Báo cáo khoa học: Reduction of S-nitrosoglutathione by human alcohol dehydrogenase 3 is an irreversible reaction as analysed by electrospray mass spectrometry pptx

Reduction of
S
-nitrosoglutathione by human alcohol dehydrogenase 3
is an irreversible reaction as analysed by electrospray mass
spectrometry
Jesper J. Hedberg*, William J. Griffiths, Stina J. F. Nilsson
†
and Jan-Olov Ho¨o¨g
From the Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden
Human alcohol dehydrogenase 3/glutathione-dependent
formaldehyde dehydrogenase was shown to rapidly and
irreversibly catalyse the reductive breakdown of S-nitroso-
glutathione. The steady-state kinetics of S-nitrosogluta-
thione reduction was studied for the wild-type and two
mutated forms of human alcohol dehydrogenase 3, muta-
tions that have previously been shown to affect the oxidative
efficiency for the substrate S-hydroxymethylglutathione.
Wild-type enzyme readily reduces S-nitrosoglutathione with
a k
cat
/K
m
approximately twice the k
cat
/K
m
for S-hydroxy-
methylglutathione oxidation, resulting in the highest cata-
lytic efficiency yet identified for a human alcohol
dehydrogenase. In a similar manner as for S-hydroxy-
methylglutathione oxidation, the catalytic efficiency of

S-nitrosoglutathione reduction was significantly decreased
by replacement of Arg115 by Ser or Lys, supporting similar
substrate binding. NADH was by far a better coenzyme than
NADPH, something that previously has been suggested to
prevent reductive reactions catalysed by alcohol dehydro-
genases through the low cytolsolic NADH/NAD
+
ratio.
However, the major products of S-nitrosoglutathione
reduction were identified by electrospray tandem mass
spectrometry as glutathione sulfinamide and oxidized
glutathione neither of which, in their purified form, served as
substrate or inhibitor for the enzyme. Hence, the reaction
products are not substrates for alcohol dehydrogenase 3 and
the overall reaction is therefore irreversible. We propose that
alcohol dehydrogenase 3 catalysed S-nitrosoglutathione
reduction is of physiological relevance in the metabolism of
NO in humans.
Keywords: alcohol dehydrogenase; glutathione-dependent
formaldehyde dehydrogenase; mass spectrometry; nitric
oxide; S-nitrosoglutathione.
The biological action of nitric oxide (NO) includes vaso-
dilation, inhibition of platelet aggregation and neurotrans-
mission [1]. In addition to the endogenous production, NO
is also a common air pollutant, a component of cigarette
smoke and is generated during metabolism of several
pharmaceutical drugs [2,3]. It has been suggested that many
intracellular processes of NO involve nitrosylation of thiols
[4]. S-nitrosothiols have been proposed to affect ventilation
[5], alter protein function [4], act as bioactivators of nitrites

and nitrates [6] or serve as a ÔpoolÕ of NO [7]. Recently,
attention has been drawn to the S-nitrosothiol of glutathi-
one (GSH), i.e. S-nitrosoglutathione (GSNO) where GSH
may alternatively act as a scavenger for NO to withstand
nitrosative stress [8] or act as a modulator of the action of
NO [9]. GSNO has been proposed to be produced under
physiologically relevant conditions [10–12] and indeed, it
has been detected in various biological systems including rat
cerebellum and primary glial cell cultures [8,9] as well as in
human airways [13].
As GSNO potentially has significant roles in various
biological processes, its fate and breakdown is of consider-
able interest. In addition to its complex chemical fate [14,15],
the enzymatic reduction of GSNO by rat alcohol dehy-
drogenase 3 (ADH3), also known as GSH-dependent
formaldehyde dehydrogenase, has been investigated [16].
More recently studies on ADH3–/– mice and yeast have
demonstrated that ADH3 is important for GSNO metabo-
lism and may regulate intracellular S-nitrosothiol levels in
these model systems [17]. Furthermore, plant ADH3 also
possesses GSNO reductase activity indicating the activity to
be general [18]. It is noteworthy with respect to ADH
activities, that reductive reactions are generally not believed
to be of physiological relevance due to the low NADH/
NAD
+
ratio in the cytosol [19,20]. Naturally this notion is
also applicable to the metabolism of GSNO by ADH3.
ADH3 belongs to the medium chain alcohol dehydro-
genase (ADH) system which, according to current nomen-

