Volume 1.17, number 1 FEBS LETTERS August 1980
`
`SPECIFIC AND NON-SPEC~IC SUCCINIC SEMIALDEHYDE REDUCTASES FROM
`RAT BRAIN: ISOLATION AND PROPERTIES
`
`Centre de Neurochimie du CNRS, and Unit& 44 de L ‘INSERM Facultk de Mtdecine, 11, Rue Humann, 67085 Strasbourg Cede&
`France
`
`Received 2 June 1980 Revised version received 24 June 1980 1. Introduction Succinic semialdehyde (SSA) is an intermediate of the 4-aminobutyrate shunt pathway (GABA-shunt). In mammals, this pathway is particularly important in the brain [l]; this was bypass calculated to account for 10% of the flux through the Krebs cycle [2]. In brain, SSA is primarly oxidized to succinate by a spe- cific dehydrogenase which has been purified and char- acterized from several species (reviewed [3]). How- ever, brain tissue can also reduce SSA to vhydroxy butyrate (GHB) [4] and recently the enzyme(s) re- sponsible have been identified as NADPH-dependent aldehyde reductases [5]. The biosynthesis of GHB in brain tissue is of great interest as this compound which occurs naturally in the brain [6] induces anaesthesia when administered to man and animals in relatively large doses [7]. Although the degree to which this reductive pathway operates in vivo is not as yet known, the recent isolation from human brain of a fairly specific SSA reductase 181 and these similar results using rat brain as the enzyme source, strongly support the hypothesis that GHB biosynthesis may be a significant pathway of pharmacological interest. 2. Materials and methods Biogenic aromatic aldehydes were obtained by incubation of the parent amines with rat liver mono- amine oxidase 191. Address correspondence to: Dr Michel MAI’I’RE, Centre de Neurochimie du CNRS, 11, Rue Humann, 67085 Strasbourg Cedex, France For qualitative assays of column eluates, 50 ti enzyme samples were pipetted into tubes of cold 100 mM potassium phosphate buffer (pH 7), con- taining 5 mM 2-mercaptoethanol and 2 X lOA M SSA or p-nitrobenzaldehyde. The reaction was started by rapid addition of NADPH to 5 X lo-” M final cont. in a 1 ml total vol. and subsequent incuba- tion for 30 min at 3?C!. Then the decrease in NADPH fluorescence in the samples was measured at excita- tion 355 mn and emission 470 nm. Qu~titative enzyme assays were performed by direct recording of the initial rate of NADPH oxida- tion at 37°C in a double beam spectrophotometer at 340 nm, assuming an absorbance of 6.22 X lo3 for NADPH. The volume and composition of the incuba- tion medium were identical to that described for the qualitative assay, except that the, pH was 6.0 and the reference cuvette contained no aldehyde substrate. The reaction was started by addition of enzyme. 2.2. Enzyme extraction and
`
`Adult Wistar rats (50) were stunned, decapitated and the brains rapidly removed and suspended in 400 ml final vol. of cold 2 mM potassium phosphate buffer (pH 7) containing 1 X lo4 M glutathione and 1 X lo4 M phenyhnethylsulphonyl fluoride. All subsequent operations were done at 4°C. The suspension was homogenized at maximum speed for 3 min in a food blender. The homogenate was then centrifuged for 1 h at 30 000 X g and concentrated KC1 solution was added to the resultant clear super- natant solution to 100 mM fmal cont.
`
`p~.~~tion
`
`~I~e~ier~North-HoI~and Biomedical Press
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 1 of 6
`
`J. F. RUMIGNY, M. MAlTRE, C. CASH and P. MANDEL
`111
`

`
`Volume 117, number 1 FEBS LETTERS August 1980 2.3.
