 |
Introduction |
Aldose reductase
(EC 1.1.1.21) is a monomeric enzyme with a molecular mass of
approximately 35 kDa that utilizes NADPH to reduce a wide range
of aromatic and aliphatic aldehydes to their corresponding alcohols.
Aldose reductase, along with aldehyde reductase (EC 1.1.1.2) and
carbonyl reductases are members of a family of aldo-keto reductases
(Turner and Flynn, 1982
; Wermuth, 1985). Unlike aldehyde reductase
(Mano et al., 1961), aldose reductase predominantly catalyzes the
reduction of aldehydes to alcohols. However, recent studies report that
aldose reductase can also oxidize some alcohol substrates (Srivastava
et al., 1984) and display activity corresponding to that of a
dihydrodiol dehydrogenase (Matsuura et al., 1987).
Animals fed naphthalene quickly develop cataracts. This
naphthalene-induced cataract has been widely utilized as a model for human senile cataract (Rossa and Pau, 1988). Ingested naphthalene is
first oxidized in liver by CYP1A1 to an epoxide that is subsequently hydrolyzed to naphthalene-1,2-dihydrodiol (ND)1
by epoxide hydrolase (van Bladeren et al., 1984; Yang et al., 1985;
Buckpitt et al., 1987). ND is further metabolized by dihydrodiol dehydrogenase (EC 1.3.1.20) to 1,2-dihydroxynaphthalene, which is
eventually autoxidized to 1,2-naphthoquinone (NQ) (Smithgall et al.,
1988). Because NQ is highly reactive and quickly forms covalent bonds
to various cellular thiols such as glutathaione, cystein, and protein
thiols (Smithgall et al., 1988), the formation of NQ in the lens is
considered the basic mechanism for the formation of naphthalene-induced
cataracts (van Heyningen and Pirie, 1967; van Heyningen, 1979; Xu et
al., 1992a). The lens protein modification by naphtoquinone is also
considered as a model for lens protein modifications, such as disulfide
bond formation, induced by various oxidative insults. Similar protein
modifications that are mediated by xanthurenic acid have been reported
in senile cataracts (Malina and Martin, 1996).
Interest in naphthalene cataracts has recently been sparked by reports
that aldose reductase inhibitors (ARI) can prevent cataract formation
in naphthalene-fed rats (Hockwin et al., 1984-1985; Tao et al., 1991;
Xu et al., 1992a). Because ARI have been observed to prevent only
cataracts induced by ND but not by NQ in lens culture, it has been
suggested that ARI prevent naphthalene cataracts by inhibiting the
conversion of ND to NQ (Xu et al., 1992b). Based on a previous report
(Sato, 1993) that dihydrodiol dehydrogenase is absent in rat lenses and
that aldose reductase appears to be the only enzyme in the rat lens
that can utilize ND as a substrate, it has been speculated that
cataract formation in naphthalene-fed rats resulted from the aldose
reductase catalyzed formation of NQ. The present study characterized
the metabolites of ND generated by treatment with rat lens aldose
reductase and has confirmed that NQ is generated from ND by aldose reductase.
| |
Materials and Methods |
Chemicals.
All chemicals used were reagent grade. The trans-ND was a
gift from Alcon Laboratories (Ft. Worth, TX). NQ, oxindole, and dimethylsulfoxide (DMSO)-d6 were obtained from
Aldrich Chemical Co. (Milwaukee, WI). 
-NAD and NADP were
purchased from Sigma (St. Louis, MO). The aldose reductase inhibitors
AL1576 and tolrestat were gifts from Alcon Laboratories and
Wyerth-Ayest Research Inc. (Princeton, NJ), respectively. Microcon-3
filters were purchased from Amicon (Beverly, MA). Preparative thin
layer chromatography (TLC) plates (ADSORBOSIL) were purchased from
Alltech (Deerfield, IL).
Rat Lens Aldose Reductase (RLAR) and Human Muscle Aldose
Reductase (HMAR).
RLAR expressed in Escherichia coli was prepared as
previously described (Old et al., 1990). HMAR was commercially obtained from Wako Chemical USA (Richmond, VA).
Dehydrogenase Assay of Aldose Reductase.
