Bioactivation of chemopreventive selenocysteine Se-conjugates and related amino acids by amino acid oxidases novel route of metabolism of selenoamino acids

Article abstract of DOI:10.1021/tx000265r

Several selenocysteine Se-conjugates have been shown to possess potent chemopreventive activity in animal models for chemical carcinogenesis. As a mechanism of action, β-elimination reactions to form chemopreventive selenols, ammonia, and pyruvate has been proposed. The enzymes involved in these β-elimination reactions, however, have been partially elucidated. Next to cysteine conjugate β-lyases, as yet unidentified non-pyridoxal-5′-phosphate-dependent enzymes also appear to be involved in cytosolic β-elimination reactions. In the present study, it was investigated whether amino acid oxidases contribute to the bioactivation of selenocysteine Se-conjugates. Using purified L-amino acid oxidase from Crotalus adamanteus as a model enzyme, significant β-elimination activities were indeed observed upon incubation with Se-methylselenocysteine (K_m, 195 μM; k_cat, 48 min^-1), Se-allylselenocysteine (K_m, 608 μM; k_cat, 34 min^-1), Se-phenylselenocysteine (K_m, 107 μM; k_cat, 57 min^-1) and Se-benzylselenocysteine (K_m, 59 μM; k_cat, 13 min^-1). For all selenocysteine Se-conjugates tested, the rate of pyruvate formation was comparable to that of hydrogen peroxide, one of the products of oxidative deamination. The fact that addition of catalase did not alter pyruvate formation indicated that the β-elimination reaction observed was not mediated by selenoxidation/syn-elimination due to the hydrogen peroxide formed via the oxidative deamination pathway. Using D-amino acid oxidase from porcine kidney and D-SeCys conjugates similar results were obtained. To delineate whether mammalian L-amino acid oxidases are also able to catalyze β-elimination of selenocysteine Se-conjugates, rat renal cytosol was fractionated and screened for β-elimination and oxidative deamination activities. One of the fractions isolated displayed oxidative deamination activity with several amino acids and cysteine S-conjugates. With selenocysteine Se-conjugates as substrates, however, this fraction displayed both oxidative deamination and β-elimination activities, when incubated in the presence of aminoxyacetic acid to block contribution of pyridoxal-5′-phosphate-dependent enzymes. The potential significance of this novel bioactivation route for the chemopreventive activity of selenocysteine Se-conjugates is discussed.

Full text of DOI:10.1021/tx000265r

996
Chem. Res. Toxicol. 2001, 14, 996-1005
Bioa ctiva tion of Ch em op r even tive Selen ocystein e
Se-Con ju ga tes a n d Rela ted Am in o Acid s by Am in o Acid
Oxid a ses Novel Rou te of Meta bolism of Selen oa m in o
Acid s
Martijn Rooseboom, Nico P. E. Vermeulen,* Nathalie van Hemert, and
J an N. M. Commandeur
Leiden/ Amsterdam Center for Drug Research (LACDR), Division of Molecular Toxicology,
Department of Pharmacochemistry, Vrije Universiteit Amsterdam, De Boelelaan 1083,
1081 HV Amsterdam, The Netherlands
Received December 29, 2000
Several selenocysteine Se-conjugates have been shown to possess potent chemopreventive
activity in animal models for chemical carcinogenesis. As a mechanism of action, â-elimination
reactions to form chemopreventive selenols, ammonia, and pyruvate has been proposed. The
enzymes involved in these â-elimination reactions, however, have been partially elucidated.
Next to cysteine conjugate â-lyases, as yet unidentified non-pyridoxal-5′-phosphate-dependent
enzymes also appear to be involved in cytosolic â-elimination reactions. In the present study,
it was investigated whether amino acid oxidases contribute to the bioactivation of selenocysteine
Se-conjugates. Using purified L-amino acid oxidase from Crotalus adamanteus as a model
enzyme, significant â-elimination activities were indeed observed upon incubation with Se-
methylselenocysteine (Km, 195 µM; kcat, 48 min-1), Se-allylselenocysteine (Km, 608 µM; kcat, 34
min-1), Se-phenylselenocysteine (Km, 107 µM; kcat, 57 min-1) and Se-benzylselenocysteine (Km,
59 µM; kcat, 13 min-1). For all selenocysteine Se-conjugates tested, the rate of pyruvate formation
was comparable to that of hydrogen peroxide, one of the products of oxidative deamination.
