Molecular Genetics and Metabolism 99 (2010) 142–148 Contents lists available at ScienceDirect Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme Biochemical profiling to predict disease severity in metachromatic leukodystrophy M.A.F. Tan a,b,c,1, M. Fuller a,b, Z.A.M.H. Zabidi-Hussin c, J.J. Hopwood a,b, P.J. Meikle a,b,*,2 a Lysosomal Diseases Research Unit, SA Pathology [at Women’s and Children’s Hospital], North Adelaide, SA 5006, Australia Department of Paediatrics, University of Adelaide, Adelaide, SA 5005, Australia c Department of Paediatrics, School of Medical Sciences, Universiti Sains Malaysia, Kubang Kerian, Kelantan 16150, Malaysia b a r t i c l e i n f o Article history: Received 14 July 2009 Received in revised form 9 September 2009 Accepted 10 September 2009 Available online 15 September 2009 Keywords: Sulfatide Mass spectrometry Skin fibroblasts Urine a b s t r a c t Metachromatic leukodystrophy is a neurodegenerative disease that is characterized by a deficiency of arylsulfatase A, resulting in the accumulation of sulfatide and other lipids in the lysosomal network of affected cells. Accumulation of sulfatide in the nervous system leads to severe impairment of neurological function with a fatal outcome. Prognosis is often poor unless treatment is carried out before the onset of clinical symptoms. Pre-symptomatic detection of affected individuals may be possible with the introduction of newborn screening programs. The ability to accurately predict clinical phenotype and rate of disease progression in asymptomatic individuals will be essential to assist selection of the most appropriate treatment strategy. Biochemical profiling, incorporating the determination of residual enzyme protein/ activity using immune-based assays, and metabolite profiling using electrospray ionization-tandem mass spectrometry, was performed on urine and cultured skin fibroblasts from a cohort of patients representing the clinical spectrum of metachromatic leukodystrophy and on unaffected controls. Residual enzyme protein/activity in fibroblasts was able to differentiate unaffected controls, arylsulfatase A pseudo-deficient individuals, pseudo-deficient compound heterozygotes and affected patients. Metachromatic leukodystrophy phenotypes were distinguished by quantification of sulfatide and other secondarily altered lipids in urine and skin fibroblasts; this enabled further differentiation of the late-infantile form of the disorder from the juvenile and adult forms. Prediction of the rate of disease progression for metachromatic leukodystrophy requires a combination of information on genotype, residual arylsulfatase A protein and activity and the measurement of sulfatide and other lipids in urine and cultured skin fibroblasts. Ó 2009 Elsevier Inc. All rights reserved. Introduction Metachromatic leukodystrophy (MLD3) is a neurodegenerative disease that is inherited in an autosomal recessive manner. The primary defect results from the decreased catalytic action of arylsulfatase A (ASA) on 3-O-sulfogalactosylceramide (sulfatide), resulting in its accumulation in the lysosomal network of cells in several peripheral organs, notably the nervous system. Whilst there is no patho* Corresponding author. Addresses: Baker IDI Heart and Diabetes Institute, 75 Commercial Road, Melbourne, Vic. 3004, Australia; P.O. Box 6492, St. Kilda Road Central, Melbourne, Vic. 8008, Australia. Fax: (+613) 8532 1100. E-mail address: peter.meikle@bakeridi.edu.au (P.J. Meikle). 1 Present address: Metabolic Services, Doping Control Centre, Universiti Sains Malaysia, 11800 Penang, Malaysia. 2 Present address: Baker IDI Heart and Diabetes Institute, Melbourne, Vic. 3006, Australia. 3 Abbreviations used: MLD, metachromatic leukodystrophy; ASA, arylsulfatase A; ASA-PD, ASA pseudo-deficiency; ASA-PD/MLD, ASA-PD compound heterozygote; PC, phosphatidylcholine; Cer, ceramide; GC, glucosylceramide; GM2, monosialoganglioside; LC, lactosylceramide, PG, phosphatidylglycerol; PI, phosphatidylinositol; rhASA, recombinant human ASA; CTH, ceramide trihexoside; SM, sphingomyelin; BMP bis(monoacylglycero)phosphate. 1096-7192/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2009.09.006 physiological effect from this accumulation in most peripheral tissues, the nervous system exhibits progressive demyelination that leads to severe impairment of neurological function with a fatal outcome [1]. MLD is classified into four clinical sub-types based on the age of onset. Clinically, the most severe sub-type or rapidly progressing type is the late-infantile form, in which death ensues 1- to 7years after diagnosis. The early- and late-juvenile types have a more protracted clinical course while the adult form is the slowest to develop neurological symptoms. The presenting clinical features differ in each of the four types of MLD even though they share the same pathophysiological defect in the nervous system [2]. Biochemical diagnosis of MLD is usually achieved through the measurement of ASA activity in peripheral blood leukocytes and/ or cultured skin fibroblasts [2]. However, diagnosis based on enzyme activity alone is complicated by the high frequency of ASA pseudo-deficiency (ASA-PD). The difficulty in diagnosis is further compounded by the low sensitivity and specificity of conventional enzymatic assays. Due to the complexity in interpreting results, a definitive diagnosis is usually only obtained after extensive testing with an array of supplementary laboratory assays [2]. 