Synthesis of a new seleninic acid anhydride and mechanistic studies into its glutathione peroxidase activity

Article abstract of DOI:10.1002/chem.200800694

Starting from low toxic salicyloylglycine, a new seleninic acid anhydride 7 that lacks Se...N or Se...O non-bonded interactions was synthesized. This compound exhibits a four-fold higher glutathione peroxidase-like (GPx-like) activity than ebselen and inhibits plant and mammalian 12/15-lipoxygenases at lower micromolar concentrations. Because of these pharmacological properties, 7 may constitute a new lead compound for the development of anti-inflammatory low-molecular-weight seleno-organic compounds. Analyzing the redox products of 7 with glutathione (GSH) and tBuOOH, we identified three potential catalytic cycles (A, B, C) of GPx-like activity that are interconnected by key metabolites. To study the relative contribution of these cycles to the catalytic activity, we prepared selected reaction intermediates and found that the activity of seleninic acid anhydride 7 and of the corresponding diselenide 11 and selenol 14 compounds were in the same range. In contrast, the GPx-like activity of monoselenide 9 was more than one order of magnitude lower. These data suggested that cycles A and B may constitute the major routes of GPx-like activity of 7, whereas cycle C may not significantly contribute to catalysis.

Full text of DOI:10.1002/chem.200800694

DOI: 10.1002/chem.200800694
Synthesis of a New Seleninic Acid Anhydride and Mechanistic Studies into Its
Glutathione Peroxidase Activity
[a]
Sun-Chol Yu,[b] Astrid Borchert,[a] Hartmut Kuhn,*[a] and IgorIvanov
Abstract: Starting from low toxic sali-
cyloylglycine, a new seleninic acid an-
hydride 7 that lacks Se···N or Se···O
non-bonded interactions was synthe-
sized. This compound exhibits a four-
fold higher glutathione peroxidase-like
(GPx-like) activity than ebselen and in-
hibits plant and mammalian 12/15-lip-
oxygenases at lower micromolar con-
centrations. Because of these pharma-
cological properties, 7 may constitute a
new lead compound for the develop-
ment of anti-inflammatory low-molecu-
lar-weight seleno-organic compounds.
Analyzing the redox products of 7 with
glutathione (GSH) and tBuOOH, we
identified three potential catalytic
cycles (A, B, C) of GPx-like activity
that are interconnected by key metabo-
lites. To study the relative contribution
of these cycles to the catalytic activity,
we prepared selected reaction inter-
mediates and found that the activity of
seleninic acid anhydride 7 and of the
corresponding diselenide 11 and sele-
nol 14 compounds were in the same
range. In contrast, the GPx-like activity
of monoselenide 9 was more than one
order of magnitude lower. These data
suggested that cycles A and B may
constitute the major routes of GPx-like
activity of 7, whereas cycle C may not
significantly contribute to catalysis.
Keywords: enzymes · inflammation ·
reaction mechanism · redox reac-
tions · selenium
Introduction
been developed as potential drugs and the seleno-organic
compound ebselen (1) has been well characterized with re-
spect to its pharmacological profile.[4] In addition, there
Selenium is an essential trace element that plays an impor-
tant role in the metabolism of pro- and eukaryotic cells.[1,2]
It constitutes a functional element of selenium-containing
enzymes, such as glutathione peroxidase (GPx), iodothyro-
nine deiodinase (ID) and thioredoxin reductase (TrxR),
which have been implicated in antioxidative defense, iodine
homeostasis, and regulation of gene expression.[1,2] Seleni-
um-containing GPx-isoforms are members of the antioxida-
tive network in mammalian cells capable of detoxifying per-
oxides at the expense of reduced glutathione, but they also
exhibit more specific functions.[2,3] Because of the beneficial
activities of GPx-isoforms, nonenzymatic GPx mimics have
[a] Dr. A. Borchert, Prof. Dr. H. Kuhn, Dr. I. Ivanov
Institute of Biochemistry
were numerous attempts to develop other seleno GPx
mimics, such as ebselen analogues,[5,6] camphor-derived sele-
namides (2),[7] dicyclo-dextrinyl diselenides,[8] diferrocenyl
diselenides (3),[9] cyclic seleninate esters (4),[10] spirodioxase-
lenanonane (5),[11] and cyclic selenenate ester (6).[12] Seleno-
organic compounds with dendrimer cores[13] exhibit a high
GPx-like activity, but owing to their low water solubility bio-
logical applications may be limited.
