Báo cáo khoa học: Brain succinic semialdehyde dehydrogenase Reactions of sulfhydryl residues connected with catalytic activity pdf

Brain succinic semialdehyde dehydrogenase
Reactions of sulfhydryl residues connected with catalytic activity
Byung Ryong Lee
1
, Dae Won Kim
1
, Joung-Woo Hong
2
, Won Sik Eum
1
, Hee Soon Choi
1
, Soo Hyun Choi
1
,
So Young Kim
1
, Jae Jin An
1
, Jee-Yin Ahn
1,
*, Oh-Shin Kwon
3
, Tae-Cheon Kang
4
, Moo Ho Won
4
,
Sung-Woo Cho
5
, Kil Soo Lee

1
, Jinseu Park
1
and Soo Young Choi
1
1
Department of Genetic Engineering and Research Institute for Bioscience and Biotechnology, Hallym University, Chunchon, Korea;
2
Department of Molecular and Cellular Biochemistry, The Ohio State University, Columbus, OH, USA;
3
Department of Biochemistry,
Kyungpook National University, Taegu, Korea;
4
Department of Anatomy, College of Medicine, Hallym University, Chunchon,
Korea;
5
Department of Biochemistry and Molecular Biology, University of Ulsan College of Medicine, Seoul, Korea
Incubation of an NAD
+
-dependent succinic semialdehyde
dehydrogenase f rom bovine brain with 4-dimethylamino-
azobenzene-4-iodoacetamide (DABIA) resulted in a time-
dependent loss of enzymatic activity. This inactivation
followed pseudo fir st-order kinetics with a second-order rate
constant of 168
M
)1
Æmin
)1
. The spectrum of DABIA-labeled

enzyme show ed a characteristic p eak of t he DABIA a lkyl-
ated sulfhydryl group chromophore at 436 nm, which was
absent from the spectrum of the native enzyme. A linear
relationship was observed between DABIA binding and t he
loss of enzyme activity, which extrapolates to a stoichio-
metry of 8.0 mol DABIA derivatives per mol enzyme
tetramer. This inactivation was prevented by p reincubating
the enzyme with substrate, succinic semialdehyde, but not b y
preincubating with coenzyme NAD
+
. After tryptic diges-
tion of the enzyme modified with DABIA, two peptides
absorbing a t 436 nm were isolated by reverse-phase HPLC.
The amino acid sequences of the DABIA-labeled pep-
tides were VCSNQFLVQR and EVGEAICTDPLVSK,
respectively. These s ites are identical to the putative active
site sequences of other brain succinic semialdehyde dehy-
drogenases. These results suggest that the c atalytic function
of succinic s emialdehyde dehydrogenase is inhibited by the
specific binding of DABIA to a cysteine residue at or near its
active site.
Keywords: brain succinic semialdehyde d ehydrogenase;
DABIA; GABA shunt; reactive cysteine residues.
c-Aminobutyric acid (GABA) is produced from glutamate
in a reaction catalyzed by glutamate decarboxylase (GAD)
and further metabolized to succinate by the successive
action of GABA transaminase (GABA-T) and succinic
semialdehyde dehydrogenase (SSADH). The carbon skel-
etal of GABA therefore enters the tricarboxylic acid in
the form of succinate. GABA metabolism has been well

characterized in the mammalian central nervous system
where GABA functions as a major inhibitory n eurotrans-
mitter.
SSADH, the final enzyme in GABA metabolism, has
been purified from rat, human an d bovine brain [1–3]. T his
enzyme is also the site of an inborn error of human
metabolism [4]. In a utosomal recessively inherited SSADH
deficiency, now identified in more than 45 patients who
manifest varying d egrees of psychomotor retardation with
speech delay, the normal o xidative pathway is blocked,
thereby resulting in the a ccumulation of succinic semialde-
hyde (SSA). Metabolic patterns in physiologic fluids derived
from patients show large increases in gamma-hydroxybu-
tyrate (GHB) [5], t he reduction p roduct of SSA by succinic
semialdehyde reductase [6]. GHB, the biochemical hallmark
of SSADH deficiency, produces central n ervous system
effects including altered motor activity and behavior
disturbances when administered to animals and humans
at pharmacologic levels [7].
Recently, an SSADH cDNA was c loned from rat brain
and human liver [8]. The mammalian SSADH bears signi-
ficant homology to bacterial NADP
+
-SSADH and con-
served regions of aldehyde dehydrogenases, suggesting that it
is a member of the aldehyde dehydrogenase superfamily.
SSADH cD NA and g enomic sequences have been used to
identify two point mutations in the SSADH genes derived
from fo ur patients [9]. Splicing mutations resulted in exon
skipping in all four cases. In addition, a frameshift a nd pre-