The fact that addition of catalase did not alter pyruvate formation indicated that the
â-elimination reaction observed was not mediated by selenoxidation/syn-elimination due to
the hydrogen peroxide formed via the oxidative deamination pathway. Using D-amino acid
oxidase from porcine kidney and D-SeCys conjugates similar results were obtained. To delineate
whether mammalian L-amino acid oxidases are also able to catalyze â-elimination of
selenocysteine Se-conjugates, rat renal cytosol was fractionated and screened for â-elimination
and oxidative deamination activities. One of the fractions isolated displayed oxidative
deamination activity with several amino acids and cysteine S-conjugates. With selenocysteine
Se-conjugates as substrates, however, this fraction displayed both oxidative deamination and
â-elimination activities, when incubated in the presence of aminoxyacetic acid to block
contribution of pyridoxal-5′-phosphate-dependent enzymes. The potential significance of this
novel bioactivation route for the chemopreventive activity of selenocysteine Se-conjugates is
discussed.
ventive agent when compared to Se-methyl-D,L-seleno-
cysteine (3). Dietary administration of 2 ppm Se-allyl-
D,L-selenocysteine resulted in an almost 90% reduction
of methylnitrosourea-induced mammary tumors in rats,
making it one of the most potent chemopreventive agents
identified. Se-(n-propyl)-D,L-selenocysteine showed the
same potency as Se-methyl-D,L-selenocysteine.
As a mechanism of action of selenocysteine Se-
conjugates (SeCys1 conjugates), bioactivation by â-elim-
ination reactions has been proposed, resulting in the
formation of pyruvate, ammonia, and chemopreventive
selenols (1, 3, 4). The enzymes involved in these bioac-
tivation reactions, however, are only partially character-
In tr od u ction
Over the past decade, numerous studies have reported
that low molecular weight selenium compounds exert
significant chemopreventive activities in animal tumor
models (1, 2). Generation of monomethylated selenium
metabolites is believed to mediate the chemopreventive
activities of selenomethionine, selenobetaine, sodium
selenite, methyl selenocyanate, and Se-methyl-L-seleno-
cysteine. The observation that longer alkyl chain homo-
logues showed a higher chemopreventive potency than
methyl selenocyanate indicated that antitumor activity
is not restricted to monomethylated selenium species, in
principle allowing the design of more powerful anticancer
selenium compounds (1). Very recently, Se-allyl-D,L-
selenocysteine appeared to be a more potent chemopre-
1
Abbreviations: AAO, amino acid oxidase; AOAA, aminooxyacetic
acid; APCI, atmospheric pressure chemical ionization; CFTE-Cys, S-(2-
chloro-1,1,2-trifluoroethyl)-L-cysteine; FPLC, fast-protein liquid chro-
matography; HPA, p-hydroxyphenylacetic acid; HRP, horseradish
peroxidase; KMB, R-keto-γ-methiolbutyrate; PLP, pyridoxal-5′-phos-
phate; SeCys, selenocysteine Se.
* To whom correspondence should be addressed. Phone: +31 20
4447590. Fax: +31 20 4447610. E-mail: vermeule@chem.vu.nl.
10.1021/tx000265r CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/29/2001

Selenocysteine Se-Conjugates and Amino Acid Oxidases
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 997
conjugates, which subsequently would be followed by a
syn-elimination reaction as described above (Figure 1,
route B).