University Medicine Berlin-CharitØ
Monbijoustrasse 2, 10117 Berlin (Germany)
Fax : (+49)30-450528905
E-mail: hartmut.kuehn@charite.de
[b] Dr. S.-C. Yu
Department of Pharmacy, Pyongyang Medical University
Ryonhwa-dong, Central district
Pyongyang (DPR of Korea)
For the catalytic activity of most of the seleno GPx
mimics, Se···N or Se···O non-bonded interactions are re-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200800694.
7066
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7066 – 7071

FULL PAPER
quired and these interactions may be established by reduc-
When normalized to the selenium content, a twofold higher
activity was determined. Moreover, we found that 7 inhibit-
ed mammalian (IC50 =8.6 for 7 versus 0.56 mm for ebselen)
and plant 15-lipoxygenases (IC50 =9.3 for 7 versus 7.3 mm for
ebselen), which suggests potential anti-inflammatory activi-
ties (Figure 1).
[14]
À
À
tive cleavage of Se N or Se O bonds. Seleninic acid an-
hydrides form a well-characterized family of stable seleni-
À
um-containing compounds. They contain covalent Se O
bonds, which cannot be converted to Se···O non-bonded in-
teractions when reacting with glutathione (GSH). Thus,
direct GPx-like activity is rather unlikely. On the other
hand, seleninic acid anhydrides can be reduced to the corre-
sponding diselenides,[15] selenenic acids, or selenol deriva-
tives, which have been reported before to function as GPx
mimics. Thus, seleninic acid anhydrides were expected to ex-
hibit GPx-like activity, but the reaction mechanisms might
differ from those of other seleno-organic compounds.
Results and Discussion
We have synthesized the novel seleno-organic compound
salicyloylglycine seleninic acid anhydride (7) and explored
the complex mechanism of its GPx-like activity. The salicy-
loylglycine moiety was selected as the substituent for two
reasons: 1) It exhibits a low in vivo toxicity and 2) the anti-
inflammatory properties of ebselene analogues were im-
proved when amino acid derived functional groups were in-
troduced.[16]
Figure 1. Inhibition of the soybean LOX I by 7 (a) and ebselen (c).
The extent of enzyme inhibition (%) calculated for the different inhibitor
concentrations is plotted against the logarithm of the inhibitor concentra-
tion
Salicyloylglycine seleninic acid anhydride (7) was synthe-
sized (Scheme 1) by reacting salicyloylglycine with selenium
tetrachloride in THF at room temperature for 1 h. The
Since 7 constitutes a seleninic acid derivative, the catalytic
mechanism was expected to be different from that of ebse-
len and other seleno-organic compounds, in which selenium
is present as monoselenide, diselenide, or selenamide. To ex-
plore the sequence of reactions involved in the catalytic
cycle(s), 7 was reacted at a concentration of 10 mm with dif-
ferent amounts of reductant. To mimic biological conditions,
the reaction was carried out in 0.1m phosphate buffer
(pH 7.4) at physiological salt concentrations and reduced
GSH was employed as a reducing agent. The reaction prod-
ucts were quantified by RP-HPLC (Table 1) and their chem-
ical structures were confirmed by selenium analysis, MS, and
77Se NMR spectroscopy after a 10 min reaction period.
When 7 was incubated with GSH at an equivalent ratio of
1:0.5, most (96%) of the starting material 7 was recovered.
Scheme 1. a) SeCl4, THF, RT, 1 h; b) H2O, RT, 18 h.
major reaction product was precipitated by addition of an
excess of water and the precipitate was backwashed with
THF. The final reaction product was purified by silica-gel
column chromatography and the analytical data (elementary
analysis, selenium content, 1H-, 13C-, and 77Se NMR spec-
troscopy, HRMS, UV/Vis spectroscopy, and RP-HPLC) indi-
cated its chemical structure.