clature, is divided into five classes in man ADH1–ADH5
[21] in which the ADH proteins and genes have been
designated the same number. All ADHs display oxidative/
reductive enzymatic activities for a variety of alcohols/
aldehydes of both endogenous and exogenous origin [22,23].
The preferred coenzymes are NAD
+
and NADH for
Correspondence to J O. Ho
¨
o
¨
g, Department of Medical Biochemistry
and Biophysics, Karolinska Institutet, SE-171 77 Stockholm,
Sweden. Fax: +468 338453, Tel.: +468 728 7740,
E-mail:
Abbreviations:ADH,alcoholdehydrogenase;ADH3,alcoholdehy-
drogenase 3/GSH-dependent formaldehyde dehydrogenase; GSNO,
S-nitrosoglutathione; HMGSH, S-hydroxymethylglutathione;
NO, nitric oxide.
*Present address: Amersham Biosciences, SE-751 84 Uppsala, Sweden.
Present address: Department of Medical Biochemistry and Micro-
biology, Uppsala University, SE-751 23 Uppsala, Sweden.
(Received 4 November 2002, revised 17 January 2003,
accepted 27 January 2003)
Eur. J. Biochem. 270, 1249–1256 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03486.x
oxidation and reduction of substrates, respectively. Most
likely, one physiological substrate for human ADH3 is the
spontaneously formed complex of formaldehyde and GSH,
S-hydroxymethylglutathione (HMGSH). On the basis that

the enzyme is considered to be the prime guardian against
formaldehyde [24–26], shows extensive evolutionary con-
servation [27] and is expressed in all tissues examined
[28,29], ADH3 is regarded as essential for basic cell
metabolism possibly also including retinoic acid production
during growth [30].
GSNO is a recent addition to the ADH3 substrate
repertoire and here we investigate the enzymatic activity of
the human form of ADH3 for this compound. Recombi-
nant expression and purification of wild-type and mutated
forms of the enzyme enabled kinetic characterization and
investigation of substrate–enzyme interactions. Electrospray
tandem MS (MS/MS) was used to identify the products of
the enzymatic reduction of GSNO, which were previously in
doubt. Finally, we conclude that the glutathione sulfinamide
formed is not a substrate for ADH3, thus ADH3 catalysed
GSNO reduction is irreversible.
Materials and methods
Enzyme purification and chemicals
Generation of expression vectors for recombinant expres-
sion of the wild-type and mutant forms of ADH3 have
previously been described [31]. The various forms of ADH3
were expressed in Escherichia coli and purified to homo-
geneity in a three-step procedure essentially as described
[31]. Protein concentrations were determined colourimetri-
cally [32] and the purity was analysed by SDS/PAGE.
NAD
+
(crystalline lithium salt) and NADH (disodium salt)
were purchased from Boehringer Mannheim. GSH, GSSG,

glutathione sulfonic acid, methylglutathione, NADP
+
,
NADPH and GSNO were from Sigma. Formaldehyde
solutions in acetonitrile were made from newly opened glass
ampoules, 20% solutions (Ladd Research Industries).
Kinetic analysis
Enzymatic activity was monitored by following the absorb-
ance change at 340 nm with a Hitachi U-3000 spectro-
photometer. For HMGSH/NAD
+
oxidation/reduction a
molar absorptivity of 6220
M
)1
Æcm
)1
wasused.ForGSNO/
NADH reduction/oxidation the additive molar absortivity
of NADH and GSNO was used, i.e. 7060
M
)1
Æcm
)1
[16].
HMGSH concentrations were calculated according to [33]
with a K
eq
of 1.77 m
M

for the adduct formation. Deter-
mination of kinetic constants for NADH and NADPH
were performed with a GSNO concentration of 0.5 m
M
.
Determination of kinetic constants for NAD
+
and
NADP
+
, were performed with GSH and formaldehyde
concentrations of 1 m
M
each. Kinetic constants for
HMGSH and GSNO were determined with 2.4 m
M
NAD
+
and 0.1 m
M
NADH, respectively. All kinetic
determinations were performed in 0.1
M
potassium phos-
phate pH 7.5. To fit lines to data points a weighted
nonlinear regression analysis program was used (
FIG P
for
Windows; Biosoft). k
cat

values are based on a molecular
mass of 40 kDa for all proteins. Standard errors for K
m
and
k
cat
values were <10% except for GSNO reduction by
ADH3–Arg115Ser and GSNO reduction using NADPH as
coenzyme. Under these conditions the enzymes could not
be saturated and kinetic constant determinations should be
regarded with caution. For mass spectrometric analyses,
reduction of GSNO was performed in 10 m
M
ammonium
bicarbonate pH 7.5, as phosphate ions severely impair mass
spectrometric analysis. Reaction velocities were not signifi-
cantly different in this buffer as compared to that used for
kinetic analyses.
MS
Electrospray mass and tandem mass spectra were recorded
on an AutoSpec-OATOFFPD high-resolution magnetic
sector-orthogonal acceleration time-of-flight (OATOF) tan-
dem mass spectrometer (Micromass). Both positive- and
negative-ion nano-electrospray spectra were recorded by
spraying sample mixtures from metal coated borosilicated
capillaries. Tandem mass spectra were recorded by selecting
the desired precursor ion with the double focussing sectors
of the instrument (MS1) focussing them into an inter-
mediate collision cell containing Xe and orthogonally
accelerating undissociated precursors and fragment ions