Next to oxidative deamination, however, it has been
demonstrated that for a limited number of substrates
AAOs may also catalyze â-eliminations (12). Using
â-chloro-D-alanine and purified hog kidney D-AAO it was
delineated that under anaerobic conditions only pyruvate
and chloride were formed. Under a 100% oxygen atmo-
sphere, almost exclusively chloropyruvate was formed,
whereas mixtures of pyruvate and chloropyruvate were
observed at intermediate oxygen levels. The total amount
of product (pyruvate and chloropyruvate) formed was
independent of the oxygen tension, which is indicative
for a common intermediate of the substrate and the
enzyme. Similar results were obtained when â-chloro-L-
alanine was incubated with L-AAO from Crotalus ada-
manteus (12). â-Elimination reactions by hog kidney
D-AAO were also observed with R-amino-â-chlorobu-
tyrate, â-bromo-D,L-alanine and â-cyano-D-alanine as
substrates (13-15). Therefore, in analogy with these
compounds a similar mechanism may be applicable in
case of SeCys conjugates as substrates (Figure 1, route
C). This pathway would result in formation of pyruvate,
ammonia and selenol, instead of the selenenic acid, which
is formed if â-elimination is mediated by the selenoxi-
dation pathway (Figure 1, route A).
In the present study, the bioactivation of chemopre-
ventive SeCys conjugates and structurally related cys-
teine S-conjugates via oxidative deamination and â-
elimination reactions was investigated, using purified
L-AAO from Crotalus adamanteus as a model enzyme.
Furthermore, it was delineated whether pyruvate formed
resulted from selenoxidation by hydrogen peroxide and/
or from direct â-elimination. Finally, the possible in-
volvement of AAO(s) in pyruvate formation from SeCys
conjugates by rat renal cytosol was investigated by
fractionation of rat renal cytosol by anion-exchange
chromatography.
F igu r e 1. Hypothetical mechanisms of oxidative deamination
and â-elimination of SeCys conjugates by amino acid oxidase.
Reaction A: oxidative deamination leading to R-keto acids,
ammonia and hydrogen peroxide. Reaction B: selenoxidation
by hydrogen peroxide and subsequent syn-elimination of selen-
oxide leading to selenenic acids and 2-aminoacrylic acid, as
described previously for flavin-containing monooxygenases (8).
2-aminoacrylic acid rapidly hydrolyses into pyruvate and am-
monia. Reaction C: direct â-elimination leading to a selenol and
2-aminoacrylic acid, the latter of which hydrolyses into pyruvate
and ammonia.
ized. Identification and localization of these enzymes may
help to elucidate in which tissues SeCys conjugates may
exert chemopreventive activities. Previously, high â-
elimination activities of SeCys conjugates were shown
in both rat and human renal cytosol (4, 5). Strong
inhibition by aminooxyacetic acid (AOAA) indicated the
involvement of pyridoxal-5′-phosphate (PLP)-dependent
enzymes, such as cysteine conjugate â-lyases (6). The
high activities observed upon incubation of SeCys con-
jugates with highly purified rat renal cysteine conjugate
â-lyase/glutamine transaminase K revealed that this
protein contributes significantly to the renal activity (7).
However, the fact that AOAA only partially inhibited
â-elimination of some SeCys conjugates, such as Se-
benzyl-L-selenocysteine, indicated that non-PLP-depend-
ent enzymes also appeared to be involved in the â-elim-
ination of SeCys conjugates in rat renal cytosol (6). As
an alternative bioactivation pathway, cytosolic selenoxi-
dation followed by syn-elimination was proposed (6). Very
recently, this hypothesis was confirmed by the observa-
tion that selenoxidation of SeCys conjugates by flavin-
containing monooxygenases in rat liver microsomes and
recombinantly expressed human flavin-containing mono-
oxygenases indeed resulted in spontaneous syn-elimina-
tion, leading to the formation of pyruvate, ammonia and
selenenic acids (8).