`
`Column chromatography
`All rinsing
`
`Protein determinations
`
`Molecular
`These
`
`weight determinations
`
`and elution buffers contained 5 mM 2.mercaptoethanol and 5 mM potassium phosphate (pH 7.2) up to the two final chromatographic steps where the pH was 7.8. The supernatant was absorbed onto a 1.6 X 13 cm column of blue Sepharose (Phar- macia). The column was rinsed with phosphate buffer (pH 7.2) containing 100 mM KCl, and then with the same buffer minus the KCl. The outlet of the blue Sepharose column was then attached to a 1.6 X 14 cm column of DEAE-cellulose (Whatman DE 52). The blue Sepharose column was then eluted directly onto the DEAE-cellulose column with 100 ml of 1 X lOA M Cibacron blue (Ciba-Geigy) dissolved in the above rinsing buffer. The dye is firmly retained as a narrow band at the top of the DEAE-cellulose column. The DEAE-cellulose column was separated from the blue Sepharose column and rinsed with buffer before elution with 400 ml linear gradient of O-200 mM KC1 in the same buffer. Fractions of 6 ml were col- lected. It is at this stage that two succinic semialde- hyde reductases are separated and thus they will now be referred to as SSR 1 and SSR 2 which is the order in which they are eluted from the DEAE-cellulose column. Both enzymes were then treated separately but almost identically. The enzymes were concen- trated to -10 ml in an Amicon cell equipped with a PM 10 membrane, then diluted to -100 ml and recon- centrated and rediluted in order to lower the KC1 concentration. They were then adsorbed onto 1.6 X 7 cm columns of QAE Sephadex (Pharmacia) columns. After rinsing with starting buffer, enzyme SSR 1 was eluted with a linear gradient of 200 ml 0- 100 mM KCl, and enzyme SSR 2 with the same volume of 50-250 mM KCl. Fractions of 4 ml were collected. The final two chromatographic steps were carried out at pH 7.8 and were identical for each enzyme. The active fractions from the QAE Sephadex columns were concentrated to 5 ml and applied to a 2.6 X 100 cm column of Sephadex G-150 (Pharmacia). Elution was carried out with potassium phosphate buffer 5 mM (pH 7.8) containing 5 mM 2.mercapto- ethanol and 6 ml fractions were collected. The active fractions were adsorbed directly onto a 0.9 X 16 cm column of 2’,5’-ADP Sepharose (Pharmacia). After rinsing with potassium phosphate buffer 5 mM (pH 7.8) containing 1 X lo4 M glutathione con- taining in addition 1 X 10e4 M glutathione, the enzymes were eluted with a linear gradient of 100 ml O-l X lo-’ M NADP and collected in 3 ml fractions. 2.4.
`
`The Folin method [lo] was used up to the DEAE- Sepharose step and thence the densitometry of the stained protein bands on SDS gels. Bovine serum albumin was used as the standard. 2.5.
`were determined using appropriate molec- ular weight markers by SDS-polyacrylamide gels [ 1 l] and non-denaturing gels of different polyacryl- amide concentrations [ 121. 2.6.
`These were measured for the two enzymes for both the forward and reverse reactions. For the for- ward reaction the standard reaction mixture was employed in 100 mM potassium phosphate buffers. For the reverse reaction, 100 mM phosphate buffer was used up to pH 8.0 and thereafter 100 mM borate buffers. The substrate concentrations were GHB = 10 mM and NADP = 5 X lo4 M. 2.7. K,,,
`For the forward reaction the Km values for SSA were measured in the standard reaction medium (pH 6) containing 5 X lo-’ M NADP. The Km values for NADPH were obtained similarly at a SSA concen- tration of 2 X lo4 M. Km values for GHB in the reverse reaction were measured in 100 mM phosphate buffer (pH 8.1) at 5 X lo4 M NADP and the values for NADP were similarly determined at 1 X lo-* M GHB. 2.8.
`For both enzymes, various aldehydes at 2 X 1 O4 M were substituted for SSA in the standard reaction medium at pH 6 and the relative initial reaction veloc- ities compared. Potential inhibitors were similarly tested using 2 X lo4 M SSA as substrate. 3. Results Fig.1 shows the separation of 3 peaks of aldehyde reductase activity by DEAE-cellulose chromato- graphy; the first peak (SSR 1) is active with both SSA and p-nitrobenzaldehyde. The second peak reduced only p-nitrobenzaldehyde whereas the third peak (SSR 2) was specific for SSA. 112
`
`pH
`
`optima
`
`values
`
`Substrate specificities and inhibition
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 2 of 6
`
`The pooled active fractions were stored at +4’C.