To compare the GPx-like activity of 7 with that of the
standard seleno-organic compound ebselen, the catalytic ac-
tivities of the two substances were assayed spectrophotomet-
rically.[17] Under our experimental conditions, 7 exhibited a
more than fourfold higher catalytic activity than ebselen
((9.98Æ0.50) for 7 versus (2.40Æ0.30) mmminÀ1 for ebselen).
Table 1. Product composition [%][a] of the GSH-dependent reduction of
7.
Ratio 7/GSH
7
8
9
10
11
13
1:0.5
1:1
1:2
1:3
1:4
96
41
1
–
–
1
11
–
–
–
1
7
1
1
1
5
24
31
30
24
–
–
–
15
18
18
13
40
77
37
29
30
38
18
25
22
22
24
1:5
1:20
–
–
–
–
[a] Determined on the basis of areas of RP-HPLC peaks (absorbance at
240 nm) considering the different molar extinction coefficients.
Chem. Eur. J. 2008, 14, 7066 – 7071
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7067

H. Kuhn et al.
Only small amounts of the corresponding monoselenoxide
8, monoselenide 9, selenenyl sulphide 10, and diselenide 11
derivatives were identified by RP-HPLC. When we in-
creased the GSH concentration to an equivalent ratio of 1:1,
the product mixture still contained large amounts of 7
(41%) and the major reaction product was identified as the
corresponding selenenyl sulphide derivative 10. Formation
of this product was confirmed by 77Se NMR spectroscopy
and the ratio of the signal intensities at d=1147.65 (7) and
467.36 ppm (10) was 1.0:1.4. In addition, smaller amounts of
the corresponding monoselenoxide 8 and monoselenide 9
derivatives were detected by RP-HPLC and MS (see the
Supporting Information).
oselenide 9, whereas the corresponding monoselenoxide 8
was hardly detectable any more.
After we had identified the reduction products of 7 at var-
ious GSH concentrations, we attempted to close the catalyt-
ic cycle by oxidizing the major reaction products 9, 10, and
11 by addition of peroxide. For this purpose, we first added
tBuOOH (final concentration 64 mm) to a 1:2 reaction mix-
ture of 7 and GSH, in which monoselenide 9 and selenenyl
sulphide 10 were the major reaction products. Analyzing the
product mixture 1 h after addition of tBuOOH, we observed
significant formation of salicyloylglycine seleninic acid anhy-
dride (7) as indicated by RP-HPLC and 77Se NMR spectros-
copy (signal at d=1148.07 ppm). The signal at d=
466.12 ppm in the spectrum was due to unconverted selene-
nylsulfide 10, which also showed up in RP-HPLC (see the
Supporting Information). In addition, significant amounts of
the corresponding monoselenoxide 8 were found, whereas
the monoselenide derivative 9 had disappeared. It should be
stressed that neither 7 nor 8 were detectable in the reaction
mixture before tBuOOH was added (Table 1). Oxidation of
selenenyl sulfide 10 by tBuOOH under our experimental
conditions appears to be a slow process since after 1 h only
about 30% of 10 was converted to 7 (catalytic cycle A in
Scheme 2). However, after 36 h we observed complete oxi-
When the GSH concentration was increased to an 1:2
equivalent ratio, we observed complete conversion of the
starting material 7. In this case, we identified selenenyl sul-
phide 10 as the major reaction product and these data were
confirmed by the appearance of a highly abundant signal at
d=466.84 ppm in the 77Se NMR spectrum. In addition,
smaller amounts of monoselenide 9 and diselenide 11 deriv-
atives (see the Supporting Information) were detected. In
contrast, the corresponding monoselenoxide 8 was not de-
tectable any more. Under these experimental conditions, we
also detected small amounts of the selenenic acid derivative
12 by MS as indicated by the molecular ion at m/z:
289.5129. When 7 was incubated with three equivalents of
GSH monoselenide 9, selenenyl sulphide 10, and diselenide
11 derivatives were identified as major reaction products by
RP-HPLC (see the Supporting Information). Under these
conditions, we also isolated an HPLC fraction containing di-
glutathione selenoxide 13 as indicated by the molecular ion
at m/z: 704.6 in the mass spectrum. A previous report[15] on
the reduction of phenyl seleninic acid with three equivalents
of reduced thiols suggested the formation of 1 mol of the
corresponding phenyl selenenyl sulphide, which was then
slowly converted to the diphenyl diselenide derivative.