In the present study, it was studied whether amino
acid oxidases (AAOs) may also be involved in cytosolic
â-elimination reactions of SeCys conjugates. AAOs typi-
cally catalyze oxidative deamination reactions of amino
acids resulting in the formation of corresponding R-keto
acids and hydrogen peroxide (Figure 1, route A). Several
racemic SeCys conjugates, such as Se-carboxymethyl-D,L-
selenocysteine, Se-(1-carboxyethyl)-, Se-(1-carboxypro-
pyl)-, and Se-(2-carboxyethyl)-D,L-selenocysteine have
been reported to be substrates for L-AAO from Crotalus
adamanteus and D-aspartate oxidase from beef kidney
(9-11). In the present study, we investigated whether
the hydrogen peroxide generated during oxidative deami-
nation of SeCys conjugates is able to selenoxidize SeCys
Exp er im en ta l P r oced u r es
R ea gen t s. Aminooxyacetic acid (AOAA; carboxymethoxyl-
amine hemihydrochloride), diphenyldiselenide, and pyruvate
were from Aldrich (Beerse, Belgium). S-Benzyl-L-cysteine, L-
phenylalanine, D-phenylalanine, and o-phenylenediamine were
purchased from J anssen Chimica (Geel, Belgium). Catalase from
beef liver (25 000 units/mg) was obtained from Boehringer
(Mannheim, Germany). Hydrogen peroxide was from J . T. Baker
(Deventer, The Netherlands). D-AAO from porcine kidney (0.14
units/mg), L-AAO type I from Crotalus adamanteus (0.46 units/
mg), p-hydroxyphenylacetic acid (HPA), Se-methyl-L-selenocys-
teine, â-chloro-L-alanine, R-keto-γ-methiolbutyrate (ΚΜΒ), and
horseradish peroxidase (HRP) type I from horseradish (116
unints/mg) were purchased from Sigma (St. Louis, MO). S-(2-
Chloro-1,1,2-trifluoroethyl)-L-cysteine (CFTE-Cys) was synthe-
sized as described previously (16). SeCys conjugates were
prepared as described previously (4, 5). Methylphenylselenide
was prepared by reduction of diphenyldiselenide with NaBH4
followed by methylation with diazomethane (4). All other
chemicals were of the highest grade commercially available.
Biotr a n sfor m a tion of SeCys Con ju ga tes a n d Cystein e
S-Con ju ga tes by L-AAO. First, incubations were performed
to investigate the time-dependence of formation of pyruvate and
hydrogen peroxide at 0.5 mM SeCys conjugate. Second, enzyme
kinetic parameters were determined by a fixed incubation time
and with varying substrate concentrations. All incubations
described below were performed in triplicate and were also

998 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Rooseboom et al.
performed in absence of L-AAO to correct for nonenzymatic
reactions.
described for L-AAO (see above). However, in these experiments
a final concentration of 1 mg/mL D-AAO from porcine kidney
was used.
Pyruvate Formation. Incubations (200 µL) containing 0.5 mM
SeCys conjugate and 1 µg of L-AAO from Crotalus adamanteus
in 100 mM potassium phosphate buffer (pH 7.2) were incubated
at 37 °C (to dissolve the substrates 5 M sodium hydroxide was
added until pH 8-9, after dissolution 5 M HCl was added until
pH 7.2). After 0, 5, 10, 15, or 20 min, 20 µL of 0.3 M zinc sulfate
was added to terminate the reaction and the samples were
placed on ice. In case of enzyme kinetic analysis, substrates were
incubated for 20 min at 10 different concentrations ranging from
0 to 500 µΜ. Enzyme kinetic parameters were obtained using
Hanes-Woolf plots. After termination of the reactions by zinc
sulfate, samples were centrifuged for 15 min at 4000g and the
amount of pyruvate in the supernatant was measured after
derivatization with o-phenylenediamine, as described previously
(17). Incubations were also performed in the presence of 1%
catalase (sufficient to scavenge at least 500 µM of hydrogen
peroxide) to delineate the role of hydrogen peroxide-mediated
selenoxidation in the pyruvate formation. To exclude contribu-
tion of contaminating PLP-dependent cysteine conjugate â-
lyases, incubations were also performed in the presence of 1 mM
AOAA.
