J. Inherit. Metab. Dis. 25 (2002) 98^106 # SSIEM and Kluwer Academic Publishers. Printed in the Netherlands.
Gas chromatographic^mass spectrometric newborn screening for propionic acidaemia by targeting methylcitrate in dried ¢lter-paper urine samples T. KUHARA1*, M. OHSE1, Y. INOUE1, T. YORIFUJI2, N. SAKURA3, H. MITSUBUCHI4,5, F. ENDO5 and J. ISHIMATU4 1
Division of Human Genetics, Medical Research Institute, Kanazawa Medical University; 2 Department of Pediatrics, Kyoto University Faculty of Medicine; 3 Department of Pediatrics, Hiroshima University School of Medicine; 4 Department of Pediatrics, Kumamoto City Hospital; 5 Department of Pediatrics, Kumamoto University School of Medicine, Japan *Correspondence: Division of Human Genetics, Medical Research Institute, Kanazawa Medical University, 1-1 Daigaku, Uchinada, Kahoku-gun, Ishikawa 920-0293, Japan. E-mail:
[email protected] MS received 18.06.01
Accepted 18.12.01
Summary: Propionic acidaemia (PCCD) or de¢ciency of propionyl-CoA carboxylase (PCC) is one of the most common organic acidaemias. Recent studies have suggested that this disease can cause somatic or cognitive deterioration even in patients without ketosis or metabolic acidosis, or in cases with unusually late onset. This suggests that for this disease a sensitive yet practical screening procedure is required to achieve early treatment. We conducted a pilot study of gas chromatographic^mass spectrometric screening of 12 000 newborns for PCCD using eluates from dried ¢lter-paper urine collected at 4^7 days of age. Methylcitrate (MC) was targeted for PCCD. For bulk screening, 2-hydroxyundecanoate was used as internal standard; for quanti¢cation, stableisotope-labelled MC was used. Urease pretreatment without fractionation allowed satisfactory recovery and reproducibility of the highly polar MC. We detected an asymptomatic male infant with distinctly elevated MC: the creatinine-corrected level relative to 2-hydroxyundecanoate was 4.8 SD above the normal mean. The MC concentration calculated using the stable-isotopelabelled internal standard was 70.6 mmol/mol creatinine 14.7 SD above the normal mean of 3.70. Parallel analysis of the dried blood spot at 4 days of age by tandem MS showed only borderline elevation of propionylcarnitine. The activity of PCC in lymphocytes was 7% of control. Gene analysis revealed 98
JIMD4041 15-05-2002
11:16
1st Proof
Screening for propionic acidaemia with ¢lter paper urine
99
that a single missense mutation, TAT to TGT, resulting in Y435C in the b chain was present in a homozygous form. Dietary treatment including carnitine supplementation decreased this infant’s MC level and to date (at 13 months of age), he shows no neurological or somatic abnormalities.
Propionic acidaemia or propionyl-CoA carboxylase de¢ciency (PCCD, McKusick 232000, 232050) is an autosomal recessive inborn error of metabolism in which the activity of propionyl-CoA carboxylase (PCC; EC 6.4.1.3) is de¢cient or greatly reduced. This disease has a wide range of clinical expression varying from severe neonatal ketoacidosis with the risk of major handicap or death to asymptomatic or mild disease that usually responds well to treatment with good long-term outcome (Fenton et al 2001). Recent studies have suggested that neurological deterioration occurs in patients even in the absence of ketosis or metabolic acidosis (Nyhan et al 1999; Ramachandran and Pietz 1995; Walter et al 1995) and that fatal cases are not limited to those of neonatal onset but also include unusual late-onset ones (Pe¤rez-Cerda¤ et al 1998). Even patients with relatively high residual PCC activity are known to develop serious symptoms (Pe¤rez-Cerda¤ et al 1998; Sethi et al 1989). Thus, a sensitive but speci¢c diagnostic method at an early stage of life seems critical to ensure the quality of life of PCCD individuals. De¢nitive diagnosis of PCCD among high-risk patients who have