mature termination was observed in one case and an
in-frame deletion in the resulting protein was detected in
the other case. Parents an d s iblings were shown to b e
Correspondence to S. Y. Choi, Department of Genetic Engineering,
Hallym, University, Chunchon 200-702, Korea.
Fax: +82 33 241 1463, Tel.: +82 33 248 2112,
E-mail: or O S. Kwon, Department of
Biochemistry, Kyungpook National University, Taegu 702-701,
Korea. Tel.: +82 53 950 6356, E- mail:
Abbreviations: DABIA, 4-dimethylaminoazobenzene-4-iodoaceta-
mide; Dys, DABIA-Cys; GAD, glutamate decarboxylase; GABA,
c-aminobutyric acid; GABA-T, c-aminobutyric acid transaminase;
GBH, gamma-hydroxybutyrate; SSA, succinic semialdehyde;
SSADH, succinic semialdehyde dehydr ogenase.
*Present address: Department o f Pathology, S chool of M edicine ,
Emory U niversity, Atlanta, G A, USA.
(Received 30 August 2004, revised 12 October 2004,
accepted 25 October 2004)
Eur. J. Biochem. 271, 4903–4908 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04459.x
heterozygous for the splicing abnormality [10]. In addition,
the intensive analysis on novel mutations in human SSADH
locus suggested that the missense mutations caused by point
mutations, s mall insertionsand s malldeletions in t he genomic
level would be causative of SSADH deficiency [11,12].
Despite the importance of SSADH in the metabolism of
GABA, the structural studies of the enzyme have not yet
been well in vestigated. We h ave purified and characterized
an NAD
+
-dependent SSADH from bovine b rain [2], and

found that the arginine residues are connected with catalytic
activity of the enzyme [13]. Recently, we reported that a
specific lysyl residue is located at or near the coenzyme
binding site of the enzyme [14]. In the present study, we
identified a regulatory site of the brain SSADH by a
combination of l abeling with 4 -dimethylaminoazobenzene-
4-iodoacetamide (DABIA) and p eptide analysis.
Materials and methods
Materials
NAD
+
, succinic semialdehyde, DABIA, EDTA, bovine
serum albumin, trypsin (treated with tosylphenylalanylchlo-
romethane), and 2-mercaptoethanol were purchased from
Sigma (St Louis, MO, USA). CM-Sepharose, Blue-Seph-
arose, and 5¢-AMP-Sepharose were obtained from Amer-
sham Bioscience (Piscataway, NJ, USA). Bovine brains
were obtained from Majang-dong Packing Company
(Seoul, Korea).
Purification of enzyme and enzymatic assays
SSADH from bovine brain was purified according to a
procedure previously described [2]. The procedure exploited
four column chromatographic steps: C M-Sepharose, Blue-
Sepharose, hydroxyapatite and 5¢-AMP-Sepharose. For
precise measurement of enzymatic act ivity, the formation of
NADH was measured by the increase in absorbance at
340 nm. All enzymatic assays were performed in duplicate
and the initial velocity data was correlated with a s tandard
assay mixture containing 30 l
M

succinic semialdehyde a nd
1m
M
NAD
+
in 0.1
M
sodium pyrophosphate (pH 8 .4) a t
25 °C. One unit of enzyme was defined as the amount of
enzyme required to reduce 1 lmol of NAD
+
per min at
25 °C. Protein concentrat ion was estimated by the Bradford
procedure with a bovine s erum albumin standard [15].
Spectroscopic studies
The purified enzyme (5 l
M
) was incubated with 300 l
M
DABIA at 25 °C, in the dark, for 30 m in. At the end of
incubation, the sample was dialyzed against 0.1
M
potas-
sium phosphate (pH 7 .0) a nd the absorption s pectra were
recorded with a Kontron UVIKON 930 double beam
spectrophotometer (Tegimenta, Rotkreuz, Switzerland) in
the range 32 5–500 n m.
Modification of succinic semialdehyde dehydrogenase
with DABIA
The purified e nzyme was dialyzed against 0.1