into the time-of-flight (TOF) analyser (MS2).
Product isolation
In an effort to isolate the major product, i.e. glutathione
sulfinamide, a reaction mixture initially containing 1 m
M
GSNO, 2 m
M
NADH and 100 lg ADH3 in 10 m
M
ammonium bicarbonate pH 7.5 was incubated at 37 °C
for 2 h to reach equilibrium. The resulting reaction products
were resolved on a strong anion exchange column
Table 1. Steady state kinetic constants for HMGSH/GSNO oxidation/reduction at pH 7.5, catalysed by wild-type and mutant forms of ADH3.
K
m
-andk
cat
-values were determined from initial velocity experiments at 25 °C with substrate concentrations varied over a 10-fold range. Values are
given as means (± SE) of three independent experiments. NAD
+
and NADH concentrations were fixed at 2.4 m
M
and 0.1 m
M
, respectively.
Enzyme
HMGSH GSNO
K
m
(l

M
)
k
cat
(min
)1
)
k
cat
/K
m
(min
)1
Æm
M
)1
)
K
m
(l
M
)
k
cat
(min
)1
)
k
cat
/K

m
(min
)1
Æm
M
)1
)
ADH3wt 2 ± 0.6 115 ± 5 58 000 27 ± 8 2400 ± 400 90 000
ADH3-Arg115Ser 270 ± 70 135 ± 10 500 700 ± 250
a
2750 ± 800
a
3900
a
ADH3-Arg115Lys 150 ± 50 150 ± 5 1000 140 ± 30 3400 ± 400 24 000
a
Values were determined under nonsaturating conditions.
1250 J. J. Hedberg et al. (Eur. J. Biochem. 270) Ó FEBS 2003
(Resource Q, 1 mL; Amersham Biosciences) with a linear
gradient from 10 m
M
to 450 m
M
ammonium bicarbonate
pH 7.5 with 20% acetonitrile, flow rate 2 mLÆmin
)1
,using
an A
¨
KTA HPLC system (Amersham Biosciences). The

elution profile was monitored at 214 nm. The identity of
the isolated products were confirmed by MS as described
above. The concentration of the collected glutathione
sulfinamide was determined spectrophotometrically at
214 nm using standardized GSH solutions as references.
Results
Steady-state kinetics
Steady-state kinetic constants for HMGSH, GSNO,
NAD
+
and NADH were determined for wild-type and
mutated forms of ADH3 (Tables 1 and 2). In addition,
kinetic constants for NADP
+
and NADPH were deter-
mined for the wild-type enzyme (Table 2). Human ADH3
was found to readily reduce GSNO with NADH as
coenzyme yielding a k
cat
/K
m
of 90 000 min
)1
Æm
M
)1
.Reduc-
tive capacity for GSNO was drastically reduced by the
substitution of Arg115 for Ser or Lys. The lowered catalytic
efficiencies were due to 25- and 5-fold increases in K

m
values
for the Arg115Ser and Arg115Lys mutants, respectively.
NAD
+
and NADH were by far the best coenzymes for
HMGSH oxidation and GSNO reduction, with K
m
values
in the micromolar range. The catalytic capacity for
HMGSH oxidation and GSNO reduction using NADP
+
and NADPH as coenzymes were two to three orders of
magnitudes lower than with NAD
+
and NADH as
coenzymes. For the two mutants, kinetic constants for
NAD
+
and NADH were unchanged as compared to the
wild-type enzyme. K
m
values for HMGSH and GSNO were
not significantly changed when using NADP
+
and
NADPH (data not shown). GSSG, GSH or methylgluta-
thione did not inhibit ADH3 significantly up to 4 m
M
(data

not shown).
Electrospray MS/MS and isolation of the products
To determine the structure of the product generated by
ADH3 catalysed reduction of GSNO, nano-electrospray
mass spectra of the reaction mixture with and without the
addition of enzyme were recorded (Fig. 1). In addition, the
same experiment was performed with and without GSH.
This allowed the source of contaminant ions to be identified
(data not shown). After an incubation period of 1 h
following addition of enzyme, the peaks corresponding to
[GSNO]H
+
(m/z 337) and [NADH]H
+
(m/z 710) were
found to decrease in intensity, and new peaks at m/z 339 and
708 to appear. In the reaction mixtures containing GSH, the
peak at m/z 708 corresponded to NAD
+
and the peak at
m/z 613 corresponded to protonated GSSG. As GSH levels
in the initial reaction mixture were increased the peak
corresponding to protonated GSSG increased in intensity.
When a 15-fold excess of GSH was applied, the peak at m/z
339 and the peak corresponding to protonated GSSG were
approximately equal in intensity. In addition, a correspond-
ing reaction mixture without enzyme was analysed yielding
similar results with respect to GSSG formation. The identity
of the compound responsible for the peak at m/z 339 was
Table 2. Steady state kinetic constants for the coenzymes NAD