To investigate the role of selenoxidation, incubations were
performed with D-AAO and D-phenylalanine in the presence of
Se-phenyl-L-selenocysteine. Incubations (1 mL) containing 0.94
mM HPA, 28.5 µg of HRP, 0.25 mM Se-phenyl-L-selenocysteine,
1 mM D-phenylalanine, and 2.5 mg D-AAO were incubated at
37 °C (to dissolve the substrates 5 M sodium hydroxide was
added until pH 8-9, after dissolution 5 M HCl was added until
pH 7.2). At several time points 200 µL aliquots were mixed with
20 µL of 0.3 M ZnSO4, placed on ice and centrifuged for 15 min
at 4000g. Pyruvate and hydrogen peroxide were measured as
described above. All incubations were performed in triplicate,
and incubations without D-AAO were performed to correct for
nonenzymatic degradation.
F r a ction a tion of Ra t Kid n ey Cytosol by An ion -Ex-
ch a n ge F a st-P r otein Liqu id Ch r om a togr a p h y (F P LC). Rat
kidney cytosol (nondialyzed) was prepared as previously de-
scribed (17) and was fractionated by FPLC using a Mono Q
anion-exchange column (Pharmacia Biotech, Uppsala, Sweden),
as described by Commandeur et al. (7). Thirty fractions of 2 mL
were collected and analyzed directly for enzyme activity. All
incubations described below were performed in triplicate and
were also performed in absence of fraction to correct for
nonenzymatic reactions.
Hydrogen Peroxide Formation. To quantify formation of
hydrogen peroxide, incubations were also performed in the
presence of 0.94 mM HPA and 28.5 µg/mL HRP, leading to the
in situ oxidation of HPA to the fluorescent product 2,2′-
dihydroxybiphenyl-5,5′-diacetate (18). The presence of HRP and
HPA did not alter the pyruvate formation. In this case, a 80 µL
incubation sample was added to a quartz cuvette and diluted
with 720 µL of 100 mM potassium phosphate buffer (pH 7.2)
and the fluorescence was detected at an excitation wavelength
of 323 nm and an emission wavelength of 400 nm. The amount
of hydrogen peroxide was analyzed using standard curves.
Activity profiles of â-elimination and oxidative deamination
were determined by incubating fractions with Se-phenyl-L-
selenocysteine. In case of â-elimination assessment, 10 µL of
FPLC-fraction was added to 50 µL of 50 mM sodium borate
buffer (pH 8.6), containing 0.6 mM Se-phenyl-L-selenocysteine
and 0.6 mM ΚΜΒ. After 20 min, 250 mL of 0.14% o-phenylene-
diamine in 3 N HCl was added. Samples were heated for 45
min at 60 °C and analyzed as described above. To investigate
the role of PLP-dependent cysteine conjugate â-lyases, incuba-
tions were also performed (a) in the presence of 1 mM AOAA
and in absence of ΚΜΒ and (b) at various ΚΜΒ concentrations
with Se-(4-methylbenzyl)-L-selenocysteine as a substrate (0.5
mM). In case of oxidative deamination assessment, incubations
(200 µL) containing 0.94 mM HPA, 5.7 µg of HRP, 0.5 mM Se-
phenyl-L-selenocysteine, and 20 µL of FPLC fraction in 50 mM
sodium borate buffer (pH 8.6) were incubated at 37 °C. After
20 min, 20 µL 0.3 M ZnSO4 was added to terminate the reaction
and the samples were placed on ice. Samples were centrifuged
for 15 min at 4000g and the amount of hydrogen peroxide was
measured as described above. The active fraction number 12,
containing L-AAO (L-R-hydroxy acid oxidase), was analyzed for
â-elimination activity in the presence of AOAA and in absence
of ΚΜΒ, and oxidative deamination activity both at pH 7.2 (100
mM potassium phosphate buffer) and 8.6 (as described above).
Substrates were L-phenylalanine, Se-phenyl-L-selenocysteine,
S-phenyl-L-cysteine, Se-benzyl-L-selenocysteine, S-benzyl-L-cys-
teine, CFTE-Cys, and â-chloro-L-alanine (to dissolve the sub-
strates in 100 mM potassium phosphate buffer, pH 7.2, 5 M
sodium hydroxide was added until pH 8-9, after dissolution 5
M HCl was added until pH 7.2). Protein concentration of the
fractions was determined with the Bio-Rad protein kit (Hercules,
CA).
