Method for Measurement of Peroxisomal Very-Long-Chain Fatty Acid β-Oxidation in Human Skin Fibroblasts Using Stable-Isotope-Labeled Tetracosanoic Acid

Peroxisomes are present in virtually every eukaryotic cell type except the mature erythrocyte. In higher eukaryotes, one of the main functions of peroxisomes is the β-oxidation of very-long-chain fatty acids (VLCFA; > 22 carbon atoms) (1). The importance of peroxisomal β-oxidation is emphasized by the existence of a variety of different diseases in which peroxisomal β-oxidation is impaired and VLCFA concentrations are increased (1)(2)(3)(4). Peroxisomal disorders can be categorized as (a) single peroxisomal enzyme deficiencies, including X-linked adrenoleukodystrophy (X-ALD) and disorders attributable to defects in one of the peroxisomal β-oxidation enzymes, such as acyl-CoA oxidase (AOX) deficiency and bifunctional protein (DBP) deficiency; and (b) disorders attributable to defects in peroxisome biogenesis. The peroxisome biogenesis disorders (PBDs) represent a continuum of clinical features ranging from the most severe form, Zellweger syndrome, through neonatal adrenoleukodystrophy to the least severe form, infantile Refsum disease. Currently, measurement of the peroxisomal fatty acid β-oxidation activity is performed with 1-[14C]-radiolabeled VLCFA substrates and one of two available methods: either in intact human skin fibroblasts cultured in monolayer (5); or in isolated fibroblasts permeabilized with digitonin (6). We investigated the feasibility of using deuterium-labeled tetracosanoic acid (D3-C24:0) as an alternative substrate to radiolabeled 1-[14C]-labeled C24:0 for the measurement of peroxisomal β-oxidation activity in cultured primary human skin fibroblasts. Before use, the purity of 24,24,24-D3-C24:0 (Larodan Fine Chemicals AB) was determined. The D3-C24:0 substrate contained ∼6% deuterium-labeled octadecanoic acid (D3-C18:0). Acetone was used to purify D3-C24:0 according to the following procedure: 4 mL of acetone was added to 20 mg of D3-C24:0. The sample was vortex-mixed vigorously, left at room temperature for 30 min, and centrifuged at 1600g for 10 min; approximately 80% of the acetone was then removed, and 3 mL of fresh acetone was added. This procedure was repeated two more times. After three washing steps with acetone, ∼80% of the acetone was removed, and the remaining acetone was evaporated at room temperature under a constant stream of nitrogen. The residue was weighed, and a stock solution of 10 mmol/L D3-C24:0 in absolute ethanol was prepared. After purification, the purity of D3-C24:0 was analyzed, and the contribution of the D3-C18:0 contaminant was determined to be <0.2%. Fibroblasts from healthy controls and patients with X-ALD were cultured in the absence or presence of 20 μmol/L D3-C24:0 in HAM-F10 tissue culture medium supplemented with 100 mL/L fetal calf serum, penicillin (100 IU/mL), streptomycin (100 IU/mL), and glutamine (2 mmol/L). Before usage, the D3-C24:0 stock solution was put in a water bath for 5 min, vortex-mixed, and diluted in HAM-F10 tissue culture medium to a final concentration of 20 μmol/L. Cells were used between passage numbers 6 and 18. For fatty acid analysis, cells were harvested with trypsin, washed twice with phosphate-buffered saline (PBS) and once with 9 g/L NaCl, dissolved in 200 μL of deionized water, and sonicated, and the protein concentration was determined. The peroxisomal β-oxidation activity was calculated by measurement of the amount of intracellular deterium-labeled hexadecanoic acid (D3-C16:0) present in nmol/mg of protein. In our method we chose D3-C16:0 as a marker for peroxisomal β-oxidation because of the availability of a D3-C16:0 internal standard, which enabled accurate calculation of the amount of D3-C16:0 present in the cells. Fatty acids were analyzed by electrospray ionization mass spectrometry using a recently described isotope-dilution method (7). For calculation of the amount of D3-C16:0, we constructed a five-point calibration curve. Of a calibration mixture containing D3-C16:0 (40 μmol/L), we added 0, 25, 50, 100, and 200 μL to 100 μL of internal standard containing deuterium-labeled behenic acid (D4-C22:0; 50.0 μmol/L), D4-C24:0 (50.0 μmol/L), and deuterium-labeled hexacosanoic acid (D4-C26:0; 1.0 μmol/L). Samples were extracted and analyzed as described previously (7). The input concentration of D3-C16:0 (in nmol) was plotted against the ratio of the peak height of D3-C16:0 to the peak height of the D4-C22:0 internal standard. The trend line and the intercept were used to calculate the D3-C16:0 concentration in the samples. The effect of incubation time on the production of D3-C16:0 from D3-C24:0 in fibroblasts from healthy individuals and patients with X-ALD is shown in Fig. 1 . At all time points