M
potassium
phosphate (pH 7.0), 1 m
M
EDTA, a nd then used immedi-
ately. A total of 200 m
M
of DABIA was freshly prepared i n
dimethylformamide and kept on ice. The final concentration
of dimethylformamide in the incubation mixture was no
more than 1% (v/v) and was found to have no effect on
enzymatic activity. The i ncubation mixture (1 mL) con-
tained S SADH (5 l
M
), DABIA (100–400 l
M
)and0.1
M
potassium phosphate (pH 7 .0). The reaction was initiated
by addition of DABIA in the dark at 25 °C. At intervals
after the initiation of inactivation, aliquots were withdrawn
for the activity assay. Whenever possible a small sample
volume (2 lL) was used t o minimize artifactual blank due
to the t ransfer of DABIA.
Protection experiments were performed in a similar
manner except that the enzyme (5 l
M
) was preincubated
with a substrate SSA (3 m
M

) or c oenzyme NAD
+
(3 m
M
)
at 25 °C for 30 min before the modification was initiated by
the addition of DABIA. The amount of DABIA b ound to
the enzyme was determined by measuring the increase in
absorbance at 436 nm using a m olar extinction coefficient
of 29 000
M
)1
Æcm
)1
[16].
Labeling and tryptic digestion of succinic semialdehyde
dehydrogenase
To identify the DABIA binding site, 2.9 mg of enzymes
(50 l
M
) w ere t reated with DABIA as d escribed previously
[14,16]. Labeling of protein was conducted for 30 min in the
dark at 25 °C. The excess reagent was removed by Sephadex
G-25 superfine gel filtration. The solution was dried,
dissolved by first adding 20 lLof50%(v/v)formicacid
and then 600 lL water, transferred to a smaller tube
suitable for enzymatic digestion a nd dried again.
DABIA-labeled protein (20 nmol) was suspended in
0.5 m L of 0.1
M

ammonium bicarbonate buffer (pH 8), and
digested with trypsin, previously treated with tosylphenyl-
alanylchloromethane, for 20 h at 37 °C. The substrate/
enzyme molar r atio was 50 : 1.
Purification of DABIA-labeled cysteine-containing
peptides
To a 0.5 mL tryptic digest, 40 lL acetic acid was added and
the precipitate was removed by centrifugation (10 000 g,20
min, 4 °C). Peptides i n the sample solution were lyophilized
and separated by reverse-phase chromatography (LKB
Instruments, Uppsala, Sweden ) using a Vydac C
18
column
(0.46 · 25 cm). The separation was perform ed with a linear
gradient from 0 to 70% B in 40 min at a flow rate of
0.8 m LÆmin
)1
.EluantA:10m
M
potassium phosphate
(pH 7 ) containing 2% dimethylformamide; eluant B:
acetonitrile containing 4% dimethylformamide.
Further purification was achieved b y rechromatograph-
ing the peptides o n a Vydac C
18
column with a linear
gradient of 0–60% B i n 60 min. Eluant A: 0.1% t rifluoro-
acetic acid (pH 2.15); eluant B: 0.1% t rifluoroacetic a cid in
acetonitrile/H
2

O (80 : 20, v/v). Cysteine derivatized with
DABIA and the corresponding phenylthiohydantoin deriv-
ative were prepared and characte rized according t o the
procedure described by Chang et al. [17]. The latter was
used as standard in the quantitative evaluation of
other phenylthiohydantions obtained after automated
Edman degradation. The absorption properties of the
4904 B. R. Lee et al.(Eur. J. Biochem. 271) Ó FEBS 2004
DABIA d erivative o f cysteine (molar extinction coefficient
e ¼ 29 000
M
)1
Æcm
)1
at 436 nm) was used to determine the
cysteine content of t he derivatized peptides.
Amino acid analysis and peptide sequencing
Peptides (6 nmol) were hydrolyzed for 24 h in 6
M
HCl,
containing 0.1% thioglycolic acid at 110 °C in vacuum.
Amino acids derivatized with phenylisothiocyanate were
identified and quantified by HPLC (Waters, Milford, MA,
USA) Pico-Tag system, using a Nova-Pak C
18
column run
at room temperature with a flow rate of 1 mLÆmin
)1
.
For the amino acid sequence analysis, the labeled peptide