+
, NADH, NADP
+
and NADPH at pH 7.5, catalyzed by wild-type and mutant forms of ADH3. K
m
and k
cat
values were determined from initial
velocity experiments at 25 °C with substrate concentrations varied over a 10-fold range. Values are given as means (± SE) of three independent experiments. HMGSH and GSNO concentrations were fixed
at 0.3 m
M
(i.e. 1 m
M
GSH and 1 m
M
formaldehyde [33]) and 0.5 m
M
, respectively. ND, not determined.
Enzyme
NAD
+
NADH NADP
+
NADPH
K
m
(l
M
)
k

cat
(min
)1
)
k
cat
/K
m
(min
)1
Æm
M
)1
)
K
m
(l
M
)
k
cat
(min
)1
)
k
cat
/K
m
(min
)1

Æm
M
)1
)
K
m
(l
M
)
k
cat
(min
)1
)
k
cat
/K
m
(min
)1
Æm
M
)1
)
K
m
(l
M
)
k

cat
(min
)1
)
k
cat
/K
m
(min
)1
Æm
M
)1
)
ADH3wt 7 ± 1.5 90 ± 5 13 000 8 ± 2.5 2700 ± 400 340 000 3600 ± 550 110 ± 16 30 >5000
a
ND 600
a
ADH3-Arg115Ser 11 ± 2.8 80 ± 7 7300 10 ± 6 1800 ± 800 180 000 ND ND ND
ADH3-Arg115Lys 11 ± 3 110 ± 20 10 000 8 ± 2 1900 ± 400 240 000 ND ND ND
a
Values were determined under nonsaturating conditions.
Ó FEBS 2003 GSNO reduction by human alcohol dehydrogenase 3 (Eur. J. Biochem. 270) 1251
determined by MS/MS. The MS/MS spectra of the peaks at
m/z 337, [GSNO]H
+
,andm/z 339 are displayed in Fig. 2.
The spectra show many similarities and are characteristic of
glutathione conjugates (Fig. 2 [34]). However, the spectra
differ in that the ion at m/z 307, specific for the glutathione

conjugation moiety is absent in MS/MS spectrum of the ion
at m/z 339. In its place a new fragment at m/z 322 is
observed.
Fig. 2. Mass fragmentation analyses of substrate and products obtained from the enzymatic reaction. Tandem mass spectrometric fragmentation
spectra of [GSNO]H
+
m/z 337 (upper), and protonated major product m/z 339 (lower). The schematic structures are depicted to the right with some
fragmentation breaks. For a more detailed description see Yang et al. [34].
Fig. 1. Nano-electrospray mass spectra of reaction mixtures. The upper spectrum shows components without the addition of enzyme (background)
and the lower spectrum shows components after the addition of ADH3. Peaks corresponding to NAD
+
, [NADH]H
+
,[GS
i
]H
+
, [GSNO]H
+
,
[GSSG]H
+
and the major product are indicated. The peaks at m/z 359 and 361 correspond to Na
+
adducts of GSNO and the Ômajor productÕ,
those at m/z 666/688 and 664/686 are associated with NADH and NAD
+
, respectively, while those at m/z 279, 280 and 316 are small contaminants.
Peaks at m/z from 524 to 587 detected in the mass spectra with enzyme are also present in the spectra without enzyme although at much lower
intensities.

1252 J. J. Hedberg et al. (Eur. J. Biochem. 270) Ó FEBS 2003
In an effort to understand the fragmentation spectrum of
the ion at m/z 339, reference spectra were recorded of
protonated glutathione sulfinic acid (generated through
acidification of the reaction mixture after incubation as
described [16]), glutathione sulfonic acid and HMGSH
(Fig. 3). From the spectra of these glutathione conjugates it
is evident that oxidation of the glutathione sulfur effects
fragmentation in such a way that the ion at m/z 307 is no
longer observed in the MS/MS spectrum. This, along with a
comparison with reference spectra of other glutathione
conjugates recorded on this instrument [34], demonstrates
that the ion at m/z 339 has the structure of a glutathione
sulfinamide (Scheme 1).
The glutathione sulfinamide was purified using an anion
exchange column (Fig. 4A) and the identity and purity was
confirmed by MS (Fig. 4B). In addition to the peak
corresponding to glutathione sulfinamide additional peaks
corresponding to NAD
+
, NADH, glutathione sulfinic acid
and GSSG were resolved. Notably, glutathione sulfinic acid,
GSSG, GSNO, NAD
+
or NADH (m/z 340, 613, 337, 708
and 710, respectively) were not found to be present in the
glutathione sulfinamide fraction, i.e. peak 3 in Fig. 4A. The
glutathione sulfinamide was not a substrate for ADH3 using
NAD
+

