Tandem Mass Spectrometric Determination of Malonylcarnitine: Diagnosis and Neonatal Screening of Malonyl-CoA Decarboxylase Deficiency

Malonic aciduria (OMIM 248360) is a rarely diagnosed autosomal-recessive inborn error of metabolism caused by congenital deficiency of malonyl-CoA decarboxylase (MCD; EC 4.1.1.9). Hypoglycemia, seizures, developmental delay, and cardiomyopathy are among the most common symptoms, but both clinical signs and the time of presentation of patients with MCD deficiency can be variable. To date, only eight cases have been reported in the literature (1)(2)(3)(4)(5)(6)(7)(8), and recently molecular defects within the MCD gene (MLYCD) have been elucidated for the first time in some of these cases (8)(9)(10)(11). At least in part, clinical heterogeneity might be caused by the fact that MCD is expressed in different compartments of the cell and that MLYCD mutations with different effects on the subcellular localization of the MCD protein may thus affect different metabolic pathways (Wightman et al., submitted for publication). To date, only symptomatic patients with MCD deficiency have been detected, and many of them were already severely handicapped at the time of diagnosis because of residues of acute metabolic crises or from episodes of cardiac decompensation, which may develop as a consequence of secondary carnitine deficiency. Typically, the key to diagnosis is the excessive amount of malonic acid in the patient’s urine, which can be detected by gas chromatography–mass spectrometry. This then leads to a confirmatory test, such as the measurement of MCD activity in cell extracts, or to molecular genetic testing. Detection of the carnitine ester of malonic acid has been mentioned previously in single cases of malonic aciduria (4)(6)(7). In the study reported here, we systematically investigated the concentration of malonylcarnitine in the blood of MCD-deficient patients by electrospray ionization tandem mass spectrometry (ESI-MS/MS). Because MCD deficiency may be amenable to dietary and medical treatment, we also evaluated whether an increase in blood malonylcarnitine is detectable in the neonatal period, which would be mandatory for early presymptomatic diagnosis. Details of the method used by our group have been reported previously (12). In summary, acylcarnitines were analyzed by ESI-MS/MS with an API 365 ESI-MS/MS system equipped with a TurboIon spray device, Series 200 lp HPLC pump, and a Series 200 autosampler (PE-SCIEX). For that purpose, 3.2-mm (1/8-inch) spots were punched from Guthrie test cards by an automated punching device and transferred to a 96-well microtiter plate. After the addition of internal standard solution (containing 7.5 pmol of l-[2H3]octanoylcarnitine), methanol was added, and after elution, the methanolic extract was evaporated to dryness. Acylcarnitines were derivatized to the corresponding butyl esters with butanolic hydrochloric acid. Samples were redissolved in acetonitrile–water–formic acid (50:50:0.025 by volume) and introduced into the ionization chamber. Acylcarnitines were measured in multiple-reaction monitoring (MRM) mode with the ion pairs 347.4/85.0 Da and 360.4/85.0 Da for l-[2H3]octanoylcarnitine and malonylcarnitine, respectively. Malonylcarnitine was quantified relative to l-[2H3]octanoylcarnitine, assuming the same extraction and ionization efficacy for both carnitine esters. Dried blood spots were obtained from two siblings with MCD deficiency at different time points. A detailed description of the clinical and biochemical data and the metabolic effect of their genotype (MLYCD 634insA/59insC) is to be published elsewhere (Wightman et al., submitted for publication). During the period of the study, both patients were in stable clinical condition without any signs of metabolic derangement. At the time of the first determination of malonylcarnitine, patient 1 was 4 years and 2 months of age, and at that time she was already on oral carnitine therapy; patient 2 was investigated for the first time at the age of 7 months before any treatment had been initiated. In addition, the neonatal Guthrie test cards of both patients were traced back, and the concentrations of acylcarnitines were also determined in these samples. To determine the reference interval for malonylcarnitine in dried blood samples, we analyzed 4333 unselected samples of healthy newborns. At the time of the first determination, the blood malonylcarnitine concentration was 2.4 μmol/L in patient 1 and 3.6 μmol/L in patient 2. At the follow-up investigations, the malonylcarnitine concentration ranged between 1.66 and 3.78 μmol/L (n = 8) in both patients. For comparison, in dried blood samples from the