was subjected to automated Edman degradation on a
Beckman Model 890M sequenator according to the manu-
facturer’s instructions.
Results
Inactivation of succinic semialdehyde dehydrogenase
by DABIA
The r elevance of sulfhydryl groups in the catalytic activity
of SSADH, was examined by reacting the enzyme with
DABIA. DABIA, a chromophoric reagent, was chosen for
our study for the following reasons: (a) it reacts with the S H
groups of SSADH under native c onditions; (b) the labeled
peptides and amino acid residues are easily monitored at
436 n m because of the large extinction coefficient of the
chromophore; (c) i n a case to purify the peptides alkylated
by DABIA, the derivatized peptides are more hydrophobic
than unlabeled peptides, allowing their separation by reverse
phase HPLC.
Incubation of SSADH with increasing concentrations of
DABIA r esulted i n a progressive d ecrease in its e nzymatic
activity (Fig. 1 ). This inactivation followed pseudo first-
order k inetics w ith c oncentrations of DABIA i n t he range
100–400 l
M
. The pseudo first-order rate constants, obtained
at each DABIA concentration, were plotted as a function of
DABIA concentration (Fig. 1, inset). The second-order rate
constant for the inactivation of DABIA was 168
M
)1
Æmin

)1
,
as determined from the slope of this plot.
In an effort to demonstrate that DABIA is bound to
sulfhydryl groups of SSADH, 5 l
M
of SSADH was
incubated with or without 400 l
M
of DABIA at p H 7.0 in
the dark for 30 min and absorption was monitored from 325
to 500 nm. The spectrum of DABIA-labeled enzyme showed
a characteristic p eak at 4 36 nm (Fig. 2 , curve 2), which was
absent from the spectrum o f the native enzyme (Fig. 2, curve
1). The absorption at 436 nm corresponds to a DABIA
alkylated sulfhydryl g roup chromophore. The value for the
incorporation of DABIA labeled on SSADH was measured
using an extinction coefficient of 29 000
M
)1
Æcm
)1
at
436 n m. DABIA gave overall incorporation values of about
7.5 m ol per enzyme tetramer, indicating that 8 mols of
sulfhydryl groups of SSADH were masked. The correlation
between DABIA i ncorporation and SS ADH e nzyme activ-
ity is shown in Fig. 3 . During the inactivation process, a
linear relationship w as obse rved betw een DABIA a nd the
loss of enzyme activity, which extrapolates to a stoichio-

metry of 8.0 mol DABIA derivatives per mol enzyme
tetramer, based on increased absorbance at 436 nm.
The inactivation studies were carried out in the presence
of substrate or coenzyme to define the site(s) modified by
DABIA. The reaction between SSADH and DABIA was
effectively prevented by incubating SSADH with the
substrate, SSA, but not with coenzyme, NAD
+
(Table 1).
Fig. 1. Determination o f the rate constant (K
obs
) f or the inactivation o f
SSADH at different concentrations of DABIA. The enzyme (5 l
M
)was
incubated with 1 00 l
M
(d), 200 l
M
(s), 300 l
M
(j) a nd 400 l
M
(h)
of DABIA in 0 .1
M
potassium phosphate ( pH 7.0) at 25 °Cinthe
dark. Aliquots withdrawn from the incubation mixtu res were tested for
enzymatic activity. T he inset sh ows the d epen dence of the observed
rate constant (K