Under our experimental conditions, the selenenyl sulphide
derivative 10 was quickly formed even at a 1:1 ratio of the
reactants (7 and GSH) and its relative abundance was in-
creased to a 1:2 ratio. The corresponding diselenide 11 was
also present in the reaction mixtures (1:2 and 1:3) and we
did not see any slow time-dependent increase in its relative
abundance. When we further increased the concentration of
the reductant in the reaction mixture (1:4 to 1:20), we did
not observe major alterations in product composition (see
the Supporting Information). In these samples, the selenenyl
sulphide 10 and diselenide 11 derivatives were identified as
major reaction products, whereas the corresponding mono-
selenide 9 and diglutathione selenoxide 13 were found in
smaller amounts. Taken together, the results shown in
Table 1 indicate that the salicyloylglycine derivatives of
monoselenoxide 8 and of monoselenide 9 are preferentially
formed at lower GSH concentrations. However, when the
thiol content was increased, corresponding selenenyl sul-
phide 10, diselenide 11, and diglutathione selenoxide 13 de-
rivatives became more abundant. Under these experimental
conditions, we still observed significant amounts of the mon-
Scheme 2. Potential catalytic cycles of the GPx-like catalytic activity of 7.
dation of 10. In contrast, the monoselenide derivative 9 was
quickly converted to monoselenoxide 8 (catalytic cycle C in
Scheme 2) as indicated by the complete disappearance of 9
after a 1 h reaction period. Addition of tBuOOH to the mix-
ture of reduction products of 7 closed the catalytic cycle of
the GPx-like activity of 7. To initiate the next cycle, we
added an excess of reduced glutathione (final concentration
130 mm) to the mixture of oxidation products and observed
the formation of 9, 10, and 11 (see the Supporting Informa-
tion). Taken together, our data suggests the existence of
three catalytic subcycles (A, B, and C in Scheme 2) for the
7068
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7066 – 7071

A New Seleninic Acid Anhydride
FULL PAPER
GPx-like activity of 7, which are interconnected by the key
metabolites 7, 8, and 10.
nide 9 when reacted with NaBH4 or GSH. Inversely, the
monoselenide derivative 9 is oxidized to the corresponding
monoselenoxide 8 when reacted with an excess tBuOOH
(see the Supporting Information). Thus, in principle, cycle C
may be involved in the GPx-like activity of 7.
Selenol derivatives of seleno-organic compounds have
previously been implicated in GPx-like reactions,[18,19] but
we did not obtain major evidence for the formation of sali-
cyloylglycine selenol 14 when the GPx-like reaction was car-
ried out in the presence of oxygen (Table 1). However,
when 7 was reduced by a molar excess of NaBH4 and the re-
action mixture was kept under an argon atmosphere, we ob-
served dominant formation of selenol 14 (Figure 2). This
After having identified the major reaction products
formed during the interaction of salicyloylglycine seleninic
acid anhydride 7 with GSH and tBuOOH, we aimed at in-
vestigating, which of these chemical entities might be in-
volved in GPx-like activity. To do so, we first prepared se-
lected key metabolites (monoselenide 9, diselenide 11, sele-
nol 14) and compared their GPx-like activities with that of
the anhydride 7. For this purpose, 7 was reduced with a 10-
mollar excess of NaBH4 under aerobic or anaerobic condi-
tions and the intermediates were isolated by RP-HPLC. To
quantify the different compounds, we determined the seleni-
um content of the HPLC fractions by atomic absorption
spectrometry and we found that the GPx-like activities of 7,
11, and 14 were in the same range (Table 2), whereas that of
Table 2. GPx-like activities of reaction intermediates.[a]
Catalyst
n [mmminÀ1
]
7
9
11
14
9.98Æ0.50
0.24Æ0.05
5.46Æ0.63
9.69Æ0.90
[a] Each measurement has been carried out in triplicate; a catalyst con-
centration of 5 mm was adjusted.
