Journal of Chromatography B, 792 (2003) 99–106
`
`www.elsevier.com/locate/chromb
`
`E ffect of valproic acid on the urinary metabolic profile of a patient
`with succinic semialdehyde dehydrogenase deficiency
`*
`b
`b
`Toshihiro Shinka
`, Masafumi Ohfu , Shinichi Hirose , Tomiko Kuhara
`aDivision of Human Genetics,Medical Research Institute,Kanazawa Medical University,1-1 Daigaku,Uchinada-machi,
`Ishikawa 920-0293, Japan
`bDepartment of Pediatrics,Fukuoka University School of Medicine,7-45-1 Nanakuma, Jonan-ku,Fukuoka 814-0180, Japan
`
`a ,
`
`a
`
`Received 8 November 2002; received in revised form 3 March 2003; accepted 3 March 2003
`
`Abstract
`
`The metabolic changes in a patient with succinic semialdehyde dehydrogenase deficiency were investigated following
`valproate administration using urease pretreatment and gas chromatography-mass spectrometry. A stable isotope dilution
`technique was used for quantification of urinary 4-hydroxybutyrate. Urinary levels of 4-hydroxybutyrate were 4-fold higher
`after 1-month valproate therapy. 4,5-Dihydrohexanoate, 2-deoxytetronate and 3-deoxytetronate were also 1.7–2.7-fold
`higher. The urinary excretions of 4-hydroxybutyrate in valproate non-medicated controls were age dependence and decreased
`with age. Relationships between 4-hydroxybutyrate excretion and 4-hydroxyvalproate or 5-hydroxyvalproate excretion were
`observed in valproate medicated controls. It seems that 4-hydroxyvalproate and 5-hydroxyvalproate as well as valproate are
`involved with increased excretion of 4-hydroxybutyrate following valproate administrations.
`© 2003 Elsevier Science B.V. All rights reserved.
`
`Keywords: Succinic semialdehyde dehydrogenase deficiency; Valproic acid
`
`1. Introduction
`
`Succinic semialdehyde dehydrogenase (SSADH)
`deficiency (McKusick 271980) is an inborn error of
`4-aminobutyrate (GABA) metabolism [1]. The clini-
`cal and biochemical findings have been summarized
`in several reports [2–4]. Increased 4-hydroxybutyrate
`(GHB) in the patients with SSADH deficiency were
`found not only in urine but also blood and cere-
`brospinal fluid using stable isotope dilution technique
`combined with gas chromatography-mass
`spec-
`
`*Corresponding author. Tel.: 181-76-286-2464; fax: 181-76-
`286-2312.
`E-mail address: shinka@kanazawa-med.ac.jp (T. Shinka).
`
`trometry (GC–MS) [5]. The diagnosis of SSADH
`deficiency is usually based on elevated concentra-
`tions of urinary GHB.
`We reported the siblings who were the second and
`third cases of SSADH deficiency in Japan [6,7]. The
`younger patient was controlling epileptiform attacks
`by taking valproate (VPA, antiepileptic drug for
`treating generalized epilepsy), but then began vig-
`abatrin (g-vinyl GABA,
`irreversible inhibitor of
`GABA-transaminase) therapy [7]. Increases in urin-
`ary excretion of GHB after administration of VPA to
`a patient with SSADH deficiency has been reported
`by Divry et al. [8]. The effects of VPA on GHB
`metabolism have been discussed and it is suspected
`that the inhibition of SSADH by therapeutic levels of
`
`1570-0232/03/$ – see front matter © 2003 Elsevier Science B.V. All rights reserved.
`doi:10.1016/S1570-0232(03)00276-9
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`100
`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`VPA results in enhanced GHB production. However,
`the mechanism of GHB accumulation is still under
`investigation. There are no reports on the relation-
`ship between GHB accumulation and VPA metabo-
`
`lism. In this report, we examined the urinary GHB
`levels in patients and controls using a sensitive
`quantification method, which consists of urease
`digestion, stable isotope dilution and GC–MS. We
`
`Fig. 1. Total ion current chromatograms of the trimethylsilyl (TMS) derivatives of the urinary metabolites in the patient with SSADH
`deficiency before and after valproate administration (relative abundance on the y-axis). (A) After VPA treatment (1 month after), (B) before
`VPA treatment), (C) control. Peaks are: (1) GHB; (2) 2,2-dimethylsuccinate (IS); (3) 4HVPA; (4) glutarate (GA); (5) 3DT; (6) 2DT;
`(7) 3KVPA; (8) 5HVPA; (9) adipate (AD); (10) 4,5DH; (a) alanine; (b) glycine; (c) b-aminoisobutyrate; (d) urea; (e) phosphate; (f) serine;
`(g) threonine; and (h) creatinine.