obs
) on DABIA concentratio n.
Fig. 2. Absorption spectra of native (curve 1) and DABIA-treated (curve
2) SSADH. At the end of incubation, the absorption spectra were
determined as desc ribed in M aterials and m ethod s.
Ó FEBS 2004 Brain succinic semialdehyde dehydrogenase (Eur. J. Biochem. 271) 4905
These results suggest that the loss of SSADH enzymatic
activity may be the result of the binding of DABIA to
specific sulfhydryl groups located at or near the substrate
binding site of SSADH.
Isolation of modified peptides
To identify the peptides modified by D ABIA, SSADH was
treatedwithDABIAanddigestedwithtrypsinasdescribed
above. After overnight trypsin digestion, the digested
sample was loaded on to a reverse-phase column (Vydac
C
18
). Two peptides, designated I and II, were detected by
monitoring the absorption spectrum at 436 nm (data not
shown), indicating that the modification induced by
DABIA was restricted to at least two amino acids in the
SSADH subunits.
Each peptide tagged with DABIA was further purified by
a second chromatography through a Hypersil ODS column
using a different solvent system as described previously
[14,16]. A fter the second chromatography, two single pure
peptides derivatized with DABIA were isolated from the
originally labeled p eptides, respectively, a s shown in F ig. 4.
Amino acid analysis and protein sequencing
The amino acid analysis and s equence of p eptides I a nd II

were examined and the observed sequences were found to be
in reasonable agreement with the a mino acid compositions
determined after the acid hydrolysis of each DABIA-labeled
peptide (data not shown).
The s toichiometric study (Fig. 3 ) and the analysis of
amino acid sequence ( Table 2 ) showed that one cysteine
residue in each peptide (peptides I and II) was labeled with
DABIA. The amino acid sequence analyses of peptides I
and II revealed that the peak fractions contained the
Fig. 3. Stoichiometry of DABIA inactivation. SSADH (5 l
M
)was
incubated with 400 l
M
of DABIA in 0.1
M
potassium phosphate
(pH 7.0) at 25 °C in the dark. The inactivation of SSADH is plotted as
a f unction o f mol DABIA incorporated per mol e nzyme.
Table 1. Inactivation of succinic semialdehyde dehydrogenase by
DABIA. Enzyme (5 l
M
), DABIA (300 l
M
), NAD
+
(3 m
M
)andsuccinic
semialdehyde (3 m

M
)wereused.Thedatarepresentthemeanofthree
independent experiments with the difference expr essed as ± deviation.
Reaction mixture
Remaining
activity (%)
Enzyme 100 ± 3
Enzyme + DABIA 15 ± 2
Enzyme + NAD
+
+ DABIA 23 ± 3
Enzyme + succinic semialdehyde + DABIA 91 ± 4
Fig. 4. Second chromatography of the tryptic peptides labeled with
DABIA. Peptides I and II were purified by HPLC using a Vydac
C
18
column and a linear gradient of ac etonitrile (0–60%) containing
5m
M
sodium phosphate ( pH 6.4) for 120 min at a flow rate of
0.5 mLÆmin
)1
. Elution was monitored at 436 nm and the DABIA-
labeled peptides ( I and II ) were s equenced by Edman degradation.
Table 2. Sequences of the cysteine-containing tryptic peptides from
succinic semialdehyde dehydrogenase. DABIA-Cys (Dys) was deter-
mined as the DABIA phenylthiohydantoin (residues indicated in
bold).
Peptide Sequence
I Val-Dys-Ser-Asn-Gln-Phe-Leu-Val-Gln-Arg

II Glu-Val-Gly-Glu-Ala-Ile-Dys-Thr-Asp-Pro-
Leu-Val-Ser-Lys
4906 B. R. Lee et al.(Eur. J. Biochem. 271) Ó FEBS 2004
amino acid s equences Val-Xaa-Ser-Asn-Gln-Phe-Leu-Val-
Gln-Arg and Glu-Val-Gly-Glu-Ala-Ileu- Xaa-Thr-Asp-Pro-
Leu-Val-Ser-Lys, respectively, where Xaa represents an
assayable phenylthiohydantoin amino acid. This residue can
be designated DABIA-Cys (Dys ),basedonaminoacid
analysis. O f i nterest, the amino acid sequences of peptides I
and II were found to be identical to regions of human
SSADH, i.e. amino acids 341–350 and 266–279, respect-
ively.
Discussion
Little is known about the chemistry of the active site of
SSADH, partly because t he crystal structure of this enzyme
is not available. Therefore, it is essential that a detailed
structural desc ription of SSADH is elucidated. Previously,
we purified the homotetramer SSADH from bovine brain
homogenate [ 2]. Recently, an investigation o n the cat alytic
role of specific amino a cid residues i n the enzyme i ndicated
the involvement of a lysyl residue in its e nzymatic activity
[14]; this conclusion was reached based on evidence
obtained by chemically modifying SSADH with pyrid-
oxal-5¢-phosphate, a specific lysine residue modifying rea-
gent. In the present study, we identified at least one
substrate binding domain in brain SSADH by combining
DABIA labeling and peptide a nalysis.
DABIA has been widely used in structural and functional
studies to selectively label reactive cysteine residues in
particular, which are ofte n directly involved in the catalytic