The reaction mixture (1 ml) contained 1 mM ND, 2 mM NAD, and either 158 pmol of RLAR or 167 pmol of HMAR in either 0.1 M potassium phosphate
buffer, pH 7.5 or 0.1 M glycine buffer, pH 9.0. The reaction was
initiated by addition of enzyme, and dehydrogenase activity was
spectrophotometrically monitored by following the increase of either
NADH or NADPH at 340 nm for 4 min with a Shimadzu spectrophotometer
(Shimadzu Corp., Kyoto, Japan).
In Vitro Incubation of Aldose Reductase with either ND or NQ.
For high-performance liquid chromatography (HPLC) analysis, the
reaction mixture (100 µl) consisted of 131 pmol of recombinant rat
lens aldose reductase (RLAR), 7 mM 2-mercaptoethanol, 2 mM NAD, 20 mM
phosphate buffer pH 7.5, and 1 mM of either ND or NQ. For the aldose
reductase inhibition studies, 10 µM of either the AL1576 or the
tolrestat were also included. After the incubation at 25°C for 30 min, the reaction was terminated by addition of 200 µl ice-cold
methanol. The mixture was then filtered through a Microcon-3 filter,
and the filtrate was analyzed by HPLC.
For liquid chromatography/mass spectrometry (LC/MS) analysis,
the incubation was conducted in the same conditions as described above
except that 10 mM NAD, 2.5 nmol of RLAR, and either 5 mM ND or 2.5 mM
NQ was used.
To investigate the stoichiometry between the oxidation of ND and the
consumption of NAD, 400 pmol RLAR was incubated for 3.3 h with 7 mM 2-mercaptoethanol, 1 mM NAD, and 50 µM ND in 20 mM phosphate
buffer, pH 7.5. The reaction was terminated by addition of 200 µl of
ice-cold methanol, and 5 nmol of oxindole was added as an internal
standard. The amounts of ND consumed were determined by HPLC, whereas
the amounts of NADH formed were determined by the increase of
absorption at 340 nm (
= 6270 M
cm1).
Preparation of the 2-Mercaptoethanol Adduct of the Enzymatic
Oxidation Product of ND for NMR Analysis.
Ten µmol ND were incubated with 52 nmol of RLAR in a reaction mixture
(2 ml) containing 20 µmol of 2-mercaptoethanol and 20 µmol of NAD
in 25 mM phosphate buffer, pH 7.5. After incubation at 25°C for
18 h, the metabolites (product I) were extracted with ethyl
acetate and purified by preparative silica gel TLC using chloroform/methanol (3:1) as a mobile phase, as described by Smithgall et al. (1988). Product I readily autoxidized, in
CHCl3 overnight at room temperature, to a
purple-colored product (product II). Product II was purified by
preparative silica gel TLC using chloroform/methanol (3:1) as a mobile phase.
Preparation of the 2-Mercaptoethanol Adduct of 1,2-NQ for NMR
Analysis.
The 2-mercaptoethanol adduct of NQ was prepared as described by
Smithgall, et al. (1988). Briefly, 9.46 ml of 50 mM phosphate buffer pH
7 containing 35 µl of 2-mercaptoethanol and 8 mg of NQ in 0.5 ml of
acetonitrile was incubated for 18 h at 25°C. Products were
extracted with ethyl acetate and purified as described above.
HPLC.
HPLC was performed using the Pharmacia Smart System (Pharmacia Biotech
Inc., Piscataway, NJ) equipped with a µRPC C2/C18 PC 3.2/3
column. The flow rate was 0.25 ml/min with a water/methanol gradient,
ramping linearly from 5 to 25% methanol for 10 min, then to 80%
methanol at 20 min, with a 5-min hold at 20 min. The eluent was
monitored at 245 nm.
LC/MS.
LC/MS experiments were performed on an HP-5989 mass spectrometer
(Hewlett-Packard, Palo Alto, CA) interfaced with an HP-1050 HPLC via an
HP-59980 particle beam LC/MS, interfaced with an in-line UV detector.