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`101
`
`Fig. 2. Mass chromatograms of the trimethylsilyl (TMS) derivatives of the urinary metabolites in the patient with SSADH deficiency. The
`ions targeted were m/z 231 for 2,2-dimethylsuccinate (IS), m/z 233 for GHB, m/z 321 for 2DT, m/z 219 for 3DT, m/z 247 for 4,5DH, m/z
`289 for 4HVPA and 5HVPA, m/z 287 for 3KVPA, m/z 261 for glutarate and m/z 275 for adipate. Peak identifications are same as Fig. 1.
`
`Fig. 3. Urinary excretion of GHB and its related metabolites in the patient with SSADH deficiency after VPA administration. Urease
`digestion method was used for sample preparation, and 2,2-dimethylsuccinate was used as an internal standard. The ions targeted were same
`as Fig. 2.
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`102
`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`also analyzed VPA metabolites in VPA medicated
`patients, and discuss the effects of VPA on GHB
`metabolism.
`
`2. Experimental
`
`2 .1. Chemicals
`
`We obtained 4-hydroxybutyric acid sodium salt
`from Tokyo Kasei Kogyo (Tokyo, Japan), and urease
`(type C-3: from Jack beans) from Sigma (St. Louis,
`MO). Gamma-butyrolactone-d (GBL-d , 99.5 atom
`6
`6
`%) was purchased from CDN (Quebec, Canada).
`GHB-d (0.5 mmol/ml) stock solution was prepared
`6
`by dissolving GBL-d in 0.1 N NaOH [9]. Other
`6
`chemicals are same as described [7].
`
`2 .2. Samples
`
`The male patient (2-month-old, the younger of the
`siblings) with SSADH deficiency has already been
`
`described [7]. Treatment with VPA was started when
`the patient was 69 days old (100 mg/kg/day). Urine
`was collected before VPA administration and on days
`4, 7, 24 and 40 after the start of administration.
`Urine specimens from patients with no metabolic
`disorders aged from 1-month-old to 7-years-old
`(VPA medicated; n523, VPA non-medicated; n520)
`were also examined. All samples were stored at
`220 8C until analysis.
`
`2 .3. Sample preparation
`
`Samples were prepared and derivatized as de-
`scribed [6,7,10] additional use of GHB-d
`as the
`6
`internal standards. In brief, 0.1 ml of urine was
`digested with 20 units of urease at 37 8C for 10 min.
`After adding 5 nmols of GHB-d and 25 nmoles of
`6
`2,2-dimethylsuccinic acid as the internal standard,
`the urine was deproteinized with 1 ml ethanol. The
`precipitate was removed by centrifugation, and then
`the supernatant was concentrated under
`reduced
`pressure and evaporated to dryness under nitrogen
`
`Fig. 4. Variations in urinary excretion of VPA metabolites and medium chain dicarboxylic acids in the patient with SSADH deficiency after
`VPA administration. Organic acid extraction method was used for sample preparation, and 2,2-dimethylsuccinate was used as an internal
`standard. The ions targeted were same as Fig. 2.
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`103
`
`gas. The residue was trimethylsilylated using 100 ml
`of BSTFA plus 10% TMCS at 80 8C for 30 min, then
`1 ml of the reaction mixture was analyzed by GC–
`MS. Urinaly organic acids were also extracted by
`organic solvent extraction method. To an amount of
`urine equivalent to 1 mmol creatinine (total volume
`1 ml), 25 nmoles of 2,2-dimethylsuccinic acid was
`added as an internal standard. After acidification to
`pH 1 with 2 N HCl, the sample was extracted three
`times with 3 ml diethyl ether. The organic phase was
`dried on anhydrous Na SO and evaporated to
`2
`4
`dryness under nitrogen stream. Trimethylsilylation
`was same as urease digestion method. Urinary
`creatinine was enzymatically measured using a Beck-
`man Synchron CX5CE auto analyzer
`(Beckman
`Instruments, Brea, CA).
`
`2 .4. Gas chromatography–mass spectrometry
`
`Samples were analyzed using a QP-5000 gas
`chromatograph–mass
`spectrometer
`(Shimadzu,
`Kyoto, Japan) with a fused-silica capillary column
`
`(J&W DB-5MS, 30 m30.25 mm30.25 mm). The
`GC–MS conditions were the same as described
`previously [7]. The temperature was programmed to
`increase at a rate of 17 8C/min from 60 to 325 8C,
`which was finally maintained for 10 min. Electron
`impact mass spectra were obtained by repetitive
`scanning at a rate of 0.25 s intervals from m/z 50 to
`650. We quantified GHB by mass chromatography.