mechanisms of active sites [16–18]. Haloacetamide deriva-
tives such a s DABIA reac t with c ysteine via a S
N
2reaction
mechanism to give the corresponding carboxamidomethyl
derivatives. The r eaction of h aloacetamide derivatives w ith
cysteine is 20–100 times as rapid as with other cysteine-
modifying reagents. In addition, the t wo benzene rings
provide DABIA with a large extinction coefficient, which
allows DABIA-labeled cysteine to be measured efficiently
and rapidly [19].
The incubation of SSADH with increasing concentra-
tions of DABIA resulted in a p rogressive reduction in
enzyme activity (Fig. 1 ). The evidence for the specific
modification o f cysteine r esidues by D ABIA was provided
by monitoring the absorption of DABIA-alkylated cyste-
ines at 436 nm ( Fig. 2). The nature of the i nhibitory effect
exerted by DABIA was studied in detail. The possibility t hat
DABIA inhibition is the result of the reaction of cysteine
residues critically connected with catalysis was investigated
by performing inhibition studies in the presence and absence
of the substrate S SA or in the presence or absence of
coenzyme NAD
+
. At pH 7.0, the inhibitory effect of
DABIA was influenced by SSA a t 3.0 m
M
(Table 1) . The
near complete protection afforded by SSA, strongly
suggests that the inactivation occurred because of an

interaction between DABIA and cysteine residues located
at or near the substrate binding site of SSADH. In marked
contrast to SSA, the coenzyme NAD
+
, did not afford any
protection against DABIA i nactivation.
Although d ifferences in the absorption spectra of native
and DABIA-labeled SSADH were observed by absorption
spectroscopy (Fig. 2), we have evide nce that the conform-
ational changes of SSADH did not occur when it reacts with
DABIA. We investigated these conformational changes
indirectly by fluorometric anisotropy, but no differences in
the anisotropies (A) of native (A ¼ 0.174) and modified
enzyme (A ¼ 0.1 79) were observed. This observation
demonstrates that the inactivation of SSADH occurs due
to the interaction between DABIA and cysteine residues o n
SSADH, and t hat it is not due to conformational changes of
SSADH.
During the inactivation process, a linear relationship was
observed between DABIA and the l oss of enzyme a ctivity,
which extrapolates to a stoichiometry of 8.0 m ol DABIA
derivatives p er mol enzyme tetramer, based on increased
absorbance at 436 n m (Fig. 3). There has been major
controversy concerning SSADH protein structure. In the
early 1970s, Cash et al. reported that SSADH was a dimeric
protein of mass i dentical subunits [20], however, in the
1980s, Ryzlak & Pietruszko reported that SSADH was a
tetrameric protein of mass nonidentical subunits [3], a
finding never repeated in the aldehyde dehydrogenase
literature. Our results support the cloning data of Chambliss

et al . [8], that SSADH is a protein of ho motetrameric
structure with mass identical subunits. Our previous puri-
fication of SSADH from bovine brain showed that the
SSADH is a tetramer c omposed of mass identical subunits,
although there is minor variation in t he molecular mass of a
single subun it [2]. This result is in keeping with t he notion
that mammalian SSADH is a homotetramer, and also
supports our conclusion that two cysteine residues per
single subunit are l abeled by DABIA.
To identify the site of inactivation, tryptic peptides
containing DABIA-labeled cysteine were prepared. The
results of sequence analysis (Table 2 ) showed that the
modified residues correspond to the cysteine residues
already identified in SSADH from human liver and rat
brain [8]. Even though the amino acid composition and
sequence of bovine brain SSADH have not been identified,
the labeled cysteine residues in peptides I and I I correspon-
ded to Cys342 and Cys272, respectively, of mammalian
brain SSADH. Cys342 is conserved in all members o f the
aldehyde dehydrogenase superfamily from bacteria to
human, and is presumed to be located at an active site
based on t he sequence homology with bovine aldehyde
dehydrogenase, the three-dimensional structure of which
has been solved [21]. This finding is consistent with our
observation that the substrate, SSA, nearly completely
blocked the DABIA-mediated in hibition of SSADH, but
coenzyme NAD
+
did not. On the other hand, Cys272,
found in peptide II, appears to be located near the coenzyme