Figure 2. Partial RP-HPLC chromatograms of reduction products formed
from 7 after addition of an excess of NaBH4. After reduction, the sample
was kept under argon (prevention of re-oxygenation of the sample) for
about 16 h and thereafter under a normal atmosphere for 7 h.
the monoselenide 9 was much lower. Unfortunately, we did
not manage to isolate pure selenyl sulphide 10 since it parti-
ally decomposes to the corresponding diselenide 11 during
the preparation procedure. Taken together, these data sug-
gest that the catalytic activity of 7 involves cycles A and B,
whereas cycle C may not play a major role.
result was confirmed by 77Se NMR spectroscopic measure-
ments, in which we detected strong signals at d=12.71 and
14.17 ppm. When the reaction mixture was kept under
argon for up to 16 h (Figure 2), only small amounts of the
corresponding diselenide 11 were found. In contrast, after
re-equilibration with air or after addition of tBuOOH, the
selenol 14 was completely oxidized to the corresponding dis-
elenide 11. This oxidation was confirmed by the appearance
of a signal at d=521.08 ppm in the 77Se NMR spectrum.
Seleninic acid anhydride 7 is partly converted at low GSH
concentrations to the corresponding monoselenoxide 8 as in-
dicated in Table 1. We hypothesized that this compound
may be further reduced to the monoselenide 9, which can
subsequently be oxidized by tBuOOH to yield 8. This rever-
sible redox reaction (cycle C) between salicyloylglycine
monoselenoxide 8 and the corresponding monoselenide 9
has not been implicated before in GPx-like activity of
seleno-organic compounds. To obtain direct experimental
evidence for the existence of that redox cycle, we followed
the interconversion of chemically synthesized 8 and 9 by
RP-HPLC. We found that the pure monoselenoxide deriva-
tive 8 was rapidly reduced to the corresponding monosele-
Despite our quantitative data on the relative catalytic ac-
tivities of major reaction intermediates, it was not possible
to reliably quantify the relative contribution of the different
catalytic subcycles A, B, and C to overall GPx-like activity
of 7. When we quantified by HPLC the metabolite pattern
during the time course of the GPx-like reaction (data not
shown) we mainly detected metabolites 10 and 11. In addi-
tion, small amounts of 8 were found suggesting minor im-
portance of subcycle C. These data are consistent with the
low catalytic efficiency of 9 (Table 2). On the other hand,
our observation that oxidation of 10 by tBuOOH yielding 7
was rather slow may point towards a concerted action of
cycles A and B and thus, catalysis may involve selenenic
acid 12, selenyl sulfide 10, and selenol 14, which have previ-
ously been implicated in the catalytic cascade of seleno-or-
ganic compounds. It should, however, be stressed that
steady-state metabolite concentrations and catalytic efficien-
cies per se do not mirror flux rates along the different cata-
lytic subcycles. To answer these questions, detailed kinetic
experimentation and mathematic modeling must be per-
formed.
Chem. Eur. J. 2008, 14, 7066 – 7071
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7069

H. Kuhn et al.
Data for 9: Yield: 1.42 mg (31%); 77Se NMR (76.20 MHz, [D6]DMSO):
d=466.84.08 ppm; HPLC: tR =35.8 min; NSI-MS: m/z: calcd for
C18H16N2O880Se: 467.2882 [M+ÀH]; found: 467.0012.
Conclusion
Salicyloylglycine seleninic acid anhydride 7 is a novel
seleno-organic compound that exhibits GPx-like activity in
the absence of Se···N or Se···O non-bonded interactions. Its
catalytic activity is fourfold higher than that of ebselen and
it inhibits lipoxygenases at lower micromolar concentrations.
The catalytic mechanism is more complex than that of previ-
ously investigated seleno-organic compounds and three po-
tential catalytic subcycles have been identified. Unfortunate-
ly, in vivo activities of seleninic acid anhydrides have never
been studied in detail and little is known about their phar-
macokinetics and systemic toxicology. However, the simple
synthetic procedure and the biochemical properties of 7 are
promising.
Data for 11: Yield: 2.05 mg (38%); 77Se NMR (76.20 MHz, [D6]DMSO):
d=521.08 ppm; HPLC: tR =37.3 min.
Preparation of selenol 14: An excess of NaBH4 (0.1 mmol) was added to
a solution of 7 (5.9 mg, 0.01 mmol) in 0.1m phosphate buffer (pH 7.4).