`The targeted ions for quantification were follows;
`m/z 233 for GHB and m/z 239 for GHB-d .6
`
`3. Results and discussion
`
`It has been reported that VPA inhibits SSADH but
`not GABA-transaminase and succinic semialdehyde
`reductase [11–13]. The ‘‘valproate-effect’’ has been
`discussed by Divry et al. and Johannessen [8,13].
`They mentioned that SSADH inhibition by valproate
`was the reason for higher production of GHB. GHB
`is converted to succinic semialdehyde by the reverse
`reaction with non-specific reductase, and inhibition
`
`Fig. 5. Calibration curve for GHB in urease digestion method. GHB-d (5 nmol) was used as an internal standard. The ions targeted for
`6
`GHB-d and GHB were m/z 239 and m/z 233, respectively.
`6
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`104
`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`of this enzyme by VPA is believed to be another
`reason for accumulation of GHB [13]. To elucidate
`the relationships between GHB accumulation and
`VPA metabolism, the urinary metabolic profiles of
`the patients were investigated following VPA treat-
`ment.
`Total ion current chromatogram and mass chro-
`matograms of urinary metabolites in our patient with
`SSADH deficiency are shown in Figs. 1 and 2. GHB,
`2-deoxytetronate
`(2DT, b-oxidation product of
`GHB), 3-deoxytetranate (3DT, a-oxidation product
`of GHB) and 4,5-dihydroxyhexanoate (4,5DH, con-
`densation product of succinic semialdehyde with a
`2-carbon fragment) were observed in the patient
`urine in large amounts [14]. The relationships be-
`tween GHB excretion and several metabolites before
`and after VPA treatment are shown in Figs. 3 and 4.
`After valproate therapy, excretion of GHB and other
`SSADH deficiency related metabolites increased by
`
`factors ranging from 1.7 to 4 times (Fig. 3). On the
`other hand, levels of 3-keto-valproate (3KVPA; b-
`oxidation product of VPA [15]) was decreased
`sharply, while
`levels
`of
`4-hydroxyvalproate
`(4HVPA) and 5-hydroxyvalproate (5HVPA), v-oxi-
`dation products of VPA [16], were elevated follow-
`ing VPA administration (Fig. 4). This suggests that
`v-oxidation of VPA was stimulated but b-oxidation
`of VPA was suppressed in the patient.
`If GHB accumulation influences fatty acid b-
`oxidation [14], it is likely that fatty acid v-oxidation
`is also stimulated and medium-chain dicarboxylic
`aciduria is observed in SSADH patients. SSADH-
`deficient patients are known to occasionally exhibit
`dicarboxylic aciduria [3]. However, dicarboxylic
`aciduria was variable and was not a parallel phenom-
`enon with GHB accumulation in our patient. Fatty
`acid b-oxidation enzymes and SSADH are located in
`mitochondria, but fatty acid v-oxidation enzymes are
`
`Fig. 6. Relationships between urinary GHB concentration and age in VPA medicated (-s-) and VPA non-medicated (-•-) patients without
`metabolic disorders. Urease digestion method was used for sample preparation, and GHB-d was used as an internal standard.
`6
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`105
`
`in cytosol. Therefore, it seems that the accumulation
`of GHB is not concern with dicarboxylic aciduria
`directly, but
`the capacity of b-oxidation in these
`patients is reduced and some factor stimulates v-
`oxidation.
`To elucidate the relationships between GHB ac-
`cumulation and VPA metabolism,
`the metabolic
`profile in the urine of the patients who have no
`metabolic disturbance were also investigated follow-
`ing VPA treatment. Although the concentrations of
`GHB in SSADH deficiency are typically 880–3630
`mmol/mol creatinine [7], but those in normal in-
`dividuals are lower than 5 mmol/mol creatinine. An
`improved method for GHB quantification, which was
`
`consisted with stable isotope dilution technique
`combined with urease digestion-GC–MS method,
`was used in this study. GHB-d was used as an
`6
`internal standard in this experiment. This method
`successfully detects GHB with a sensitivity of 0.05
`nmol/0.1 ml urine (Fig. 5). Detection limit was
`improved about 10 times compared with former
`report [7] and recovery of GHB from urine is more
`than 93% and reproducibility (C.V.%) was 8% at 5
`nmol. The concentrations of GHB in control are
`shown in Fig. 6. Age dependent excretion of GHB
`was observed in VPA non-medicated patients without
`metabolic disorders, Spearman’s correlation coeffi-
`cient, P50.024 [17]. On the other hand, no statisti-
`
`Fig. 7. Relationships between urinary GHB concentration and urinary 4HVPA (-j-) or 5HVPA (-h-) levels in VPA medicated patients
`without metabolic disorders. Urease digestion method was used for sample preparation. The ions targeted were m/z 289 for 4HVPA and
`5HVPA, and m/z 239 for GHB-d (internal standard).