The mass spectrometer was scanned from 100 to 300 m/z at 0.8 s/scan. The ion source temperature was 250°C. HPLC separations
were performed on an HP ODS Hypersil column (5 µm pore size, 100 × 2.1 mm). A water/methanol gradient was used as described above, at a
flow rate of 0.4 ml/min.
NMR Analysis.
Approximately 0.5 mg of the purified samples, dissolved in 120 µl of
DMSO-d6, was analyzed on a VXR-500S/Unity
spectrometer operating at 500 MHz for 1H NMR and
125 MHz for 13C NMR. Structures of the
metabolites were determined using homonuclear correlation spectroscopy
(COSY), heteronuclear multiple quantum coherence (HMQC) and
heteronuclear shift correlations via multiple bond connectivities
(HMBC). The HMBC was set up with a narrow F1 sweep width, a reduction
of the number of F1 increments to 256, a change in the long range delay
to 6 Hz, and the application of linear prediction to the processing of
the final spectrum.
Electron Ionization (EI)-MS.
EI-MS data were collected on a Finnigan TSQ-70 (Finnigan-MAT, San Jose,
CA). Introduction was via solid probe. The ion source temperature was
150°C. Full scan data were recorded from m/z 50 to 550 at 0.5 s/scan in electron ionization mode at 70 eV.
Infrared (IR) Analysis.
Fourier-transformed infrared spectra were obtained on samples in KBr
pellets with a Mattson Galaxy 5000 instrument from Mattson Instruments
(Madison, WI).
| |
Results |
Effect of pH on the Dehydrogenase Activity of Aldose Reductase.
Effect of pH on dehydrogenase activity of RLAR with ND as substrate was
examined in the pH range of 6.5 to 10 (Fig.
1). Like many other NAD-dependent
dehydrogenases, the dehydrogenase activity of aldose reductase was
higher under basic conditions, with a maximum at pH 9. Dehydrogenase
activity was observed to be 3- to 5-fold lower at physiological pH.
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 1.
Effect of pH on dehydrogenase activity of
RLAR with ND as substrate.
Twenty millimolar phosphate buffer was used for the pH range of 6.5 to
8.5, while 50 mM glycine buffer was used for pH 9 to 10.
|
|
Kinetic Properties of Aldose Reductase with NR as Substrate.
Kinetic analysis of dehydrogenase activity was conducted at pH 7.5 and
9 (Table 1). For RLAR, NAD was a better
cofactor than NADP, with Km values for NAD
2- to 4-fold lower than that with NADP. The
Vmax values were also higher, with NAD
rather than NADP as a cofactor.
Vmax/Km values
for RLAR, a marker for turnover rate, was 10- to 20-fold higher with
NAD than with NADP. Compared with RLAR, dehydrogenase activity of HMAR
with ND as substrate was much lower. At pH 7.5, HMAR displayed only
trace amounts of dehydrogenase activity with NAD as a cofactor, and at
pH 9, the Vmax/Km
of HMAR was less than 10% that of RLAR. Like the rat enzyme, the
Km of HMAR with NAD was smaller than with
NADP. However, unlike rat enzyme, the
Vmax/Km of HMAR
was slightly higher with NADP than with NAD.
1,2-Naphthoquinone Formation by Aldose Reductase In Vitro.
To investigate whether ND is metabolized to NQ by aldose reductase, ND
was incubated with RLAR in the presence of NAD and 2-mercaptoethanol.
In the absence of RLAR, the HPLC chromatogram of the incubation mixture
displayed a peak corresponding only to unmetabolized ND (Fig.
2A). Incubation of aldose reductase with ND clearly resulted in the formation of new metabolite peaks (Fig.
2B). The major metabolite peak of ND exactly corresponded to the major
metabolite peak obtained when NQ was incubated under identical
conditions (Fig. 2C). Addition of the aldose reductase inhibitors,
AL1576 and tolrestat into the incubation mixtures significantly reduced
the size of these peaks (Fig. 2; D and E).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2.
HPLC profile of metabolites of ND catalyzed
by RLAR.