binding site. Sequence similarity between Cys272 and the
aldehyde dehydrogen ase s uperfamily shows that Cys272 is
located 12 amino acids away from Gly284, which is widely
accepted to be a coenzyme binding site. In addition,
coenzyme binding sites in bovine a ldehyde dehydrogenase
have been shown to lie in an a-helical structure near the
surface of the enzyme [21]. If the three-dimensional structure
of SSADH is not much different from the structure of
aldehyde dehydrogenase, the DABIA molecule bound to
Cys272 seems to be located too far away from the coenzyme
binding site to interfere the NAD
+
–protein interaction.
In summary, the study presented here establishes that
brain SSADH is inhibited by t he binding of DABIA to
specific cysteine residues at or near the active site of the
protein. Knowledge o f the interaction between DABIA and
Ó FEBS 2004 Brain succinic semialdehyde dehydrogenase (Eur. J. Biochem. 271) 4907
SSADH may provide insights into approaches for the
design of a new class of regulators, which do not resemble
SSADH substrates.
Acknowledgements
This work was suppo rted by the 21st Century Brain Frontier Research
Grant (M103KV010019-03K2201-019 10), and National Research
Laboratory (NRL) Grant (M1-9911-00-0025 ) from the Ministry of
Science and Technology, and in part by the Research Grant from
Hallym University.
References
1. Chambliss, K .L. & G ibson, K.M. (1992) Succinic s emialdehyde
dehydrogenase from mamm alian brain: subunit analysis using

polyclonal antiserum. Int. J. Biochem. 24, 1493–1499.
2. Lee, B.R., H on g, J.W., Yoo, B .K., Lee, S.J., Cho, S.W. & Choi,
S.Y. (1995) Bovine brain succinic semialdehyde dehydrogenase:
purification, kinetics and reactivity of lysyl residues connected with
catalytic activity. Mol. Cells 5, 611–617.
3. Ryzlak, M.T. & Pietruszko, R. (1988) Human brain Ôhigh KmÕ
aldehyde dehydrogenase: purification, characterization, and
identification as N AD
+
-dependent succinic semialdehyde dehy-
drogenase. Arch. B iochem. B iophys. 266, 386–396.
4. Jakobs, C., Jaeken, J. & Gibson, K.M. (1993) Inherited dis orders
of GABA metabolism. J. Inherit. Metab. Dis. 16, 704–715.
5. Hogema, B.M., Akaboshi, S., Taylor, M., Solomons, G.S.,
Jakobs, C., S chutgens, R.B ., W ilcken, B., Worthington, S.,
Maropoulos, G., Grompe, M. & Gibson , K.M. ( 2001) Prenatal
diagnosis o f su ccin ic semialdehyde dehy drogenase deficiency:
Increased accuracy employing DNA, enzyme, and metabolic
analyses. Mol. Genet. Metab. 72, 218–222.
6. Cho, S.W., Son g, M.S., Kim, Y.G., Kang, W.D., Choi, E .Y. &
Choi, S.Y. (1993) Kinetics and m echanism of an NADPH-
dependent succinic semialdehyde reductase from bovine brain.
Eur. J. Biochem. 21 1 , 757–762.
7. Snead, O.C. (19 78) G amma hydroxybutyrate in the monkey. I.
Electroencephalographic, behavioral, and pharmacokinetic stud-
ies. N eurology 28 , 636–642.
8. Chambliss, K.L ., Caudle, D.L ., Hinson, D.D., Moomaw, C.R.,
Slaughter, C.A., Jakobs, C. & Gibson, K.M. (1995) Molecular
cloning of the mature NAD(+)-dependent succinic semialde-
hyde dehydrogenase from rat and human. cDNA isolation, evo-