143 M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 In common with most lysosomal storage disorders, the prognosis for MLD patients is poor because diagnosis is usually made after the onset of substantial clinical symptoms. Currently, the only effective treatment available for MLD is hematopoietic stem cell transplant [3] which has an associated high mortality rate. Positive outcomes have been reported in the adult and juvenile forms of MLD following successful engraftment [4,5]. Results from hematopoietic stem cell transplants also showed that the highest efficacy was obtained when the transplant was carried out before the onset of clinical symptoms; this may be possible with a newborn screening program, as suggested previously [6,7]. However, early and accurate diagnosis for MLD, as well as the prediction of disease severity, will be important in the selection and timing of treatment strategies. To address this need we evaluated measurements of ASA activity and protein as biochemical markers using immunebased assays. We also established a profile of sulfatide and other lipid species in the urine and skin fibroblasts from the major forms of MLD, ASA-PD, ASA-PD compound heterozygotes (ASA-PD/MLD), and unaffected controls that have been subjected to sulfatide loading. This approach has enabled discrimination between the lateinfantile MLD (severe form), adult-onset and ASA-PD and shows promise for improved clinical assessment of MLD. Table 2 Skin fibroblast cell lines used to evaluate multiple parameters for the prediction of clinical severity in MLD. Cell line S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S15 S16 S17 S18 S19 S20 S21 S22 a b Materials and methods Patient samples Urine and skin fibroblasts from MLD patients, ASA-PD and ASAPD/MLD individuals were provided by the National Referral Laboratory for the Diagnosis of Lysosomal, Peroxisomal and Related Genetic Disorders, SA Pathology [Women’s and Children’s Hospital campus], Adelaide, SA, Australia. Urine samples from 11 MLD patients, six ASA-PD individuals and 18 unaffected controls (six samples each from adults, juveniles and infants) were analyzed. Samples from healthy controls were obtained with informed consent. The use of all samples was approved by the Research Ethics Committee of the Women’s and Children’s Hospital. Details and biochemical data of patients are shown in Table 1. Skin fibroblasts Age (years) a NA NA NA NA NA NA NA NA NA 31 19 17 8 7 11 8 6 1 3 2 3 4 Type Genotypeb Control Control Control Control Control ASA-PD ASA-PD ASA-PD/MLD ASA-PD/MLD Adult Adult Adult Juvenile Juvenile Juvenile Juvenile Juvenile Late-infantile Late-infantile Late-infantile Late-infantile Late-infantile ASA-PD/ASA-PD ASA-PD/ASA-PD ASA-PD/T274M ASA-PD/D169N SDEX2/I179S P426L/unknown SDEX2/unknown SDEX2/P426L G345C/T274M R244C/R288C SDEX2/unknown SDEX2/P426L, Y429S T274M/T274M DelCCT(EX7)/DelCCT(EX7) SDEX2/SDEX2 SDEX2/SDEX2 D335V/P377L NA, age not available. SDEX2, single deletion at exon 2; delCCT(EX7), deletion of CCT at exon 7. from 18 unaffected controls, two ASA-PD, two ASA-PD/MLD, three adult MLD, five juvenile MLD and five late-infantile MLD patients were used in this study. Details and biochemical data of these patients are shown in Table 2. Genotyping was carried out in the National Referral Laboratory for the Diagnosis of Lysosomal, Peroxisomal and Related Genetic Disorders. Reagents All solvents were of HPLC grade. Sulfatide (bovine brain) and phosphatidylcholine (PC 28:0) were purchased from Sigma–Aldrich, St. Louis, MO, USA. Ceramide (Cer 17:0), deuterated gluco- Table 1 Biochemical data of individuals providing urine samples in this study. Samplea UA1–6 (n = 6) UJ1–6 (n = 6) UI1–6 (n = 6) U1 U2 U3 U4 U5 U6 U7 U8 U9 U10 U11 U12 U13 U14 U15 U16 U17 Sex Age (years) Type ASA activityb Creatinine (mM) Genotype M F M M M M F M F F M M F F M F M 1–4 10–16 27–51 23 27 31 9 29 26 5 4 9 2 2 2 12 10 11 4 13 Control adult Control juvenile Control infant ASA-PD ASA-PD ASA-PD ASA-PD ASA-PD ASA-PD Late-infantile Late-infantile Late-infantile Late-infantile Late-infantile Late-infantile Juvenile Juvenile Juvenile Juvenile Juvenile 0.34 0.66 0.66 0.45 0.63 0.60c 0.30 0.04 0.30 0.15 0.11 0.14 0.12 0.38 NAe 0.20 NAe 5.9 8.8 3.0 1.2 12.0 8.1 1.6 2.4 6.6 4.7 1.4 3.7 4.7 9.4 10.7 0.9 11.3 ASA-PD/? ASA-PD/ASA-PD ASA-PD/ASA-PD ASA-PD/T391S ASA-PD/ASA-PD ASA-PD/ASA-PD + Y39C T274M/T274M A212 V/Y39C Not done Not done R84Q/R114X SDEX2/1027delCd SDEX2/P426L Not done Y306H/1622G > A(IVS5) InsG97–99/P426L Y306H/1622G > A(IVS5) a Urine samples from: UA1–6, unaffected adults; UJ1–6, unaffected juveniles; UI1–6, unaffected infants; U1–U6, ASA-PD individuals; U7–U12, late-infantile MLD; and U13– U17, juvenile MLD. b Blood leukocyte ASA activity measured using artificial substrate (control range 1–5 nmol/min/mg protein). c ASA activity (nmol/min/mg protein) measured in skin fibroblasts using artificial substrate (control range 6–50 nmol/min/mg protein). d SDEX2, single deletion at exon 2. e NA, not available. 