A, represents the chromatogram obtained from the reaction mixture
containing 7 mM 2-mercaptoethanol, 2 mM NAD and 1 mM ND in 20 mM
phosphate buffer, pH 7.5, but not RLAR. B, chromatogram B represents
the complete mixture containing 2-mercaptoethanol, NAD, ND, and 131 pmol of RLAR in 20 mM phosphate buffer, pH 7.5. C, represents the same
complete mixture where NQ was substituted for ND. D and E,
chromatographs represent the complete reaction mixtures containing ND
as substrate and 1 µM aldose reductase inhibitors AL 1576 and
tolrestat, respectively. Incubation was carried out at 25°C for 30 min.
|
|
With this in vitro incubation system, some unidentified peaks at around
3, 4.5, and 15 min were occasionally observed. However, because the
appearance of these peaks was not consistent, the identification of
these peaks was not pursued in this study.
Stoichiometry between Oxidation of Napthalene-1,2-Dihydrodiol and
the Formation of NADH.
To estimate the number of NAD molecules required to oxidize one
molecule of ND by RLAR, the consumption of ND and the formation of NADH
was quantified after incubation of the enzyme with ND and NAD. The
decrease of the substrate, ND, concentration and the increase of NADH
concentration were 19.2 ±1.6 and 19.3 ± 0.60 µM (n = 6),
respectively, indicating that one molecule of NAD is utilized to
oxidize one molecule of ND.
LC/MS.
The major product formed from the enzymatic oxidation of ND was
isolated by HPLC and analyzed by particle beam LC/MS (Fig. 3). The fragmentation pattern observed
for the metabolite of ND was identical with that of NQ. Both spectra
contained a molecular ion of m/z 236, corresponding to
4-(2-hydroxyethylsulfanyl)-1,2-naphthalenediol. Loss of the ethanol
side chain yielded a fragment ion at m/z 191. Further loss
of a hydroxyl group yielded an ion at m/z 174. Loss of the
2-hydroxyethysulfanyl group from the parent ion gave fragment ions at
m/z 77 and 159. A fragmentation associated with cleavages between C-1 and C-2 and C-4 and C-4a yielded ions at m/z 105 and 131 at m/z. A cleavage between ring A and B yielded ions
at m/z 76 and 160.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 3.
LC/MS spectra of the 2-mercaptoethanol
adducts of 1,2-naphthoquinone formed either from the oxidation of ND
catalyzed by aldose reductase in the presence of 2-mercaptoethanol (A)
or from the same reaction with o-naphthoquinone as substrate
(B).
Structure of the parent compound (m/z 236) and probable
fragment ions are shown at the top.
|
|
NMR.
The metabolite of ND formed by treatment with RLAR was analyzed by both
500 MHz 1H NMR and 125 MHz
13C NMR (Figs. 4
and 5; Table
2). Both 1H and
13C spectra obtained for the ND metabolite were
identical with those of the mercaptoethanol adducts of NQ. The
1H NMR spectra showed triplets at 3.27 and 3.75 ppm, corresponding to the methylene protons of the mercaptoethanol
substituents at C-9 and C-10, as well as a triplet at 5.19, which are
due to a hydroxyl proton at C-10. The aromatic region of each
1H NMR spectrum showed the presence of four
aromatic protons at 7.67, 7.80, 7.88, and 8.02 ppm, corresponding to
protons at C-7, C-6, C-5, and C-8, respectively, as well as a sharp
singlet resonance at 6.49 ppm corresponding to an isolated vinyl proton
at C-3. The 13C NMR spectrum showed the presence
of two dicarbonyls at resonances 175.71 and 178.78 ppm, corresponding
to C-2 and C-1. The 13C chemical shifts of the
aromatic carbons at C-5, C-6, C-7, C-8, C-8a, and C-4a are 124.9, 135.05, 131.09, 128.24, 130.48, and 133.21, respectively, while the
13C chemical shifts of the C-9 and C-10 carbons
are 33.98 and 58.32 ppm, respectively. The 13C
chemical shift of 157.7 ppm indicated the attachment of an electron withdrawing group, -SR. The site of attachment of the side chain at
C-4 was identified by HMBC. The 13C chemical
shift of the C-4 carbon (157.71 ppm) was coupled to the proton at C-9,
C-5, and C-3, while the 13C chemical shift of the
carbon at C-3 was coupled to the proton at C-3 (6.50 ppm). The
1H NMR and 13C NMR results
supported the hypothesis that the product formed from ND by
treatment with aldose reductase and from NQ are both 4-(2-hydroxyethylsulfanyl)-1,2-dihydro-1,2-naphthalenedione (product I). IR spectroscopy further confirmed the presence of carbonyls in
product I, and this IR spectrum was identical with that of the
2-mercaptoethanol adduct of NQ (Fig.