`

`
`Volume 117, number 1 FEBS LETTERS August 1980
`
`Activity,
`
`srbitrary units
`
`fraction XV
`
`Fig.]. DEAE-cellulose chromatography: (0) Activity with succinic semialdehyde; (r) Activity with 4nitrobenzaldehyde; (o) Azs4 Table 1 Purification of two SSA reductases from 50 rat brains Fraction Volume Units Total Protein units/ Yield Purification (ml) /ml units (mglml) mg protein (%I (-fold) SSR 2 Supernatant + Blue Sepharose DEAE-cellulose QAE-Sephadex Sephadex G-150 ADP-Sepharose SSR 1 Supernatant + Blue Sepharose DEAE-cellulose QAE-Sephadex Sephadex G-150 ADP-Sepharose 268 49.2 13 186 9.21 5.34 (100) - 95 26.9 2 556 0.105 256 19.4 48 59 35.3 2 083 0.0279 1 265 15.8 237 57 19.0 1 083 0.0104 1 827 8.2 342 10.5 60.3 633 0.0056 10 768 4.8 2016 280 56.5 15 820 7.21 7.84 (100) - 50 270 13500 0.056 4 821 85 615 50 197.6 9 880 0.017 11 624 62.5 1 483 90 77.6 6 984 0.005 15 520 44 1 980 18.7 360 6 732 0.020 18 000 42.5 2 297 Units are lo-’ mol NADPH oxidized/mm at 37°C
`
`113
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 3 of 6
`
`

`
`Volume 117, number 1 FEBS LETTERS August 1980 Table 1 summarizes the results of a typical purifi- cation. Both purified enzymes migrated as single pro- tein bands on SDS and nondenaturing polyacrylamide gels. The molecular weights determined by SDS gel electrophoresis are, respectively, 50 000 for enzyme SSR 1 and 43 000 for enzyme SSR 2. The molecular weights of the native enzymes determined on poly- a&amide gels are, respectively, 54 000 and 45 000. The pH optima of enzyme SSR 1 are, respectively, 5.5 for SSA reduction and 8.7 for GHB dehydrogena- tion (reverse reaction). For enzyme SSR 2, these values are, respectively, 5.0 and 8.1. The Km values determined under the above conditions are given in table 2. Table 3 shows the relative activities with various aldehyde substrates and table 4 compares the effect of various potential inhibitors on the two enzymes. Table 2 Km values expressed in molarities SSA NADPH GHB NADP SSR 1 1.4 x lo+ 2.6 x 1O-6 1.5 x 1o-2 2.2 x 10-s SSR 2 2.8 x 10-s 2.4 x lO+ 1.2 x 1o-2 1.4 x 10-G Table 3 Substrate specificities Substrate Relative activities SSR 2 SSR 1 (Specific enzyme) (Non-specific enzyme) Succinic semialdehyde 100 46.5 4Catboxy benzaldehyde nd 100 2-Methyl succinic semialdehyde 76.5 85 4-Nitrobenzaldehyde nd 62 3-Pyridine carboxaldehyde nd 32 Glyoxal nd 12.5 D-Lactaldehyde nd 9.2 3,4-Dihydroxyphenyl-acetaldehyde nd 8.9 D-LGlyceraldehyde nd 3.4 Iso-phthalaldehyde nd 3.1 5-Hydroxyindolacetaldehyde nd nd Propionaldehyde nd nd Indolacetaldehyde nd nd 3-Methoxy4-hydroxy-phenylglycolaldehyde nd nd Hydroxyphenylglycolaldehyde nd nd 4-Anisaldehyde nd nd n-Valeraldehyde nd nd Phenyl-methyl-ketone nd nd Acetaldehyde nd nd Benzaldehyde nd nd 4-Hydroxybenzaldehyde nd nd Glucose nd nd Lactose nd nd Arabinose nd nd Succinic semialdehyde/NADH 20 nd nd, not detected under our assay conditions 114
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 4 of 6
`
`

`
`x
`
`Volume 117, number 1 FEBS LETTERS August 1980 Table 4 Comparative inhibition study I~ibito~ Percent ~hibition SSR 2 Specific enzyme) SSR 1 (Nonspecific enzyme) Barbiturates Barbital 0 Pentobarbital 0 Phenobarbital 0 Amobarbital 0 Miscellaneous 4-Hydroxybenzaldehyde 55 Oxalate 0 Dipheny~yd~toinb 0 ~~orprorn~~e 29 Pyrazol 0 Diazepam 0 4-Hydroxybutyrate 20 ChloraI 0 Valproate 10 a Final cont. 10e3 M; b Final cont. 2