developed metabolic ketoacidosis has been accomplished by GC-MS analysis of urine (Chalmers and Lawson 1982; Goodman and Markey 1981). However, presymptomatic diagnosis or neonatal diagnosis of PCCD using urine samples has not been performed extensively. Methylcitrate (MC) is the one of the most reliable targets for PCCD but is highly water-soluble. The recovery of MC is high in the simpli¢ed urease pretreatment and GC-MS method described previously (Kuhara and Matsumoto 1995; Matsumoto and Kuhara 1996). We applied this noninvasive diagnostic procedure in a pilot study of 12 000 neonates, and we evaluated this method with respect to its sensitivity for detecting asymptomatic or mildly affected PCCD neonates and its diagnostic accuracy. MATERIALS AND METHODS Chemicals: Methylcitrate and (methyl-2H3)methylcitrate (d3-MC) were purchased from Cambridge Isotope Laboratory (Andover, MA, USA) and (methyl-2H3)creatinine (d3-creatinine) from Isotec Inc. (Miamisburg, OH, USA). The purity of these stable-isotope-labelled compounds, used as internal standards, was higher than 99% as judged by the lack of additional peaks on GC-MS. Urease type C-3 was obtained from Sigma (St Louis, MO, USA). Urine samples and pretreatment: The urine specimens were taken for the pilot study for newborn screening of 22 target diseases from 12 000 neonates with the informed consent of parents on days 4^7, when blood was also taken for the Guthrie test. The procedure for sample preparation of liquid urine or ¢lter-paper urine and J. Inherit. Metab. Dis. 25 (2002)
100
Kuhara et al
GC-MS measurement was based on the method of Matsumoto and Kuhara described previously (Kuhara and Matsumoto 1995; Matsumoto and Kuhara 1996). Creatinine determination: As previously described by Shoemaker and Elliott (1991) the urinary creatinine value obtained using d3-creatinine by the method is the sum of creatinine and creatine, and is termed here ‘total creatinine’. We also determined urinary creatinine using a conventional Beckman CX5 autoanalyser, and evaluated urinary metabolite levels on both bases: creatinine obtained from autoanalyser and total creatinine calculated from GC-MS with d3-creatinine. MC level and concentration: For quanti¢cation, a ¢xed amount of each compound was added as an internal standard: 100 nmol of d3-creatinine and 25 nmol of 2-hydroxyundecanoate (2HUD). For routine screening 2HUD was used as internal standard and the ions of [M^CO2TMS^HOTMS]þ at m/z 287 and [M^CO2TMS]þ at m/z 229 were monitored for MC and 2HUD, respectively. Endogenous MC gives two chromatographic peaks. The combined areas of these, divided by the area of the 2HUD m/z 229 peak and the molar concentration of creatinine in the extract gave a measure here termed ‘MC level’. Elevated levels of MC were quanti¢ed using d3-MC (5 nmol) as internal standard and monitoring the ions at m/z 287 for MC and at m/z 290 for d3-MC to give the ‘MC concentration’. Both ‘levels’ and concentrations of MC were expressed relative to creatinine or total creatinine. GC-MS measurement: GC-MS measurement was carried out under the conditions described previously (Kuhara and Matsumoto 1995; Matsumoto and Kuhara 1996). Measurement of propionylcarnitine in blood was kindly performed by Dr Y. Shigematsu, Fukui Medical University, Fukui, Japan (Shigematsu et al 1999). PCC activity: PCC activity in lymphocytes was determined as described previously (Oizumi and Hayakawa 1990). RT^PCR/sequencing: Peripheral blood mononuclear cells were collected from 1 ml of heparinized blood by Ficoll-Paque (Amersham-Pharmacia, UK) centrifugation and then total RNA was isolated from the cells using Sepasol RNA isolation reagent (Nacalai Tesque, Japan). The RNA was reverse transcribed with SuperScript II reverse transcriptase as recommended by the supplier (Gibco-BRL, MD, USA) and the cDNA product was resuspended in 20 ml of DEPC-treated water. The whole coding