lutionary homology, and tissue expression. J. Biol. Chem. 270,
461–467.
9. Blasi,P.,Boyl,P.P.,Ledda,M.,Novelleto,A.,Gibson,K.M.,
Jakobs, C., Homega, B., Akaboshi,S.,Loreni,F.&Malaspina,P.
(2002) Structure of h uman suc cinic se mialde hyde dehydroge nase
gene: Identification of promoter region and alternatively processed
isoforms. Mol. Genet. Met ab. 76, 348–362.
10. Chambliss, K.L., Hinson, D.D., Trettel, F., Malaspina, P.,
Novelletto, A., Jakobs, C. & Gibson, K.M. (1998) Two exon-
skipping mutations as the molecular basis of succinic
semialdehyde dehydrogenase deficiency (4-hydroxybutyric acid-
uria). Am.J.Hum.Genet.63, 399–408.
11. Aoshima, T., K ajita, M., Sekido, Y., Ishiguro, Y., Tsuge, I.,
Kimura,M.,Yamaguchi,S.,Watanabe,K.,Shimokata,K.&
Toshimitsu, N. (2002) Mutation analysis in a p atient with succinic
semialdehyde deh ydrogen ase deficiency: a c ompound hetero-
zygote with 103–121del and 146 0T>A of the ALDH5A1 ge ne.
Human Heredity 53 , 42–44.
12. Akaboshi, S., Hogema, B.M., Novelletto, A., Malaspina, P.,
Salomons, G.S., Maropoulos, G.D., Jakob s, C., Grompe, M. &
Gibson, K.M. (2003) Mutational spectrum of the s uccinic semi-
aldehyde dehydrogenase (ALDH5A1) ge ne and functional ana-
lysis o f 27 novel disease-causing mutations in patients with
SSADH deficiency. Human M utatio n 22, 442–450.
13. Bahn,J.H.,Lee,B.R.,Jeon,S.G.,Jang,J.S.,Kim,C.K.,Jin,L.H.,
Park,J.,Cho,Y.J.,Cho,S.W.,Kwon,O.S.&Choi,S.Y.(2000)
Brain succinic s emialdehyde dehydrogenase: r eaction of arginine
residues connected w ith catalytic activities. J. Biochem. Mol. Biol.
33, 3 17–320.
14. Choi,S.Y.,Bahn,J.H.,Lee,B.Y.,Jeon,S.G.,Jang,J.S.,Kim,

C.K.,Jin,L.H.,Kim,K.H.,Park,J.S.,Park,J.&Cho,S.W.
(2001) Brain succinic semialdehyde dehydrogenase: identification
of reactive lysyl residues labeled with pyridoxal-5¢-phosphate.
J. Neurochem. 76, 919–925.
15. Bradford, M.M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-d ye binding. Anal. Biochem. 72, 248–254.
16. Kim, Y.T. & Churchich, J.E. (1989) S equence of t he c ysteinyl-
containing pe ptides of 4-aminobutyrate aminotransferase. Iden-
tification of sulfhydryl residues i n volved in intersubunit linkage.
Eur. J. Biochem. 181, 397–401.
17. Chang, J.Y., Knecht, R. & Braun, D.G. (1983) A new method for
the selective isolation of cysteine-containing peptides. Specific
labeling of the thiol group with a hydrophobic chromophore.
Biochem. J. 211, 163–171.
18. Cardamone, M., Aslunan, K., Brandon, M.R. & Puri, N.K.
(1993) Identification of cys-containing peptides during
peptide mapping of recombinant proteins. Pept. Res. 6, 2 42–248.
19. Lundblad, R.L. (1995) The modification of cysteine. In Techniques
in Protein Modification pp. 63–83. C RC Press, Boca Raton, FL.
20. Cash, C .D., Ma itre, M . & Mandel, P. ( 1979) Purificati on f rom
human brain and some properties of two NADPH-linked akde-
hyde reductases which reduce succinic semialdehyde to 4- hydro-
xybutyrate. J. Neurochem. 33 , 1169–1175.
21. Steinmetz, C.G., Xie, P., Weiner,H.&Hurley,T.D.(1997)
Structure of m itochondrial aldehyde d ehydrogenase: the genetic
component of ethanol a ve rsion. Structure 5, 701–711.
4908 B. R. Lee et al.(Eur. J. Biochem. 271) Ó FEBS 2004