as coenzyme nor did it inhibit octanol or HMGSH
oxidation by ADH3 to any extent at concentrations up to
100 l
M
(data not shown). During the HPLC purification,
glutathione sulfinic acid was detected (Fig. 4). However, this
compound was never observed during the initial reaction
analysis as described above. Probably this compound is
spontaneously formed during the exceptionally long incu-
bation used to reach equilibrium. The formation of this
glutathione sulfinic acid can be enhanced by addition of acid
as described [16].
Discussion
Human ADH3 readily catalyses the reduction of GSNO.
The two glutathione conjugates HMGSH and GSNO are
by far the best substrates identified for ADH3. Further, the
catalytic efficiency for GSNO reduction is almost twofold
higher than for HMGSH oxidation with the result that
GSNO is the best substrate for the human ADH3 yet
identified. Notably, GSNO concentrations have been
reported to reach micromolar levels under certain circum-
stances [8,13]. These observations, together with the fact that
ADH3 is expressed in all tissues, lend support to the notion
that ADH3 may serve as a GSNO metabolizer in humans.
It has been suggested however, that due to the ÔnormallyÕ
low NADH/NAD
+
ratio in the cytoplasm [19], reductive
reactions exerted by ADHs are not favourable [20].
Analysis of the major product by electrospray tandem

MS demonstrated that in addition to GSSG, the other
Fig. 3. Tandem mass spectrometric fragmentation spectra of protonated glutathione sulfinic acid m/z 340 (upper) and protonated HMGSH m/z 338
(lower). The schematic structures are depicted to the right with some fragmentation breaks. For a more detailed description see Yang et al. [34].
Scheme 1. Transformation of GSNO to glutathione sulfinamide by
ADH3.
Ó FEBS 2003 GSNO reduction by human alcohol dehydrogenase 3 (Eur. J. Biochem. 270) 1253
major product is glutathione sulfinamide (Scheme 1). This
finding is in line with the previously suggested but not
confirmed structure of Jensen and coworkers [16]. The
isolated glutathione sulfinamide did not serve as substrate or
inhibitor for ADH3. Other sulfinamides have been shown to
be formed spontaneously by rearrangement of the corres-
ponding semimercaptale [35], a mechanism where the S¼O
oxygen originates from the solvent. It is most likely that
the isolated glutathione sulfinamide is also formed via a
semimercaptale intermediate (Scheme 1). Hence, the over-
all reduction of GSNO catalysed by ADH3 is clearly an
irreversible reaction. With these observations taken into
account, the reaction may therefore be forced in the
reductive direction thereby ÔoverridingÕ the influence of a
low cytoplasmic NADH/NAD
+
ratio and lending support
to the physiological relevance.
There are circumstances under which the NADH/NAD
+
ratio is significantly changed. For instance, during ethanol
intake increased levels of enzyme coenzyme complexes in rat
hepatocytes have been observed [36] and ethanol consump-
tion has also been shown to increase the output of lactate

from liver resulting in an increased NADH/NAD
+
ratio in
kidney [37]. It is possible that such a change may also occur
in other tissues, i.e. where GSNO is produced, and thereby
influence its metabolism. Naturally, such a change in redox
potential would presumably also influence HMGSH meta-
bolism. Similar reasoning has included other ADHs and
their metabolism of various endogenous substrates, e.g.
certain serotonin metabolites, during ethanol intake [23].
The observation of GSSG as a product is in line with
findings of previous investigators. However, for the bacter-
ial ADH3, Liu et al. do not report any detection of
glutathione sulfinamide but propose that the intermediate
semimercaptale reacts with NADH to form glutathione
amine, which is subsequently oxidized to GSSG by an
additional GSH [17]. We do not detect any glutathione
amine for the human ADH3-catalysed reaction. Moreover,
Jensen et al. proposes that for the mammalian enzyme,
GSSG is only a minor product [16]. We find that when
excess GSH is added, approximately equal amounts of
GSSG and glutathione sulfinamide are formed, and so both
products are probably physiologically relevant. Of note is
that GSSG is one of the major products spontaneously
formed from GSH and GSNO [14] and so it is conceivable
that the GSSG formed is not generated from the enzyme
catalysed reaction but rather is formed spontaneously
during the incubation. This was also confirmed in a control
experiment without enzyme.
Fig. 4. Isolation and analysis of reaction