144 M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 cerebroside (GC 16:0-d3), monosialogangliosides (GM2 22:1 and 24:1), deuterated lactosylceramide (LC 16:0-d3) and sulfatide 16:0 were purchased from Matreya Inc., Pleasant Gap, PA, USA. Phosphatidylglycerol (PG 14:0/14:0) and phosphatidylinositol (PI 16:0/16:0) were from Avanti Polar Lipids, Inc., Alabaster, AL, USA. Anti-ASA polyclonal antibody and recombinant ASA enzyme were prepared as described [7]. Skin fibroblast culture and preparation of cell lysates Human skin fibroblast cell lines were established from primary skin biopsies and cultured as described previously [8]. Cell extracts were prepared by sonication for 20 s in 200 lL of either 0.1 M Tris/ HCl, 0.25 M NaCl, pH 7.2, containing 0.1% NP-40 for enzyme assays, or in sterile water for lipid analysis. For lipid profiling experiments cell lines were cultured (in triplicate, 75 cm2 flasks) in BME with 10% FCS to seven-days post-confluence, as described previously [8]. The cell lines were washed with PBS (3  5.0 mL) and cultured with BME (8.0 mL) containing 5% HIFCS, 1% penicillin/streptomycin and bovine brain sulfatide (64 lM). The cell lines were incubated (5% CO2, 37 °C) for a further eight days, harvested and cell lysates prepared as described above. Total protein was estimated using the bicinchoninic acid method [9]. Lipids were extracted from each skin fibroblast lysate for electrospray ionization-tandem mass spectrometry analysis using the method of Folch [10]. Immune-capture ASA activity and protein assays ASA activity and protein were measured on seven-day post confluent cell lysates using two immuno-based assays developed as described previously [7]. ASA activity and protein in skin fibroblast lysates were determined using different volumes of lysate: 2–5 lL for unaffected; 25 lL for ASA-PD lysates; and 40 lL for MLD lysates. Briefly, microtiter plates were coated with anti-ASA polyclonal antibody at 5 mg/L in 0.1 M NaHCO3 (100 lL/well, 16 h, 4 °C). Wells were washed twice with DELFIAÒ wash buffer (Perkin-Elmer Life Sciences, Turku, Finland). Skin fibroblast lysate was then added and the volume in the wells was made up to 100 lL by the addition of 0.1 M NaCOOCH3/HCOOCH3, 0.1% heat-treated BSA, pH 5.0, and incubated (4 °C, 16 h). Wells were washed (6) with wash buffer (0.1 M NaCOOCH3/HCOOCH3, pH 5.0); 100 lL of 5 mM 4-methylumbelliferyl sulfate (4-MUS) in 0.2 M sodium acetate buffer, pH 5.0, with 0.1% heat-treated BSA was then added to each well. The plate was incubated for 24 h at 37 °C. The reaction was stopped with the addition of 0.2 M glycine/NaOH buffer, pH 10.7 (100 lL), and fluorescence was read on a on a Wallac Victor 2 1420 multilabel counter. A calibration curve using recombinant human ASA (rhASA) (2.0 pg/well to 2.0 ng/well) was included in each assay. rhASA activity was determined with the fluorogenic substrate 4methylumbelliferyl sulfate [11] before each assay. ASA activity was calculated from the calibration curve using Multicalc data analysis software. To determine the ASA protein, skin fibroblast lysate was added to pre-coated (anti-ASA polyclonal antibody, 5.0 lg/mL, 4 °C, 16 h) wells; the volume in the wells was then made up to 100 lL by the addition of DELFIAÒ assay buffer (Perkin-Elmer Life Sciences, Turku, Finland) and incubated (4 °C overnight). The plate was washed (6) with DELFIAÒ wash buffer (Perkin-Elmer Life Sciences, Turku, Finland) before assay buffer (100 lL) containing 0.2 lg/mL Eu3+-labeled anti-ASA polyclonal antibody was added to each well. After incubation (4 °C, 16 h), the plate was washed (6) and 200 lL DELFIAÒ enhancement buffer (Perkin-Elmer Life Sciences, Turku, Finland) was added to each well. The plate was shaken (15 min) and fluorescence read on a Wallac Victor 2 1420 multilabel counter. A calibration curve (2.0 pg/well to 2.0 ng/well) was included in each assay. ASA concentrations were calculated from the calibration curve using Multicalc data analysis software. The concentration of the pure rhASA used as a calibrator for the immune assays was determined using the bicinchoninic acid method [9]. Calibrators and quality control materials were prepared by diluting rhASA in working buffer (0.1 M NaCOOCH3/HCOOCH3, 0.1% heat-treated BSA, pH 5.0) or DELFIAÒ assay buffer (Perkin-Elmer Life Sciences, Turku, Finland) to measure ASA activity and protein, respectively. Extraction of lipids Urine was subjected to lipid extraction using the method of Bligh and Dyer [12]. Briefly, each urine (1.5 mL) was extracted using chloroform/methanol (1:2 [v/v], 5.6 mL) containing