6). The infrared spectrum of product I
displayed a band at 1643 cm1, which is
characteristic of the stretching frequency of a carbonyl of aryl or
unsaturated ketones and similar to that (1670 cm1) reported for NQ (Das et al., 1965).
The broad band at 3433 cm1 reflected the O-H
stretching frequency of alcohols, and the band at 1050 cm1 was consistent with the asymmetric C-C-O
stretching vibration of primary alcohols. The band at 3070 cm1 was consistent with aromatic C-H stretching
or olefinic stretching frequencies, and the bands at 2930 cm1 and 2850 cm1
corresponded to asymmetric and symmetric C-H stretching frequencies.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 4.
500 MHz 1H NMR
spectra of the 2-mercaptoethanol adducts formed either from the
metabolite of ND (A) or from the reaction of o-naphthoquinone (B) in
the presence of 2-mercaptoethanol.
Chemical shifts are given in ppm relative to tetramethylsilane.
Impurity, DMSO and water peaks are marked X, D, and W, respectively.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 5.
125 MHz 13C NMR
spectra of the 2-mercaptoethanol adducts formed either from the
metabolite of ND (A) or from the reaction of NQ (B) in the presence of
2-mercaptoethanol.
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 6.
Fourier transforms IR spectra of the
2-mercaptoethanol adducts formed from either the enzymatic oxidation of
ND (A) or from the reaction of NQ (B) in the presence of
2-mercaptoethanol.
|
|
The metabolite (product I) from ND was readily autoxidized to form a
purple-colored compound (product II). This oxidation product was also
analyzed by 1H and 13C NMR
(Fig 7; Table
3). The 1H NMR
spectrum showed the presence of triplets at 3.22 and 4.65 ppm,
corresponding to the methylene protons of the mercaptoethanol substituents at C-3 and C-2. Four aromatic protons were present at
7.57, 7.71, 7.75, and 7.90 ppm corresponding to protons at C-7, C-5,
C-6, and C-8, respectively. The 13C NMR spectrum
showed the presence of the dicarbonyl of the o-quinone as
two resonances at 175.34 and 177.34 ppm. Six aromatic carbon chemical
shifts were observed between 122 and 136 ppm, and the two methylene
carbons at C-2 and C-3 were present at 67.63 and 23.22 ppm,
respectively. Carbon 13 chemical shift data indicated that the product
was a three-ring system. Evidence for ring closure was based on 1) the
13C chemical shifts of the carbons corresponding
to C-4a and C-10a at 111.33 and 157.93 ppm, respectively, indicating
the attachment of an electron withdrawing group at C-10a; 2) the loss
of the C-10a hydrogen and C-2 hydroxyl hydrogen in the
1H NMR spectrum of the product; 3) the loss of
the C-10a protonated carbon (C-3 in product I) at 120.04 ppm, as
indicated by HMQC and the appearance of a new quaternary carbon at
C-10a (note that the chemical shift of the C-4b carbon, as well as
other carbons, remained unchanged); and 4) the HMBC data showing
connectivity between C-2 and C-10a and between C-3 and C-4a.
1H NMR and 13C NMR
identified the autoxidation product of product I as
2,3-dihydro-1-oxa-4-thia-9,10-phenanthrenedione. Furthermore, the
presence of carbonyls in product II was confirmed by IR spectroscopy.
The IR spectrum of product II gave a band at 1649 cm1 (Fig. 8)
corresponding to a carbonyl stretching frequency for aryl ketones. The
broad O-H stretching band observed at 3433 cm1
in product I diminished to the level of residual
H2O. The band at 1092 cm1
was consistent with C-O-C stretching vibration of ethers.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 7.
500 MHz 1H and 125 MHz 13C NMR spectra of the autoxidation product
(product II) of product I.