region of the PCCB gene was ampli¢ed as three overlapping fragments in 25 ml reactions (2 ml of cDNA in 10 mmol/L Tris-HCl, pH 8.3, 50 mmol/L KCl, 15 mmol/L MgCl2, 0.01% (w/v) gelatin, 25 pmol of each primer, 1.25 U AmpliTaq Gold DNA polymerase (Perkin-Elmer, NJ, USA)) with initial denaturation at 94 C for 10 min followed by 35 cycles of denaturation at 94 C for 30 s, annealing at 60 C for 30 s and extension at 72 C for 2 min. The sequences of the primers were as follows: 50 -GTACTCAGGTGCGCCGGTAG-30 /50 CTTCCATTGATTCGGCCTCG-30 ; 50 -GGAGACAGCGTGGTCACTGG-30 /5’GGTCACTGGGATCGTGGCAC-30 ; 50 -ATGCCTTGTGTAATCTCCGG-30 /50 J. Inherit. Metab. Dis. 25 (2002)
Screening for propionic acidaemia with ¢lter paper urine
101
GGCAGACAGTAATTCAGTTCTTGG-30 . PCR products were cloned into the pCR2.1TOPO vector using a TOPO-TA cloning kit (Invitrogen, CA, USA) and sequenced with an ABI PRISM 310 Genetic Analyzer (Perkin^Elmer) using a BigDyeTM Terminator Cycle Sequencing kit (Perkin^Elmer). RESULTS AND DISCUSSION Determination of MC: In patients with PCCD, propionyl-CoA undergoes altered metabolism to give MC (Ando et al 1972a), 3-hydroxypropionate (Ando et al 1972b) and other products (Fenton et al 2001; Lehnert et al 1994). We targeted MC for screening of PCCD. The di¡erential diagnosis of disorders with elevated MC depends the presence or absence of methylmalonate, 3-hydroxyisovalerate or 3-methylcrotonylglycine, which can also be determined by this GC-MS screening method (Kuhara 2001). MC ‘levels’ in dried ¢lter paper urine from healthy neonates are not normally distributed. Consequently, the data were log10-transformed before statistical analysis to give mean of 1:40 and SD 0.34 for creatinine-based results and mean of 1:57 and SD 0.36 for total creatinine-based results (Figures 1a and 1b). Quanti¢cation of MC by stable-isotope dilution method gave log10-transformed means of 0.569 (corresponding to 3.70 mmol/mol creatinine) with SD 0.0873, and 0.425 (corresponding to 2.66 mmol/mol total creatinine) with SD 0.132 (Figures 1c and 1d). Results of screening: During the pilot study, a presymptomatic male PCCD neonate was detected among 12 000 neonates. MC was highly elevated and clearly diagnostic. The MC ‘levels’ in his urine taken at 4 days of age were 1.74 (0:24 ¼ log10 (1.74), > mean plus 4.8 SD) and 0.91 (0:04 ¼ log10 (0.91), > mean plus 4.2 SD) respectively for creatinine and total creatinine-based results. The concentration of MC was 70.6 mmol/mol creatinine and 36.0 mmol/mol total creatinine. These were mean plus 14.7 SD (per creatinine) and 8.6 SD (per total creatinine), respectively. None of the other metabolites associated with increased MC excretion (glycine, lactate, 3-hydroxyisovalerate, 3-methylcrotonylglycine, or methylmalonate) was increased. A small increase of 3-hydroxypropionate was seen at 12 weeks of age. The MC concentrations in the urine of his father and mother were 1.4 and 1.7 mmol/mol creatinine, within our normal range (1.29^4.71) mmol/mol creatinine) for subjects > 4 years of age. Case history: This male infant is the ¢rst child of healthy, nonconsanguineous Japanese parents. He was born at 39 weeks of gestation. The birth and neonatal course were unremarkable and the neonate was discharged on the 5th postnatal day with a body weight of 3130 g. After written informed consent was obtained from his parents, his urine specimen was collected at 4 days of age for the pilot study of GC-MS newborn screening for IEM conducted by Kanazawa Medical University. The GC-MS analysis of his urine revealed an elevated MC level, indicating that J. Inherit. Metab. Dis. 25 (2002)
102
Kuhara et al
Figure 1 Histograms of urinary methylcitrate ‘level’ for 133 healthy newborns. 2-Hydroxyundecanoate was used as internal standard and the results are expressed per creatinine by Beckman autoanalyser (a), or total creatinine determined by GC-MS with labelled creatinine (b).