the following internal standards: Sulf 16:0 (2.5 nmol); Cer 17:0; GC 16:0-d3; GM2; LC 16:0-d3; PC 28:0; PG 28:0; and PI 32:0 (400 pmol/each). The mixtures were shaken (150 rpm, 10 min) and left to stand at room temperature for 1 h. Phase partitioning was promoted by the addition of 1.9 mL each of chloroform and Milli-QÒ water. The mixtures were shaken again before centrifugation (830g, 2 min). The upper phase was transferred to a new tube for ganglioside extraction. The lower phase was washed by the addition of 0.5 mL of Bligh and Dyer synthetic upper phase, vortexed and centrifuged (830g, 2 min); the upper phase was discarded and the lower phase dried under a gentle stream of nitrogen. The dried lipid extract was resuspended in 200 lL of methanol containing 10 mM NH4COOH prior to analysis by electrospray ionization-tandem mass spectrometry. Skin fibroblast lysates were subjected to lipid extraction using the Folch method [10]. Each skin fibroblast lysate (100 lL) was extracted using 2.0 mL of chloroform/methanol (2:1 [v/v]) containing the following internal standards: Sulf 16:0 (2.5 nmol); and 400 pmol/each of Cer 17:0; GC 16:0-d3; GM2; LC 16:0-d3; PC 28:0; PG 28:0; and PI 32:0. The mixtures were shaken at 150 rpm for 10 min and left to stand at room temperature for 1 h. Phase partitioning was promoted by the addition of 0.4 mL Milli-QÒ water. Tubes were shaken again before centrifugation (830g, 10 min). The upper phase was transferred and set aside for ganglioside extraction; the lower phase was washed by the addition (0.5 mL) of synthetic upper phase, vortexed and centrifuged (830g, 2 min); the upper phase was then discarded. The lower phase was dried under a gentle stream of nitrogen. The dried lipid extract was resuspended in 200 lL methanol containing 10 mM NH4COOH prior to analysis by electrospray ionization-tandem mass spectrometry. To isolate gangliosides the upper phase of either the Bligh and Dyer or Folch extraction methods were loaded onto a primed (3  1.0 mL methanol followed by 3  1.0 mL Milli-QÒ water) ISOLUTE-96 C18 column (25 mg). After the upper phase had completely entered the solid phase, the column was washed with Milli-QÒ water (3  1.0 mL). The gangliosides were eluted with methanol (2  1.0 mL) and dried under a gentle stream of nitrogen. The dried lipid extract was resuspended in 200 lL methanol containing 10 mM NH4COOH prior to analysis by electrospray ionization-tandem mass spectrometry. Lipid profiling by electrospray ionization-tandem mass spectrometry Mass spectrometric analysis was performed on a PE Sciex API 3000 triple-quadrupole mass spectrometer with a turbo-ionspray source (200 °C) and Analyst 1.1 data system. Samples (20 lL) were injected into the electrospray source with a Gilson 233 autosampler using methanol as the carrying solvent at a flow rate of 80 lL/min. N2 was used as the collision gas at a pressure of 2  10 5 Torr. Quantification of individual lipid species was performed using mul- 145 M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 tiple-reaction monitoring (MRM). Each ion pair was monitored for 100 ms with a resolution of 0.7 amu at half-peak height and averaged from continuous scans over the injection period. Concentrations of each molecular species were calculated by relating the peak height of each species to the peak height of the corresponding internal standard. The total amount of each lipid type was calculated by summing the individual species in each class. Sulfatide analysis was performed in negative ion mode, as reported by Whitfield et al. [13] using MRM ion pairs. A total of 19 molecular species were monitored, representing the 18 most abundant species of sulfatide and sulfatide 16:0 which was used as the internal standard. Analysis of phospholipids was also performed in negative ion mode using MRM ion pairs base on the [M H] and the [fatty acidH] ions. The MRM method used to determine the phospholipid species measured both phosphatidylglycerol (PG) and bis(monoacylglycero)phosphate (BMP), which are structural isomers [14], hence the amount measured can represent both or either species. A total of 13 PG/BMP species and 10 phosphatidylinositol (PI) species were monitored using PG 14:0/14:0 and PI 16:0/16:0 as the respective internal standards. Seven Cer species, six GC species, six LC, six ceramide trihexoside (CTH) species, seven PC species and three sphingomyelin (SM) species were monitored as described in Fuller et al. [15]. Cer 17:0, GC 16:0-d3, LC 16:0-d3 and PC 28:0 were used as internal standards. CTH species were related to the LC 16:0-d3 internal standard and sphingomyelin to the PC 28:0 internal standard. Statistical analysis The Mann–Whitney U-test was used to evaluate the ability of each lipid analyte to differentiate between: (i) MLD and unaffected controls; (ii) ASA-PD (together with ASA-PD/MLD) and unaffected controls and (iii) ASA-PD (together with ASA-PD/MLD) and MLD patients. Correlation coefficients (Spearman’s rho) were used to measure the relationship between lipid markers