products. (A) After incubating the ADH3-
catalysed GSNO reduction for 2 h at 37 °C
(see Materials and methods), the components
were resolved on a strong anion exchange
column (Resource Q) using 10 m
M
ammo-
nium bicarbonate pH 7.5, 20% acetonitrile as
the initial solvent and 1
M
ammonium bicar-
bonate pH 7.5, 20% acetonitrile as the elution
solvent (buffer B). The elution positions (1)
NAD
+
(2) NADH (3) glutathione sulfina-
mide (4) glutathione sulfinic acid and (5)
GSSG are indicated by arrows. (B) MS of the
fraction corresponding to peak 3 in (A) dem-
onstrating the identity and purity of the
glutathione sulfinamide. Notably, no con-
taminants of GSSG, GSNO, NAD
+
or
NADH (m/z 613, 337, 708 and 710, respect-
ively) were detected.
1254 J. J. Hedberg et al. (Eur. J. Biochem. 270) Ó FEBS 2003
The effect of substituting Arg115 by Ser or Lys in ADH3
has previously been investigated [31]. For HMGSH, an
exact positioning in the active site is essential for efficient

catalysis. The substrate binding is accomplished in part by a
charge interaction between Arg115 and the glycine carb-
oxylate group in the GSH moiety. However, both mutants
studied stress the requirement for exact positioning of
interacting groups in the substrate and the polypeptide
chain. Like the results for HMGSH, GSNO reduction was
impaired by the replacement of Arg115 by both Ser and Lys
due to increased K
m
values. These findings show that
GSNO and HMGSH bind in similar, if not identical,
manners in the ADH3 active site cleft. Moreover, HMGSH,
GSNO, GSH and methylglutathione show extensive struc-
tural similarities. Still, GSH or methylglutathione did not
inhibit the enzyme, which further illustrates the exactness
with which the HMGSH or GSNO molecule interacts with
the substrate pocket.
Addition of millimolar concentrations of NADPH to
cytosol fractions of rat liver has been reported to increase
the rate of GSNO disappearance [38]. NADPH could
indeed be utilized as coenzyme, but with three orders of
magnitude lower efficiency, primarily due to higher K
m
as
compared to NADH (Table 2). NADPH concentrations
have been estimated to be in the range of 50–100 l
M
in the
hepatocyte cytosol and even lower in other tissues, including
heart and brain [39], indicating that NADPH is unlikely to

be utilized in vivo by ADH3 as coenzyme for GSNO
reduction.
In conclusion, human ADH3 shows high catalytic
capacity for GSNO reduction with NADH as coenzyme.
Mutational analysis demonstrates that GSNO binds in a
similar manner to HMGSH to ADH3. As the major
products formed are GSSG and glutathione sulfinamide,
neither of which function as substrate or inhibitor for the
enzyme, it is now clear that the overall reaction is
irreversible. We propose that human ADH3 may serve as
a GSNO metabolizer in vivo.
Acknowledgements
We thank B. Agerberth and M. Tollin for the use of equipment and
valuable help with HPLC. This work was supported by the Alcohol
Research Council of the Swedish Alcohol Retailing Monopoly, the
Swedish Match and Karolinska Institutet.
References
1. Moncada, S. (1994) Nitric oxide. J. Hypertens. Suppl. 12, S35–S39.
2. IARC (1986) Tobacco Smoking. International Agency for
Research on Cancer, Lyon.
3. Graedel, T. E. (1988) Ambient levels of anthropogenic emissions
and their atmospheric transformation products. In Air Pollution,
the Automobile and Public Health (Watson, A.Y., Bates, R.R. &
Kennedy, D., eds), pp. 133–160. Health Effects Institute, Wash-
ington DC.
4. Stamler, J.S. (1994) Redox signaling: nitrosylation and related
target interactions of nitric oxide. Cell 78, 931–936.
5. Lipton, A.J., Johnson, M.A., Macdonald, T., Lieberman, M.W.,
Gozal, D. & Gaston, B. (2001) S-Nitrosothiols signal the venti-
latory response to hypoxia. Nature 413, 171–174.