J. Inherit. Metab. Dis. 25 (2002)
Screening for propionic acidaemia with ¢lter paper urine
103
Figure 1 Histograms of log10 transformed methylcitrate level per creatinine (c) and per total creatinine (d)
J. Inherit. Metab. Dis. 25 (2002)
104
Kuhara et al
he had PCCD. The other laboratory data, routine blood counts, blood gas examination, blood glucose and ammonia were all within the normal ranges. At 3 months of age, he was put on a therapeutic formula restricted in amino acids (isoleucine, methionine, threonine and valine), with the amounts of natural and total protein kept below 1 g/kg per day and 2 g/kg per day respectively. Carnitine was added at 7 months of age. At 1 year 1 month of age, his weight and height were at the 75th centile for age. He had good development in terms of gross and ¢ne motor function, speech development, social behaviour, sensory screening, and emergence of autonomy and independence. Enzyme activity and mutation analysis: PCC is composed of nonidentical a and b subunits encoded by genes on chromosomes 13q32 and 3q13.3 respectively. Mutations a¡ecting the a subunit result in more severe phenotype than those a¡ecting the b subunit (Fenton et al 2001). The PCC activity in lymphocytes in the index case was 3.07 pmol/min per 106 cells, which was 6.7% of the control value. The activity in his parents was within the normal range, suggesting that the patient had a b subunit defect (Wolf and Rosenberg 1978). Sequencing analysis revealed a homozygous missense mutation, Y435C/Y435C (TAT ! TGT) in the PCCB gene. Both parents were heterozygous for this mutation and there were no other mutations in the coding region of the gene. Propionylcarnitine: Propionylcarnitine is increased in blood and urine of patients with PCCD or methylmalonic acidaemia (Di Donato et al 1984). Therefore, these diseases are screened by targeting propionylcarnitine in blood by tandem mass spectrometry (Millington et al 1990; Rashed et al 1997). The level of propionylcarnitine in the ¢lter-paper blood sample taken from our patient at 4 days of age was determined by the modi¢ed tandem MS method (Shigematsu et al 1999). The specimen had been stored at 4 C and was analysed when the patient was 5 months of age. The propionylcarnitine concentration was 3.41 nmol/ml, 2.1 SD above normal mean of 1:28 1:01 (n ¼ 2724). A blood sample taken at 2 months of age had a propionylcarnitine concentration of 3.14 nmol/ml. The molar ratios of propionylcarnitine to acetylcarnitine in these two samples were 0.287 and 0.258, compared to mean control of 0.053 and the screening cut-o¡ of 0.21. This contrasts with the highly elevated and clearly diagnostic concentration of MC in the urine at 4 days of age. We speculate that in milder PCCD cases, di¡erentiation from healthy control might be more di⁄cult using the blood tandem MS method than using the simpli¢ed urease/GC-MS method. When the tandem MS dried blood spot screening shows borderline elevations of propionylcarnitine, GC-MS urine screening should be carried out. Our procedure enables simultaneous analysis of urinary MC and other metabolites in 0.01^0.1 ml of liquid urine samples or urine-soaked ¢lter paper strips. The recovery of polar MC is high and sample preparation is simple. The procedure was shown to be highly sensitive and to enable correct chemical diagnosis of PCCD even for a presymptomatic or mild case, which allows early treatment, genetic counselling and/or prenatal diagnosis for the next pregnancy. J. Inherit. Metab. Dis. 25 (2002)
Screening for propionic acidaemia with ¢lter paper urine
105
ACKNOWLEDGEMENTS The authors are grateful to Dr I. Matsumoto (Professor Emeritus, Kanazawa Medical University), Dr S. Sakamoto (Professor Emeritus, The University of Tokyo) and Dr M. Tetso (Tetso Hospital, Nagasaki, Japan) for their continuing interest and encouragement. This study was supported by a grant from the JAOG Ogyaa Donation Foundation, a 1999^2000 Grant-in-Aid for Scienti¢c Research (11672312) from the Ministry of Education, Science and Culture of Japan, Health Sciences Research Grants for ‘‘Seiiku Iryo Kenku’’ and Research on Children and Families (H13-Kodomo-031) from the Ministry of Health and Welfare of Japan, and a grant for project research from the High-Technology Center of Kanazawa Medical University (H2001-P3). The authors express thanks to Mrs T. Sakaida for her assistance in preparing the manuscript.