and disease severity. Results Urine analysis All species of sulfatide were elevated in MLD patients above the levels observed in the controls and the ASA-PD individuals. Fig. 1 shows the total sulfatide levels calculated from the sum of the indi- Skin fibroblast analysis The levels of ASA protein and activity were correlated in the control and patient groups with a Spearman’s rho correlation coefficient of 0.944 (p < 0.001). The exceptions to this were the lateinfantile patients who all had low to undetectable activity levels but had protein levels ranging from 0.1 to 1.8 ng/mg cell protein (Fig. 3). The amount of ASA protein detected was approximately: (i) 10% and 5% of control in ASA-PD and ASA-PD/MLD skin fibroblasts, respectively; (ii) 2% and 3.5% of control in adult and juvenile MLD skin fibroblasts, respectively (the higher percentage of ASA protein detected in the juvenile group was due to patient S15 (29.1 ng/mg cell protein)); (iii) less than 2% of control in late-infantile MLD skin fibroblasts. Patients S20 and S21 had no detectable ASA protein (below the assay detection limit of 95 pg/mg cell protein). The amount of ASA activity detected was approximately: (i) 18% and 8.5% of control in ASA-PD and ASA/MLD skin fibroblasts, respectively; (ii) 3% and 1.5% of control in adult and juvenile MLD skin fibroblasts, respectively; (iii) less than 1% of control in late-infantile MLD skin fibroblasts. Patients S19, S20 and S21 had 100 sulfatide (pmol/nmol PC) 100 sulfatide (pmol/nmol PC) vidual sulfatide species: the juvenile and late-infantile MLD patients can be clearly distinguished from controls and ASA-PD individuals; however the sulfatide levels were not different between the late-infantile and juvenile MLD groups. Whilst the total sulfatide levels in the ASA-PD group were significantly elevated above those observed in the adult and juvenile controls (Mann– Whitney U, p < 0.05), this group was not completely discriminated using total sulfatide alone (Fig. 1). To improve the discrimination of the MLD groups a range of other glycosphingolipids and phospholipids were measured in the urine samples. Total ceramide showed a significant difference between the control (n = 18, median value = 38 pmol/nmol PC) and MLD (n = 11, median value = 83 pmol/nmol PC) groups with a Mann–Whitney U-value of 22 (p = 0.001). Further, a number of species of PG/BMP also showed significant differences between the control and MLD groups, with PG/BMP 18:1/18:1 and 18:1/18:2 showing almost complete discrimination with Mann–Whitney U-values of 2 (p < 0.001). Importantly, the level of PG/BMP 18:1/18:1 was higher in the lateinfantile patients compared to the juvenile patients (Fig. 2) which gave improved discrimination over the sulfatide levels alone (Fig. 1). There was no significant difference between the levels of GC, LC, CTH, GM3, SM or PI in the control and MLD patient groups (data not shown). 10 1 10 1 0.1 0.1 control control control ASAadult juv. infant PD MLD MLD juv. late-inf. Fig. 1. Urinary sulfatide in control, ASA-PD and MLD patients. Lipid extracts of urine (1.5 ml) from control adults (h), juveniles (4) and infants (s), from ASA-PD individuals (+) and from juvenile ( ) and late-infantile ( ) MLD patients were analyzed for total sulfatide and total phosphatidylcholine by electrospray ionization-tandem mass spectrometry. Total pmol of sulfatide is expressed per nmol of total phosphatidylcholine. 0 50 100 150 200 PG/BMP 18:1/18:1 (pmol/nmol PC) Fig. 2. Urinary sulfatide and PG/BMP in control, ASA-PD and MLD patients. Lipid extracts of urine (1.5 ml) from control adults (h), juveniles (4) and infants (s), from ASA-PD individuals (+) and from juvenile ( ) and late-infantile ( ) MLD patients were analyzed for total sulfatide and PG/BMP 18:1/18:1 by electrospray ionization-tandem mass spectrometry. Total sulfatide is plotted against PG/BMP 18:1/18:1. 146 sulfatide (nmol/mg cell protein) M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 ASA activity (pmol/min/mg) 10000 1000 100 10 1 0.1 1 10 100 ASA protein (ng/mg) 1000 Fig. 3. ASA protein and activity in cultures skin fibroblasts. Skin fibroblast lysates from control individuals (h), ASA-PD (+), ASA-PD/MLD (), adult MLD ( ) juvenile MLD ( ) and late-infantile MLD ( ) were assayed for ASA protein and activity as described. ASA protein is plotted against ASA activity. no detectable ASA activity (below the assay detection limit of 1.0 pmol/min/mg cell protein). Upon sulfatide loading of the cultured skin fibroblasts we observed an accumulation of sulfatide within the cells that was inversely proportional to the residual ASA activity. This identified all MLD patients as being defective in sulfatide degradation, and three of the five late-infantile MLD patients could be identified as more severely affected than the juvenile or adult MLD patients (Fig. 4). While the juvenile patients showed a significant increase in sulfatide accumulation over the adult patients (p = 0.05) the difference was not sufficient to allow unambiguous