The mixture was kept under argon for 1 h and then 14 was purified by
RP-HPLC. The HPLC solvent was evaporated in vacuum and the solid
residue was reconstituted in DMSO under an argon atmosphere. White
solid; yield: 2.13 mg (78%); HPLC: tR =37.8 min; 77Se NMR
(76.20 MHz, [D6]DMSO): d=12.71, 14.17 ppm.
Measurements of glutathione peroxidase (PPx-like) acitivity: The gluta-
thione peroxidase activity (GPx-like activity) of the test compounds (eb-
selen, 7, 9, 11, 14) was assayed spectrophotometrically by measuring the
decrease in absorbance at 340 nm (coupled optical test). The catalytic re-
action was run at 378C in a 1 mL reaction mixture consisting of 100 mm
Tris-HCl buffer (pH 7.4) containing 5 mm ethylenediaminetetraacetic
acid, 0.1% Triton X-100, 3 mm GSH, 0.2 mm NADPH (nicotinamide ade-
nine dinucleotide phosphate), 1 U of glutathione reductase, and 5 mm of
the test compound. The assay sample was equilibrated for 10 min in the
absence of peroxide substrate and the GPx-like reaction was initiated by
addition of 0.5 mm tBuOOH. The time-dependent decrease in absorbance
at 340 nm (10 to 70 s after the addition of tBuOOH) was recorded by
using a Shimazu U-2102 PC UV/VIS spectrophotometer. During the
GPx-like reaction reduced GSH is oxidized and the resulting disulfide
(GSSG) is back-reduced by the glutathione reductase reaction consuming
stoichiometric amounts NADPH. Since NADPH exhibits a local absorb-
ance maximum at 340 nm, but its oxidized counterpart (NADP) does
not, NADPH oxidation can be quantified by measuring the decrease in
absorbance at 340 nm. A blank assay (solvent control) was run in the ab-
sence of catalysts and this control rate was subtracted. The rate of
NADPH oxidation was calculated by using a molar absorbance coeffi-
cient for NADPH of 6.22”103 mÀ1 cmÀ1. Measurements were carried out
in triplicate and ebselen was used as a reference compound. The test
compounds were added as DMSO solutions at a final concentration of
5 mm.
Experimental Section
General information: Commercial reagents salicyloylglycine (o-hydroxy
hippuric acid) (Merck) and selenium tetrachloride (Sigma-Aldrich) were
used as received. All solvents and reagents used were of extra pure grade
and purchased from Merck, Aldrich, or Roth (Germany). IR spectra
were recorded on a Nicolet Magma IR 750. Kinetic measurements were
performed on a Shimadzu UV-2102 spectrophotometer. H and 13C NMR
1
spectra were recorded on a Brucker Biospin AV 400 (1H: 400 MHz; 13C:
100.5 MHz) by using [D6]DMSO as a solvent. Chemical shift values (d)
are reported in ppm downfield from TMS (d=0.0 ppm) as internal stan-
dard. Data are reported as follows: chemical shift (d/ppm), multiplicity
(s=singlet, d=doublet, t=triplet, m=multiplet), integration, coupling
constant (Hz). 77Se NMR spectra were recorded by using
a JEOL
JNMLA 400 instrument (76.20 MHz) in [D6]DMSO-dimethylselenide as
external standard. HPLC analysis was carried out on a Shimadzu LC-
10Avp liquid chromatograph connected to SPD-10Advp UV detector.
RP-HPLC analysis was performed on a Nucleosil C18-column; 250/
4 mm, 5 mm particle size (Machery-Nagel, Düren, Germany) by using a
linear gradient of MeOH in a water/MeOH solvent system (from 5 to
100%) with a flow rate of 1 mLminÀ1. HRMS was carried out on a Finni-
gan LTQST NSI.