REFERENCES Ando T, Rasmussen K, Wright JM, Nyhan WL (1972a) Isolation and identi¢cation of methylcitrate, a major metabolic product of propionate in a patient with propionic acidemia. J Biol Chem 247: 2200^2204. Ando T, Rasmussen K, Nyhan WL, Hull D (1972b) 3-Hydroxypropionate: signi¢cance of b-oxidation of propionate in patients with propionic acidemia and methylmalonic acidemia. Proc Natl Acad Sci USA 69: 2807^2811. Chalmers RA, Lawson AM (1982) Organic Acids in Man. London: Chapman and Hall. Di Donato S, Rimoldi M, Garavaglia B, Uziel G (1984) Propionylcarnitine excretion in propionic and methylmalonic acidurias: a cause of carnitine de¢ciency. Clin Chim Acta 136: 13^21. Fenton WA, Gravel RA, Rosenblatt DS (2001) Disorders of propionate and methylmalonate metabolism. In Scriver CR, Beaudet AL, Sly WS, Valle D, eds; Childs B, Kinzler KW, Vogelstein B, assoc. eds. The Metabolic and Molecular Bases of Inherited Disease, 8th edn. New York: McGraw-Hill, 2165^2193. Goodman SI, Markey SP (1981) Diagnosis of organic acidemias by gas chromatography^ mass spectrometry. Lab Res Methods Biol Med 6: 1^158. Kuhara T (2001) Diagnosis of inborn errors of metabolism using ¢lter paper urine, urease treatment, isotope dilution and gas chromatography^mass spectrometry. J Chromatogr B Biomed Sci Appl 758: 3^25. Kuhara T, Matsumoto I (1995) A simultaneous gas chromatographic mass spectrometric analysis of urinary metabolitesQapplication to the neonatal mass screening. Proc Jap Soc Biomed Mass Spectrom 20: 45^51. Lehnert W, Sperl W, Suormala T, Baumgartner ER (1994) Propionic acidemia: clinical, biochemical and therapeutic aspectsQexperience in 30 patients. J Pediatr 153: S69^S80. Matsumoto I, Kuhara T (1996) A new chemical diagnostic method for inborn errors of metabolism by mass spectrometry. Mass Spectrom Rev 15(1): 43^57. Millington DS, Kodo N, Norwood DL, Roe CR (1990) Tandem mass spectrometry: a new method for acylcarnitine pro¢ling with potential for neonatal screening for inborn errors of metabolism. J Inherit Metab Dis 13: 321^324. Nyhan WL, Bay C, Beyer EW, Mazi M (1999) Neurologic nonmetabolic presentation of propionic acidemia. Arch Neurol 56: 1143^1147. Oizumi J, Hayakawa K (1990) High-performance liquid chromatographic determination of four kinds of biotin-containing carboxylase. J Chromatogr 529: 55^63.
J. Inherit. Metab. Dis. 25 (2002)
106
Kuhara et al
Pe¤rez-Cerda¤ C, Merinero B, Mart|¤ M, et al (1998) An unusual late-onset case of propionic acidemia: biochemical investigations, neuroradiological ¢ndings and mutation analysis. Eur J Pediatr 157: 50^52. Ramachandran R, Pietz J (1995) Propionic acidemia without acidemia: a case report. J Perinatol 15(1): 71^73. Rashed MS, Bucknall MP, Little D, et al (1997) Screening blood spots for inborn errors of metabolism by electrospray tandem mass spectrometry with a microplate batch process and a computer algorithm for automated £agging of abnormal pro¢les. Clin Chem 43: 1129^1141. Sethi KD, Ray R, Roesel RA, et al (1989) Adult-onset chorea and dementia with propionic acidemia. Neurology 39: 1343^1345. Shigematsu Y, Hata I, Kikawa Y, et al (1999) Modi¢cations in electrospray tandem mass spectrometry for a neonatal-screening pilot study in Japan. J Chromatogr B Biomed Sci Appl 731: 97^103. Shoemaker JD, Elliott WH (1991) Automated screening of urine samples for carbohydrates, organic and amino acids after treatment with urease. J Chromatogr 562: 125^138. Walter JH, Wraith JE, Cleary MA (1995) Absence of acidosis in the initial presentation of propionic acidaemia. Arch Dis Child 72: F197^F199. Wolf B, Rosenberg LE (1978) Heterozygote expression in propionyl CoA carboxylase de¢ciency: differences between major complementation group. J Clin Invest 62: 931^936.
J. Inherit. Metab. Dis. 25 (2002)