classification. ASA-PD/ MLD heterozygotes also showed elevated levels of sulfatide, intermediate between the levels in control and MLD adults. Similar to the urine analysis, most species of PG/BMP were elevated in the MLD skin fibroblasts relative to the control skin fibroblasts, with PG/BMP 18:1/18:1, 18:1/20:4 and 18:1/22:6 having Mann–Whitney U-values of 4, 3 and 4, respectively (p < 0.01). Fig. 5 shows the levels of PG/BMP 18:1/22:6 relative to total sulfatide. Discussion sulfatide (nmol/mg cell protein) Currently, classification and prognosis of MLD patients is based primarily on age at diagnosis. However, with the continued development and implementation of newborn screening strategies for lysosomal storage disorders [16–18], clinicians will soon be faced 300 200 100 0 control MLD ASA- ASAPD. PD/MLD adult MLD MLD juv. late inf. Fig. 4. Sulfatide accumulation in cultured skin fibroblasts. Skin fibroblasts from control individuals (h), ASA-PD (+), ASA-PD/MLD (), adult MLD ( ) juvenile MLD ( ) and late-infantile MLD ( ) were cultured in BME containing 64 lM bovine sulfatide for eight days then harvested and total sulfatide determined by electrospray ionization-tandem mass spectrometry as described. 300 200 100 0 0 1 2 3 PG/BMP 18:1/22:6 (nmol/mg cell protein) Fig. 5. Sulfatide and PG/BMP accumulation in cultured skin fibroblasts. Skin fibroblasts from control individuals (h), ASA-PD (+), ASA-PD/MLD (), adult MLD ( ) juvenile MLD ( ) and late-infantile MLD ( ) were cultured in BME containing 64 lM bovine sulfatide for eight days then harvested and the sulfatide and PG/BMP 18:1/22:6 determined by electrospray ionization-tandem mass spectrometry as described. Total sulfatide was plotted against PG/BMP 18:1/22:6. with the prospect of counseling families and treating asymptomatic MLD newborns. While stem cell transplantation – with its associated risks – is currently available [3,19,20], other treatment options are under development, including enzyme replacement [21] and gene [22] therapies. Selection and timing of the appropriate treatment modality will require accurate prognosis for each newborn. Identification and characterization of ASA-PD and ASAPD/MLD compound heterozygotes and adult forms of the disease will also be important to provide appropriate counseling to those families. Prognosis of MLD patients based on genotype is possible in some patients with common mutations that have been well characterized in terms of the ASA protein and activity resulting from the mutant alleles. Polten et al. reported on two common alleles, allele I, which associated with the severe form of the disease and allele A, that was associated with a more attenuated form of the disease [23]. Genotype/phenotype correlations have also been reported in late-onset MLD; patients homozygous for the P426L mutation showed reduced peripheral nerve conduction velocities and less residual ASA activity compared with I179S heterozygotes [24]. However, genotype alone does not provide a prognosis in approximately one-half of patients, i.e. those who possess rare or private mutations [25]. While genotype/phenotype correlations are continuing to be developed [25], novel mutations will invariably lead to deficiencies in this approach and highlight the need for alternate prognostic markers. Both ASA activity and protein are potential markers for the development of newborn screening programs and, while the results of this study support these approaches, it is clear that differentiation of ASA-PD newborns and prediction of phenotype for MLD affected newborns will require additional information. Lugowska et al. demonstrated that the quantitative measurement of sulfatide in urine allows differentiation between MLD heterozygotes, MLD/ PD-heterozygotes and MLD homozygotes [26]. Using the more sensitive mass spectrometric analysis Whitfield et al. was able to demonstrate elevated urinary sulfatide in MLD patients and some ASAPD individuals relative to healthy individuals [13]. While we also observed this difference we were unable to differentiate late-infantile MLD from the later-onset juvenile form using urinary sulfatide levels alone (Fig. 1). This may relate to the fact that the juvenile patients were older at the time of urine collection and so both the lateinfantile and juvenile groups were symptomatic at the time of sampling and displayed elevated urinary sulfatide. Sulfatide and other lipids in urine result from renal cells found in the urinary sediment [13,15]. Lysosomal storage in cells results in alterations in lipid M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 metabolism leading to secondary changes in lipid levels [27]. We therefore measured multiple lipid species in the urine of MLD patients with a view to identifying secondary lipid changes that may prove useful in disease prognosis. We observed that the best discrimination of late-infantile and juvenile MLD patients was achieved using a combination of sulfatide and PG/BMP 18:1/18:1 (Fig. 2), with only a single juvenile patient (U16, 