`10T4 M 80 16 72 70 68 0 63 43 0 25 21 19 96 4. Discussion Enzyme SSR 1 exhibits a broad substrate specificity and is strongly ~bited by b~biturates and certain ~ticon~s~t drugs. It is thus similar to many NADPHdependent aldehyde reductases (reviewed [ 131). However, enzyme SSR 2 shows a fairly high degree of specificity for SSA, and unlike the first enzyme, it is not appreciably inhibited by the various hypnotics/anticonvulsants tested, including valproate. This enzyme is thus similar to the SSA reductase isolated from human brain [S]. As valproate is an inhibitor of rat brain SSADH [14], its adm~istration might raise the endogenous level of SSA. If this is the case, it could explain the recent finding that adm~~tration of valproate to rats brings about some increase in cerebral GHB levels [ 151 since we have shown that one of the enzymes capable of reducing SSA to GHB is not significantly inhibited by this drug. However, it is unlikely that GHB forma- tion contributes to the anticonvulsant effect of val- proate, since administration of this drug to rats has been shown to antagonise the epileptiform electro- corticogram patterns elicited by GHB administration Pa In view of the multiple effects of GHB on the cen- tral nervous system (reviewed 1171) the fmding of an enzyme in brain which is apparently specific for its biosynthesis is of great interest. This work has been partially supported by a grant from INSERM (ATP no. 81.79.113). References [i ] Baxter, C. F. (1970) Handbook of neurochemistry (La&ha, A. ed) vol. 3, pp. 289-353, Plenum Press, New York. [2] Patel, A. J., Balazs, R. and Richter, D. (1970) Nature 226,1160-1161. [3] Cash, C., Ma&e, M., Ciesielski, L. and Mandel, P. (1979) GABA - Biochemistry and CNS functions (Mandel, P. and DeFeudis, F. eds) pp. 93-100, Plenum Press, New York. [4] Fishbein, W. and Bessman, P. (1964) J. Biol. Chem. 239, 357-361. [S] Tabakoff, B. and Von Wartburg, J. P. (1975) Biochem. Biophys. Res. Commun. 63,957-966. 161 Doherty, J. D., Hattox, S. E., Snead, 0. C. and Roth, R. H. (1978) J. Pharmac. Exp. Ther. 207,130-l 39. [7] Laborit, H., Jouany, J. M., G&d, J. and Fabiani, P. (1961) Neuro~sychoph~ma~l. 2,490-497. 115
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 5 of 6
`
`

`
`Volume 117, number I [8] Cash, C. D., Maitre, M. and Mandel, P. (1979) J. Neuro- them. 33,1169-1175. [9] Tabakoff, B., Anderson, R. and ABvisatos, S. (1973) Mol. Pharmacol. 9,428-437. [lo] Lowry, 0. H., Rosebrough, N. J., Farr, A.
`and RandaB, R. J. (1951) J. Biol. Chem. 193,265-279. [ 1 l] Weber, K. and Osborn, M. (1969) J. Biol. Chem. 294, 4406-4412. [ 121 Hedrik, J. L. and Smith, A. J. (1968) Arch. Biochem. Biophys. 126,155-164. FEBS LETTERS August 1980 [ 131 Tipton, K. F., Houslay, M. D. and Turner, A. J. (1977) Essays in neurochemistry and neuropharmacology (Youdim, M. B. H. et al. eds) vol. 1, pp. 103-138, Wiley-Interscience, London. [14] Harvey, P. K. P., Bradford, H. F. and Davidson, A. N. (1975) FEBS Lett. 52,251-254. (151 Snead, 0. C., Bearden, L. J. and Pegram, V. (1980) Neuropharmacol. 19,47-52. [ 161 Godshalk, M., Dzolzic, M. R. and Bonta, I. C. (1976) Neurosci. Lett. 3, 145-150. [17] Snead, 0. C. (1977) Life ScI. 20,1935-1944.
`
`L.
`
`116
`
`PAR1035
`IPR of U.S. Patent No. 8,772,306
`Page 6 of 6