6. Ignarro, L.J., Lippton, H., Edwards, J.C., Baricos, W.H., Hyman,
A.L., Kadowitz, P.J. & Gruetter, C.A. (1981) Mechanism of
vascular smooth muscle relaxation by organic nitrates, nitrites,
nitroprusside and nitric oxide: evidence for the involvement of
S-nitrosothiols as active intermediates. J. Pharmacol. Exp. Ther.
218, 739–749.
7. Stamler, J.S., Jaraki, O., Osborne, J., Simon, D.I., Keaney, J.,
Vita, J., Singel, D., Valeri, C.R. & Loscalzo, J. (1992) Nitric oxide
circulates in mammalian plasma primarily as an S-nitroso adduct
of serum albumin. Proc. Natl Acad. Sci. USA 89, 7674–7677.
8. Chatterjee, S., Noack, H., Possel, H. & Wolf, G. (2000) Induction
of nitric oxide synthesis lowers intracellular glutathione in
microglia of primary glial cultures. Glia 29, 98–101.
9. Kluge,I.,Gutteck-Amsler,U.,Zollinger,M.&Do,K.Q.(1997)
S-nitrosoglutathione in rat cerebellum: identification and quanti-
fication by liquid chromatography-mass spectrometry. J. Neuro-
chem. 69, 2599–2607.
10. Kharitonov, V.G., Sundquist, A.R. & Sharma, V.S. (1995)
Kinetics of nitrosation of thiols by nitric oxide in the presence of
oxygen. J. Biol. Chem. 270, 28158–28164.
11. Gow, A.J., Buerk, D.G. & Ischiropoulos, H. (1997) A novel
reaction mechanism for the formation of S-nitrosothiol in vivo.
J. Biol. Chem. 272, 2841–2845.
12. Tsikas, D., Sandmann, J., Denker, K. & Fro
¨
lich, J.C. (2000) Is
S-nitrosoglutathione formed in nitric oxide synthase incubates?
FEBS Lett. 483, 83–85.
13. Gaston, B., Reilly, J., Drazen, J.M., Fackler, J., Ramdev, P.,
Arnelle, D., Mullins, M.E., Sugarbaker, D.J., Chee, C., Singel,

D.J., Loscalzo, J. & Stamler, J.S. (1993) Endogenous nitrogen
oxides and bronchodilator S-nitrosothiols in human airways.
Proc. Natl Acad. Sci. USA 90, 10957–10961.
14. Singh, S.P., Wishnok, J.S., Keshive, M., Deen, W.M. & Tan-
nenbaum, S.R. (1996) The chemistry of the S-nitrosoglutathione/
glutathione system. Proc.NatlAcad.Sci.USA93, 14428–14433.
15. Wong, P.S., Hyun, J., Fukuto, J.M., Shirota, F.N.,
DeMaster, E.G., Shoeman, D.W. & Nagasawa, H.T. (1998)
Reaction between S-nitrosothiols and thiols: generation of
nitroxyl (HNO) and subsequent chemistry. Biochemistry 37, 5362–
5371.
16. Jensen, D.E., Belka, G.K. & Du Bois, G.C. (1998) S-nitroso-
glutathione is a substrate for rat alcohol dehydrogenase class III
isoenzyme. Biochem. J. 331, 659–668.
17. Liu,L.,Hausladen,A.,Zeng,M.,Que,L.,Heitman,J.&Stamler,
J.S. (2001) A metabolic enzyme for S-nitosothiol conserved from
bacteria to humans. Nature 410, 490–494.
18. Sakamoto, A., Ueda, M. & Morikawa, H. (2002) Arabidopsis
glutathione-dependent formaldehyde dehydrogenase is an
S-nitrosoglutathione reductase. FEBS Lett. 515, 20–24.
19. Bu
¨
cher,T.(1970)The State of the DPN System in Liver (Sund, H.,
eds), pp. 439. Springer-Verlag, Berlin.
20. Svensson, S., Some, M., Lundsjo
¨
, A., Helander, A., Cronholm, T.
&Ho
¨
o

¨
g, J O. (1999) Activities of human alcohol dehydrogenases
in the metabolic pathways of ethanol and serotonin. Eur. J. Bio-
chem. 262, 324–329.
21. Duester, G., Farre
´
s, J., Felder, M.R., Holmes, R.S., Ho
¨
o
¨
g, J O.,
Pare
´
s,X.,Plapp,B.V.,Yin,S.J.&Jo
¨
rnvall, H. (1999)
Recommended nomenclature for the vertebrate alcohol dehydro-
genase gene family. Biochem. Pharmacol. 58, 389–395.
22. Edenberg, H.J. & Bosron, W.F. (1997) Alcohol dehydrogenases.
In Comprehensive Toxicology (Guengerich, F. P., eds) pp. 119–
131. Pergamon Press, Inc., New York.
23. Ho
¨
o
¨
g, J O., Hedberg, J.J., Stro
¨
mberg, P. & Svensson, S. (2001)
Mammalian alcohol dehydrogenase – functional and structural
implications. J. Biomed. Sci. 8, 71–76.