4-years of age) falling into the late-infantile group. Mass spectrometric analysis did not distinguish between the structural isomers of PG and BMP; however, we have recently identified BMP, a lysosomal-specific lipid, as being altered in cultured skin fibroblasts from a number of different lysosomal storage disorder patients [14]. It is worth noting that the urine samples analyzed in this study were taken from symptomatic patients who were therefore likely to have significant storage of sulfatide within renal cells; in asymptomatic patients one might expect a greater difference between the early-onset late-infantile form and the later-onset juvenile and adult forms. However, such studies must await the collection of suitable samples following early diagnosis of MLD patients. While urine is easy to collect it has limited prognostic capacity; in contrast, cultured skin fibroblasts represent a specimen that can be used to characterize and compare patients regardless of patient age and disease severity at the time of collection; such cells can be cultured under identical conditions to provide a common metabolic base upon which to identify subtle differences in ASA expression/activity as well as in lipid metabolism. We performed ASA protein and activity measurements on cultured skin fibroblasts and observed a correlation between protein and activity (Fig. 3). Using a combination of ASA protein and activity, unaffected controls, ASA-PD and ASA-PD/MLD compound heterozygotes could be distinguished from each other (Fig. 3). However, despite the differentiation of the unaffected groups, these two parameters alone did not clearly discriminate between the different MLD phenotypes, although three of the five late-infantile MLD patients who had a severe phenotype (S19, S20 and S21) were separated from the juvenile and adult phenotypes. The inability to assign phenotype based solely on ASA protein and activity may result from a number of factors which influence in vivo ASA activity. These may include the incorrect targeting of the mutant enzyme to the lysosomal compartment or the incorrect conformation that does not allow interaction with the activator protein, saposin C. To overcome the limitations of in vitro ASA assays we have performed in vivo measurements of sulfatide metabolism by loading sulfatide into the skin fibroblasts and measuring its accumulation. At the same time we were able to measure the secondary effects of sulfatide accumulation on lipid metabolism. Sulfatide accumulation alone differentiated three of the five late-infantile patients from the later-onset forms (Fig. 4); however, when combined with the PG/BMP measurements we were able to differentiate all lateinfantile patients from the juvenile and adult forms (Fig. 5). The classification of MLD based on age at presentation does not represent distinct groups but rather a continuum and we note that the late-infantile patient who had low sulfatide and PG/BMP (Fig. 5) was the oldest in this group (4-years of age at diagnosis, S22, Table 2) while the youngest in the juvenile group was only 6-years of age at diagnosis. Our results suggest that a combined approach using genotype determination, where appropriate, followed by urine analysis combined with ASA and lipid measurements on cultured skin fibroblasts will provide the most comprehensive prediction of the age of disease onset and rate of progression. Acknowledgments We wish to acknowledge the Universiti Sains Malaysia for their sponsorship of a scholarship for MAFT. This work was supported in 147 part by the National Heath & Medical Research Council (Canberra, Australia). We are particularly grateful to the MLD patients and their families and to Dr. Michael Fietz and the staff of the National Referral Laboratory for the Diagnosis of Lysosomal, Peroxisomal and Related Genetic Disorders, SA Pathology for providing urine and fibroblast samples. References [1] E.H. Kolodny, A.L. Fluharty, Metachromatic leukodystrophy and multiple sulfatase deficiency: sulfatide lipidosis, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Basis of Inherited Disease, McGraw-Hill, New York, 1995, pp. 2693–2739. [2] K. von Figura, V. Gieselmann, J. Jaeken, Metachromatic leukodystrophy, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Basis of Inherited Disease, McGraw-Hill, New York, 2001, pp. 3695–3724. [3] W. Krivit, Allogeneic stem cell transplantation for the treatment of lysosomal and peroxisomal metabolic diseases, Springer Semin. Immunopathol. 26 (2004) 119–132. [4] W. Krivit, C. Peters, E.G. Shapiro, Bone marrow transplantation as effective treatment of central nervous system disease in globoid cell leukodystrophy, metachromatic leukodystrophy, adrenoleukodystrophy, mannosidosis, fucosidosis, aspartylglucosaminuria, Hurler, Maroteaux-Lamy, and Sly syndromes, and Gaucher disease type III, Curr. Opin. Neurol. 12 (1999) 167– 176. [5] P. Kapaun, R.W. Dittmann, B. Granitzny, W. Eickhoff, H. Wulbrand, E. Neumaier-Probst, A. Zander, A. Kohlschuetter, Slow progression of juvenile metachromatic leukodystrophy 6 years after bone marrow transplantation, J. Child Neurol. 