of cells are needed for the analysis. We describe a new approach for the enzymatic diagnosis of GAMT deficiency. The method obviates the need for radioactive compounds because it uses stable-isotope-labeled substrates. Analysis of product formation is performed by stable-isotope-dilution gas chromatography–mass spectrometry (GC-MS). The high sensitivity of the assay enables reliable diagnosis of GAMT deficiency from a smaller number of cells. GAMT-deficient cultivated Epstein-Barr-virus-transformed lymphoblasts and skin fibroblasts were available from two and five patients with GAMT deficiency, respectively, as confirmed by increased concentrations of guanidinoacetate in body fluids and by mutation studies. Lymphoblast cell lines were also obtained from two carriers of GAMT deficiency. Control lymphoblasts and fibroblasts were available from pediatric patients without a diagnosis of inborn error of metabolism and from controls, and had originally been obtained for carrier screening. No defects were found, and the samples were anonymized. Lymphoblasts and fibroblasts were harvested from culture flasks and washed twice with Hanks Balanced Salt Solution (GibcoBRL). Pellets were resuspended in 200 mmol/L Tris-HCl (pH 8.5) and kept on ice. The cells were sonicated (3 times; 10 s each time at standard capacity) on ice with a Soniprep 150 Ultrasonic Desintegrator (MSE). Lysates were centrifuged for 5 min at 4 °C and 8800g (Hettich centrifuge Micro22R). After the protein concentrations were measured (bicinchoninic acid protein assay; Sigma), the supernatants were used for the enzyme assay. Tris-HCl and dithiothreitol were purchased from Merck. [2H3]-S-Adenosylmethionine (SAM-d3; >99 atom-% D) and [2H3]creatine (>99 atom-% D) were purchased from C/D/N Isotopes. [13C2]Guanidinoacetate was synthesized as described previously (8). The reaction mixture contained 100 mmol/L Tris-HCl (pH 8.5), 2 mmol/L dithiothreitol, 1.5 mmol/L SAM-d3, 1 mmol/L [13C2]guanidinoacetate, and cell homogenates containing 0.5–1 mg of protein for lymphoblasts and 0.1–0.4 mg (corresponding to less than one-half of a 75-cm2 culture flask) for fibroblasts in a final volume of 0.2 mL. Assay blanks were incubated with distilled water instead of [13C2]guanidinoacetate. The reaction mixture was incubated for 2 h at 37 °C. One 85-μL sample was taken at the start and one sample at the end of the incubation. The samples were stored at −20 °C until they were prepared for analysis of labeled creatine by GC-MS. As an internal standard, we added 62.5 pmoles of [2H3]creatine to the samples. Analysis of the formed labeled creatine was based on a method published for guanidinoacetate (8). To an 85-μL sample, we added 50 μL of saturated aqueous sodium bicarbonate, 50 μL of hexafluoroacetylacetone, and 500 μL of toluene. The mixture was heated to 80 °C for 2 h with continuous stirring and allowed to cool. From the upper toluene phase, 100 μL was transferred to another test tube and evaporated to dryness with nitrogen. Pentafluorobenzyl derivatives were formed by treating the residue with 10 μL of triethylamine and 100 μL of 70 mL/L pentafluorobenzyl bromide in acetonitrile at room temperature for 15 min. After the addition of 200 μL of 0.5 mol/L HCl, the formed derivatives were extracted with 1 mL of hexane. From this hexane extract, 1 μL was injected into the GC-MS instrument. GC-MS analyses were performed with a Hewlett Packard 5890 gas chromatograph connected to a Hewlett Packard mass spectrometer type Engine. Chromatographic separation was achieved on a SGE BPX-70 capillary column [25 m × 0.32 μm (i.d.); film thickness, 0.25 μm], coated with a very polar phase. Samples were injected splitless at a temperature of 260 °C. After 1 min at 80 °C, the column temperature was increased at a rate of 10 °C/min to 150 °C, followed by an increase at 30 °C/min to 260 °C. The temperature of the transfer line to the mass spectrometer was set at 300 °C. The column was inserted directly into the ion source, which was set at 200 °C. The quadrupole temperature was 150 °C. Methane containing 5% ammonia was used as moderating gas at an optimized gas pressure. The mass spectrometer was operated under electron capture negative chemical ionization in the single-ion monitoring mode. The negative ions measured for creatine were: m/z − 305 ([2H3]creatine, as internal standard) and m/z − 307 ([13C2]creatine-d3). Formation of [13C2-2H3]creatine during the incubation of control human lymphoblasts with SAM-d3 was found to be linear at incubation times up to 2 h (results not shown). Activity was determined by calculation of the amount of labeled creatine formed (c2h − c0) per hour per milligram of protein used, where c2h is the concentration at 2 h, and c0 is the initial concentration. Under the chosen assay conditions, the amount of creatine formed was <1 nmol. To investigate the possible occurrence of product inhibition, the assay was performed in the presence of various amounts of unlabeled creatine and S-adenosylhomocysteine (Fig. 1 ). Greater than 10% inhibition of [13C2-2H3]creatine formation was observed at added creatine or S-adenosylhomocysteine concentrations ≥25 nmol. Inhibition of [13C2-2H3]creatine formation by exogenous creatine and S-adenosylhomocysteine in control human lymphoblasts. The intraassay variation (CV) of the GAMT assay was determined by analyzing a single cell line eight times within one series and was 9% and 10% for fibroblasts and lymphoblasts, respectively. To determine the interassay variation, one cell line was analyzed in five distinct series and was 12% for both lymphoblasts and fibroblasts. GAMT activities in lymphoblasts and fibroblasts from controls, carriers, and patients with GAMT deficiency are shown in Table 1 . In all patients, activity was not detectable, pointing to a complete deficiency of the enzyme in lymphoblasts or fibroblasts. GAMT activity in lymphoblasts and fibroblasts from controls, carriers, and individuals with GAMT deficiency. GAMT activity in lymphoblasts and fibroblasts from controls, carriers, and individuals with GAMT deficiency. The method we describe here for the enzymatic diagnosis of GAMT deficiency is highly specific and very sensitive. The use of two stable-isotope-labeled substrates, [13C2]guanidinoacetate and SAM-d3 leads to the formation of a labeled product, [13C2-2H3]creatine, that is 5 mass units heavier than endogenous unlabeled creatine. As internal standard for the exact quantification of product formation [2H3]creatine is used. This internal standard gives no interference in the mass spectrum of the product, [13C2-2H3]creatine. We have also shown that GAMT can be inhibited by high concentrations of enzyme product. However, under standard assay conditions, no product inhibition occurs because the amount of [13C2-2H3]creatine formed is <1 nmol, and significant inhibition only occurs at product concentrations ≥25 nmol. Using 500-1000 μg of cellular protein, we were able to establish reference values. The activities that we found were different from the published values. This may be ascribed to the purification of the cell homogenate, which was used by Ilas et al. (11). In our procedure, no dialysis is needed. We applied our method to cell lines obtained from patients with a confirmed GAMT deficiency and found an absence of GAMT activity in both fibroblasts and lymphoblasts, demonstrating the applicability of the method for enzymatic confirmation of the defect. In heterozygous siblings of two patients, GAMT activity was within reference values, showing that this method cannot be used to detect carriers of GAMT deficiency. Prenatal diagnosis using this enzyme assay in cultured amniocytes may be possible. In summary, we describe a new approach to the enzymatic diagnosis of GAMT deficiency that obviates the use of radioactive substrates and is very sensitive. We thank Wjera Wickenhagen for expert technical assistance.