Catalytic mechanism of the GPx-like activity of salicyloylglycine seleninic
acid anhydride (7): To investigate the catalytic mechanism of the GPx-
like activity of salicyloylglycine seleninic acid anhydride (7), we first ana-
lyzed the composition of the reduction products of 7 formed during the
reaction with different amounts of reduced GSH. Next, selected key me-
tabolites were prepared and re-oxidized with tBuOOH. Reaction prod-
ucts were analyzed by RP-HPLC on a Shimadzu LC10 HPLC system
that was connected to a SPD-M19AVP diode array detector. Analytes
were resolved on a CC 250/4 Nucleosil 120–5 C18 column (Macherey and
Nagel, Düren, Germany) with a linear water/methanol gradient contain-
ing 0.1% acetic acid. For analysis, the column was equilibrated for
15 min with a water/methanol mixture (95:5, by volume) containing
0.1% acetic acid. Aliquots (20–50 mL) of the reaction mixtures were in-
jected and the reaction products were separated by the following analyti-
cal profile: 5 min isocratic elution at a water/methanol ratio of 95:5
(0.1% acetic acid). This pre-elution phase was followed by a linear gradi-
ent elution of increasing methanol concentration starting at 5% and
reaching 100% after 30 min (0.1% acetic acid). This gradient elution
phase was followed by a 10 min isocratic elution at 100% methanol
(0.1% acetic acid). To start the next analysis, the column was re-equili-
brated at starting conditions (5% methanol in water containing 0.1%
acetic acid). Major eluting compounds were identified by retention times
in RP-HPLC, by their UV-spectral properties (diode array detector) and
NSI-MS.
Synthesis of salicyloylglycine seleninic acid anhydride (7): Selenium tetra-
chloride (1.107 g, 5 mmol) was added to a solution of salicyloylglycine
(0.975 g, 5 mmol) in THF (10 mL) and the mixture was stirred for 1 h at
room temperature. After this time, water (100 mL) was added to the re-
action mixture and the sample was stirred at room temperature for an
additional 18 h. The white precipitate was filtered, back-washed with
THF (3”30 mL), and air-dried at 308C to give salicyloylglycine seleninic
acid anhydride (7). Yield: 0.939 g (1.58 mmol); m.p. 175–1788C
(decomp.); 1H NMR (400 MHz, [D6]DMSO): d=3.99 (d, J=5.8 Hz, 4H;
CH2), 7.11 (d, J=8.0 Hz, 2H; 3-H), 7.81 (dd, J=10.0 Hz, 2H; 4-H), 8.32
(d, J=2.0 Hz, 2H; 6-H), 9.21 (t, J=11.0 Hz, 2H; NH), 12.63 ppm (s, 2H;
COOH); 13C NMR (100.5 MHz [D6]DMSO): d=041.31 (CH2), 116.41 (1-
C), 117.96 (3-C), 127.31 (6-C), 131.17 (4-C), 139.01 (5-C), 161.54 (2-C),
167.22 (C=O), 170.95 ppm (COOH); 77Se NMR (76.20 MHz): d=1182.08
([D6]DMSO), 1150.08 ppm (NaOH); IR (KBr): 3303, 3071, 2925, 2706,
2577, 1715, 1599, 1551, 1426, 1300, 1249, 1067, 831, 671, 545 cmÀ1; UV/Vis
(0.1n NaOH): lmax =227.20, 263.80 nm; NSI-MS: m/z: calcd for
C18H15N2O1180Se2: 594.9012 [M+ÀH]; found: 594.9009; elemental analysis
calcd (%) for C18H15N2O11Se2: C 36.38, H 2.71, N 4.71; found: C 36.51, H
2.34, N 4.45.
Inhibition of 15-lipoxygenase isoforms: The pure rabbit reticulocyte 12/
15-lipoxygenase (2 mL of a 1.7 mgmLÀ1 enzyme solution) or soybean
LOX-1 (Merck, 10 mm, 20 mL) were preincubated with the inhibitors
(final concentration ranging between 0.05–100 mm) either at 228C in 0.1m
phosphate buffer (pH 7.4) or at 378C in 0.1m borate buffer (pH 9.0) for
5 min. The reaction was started by the addition of linoleic acid as the
substrate (20 mL of a methanolic stock solution, 100 mm final concentra-
Preparation of monoselenide 9 and diselenide 11: An excess of NaBH4
(0.1 mmol) was added to a stirred solution of 7 (5.90 mg, 0.01 mmol) in
0.1m phosphate buffer (pH 7.4). The mixture was vigorously mixed for
36 h under air, acidified with 0.1 mL AcOH (5m) to pH 4.0, and purified
by RP-HPLC to yield 11 and 9.