24. Uotila, L. & Koivusalo, M. (1974) Formaldehyde dehydrogenase
from human liver. Purification, properties, and evidence for the
formation of glutathione thiol esters by the enzyme. J. Biol. Chem.
249, 7653–7663.
Ó FEBS 2003 GSNO reduction by human alcohol dehydrogenase 3 (Eur. J. Biochem. 270) 1255
25. Deltour, L., Foglio, M.H. & Duester, G. (1999) Metabolic defi-
ciencies in alcohol dehydrogenase Adh1, Adh3, and Adh4 null
mutant mice – overlapping roles of Adh1 and Adh4 in ethanol
clearance and metabolism of retinol to retinoic acid. J. Biol. Chem.
274, 16796–16801.
26. Hedberg, J.J., Ho
¨
o
¨
g, J O., Nilsson, J.A., Xi, Z., Elfwing, A
˚
.&
Grafstro
¨
m, R.C. (2000) Expression of alcohol dehydrogenase 3 in
tissue and cultured cells from human oral mucosa. Am. J. Pathol.
157, 1745–1755.
27. Danielsson, O., Shafqat, J., Estonius, M. & Jo
¨
rnvall, H. (1994)
Alcohol dehydrogenase class III contrasted to class I. Charac-
terization of the cyclostome enzyme, the existence of multiple
forms as for the human enzyme, and distant cross-species
hybridization. Eur. J. Biochem. 225, 1081–1088.
28.Estonius,M.,Svensson,S.&Ho

¨
o
¨
g, J O. (1996) Alcohol
dehydrogenase in human tissues: localisation of transcripts
coding for five classes of the enzyme. FEBS Lett. 397, 338–342.
29. Uotila, L. & Koivusalo, M. (1997) Expression of formaldehyde
dehydrogenase and S-formylglutathione hydrolase activities in
different rat tissues. Adv. Exp. Med. Biol. 414, 365–371.
30. Molotkov, A., Fan, X., Deltour, L., Foglio, M.H., Martras, S.,
Farre
´
s, J., Pare
´
s, X. & Duester, G. (2002) Stimulation of retinoic
acid production and growth by ubiquitously expressed alcohol
dehydrogenase Adh3. Proc. Natl Acad. Sci. USA 8, 5337–5342.
31. Hedberg, J.J., Stro
¨
mberg, P. & Ho
¨
o
¨
g, J O. (1998) An attempt to
transform class characteristics within the alcohol dehydrogenase
family. FEBS Lett. 436, 67–70.
32. Bradford, M.M. (1976) A rapid and sensitive method for the
quantitation of microgram quantities of protein utilizing the
principle of protein-dye binding. Anal. Biochem. 72, 248–254.
33. Sanghani, P.C., Stone, C.L., Ray, B.D., Pindel, E.V., Hurley, T.D.

& Bosron, W.F. (2000) Kinetic mechanism of human glutathione-
dependent formaldehyde dehydrogenase. Biochemistry 39, 10720–
10729.
34. Yang, Y., Rafter, J., Gustafsson, J A
˚
., Sjo
¨
vall, J. & Griffiths, W.J.
(1997) Differentiation of isomeric mercapturic acid pathway
metabolites of benzo[a]pyrene. Eur. Mass. Spec. 3, 396–399.
35. Kazanis, S. & McClelland, R.A. (1992) Electrophilic intermediate
in the reaction of glutathione and nitrosarenes. J. Am. Chem. Soc.
114, 3052–3059.
36. Cronholm, T. (1987) Effect of ethanol on the redox state of the
coenzyme bound to alcohol dehydrogenase studied in isolated
hepatocytes. Biochem. J. 248, 567–572.
37. Cherrik, G.R. & Leevy, C.M. (1965) The effect of ethanol meta-
bolism on levels of oxidized and reduced nicotinamide-adenine
dinucleotide in liver, kidney and heart. Biochem. Biophys. Acta
107, 29–37.
38. Jensen, D.E. & Belka, G.K. (1997) Enzymic denitrosation of 1,3-
dimethyl-2-cyano-1-nitrosoguanidine in rat liver cytosol and the
fate of the immediate product S-nitrosoglutathione. Biochem.
Pharmacol. 53, 1279–1295.
39. Sies, H. (1982) Nicotinamide Nucleotide Compartmentation. In
Metabolic Compartmentation (Sies, H., ed.), pp. 205–231.
Academic Press Ltd, London.
1256 J. J. Hedberg et al. (Eur. J. Biochem. 270) Ó FEBS 2003