healthy controls, malonylcarnitine ranged from “not detectable” to 0.30 μmol/L (99.5th centile; n = 4333). In the neonatal test card of patient 1, which had been stored for 4 years and 4 months, malonylcarnitine was not detectable. In contrast, in the test card of patient 2, stored for 9 months, the malonylcarnitine concentration was 0.40 μmol/L. The linearity and precision of the assay and the recovery of malonylcarnitine could not be calculated because malonylcarnitine is not commercially available. The detection limit (signal + 3 SD of a sample free of analyte) for malonylcarnitine was 0.08 μmol/L, and the quantification limit (signal + 6 SD of a sample free of analyte) was 0.13 μmol/L. To determine the stability of malonylcarnitine in dried blood samples stored at room temperature, we reanalyzed eight samples from patients with MCD deficiency after various storage times at room temperature. All samples showed a decrease in malonylcarnitine concentration. Regression analysis with curve fitting revealed significant agreement (P <0.001) with a logarithmic model (y = eb*t × 100%). In this equation, y is the concentration of malonylcarnitine relative to the first measurement, t is the time of storage between the first and the consecutive measurement, and the constant b was calculated to be −0.0028. With this model, the half-life (t1/2) of malonylcarnitine in dried blood on filter paper stored at room temperature could be calculated to be 248 days (Fig. 1 ). Stability of malonylcarnitine in dried blood samples from patients with MCD deficiency. The X depict the repeated determinations of malonylcarnitine in eight dried blood samples after various periods of storage at room temperature. The curve shows the calculated function for the stability of malonylcarnitine: y = e−0.0028*t × 100%, where y is the concentration of malonylcarnitine relative to the initial measurement, and t is the time of storage at room temperature in days. With this equation and the malonylcarnitine concentration measured in the screening card of patient 2 after 296 days of storage, the initial malonylcarnitine concentration was calculated to be 0.92 μmol/L. With a presumptive malonylcarnitine concentration of ∼1 μmol/L in patients with MCD deficiency during the first days of life, the decrease in malonylcarnitine can be calculated. Concentrations should be expected to drop below the quantification limit after ∼730 days. Even with a 10-fold higher initial concentration (10 μmol/L), malonylcarnitine would no longer be detectable after ∼1500 days. These considerations make it plausible that we could not detect malonylcarnitine in the screening card of patient 1 (after storage for 1514 days). The diagnostic specificity of the test cannot yet be finally evaluated, although our preliminary data from 4333 newborns and the two patients with MCD deficiency seem to show that ESI-MS/MS has a high specificity for the detection of MCD deficiency. The highest concentration of malonylcarnitine in the healthy newborns was 0.43 μmol/L, whereas the malonylcarnitine concentrations in dried blood samples from the two patients with MCD deficiency was 3.9- to 8.8-fold higher than the highest value of the reference population. Several groups have reported that MCD-deficient patients benefit from a low-fat, high-carbohydrate diet, with the effects being improved urinary organic acid excretion and the avoidance of hypoglycemic episodes (2)(3). Both our patients and cases from the literature (4) also showed a dramatic effect of carnitine treatment, which demonstrates that cardiac decompensation can be avoided in MCD-deficient patients. Finally, by analogy with fatty acid oxidation disorders such as medium-chain acyl-CoA dehydrogenase deficiency (13), one may conclude that knowledge of the diagnosis will improve outcome because early intervention is then possible in catabolic episodes that otherwise may lead to metabolic derangement. It therefore seems reasonable to suppose that early diagnosis and treatment can reduce the morbidity and mortality associated with MCD deficiency and to include MCD deficiency in neonatal metabolic screening programs. In this study, the malonylcarnitine concentration in the neonatal screening test card was increased in only one of the two patients investigated, but this was a problem of stability after sample storage for several years. We conclude that (a) malonylcarnitine is detectable by ESI-MS/MS, (b) malonylcarnitine is increased in the blood of patients with malonic aciduria, and (c) ESI-MS/MS-based neonatal screening programs should be able to detect patients with MCD deficiency, a treatable metabolic disorder, before the development of symptoms.