`6
`
`AMN1007
`IPR of Patent No. 8,772,306
`
`

`
`106
`
`T.Shinka et al. / J.Chromatogr.B 792 (2003) 99–106
`
`cally significant age dependence was observed in
`VPA medicated patients.
`We compared the relationship between GHB and
`VPA metabolites in VPA medicated control. The
`levels of 4HVPA and 5HVPA excretion had both
`statistically significant relationships with GHB (Fig.
`7).
`Our data reveals that accumulation of 4HVPA and
`5HVPA is correlated with GHB levels. These VPA
`metabolites are expected to have the competitive
`inhibition on GHB non-specific reductase because
`their chemical structures are very similar to that of
`GHB. The mechanism of GHB accumulation in
`SSADH deficient patient after VPA administration is
`thought
`to be follows: VPA inhibits the SSADH
`reaction and succinic semialdehyde levels are in-
`creased, but VPA dose not inhibit succinic semial-
`dehyde reductase and thus SSA is converted to GHB.
`This pathway is reversible, but the conversion of
`GHB to SSA is catalyzed by cytosolic non-specific
`reductase. This non-specific reductase is competitive-
`ly inhibited by 4HVPA and 5HVPA, resulting in
`GHB accumulation.
`
`Acknowledgements
`
`This study was partly supported by a grant for
`Child Health and Development (13C-4) from the
`Ministry of Health, Labor and Welfare of Japan.
`
`References
`
`[1] C . Jakobs, M. Bojasch, E. Moench, D. Rating, H. Siemes, F.
`Hanefeld, Clin. Chim. Acta 111 (1981) 169.
`[2] C . Jakobs, J. Jaeken, K.M. Gibson, J. Inherit. Metab. Dis. 16
`(1993) 704.
`[3] Z . Rahbeeni, P.T. Ozand, M. Rashed, G.G. Gascon, M. Al
`Nasser, A. Al Odaib, M. Amoudi, M. Nester, S. Al Garawi,
`J. Brismar, Brain Dev. 16 (1994) 64.
`[4] K .M. Gibson, G.F. Hoffmann, A.K. Hodson, T. Bottiglieri,
`C. Jakobs, Neuropediatrics 29 (1998) 14.
`[5] K .M. Gibson, S. Aramaki, L. Sweetman, W.L. Nyhan, D.C.
`DeVivo, A.K. Hodoson, C. Jakobs, Biomed. Environ. Mass
`Spectrom. 19 (1990) 89.
`[6] T . Kuhara, J. Chromatogr. B 758 (2001) 3.
`[7] T . Shinka, Y. Inoue, M. Ohse, A. Ito, M. Ohfu, S. Hirose, T.
`Kuhara, J. Chromatogr. B 776 (2002) 57.
`[8] P . Divry, P. Baltassat, M.O. Rolland, J. Cotte, M. Hermier,
`M. Duran, S.K. Wadman, Clin. Chim. Acta 129 (1983) 303.
`[9] L .A. Ciolino, M.Z. Mesmer, R.D. Satzger, A.C. Machal,
`H.A. McCauley, A.S. Mohrhaus, J. Forensic Sci. 46 (2001)
`1315.
`[10] I . Matsumoto, T. Kuhara, Mass Spectrom. Rev. 15 (1996)
`43.
`[11] J .W. Van Der Laan, T.H. De Boer, J. Bruinvels, J. Neuro-
`chem. 32 (1979) 1769.
`[12] J .F. Rumigny, M. Maitre, C. Cash, P. Mandel, FEBS Lett.
`117 (1980) 111.
`[13] C .U. Johannessen, Neurochem. Int. 37 (2000) 103.
`[14] G .K. Brown, C.H. Cromby, N.J. Manning, R.J. Pollitt, J.
`Inherit. Metab. Dis. 10 (1987) 367.
`[15] I . Matsumoto, T. Kuhara, M. Yoshino, Biomed. Mass
`Spectrom. 3 (1976) 235.
`[16] T . Kuhara, I. Matsumoto, Biomed. Mass Spectrom. 1 (1974)
`291.
`[17] J .J. Pitt, R. Hawkins, M. Cleary, M. Eggington, D.R.
`Thorburn, L. Warwick, J. Inherit. Metab. Dis. 20 (1997) 39.
`
`AMN1007
`IPR of Patent No. 8,772,306