Chemical shifts are given in ppm relative to tetramethylsilane.
Impurity, DMSO, and water peaks are marked X, D, and W, respectively.
|
|
MS.
The metabolite (product I) from ND and its oxidized product (product
II) were further analyzed by EI-MS (Fig.
9). Because a conversion between
carbonyl and aromatic hydroxyls by oxidation/reduction can occur in an
EI-MS (Das et al., 1965; Aplin and Pike, 1966), ions at m/z
232 and 236, in addition to the molecular ion (m/z 234) of
product I, were observed by EI-MS. An ion at 234 was also observed for
product II in addition to the molecular ion, m/z 232.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 9.
EI mass spectra of the 2-mercaptoethanol
adducts (product I) formed from the enzymatic oxidation of ND in the
presence of 2-mercaptoethanol (A) and the autoxidation product (product
II) of product I (B).
|
|
| |
Discussion |
Aldose reductase (EC 1.1.1.21) is an NADPH-dependent enzyme that
reduces various aldehyde substrates. Aldose reductase also possesses
dihydrodiol dehydrogenase (EC 1.3.1.20) activity (Matsuura et al.,
1987) and can utilize ND as substrate (Sato, 1993). The present data
confirms the identity of the metabolite formed from the oxidation of ND
catalyzed by aldose reductase as 1,2-naphthoquinone (NQ).
Evidence supporting the identity of the metabolite of ND by treatment
with aldose reductase as 1,2-naphthoquinone (NQ) comes from incubation
studies of aldose reductase with ND, NAD, and 2-mercaptoethanol.
Incubation resulted in the formation of a new metabolite that was
reduced by the presence of aldose reductase inhibitors in the
incubation medium. HPLC analysis indicated that this major peak
corresponded to the major peak obtained when NQ was substituted for ND
in the enzymatic reaction. Also, the LC/MS spectrum of the ND
metabolite peak is identical with that of the major peak from NQ.
Catechol formation from dihydrodiols has been detected in the oxidation
of 3,5-cyclohexadiene-1,2-diol catalyzed by rat liver dihydrodiol
dehydrogenase (Ayengar et al., 1959). However, catechols are extremely
unstable in solutions. 4-(2-hydroxyethylsulfanyl)-1,2-naphthalenediol is also readily autoxidized to
4-(2-hydroxyethylsulfanyl)-1,2-dihydro-1,2-naphthalenedione (product I). Although
4-(2-hydroxyethylsulfanyl)-1,2-dihydro-1,2-naphthalenedione is either
reduced to the corresponding catechol or further oxidized during EI-MS,
only the catechol was observed by particle beam LC/MS. Therefore, it is
difficult to distinguish between carbonyls and aromatic hydroxyl groups
by MS because of the potential conversion of the two groups by
oxidation/reduction (Das et al., 1965; Aplin and Pike, 1966). Further
autoxidation of
4-(2-hydroxyethylsulfanyl)-1,2-dihydro-1,2-naphthalenedione was
observed to yield 2,3-dihydro-1-oxa-4-thia-9,10-phenanthrenedione, which was also reduced to
4-(2-hydroxyethylsulfanyl)-1,2-dihydroxy-1,2-naphthalene by EI-MS and
particle beam LC/MS (data not shown).
The conversions of dihydrodiols to catechols are generally catalyzed by
dihydrodiol dehydrogenase. Dihydrodiol dehydrogenase utilizes NADP to
catalyze a 2e oxidation of ND to yield the
corresponding ketol, which is autoxidized to NQ (Smithgall et al.,
1986). Because aldose reductase utilizes only one molecule of NAD to
consume one molecule of ND, aldose reductase, like dihydrodiol
dehydrogenase, appears to catalyze the
2eoxidation of ND to a ketol rather than a
4e oxidation to directly form NQ. However, RLAR
displayed a different preference for cofactor from dihydrodiol
dehydrogenase. Although aldose reductase requires NADPH rather than
NADH to reduce aldehyde substrates to their corresponding alcohols
(Hayman and Kinoshita, 1964; Wermuth, 1985), the oxidation of ND
by rat lens aldose reductase is better with NAD than with NADP as
cofactor. In contrast to stereospecific dehydrogenase activity of
dihydrodiol dehydrogenase, the oxidation of ND by aldose reductase may
not be stereospecific. The consumption of more than 50% ND when low
concentrations of racemic ND were incubated for prolonged periods (data
not shown) suggests that aldose reductase recognizes both the
S,S- and R,R-stereoisomers of ND.