7070
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 7066 – 7071

A New Seleninic Acid Anhydride
FULL PAPER
tion, assay volume 1 mL). The inhibitors were dissolved in DMSO. The
catalytic activity of the enzyme was assayed spectrophotometrically fol-
lowing the increase in absorbance at 235 nm over a time interval of 30 s.
[8] S. W. Lv, X. G. Wang, Y. Mu, T. Z. Zang, Y. T. Ji, J. Q. Liu, J. C.
Shen, G. M. Luo, FEBS J. 2007, 274, 3846–3854.
[9] G. Mugesh, A. Panda, H. B. Singh, N. S. Punekar, R. J. Butcher,
Chem. Commun. 1998, 2227–2228.
[10] T. G. Back, Z. Moussa, J. Am. Chem. Soc. 2002, 124, 12104–12105.
[11] T. G. Back, Z. Moussa, M. Parvez, Angew. Chem. 2004, 116, 1288–
1290; Angew. Chem. Int. Ed. 2004, 43, 1268–1270.
[12] S. S. Zade, H. B. Singh, R. J. Butcher, Angew. Chem. 2004, 116,
4613–4615; Angew. Chem. Int. Ed. 2004, 43, 4513–4515.
[13] X. Zhang, H. Xu, Z. Dong, Y. Wang, J. Liu, J. Shen, J. Am. Chem.
Soc. 2004, 126, 10556–10557.
Acknowledgements
Financial support from the Alexander von Humboldt Foundation (S.C.Y)
and the European Commission (LSHM-CT-2004-0050333 and MIFI-CT-
2006-021230) is acknowledged. The authors thank Drs. R. Zeisberg, H.
Poleschner, and M. von Loewis for recording spectra.
[14] C. A. Bayse, R. A. Baker, K. N. Ortwine, Inorg. Chim. Acta 2005,
358, 3849–3854.
[15] J. L. Kice, T. W. S. Lee, J. Am. Chem. Soc. 1978, 100, 5094–5102.
[16] A. Welter, C. Lambert, N. Dereu, A. Huether, E. Etschenb (A. Nat-
termann und Cie. GmbH), DE-B 3515274, 1985.
[17] P. Scheerer, A. Borchert, N. Krauss, H. Wesner, C. Gerth, W.
Hoehne, H. Kuhn, Biochemistry 2007, 46, 9041–9049.
[18] S. K. Tripathi, U. Patel, D. Roy, R. B. Sunoj, H. B. Singh, G. Wolmer-
schaeuser, R. J. Butcher, J. Org. Chem. 2005, 70, 9237–9247.
[19] Y. Saito, D. Umemoto, A. Matsunaga, T. Sato, M. Chikuma,
Biomed. Res. Trace Elem. 2006, 17, 423–426.
[1] L.V. Papp, J. Lu, A. Holmgren, K.K. Khanna, Antioxid. Redox Sig-
naling 2007, 9, 775–806.
[2] a) M. Birringer, S. Pilawa, L. Flohe, Nat. Prod. Rep. 2002, 19, 693–
718; b) P. Brenneisen, H. Steinbrenner, H. Sies, Mol Aspects Med.
2005, 26, 256–267.
[3] J. R. Drevet, Mol. Cell. Endocrinology 2006, 250, 70–79.
[4] a) M. Parnham, H. Sies, Expert Opin. Invest. Drugs 2000, 9, 607–
619; b) T. Schewe, Gen. Pharmacol. 1995, 26, 1153–1169.
[5] S. S. Zade, S. Panda, S. K. Tripathi, H. B. Singh, G. Wolmerschaeus-
er, Eur. J. Org. Chem. 2004, 3857–3864.
[6] K. P. Bhabak, G. Mugesh, Chem. Eur. J. 2007, 13, 4594–4601.
[7] T. G. Back, B. P. Dyck, J. Am. Chem. Soc. 1997, 119, 2079–2083.
Received: April 11, 2008
Published online: July 4, 2008
Chem. Eur. J. 2008, 14, 7066 – 7071
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7071