investigated, the amount of D3-C16:0 in the X-ALD cell lines was markedly lower than that in the control cell lines. Because D3-C16:0 is an intermediate of peroxisomal β-oxidation and not an end product, the amount of D3-C16:0 measured in the cells at the different time points reflects the flux through the β-oxidation pathway and hence is an indicator of the overall activity of the pathway. After 2–3 days, the amount of D3-C16:0 present in control and X-ALD cell lines plateaued, indicating that a steady state was reached. To exclude depletion of the substrate in the medium, we measured the amount of D3-C24:0 present in the medium after 72 h. The medium of the control cells still contained >80% of the initial D3-C24:0 concentration. On the basis of the data presented in Fig. 1 , we selected a 3-day incubation period with 20 μmol/L D3-C24:0 for subsequent studies. Effect of incubation time on D3-C16:0 concentrations. Primary human skin fibroblasts from healthy individuals (▪) and X-ALD patients (○) were cultured under standard tissue culture conditions, for the time indicated, in the presence of 20 μmol/L D3-C24:0, and harvested. The amount of D3-C16:0 present in the cells was then measured. P values were calculated by the two-tailed Student t-test. Values are the mean (SD; error bars). ∗, P <0.01. The intraassay CV was determined by the following procedure: the cells were divided into five separate tissue culture flasks, treated with 20 μmol/L D3-C24:0, and after 3 days, the amount of D3-C16:0 present in the cells was measured. The intraassay CV obtained was 5.8%. The interassay CV, determined by assaying control cell lines during 5 separate weeks, was 8.8%. The D3-C16:0 concentrations present after 3 days of incubation with 20 μmol/L D3-C24:0 of fibroblasts from controls and patients with different peroxisomal β-oxidation disorders, including X-ALD, AOX deficiency, and DBP deficiency, are summarized in Table 1 . In addition, cells from different PBD patients were analyzed as well. D3-C16:0 concentrations in skin fibroblasts from controls and patients after 3 days of incubation with D3-C24:0. D3-C16:0 concentrations in cultured skin fibroblasts after 3 days of incubation with 20 μmol/L D3-C24:0. D3-C16:0 concentrations in control fibroblasts were used to calculate the relative amounts in patient cell lines. P values were calculated by use of the two-tailed Student t-test. D3-C16:0 concentrations in skin fibroblasts from controls and patients after 3 days of incubation with D3-C24:0. D3-C16:0 concentrations in cultured skin fibroblasts after 3 days of incubation with 20 μmol/L D3-C24:0. D3-C16:0 concentrations in control fibroblasts were used to calculate the relative amounts in patient cell lines. P values were calculated by use of the two-tailed Student t-test. In none of the six AOX- or DBP-deficient cell lines could D3-C16:0 be detected. The cells had taken up the D3-C24:0 substrate, as we concluded from measurement of intracellular D3-C24:0 concentrations. These data indicate that β-oxidation of C24:0 to C16:0 takes place exclusively in peroxisomes and not in mitochondria. In fibroblasts derived from PBD patients, the amount of D3-C16:0 was only 5% of the amount in control fibroblasts (Table 1 ). Among the different PBD patient cell lines analyzed, however, we observed variation in residual peroxisomal β-oxidation activity, as indicated by the amount of intracellular D3-C16:0 present. No D3-C16:0 was detectable in eight cell lines derived from patients with the severe Zellweger phenotype, whereas D3-C16:0 was detectable in four cell lines derived from patients with the milder neonatal adrenoleukodystrophy or infantile Refsum disease phenotypes. These observations are in agreement with a previous study that reported the predictive value of dihydroxyacetonephosphate acyltransferase (DHAPAT) activity and residual peroxisomal VLCFA β-oxidation activity, measured with 1-[14C]-C24:0 as substrate, for the life expectancy of PBD patients (8). Fibroblasts derived from patients with X-ALD had the highest (15%) relative amount of D3-C16:0 formed (Table 1 ). The mean (SD) amount of D3-C16:0 in X-ALD fibroblasts was 0.60 (0.42) nmol/mg of protein. Within the group of 12 X-ALD patients included in the analysis, no correlation was observed between the peroxisomal β-oxidation activity and the phenotype of the patient. In conclusion, we have developed an easy, sensitive, nonradioactive method for analysis of peroxisomal β-oxidation activity in fibroblasts. We thank Herman ten Brink and Rob Ofman for helpful technical suggestions and discussion. This work was supported by grants from the Netherlands Organization for Scientific Research (NWO-MW: No. 903-42-077), the European Leukodystrophy Association, and European Union Project LSHM-CT-2004-502987.