The formation of NQ is the key process in naphthalene-induced cataract
formation. Because naphthalene dihydrodiol dehydrogenase activity in
rat lens is associated with aldose reductase (Sato, 1993), aldose
reductase likely produces NQ, the metabolite responsible for
naphthalene cataract in rat. Recently, Lee and Chung (1998) have
reported that lenses obtained from transgenic mice that overexpress aldose reductase quickly developed cataracts when incubated with ND in
vitro. This observation confirms the involvement of aldose reductase in
naphthalene cataract formation in rats. The present data confirming the
identity of the metabolite of ND formed by treatment with aldose
reductase as NQ supports the role of aldose reductase in naphthalene
cataract formation.
Although the evidence strongly suggests the importance of aldose
reductase in the formation of naphthalene cataract in rats, this must
be limited to rats at present. The homologies between the animal and
human aldose reductases are generally high (more than 80%). Both rat
and human enzymes display the similar activities with the common
aldehyde substrates (Sato et al., 1995). However, there are some
differences. For example, 17
-hydroxysteroid is a good substrate for
both human and bovine aldose reductases, but not for rat enzyme. ND may
be one of them. Compared with rat lens enzyme, dehydrogenase activity
with ND of human aldose reductase was much lower. Moreover, the human
lens contains dihydrodiol dehydrogenase that is lacking in rat lens
(Hara et al., 1991). It is unlikely that aldose reductase plays a more
dominant role than dihydrodiol dehydrogenase in the formation of
catechols from dihydrodiols in the human lens.
Figure 10 summarizes the mechanism
for the aldose reductase catalyzed oxidation of ND. Aldose reductase
catalyzes a 2e oxidation of ND to a corresponding ketol,
which is subsequently autoxidized to NQ. In the presence of
2-mercaptoethanol, NQ reacts with 2-mercaptoethanol to form
4-(2-hydroxyethylsulfanyl)-1,2-naphthalenediol. This catechol is
further autoxidized to the corresponding quinone (product I), which
undergoes an intramolecular cyclization to yield another catechol. This
catechol is further autoxidized to yield
2,3-tetrahydro-1-oxa-4-thia-9,10-phenanthrenedione (product II), which
again may be further autoxidized in the presence of 2-mercaptoehanol.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 10.
Formation of the 2-mercaptoethanol NQ
adducts initiated by the oxidation of ND catalyzed by aldose
reductase.
|
|
Whereas catechols are reducing agents, quinones are potent autoxidants.
Therefore, the formation of catechols from aromatic dihydrodiols by
aldose reductase can lead to the generation of H2O2 and quinones, which
may deplete ascorbate and cellular thiols (glutathione, cysteine,
and protein thiols) in the lens (Rees and Pirie, 1967; van Heyningen
and Pirie, 1967; van Heyningen, 1976). In naphthalene-fed animals, the
depletion of ascorbate or reduced form of glutathione also accelerates
protein modification by NQ to form naphthalene cataracts.
We thank SAIC Frederick (a Division of Science Applications
International Corporation, National Cancer Institute-Frederick Cancer
Research and Development Center, Frederick, MD) for obtaining the NMR,
IR, and EI-MS spectra.
Abbreviations used are:
ND, naphthalene-1,2-dihydrodiol;
NQ, naphthoquinone;
DMSO, dimethylsulfoxide;
TLC, thin layer chromatography;
RLAR, recombinant
rat lens aldose reductase;
HMAR, recombinant human muscle aldose
reductase;
HPLC, high-performance liquid chromatography;
LC, liquid
chromatography;
ARI, aldose reductase inhibitors;
EI, electron
ionization;
MS, mass spectrometry;
COSY, homonuclear correlation
spectroscopy;
HMQC, heteronuclear multiple quantum coherence;
HMBC, heteronuclear shift correlations via multiple bond connectivities;
IR, infrared.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