14 (1999) 222–228. [6] Y. Li, C.R. Scott, N.A. Chamoles, A. Ghavami, B.M. Pinto, F. Turecek, M.H. Gelb, Direct multiplex assay of lysosomal enzymes in dried blood spots for newborn screening, Clin. Chem. 50 (2004) 1785–1796. [7] M.A. Tan, C.J. Dean, J.J. Hopwood, P.J. Meikle, Diagnosis of metachromatic leukodystrophy by immune quantification of arylsulphatase A protein and activity in dried blood spots, Clin. Chem. 54 (2008) 1925–1927. [8] J.J. Hopwood, V. Muller, J.R. Harrison, W.F. Carey, H. Elliott, E.F. Robertson, A.C. Pollard, Enzymatic diagnosis of the mucopolysaccharidoses: experience of 96 cases diagnosed in a 5-year period, Med. J. Aust. 1 (1982) 257–260. [9] P.K. Smith, R.I. Krohn, G.T. Hermanson, A.K. Mallia, F.H. Gartner, M.D. Provenzano, E.K. Fujimoto, N.M. Goeke, B.J. Olson, D.C. Klenk, Measurement of protein using bicinchoninic acid, Anal. Biochem. 150 (1985) 76–85. [10] J. Folch, M. Lees, G.H. Sloane Stanley, A simple method for the isolation and purification of total lipides from animal tissues, J. Biol. Chem. 226 (1957) 497– 509. [11] E.H. Kolodny, R.A. Mumford, Arylsulfatases A and B in metachromatic leukodystrophy and Maroteaux–Lamy syndrome: studies with 4methylumelliferyl sulfate, Adv. Exp. Med. Biol. 68 (1976) 239–251. [12] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. [13] P.D. Whitfield, P.C. Sharp, D.W. Johnson, P. Nelson, P.J. Meikle, Characterization of urinary sulfatides in metachromatic leukodystrophy using electrospray ionization-tandem mass spectrometry, Mol. Genet. Metab. 73 (2001) 30–37. [14] P.J. Meikle, S. Duplock, D. Blacklock, P.D. Whitfield, G. Macintosh, J.J. Hopwood, M. Fuller, Effect of lysosomal storage on bis(monoacylglycero)phosphate, Biochem. J. 411 (2008) 71–78. [15] M. Fuller, P.C. Sharp, T. Rozaklis, P.D. Whitfield, D. Blacklock, J.J. Hopwood, P.J. Meikle, Urinary lipid profiling for the identification of fabry hemizygotes and heterozygotes, Clin. Chem. 51 (2005) 688–694. [16] M.H. Gelb, F. Turecek, C.R. Scott, N.A. Chamoles, Direct multiplex assay of enzymes in dried blood spots by tandem mass spectrometry for the newborn screening of lysosomal storage disorders, J. Inherit. Metab. Dis. 29 (2006) 397– 404. [17] P.J. Meikle, D.J. Grasby, C.J. Dean, D.L. Lang, M. Bockmann, A.M. Whittle, M.J. Fietz, H. Simonsen, M. Fuller, D.A. Brooks, J.J. Hopwood, Newborn screening for lysosomal storage disorders, Mol. Genet. Metab. 88 (2006) 307–314. [18] X.K. Zhang, C.S. Elbin, W.L. Chuang, S.K. Cooper, C.A. Marashio, C. Beauregard, J.M. Keutzer, Multiplex enzyme assay screening of dried blood spots for lysosomal storage disorders by using tandem mass spectrometry, Clin. Chem. 54 (2008) 1725–1728. [19] N. Cartier, P. Aubourg, Hematopoietic stem cell gene therapy in Hurler syndrome, globoid cell leukodystrophy, metachromatic leukodystrophy and Xadrenoleukodystrophy, Curr. Opin. Mol. Ther. 10 (2008) 471–478. [20] T.M. Pierson, C.G. Bonnemann, R.S. Finkel, N. Bunin, G.I. Tennekoon, Umbilical cord blood transplantation for juvenile metachromatic leukodystrophy, Ann. Neurol. 64 (2008) 583–587. [21] U. Matzner, R. Lullmann-Rauch, S. Stroobants, C. Andersson, C. Weigelt, C. Eistrup, J. Fogh, R. D’Hooge, V. Gieselmann, Enzyme replacement improves ataxic gait and central nervous system histopathology in a mouse model of metachromatic leukodystrophy, Mol. Ther. 17 (2009) 600–606. [22] N. Iwamoto, A. Watanabe, M. Yamamoto, N. Miyake, T. Kurai, A. Teramoto, T. Shimada, Global diffuse distribution in the brain and efficient gene delivery to the dorsal root ganglia by intrathecal injection of adeno-associated viral vector serotype 1, J. Gene Med. 5 (2009) 498–505. 148 M.A.F. Tan et al. / Molecular Genetics and Metabolism 99 (2010) 142–148 [23] A. Polten, A.L. Fluharty, C.B. Fluharty, J. Kappler, K. Von Figura, V. Gieselmann, Molecular basis of different forms of metachromatic leukodystrophy, N. Engl. J. Med. 324 (1991) 18–22. [24] H. Rauschka, B. Colsch, N. Baumann, R. Wevers, M. Schmidbauer, M. Krammer, J.C. Turpin, M. Lefevre, C. Olivier, S. Tardieu, W. Krivit, H. Moser, A. Moser, V. Gieselmann, B. Zalc, T. Cox, U. Reuner, A. Tylki-Szymanska, F. Aboul-Enein, E. LeGuern, H. Bernheimer, J. Berger, Late-onset metachromatic leukodystrophy: genotype strongly influences phenotype, Neurology 67 (2006) 859–863. [25] A. Biffi, M. Cesani, F. Fumagalli, U. Del Carro, C. Baldoli, S. Canale, S. Gerevini, S. Amadio, M. Falautano, A. Rovelli, G. Comi, M.G. Roncarolo, M. Sessa, Metachromatic leukodystrophy – mutation analysis provides further evidence of genotype–phenotype correlation, Clin. Genet. 74 (2008) 349–357. [26] A. Lugowska, A. Tylki-Szymanska, J. Berger, B. Molzer, Elevated sulfatide excretion in compound heterozygotes of metachromatic leukodystrophy and ASA-pseudodeficiency allele, Clin. Biochem. 30 (1997) 325–331. [27] S.U. Walkley, Secondary accumulation of gangliosides in lysosomal storage disorders, Semin. Cell Dev. Biol. 15 (2004) 433–444.