Id en tifica tion of P r od u cts of L-AAO a n d Se-P h en yl-L-
selen ocystein e a n d Assessm en t of Stoich iom etr y. The
stoichiometry of products was determined by performing a large-
scale incubation of L-AAO and Se-phenyl-L-selenocysteine as
model substrate. Incubations (3 mL) containing 0.94 mM HPA,
85.7 µg of HRP, 0.5 mM Se-phenyl-L-selenocysteine and 15 µg
of L-AAO in 100 mM potassium phosphate buffer (pH 7.2) were
incubated at 37 °C (to dissolve the substrates 5 M sodium
hydroxide was added until pH 8-9, after dissolution 5 M HCl
was added until pH 7.2). At several time points 450 µL aliquots
were mixed with 50 µL of 0.3 M ZnSO4, placed on ice and
centrifuged at 4000g for 15 min. Incubations were performed
in triplicate, and incubations without L-AAO were performed
to correct for nonenzymatic degradation. The amount of pyru-
vate and hydrogen peroxide was measured as described above.
Ammonia was measured using an ammonium kit of Sigma (St.
Louis, MO) based on the reductive amination of 2-oxoglutarate
by glutamate dehydrogenase and NADPH. Consumption of Se-
phenyl-L-selenocysteine was measured by a modified method of
Wang et al. (19), using GC-ECD, as follows: To a solution
containing 60 µL of sample, 10 µL of 4 mM L-phenylalanine (as
an internal standard), 30 µL of 2,2,2-trifluoroethanol, and 10
µL of pyridine was added 10 µL of ethylchloroformate. After 2
min, 500 µL of heptane was added and the derivatized products
were extracted and analyzed by GC-ECD as described below.
An a lytica l Equ ip m en t. GC-MS. GC-MS analyses were
performed on a HP 5890 gas chromatograph equipped with a
25 m BPX5 column (0.25 mm i.d., 0.25 µm film thickness, SGE,
Amstelveen, The Netherlands) coupled to a Hewlett-Packard
MSD 5970 mass spectrometer (ion source, electron impact
ionization, electron energy of 70 eV). Temperatures of the
injection port and transfer line were 270 °C. The column
temperature was programmed from 60 °C (2 min) to 270 °C (20
°C/min) and kept at 270 °C for 10 min. Full scanning analyses
were performed in the range of m/z 30-650. Using these GC-
MS conditions, the following retention times and mass spectra
were recorded:
Selenium-containing products were analyzed by GC-MS and
LC-APCI-MS/MS, as described below. For GC-MS analysis,
samples were acidified by 2 N hydrochloric acid, extracted by
ethyl acetate, and methylated by ethereal diazomethane. For
LC-APCI-MS/MS-analysis, samples were deproteinized by zinc
sulfate after which the supernatant is applied directly to LC-
APCI-MS/MS, as described below (see Analytical Equipment).
Selenium-containing products were also determined in incuba-
tions in absence of HPA and HRP, and in incubations in the
presence of 1% catalase.
In cu ba tion s w ith D-AAO. Specific activities of oxidative
deamination and â-elimination of Se-(n-propyl)-D-selenocysteine
and Se-(4-methylphenyl)-D-selenocysteine were determined as
Methylphenylselenide. Phenylselenol was detected as the
methylated derivative due to diazomethane treatment. Reten-

Selenocysteine Se-Conjugates and Amino Acid Oxidases
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 999
tion time, 5.4 min; mass spectrum m/z (relative intensity,
selenium isotope, assignment) 172 (85, 80Se, M•+), 157 (100, 80Se,
M•+ - CH3), 117 (20), 91 (45), 77 (47, C6H5+), 51 (56).
Methylated Phenylselenoacetic Acid. Phenylselenoacetic acid
was detected as the methylated derivative due to diazomethane
treatment. Retention time, 8.2 min; mass spectrum m/z (relative
intensity, selenium isotope, assignment) 230 (98,80Se, M•+), 171
(59, 80Se, M•+ - COOCH3), 157 (61, 80Se, C6H5Se+), 91 (100), 77
(52, C6H5+).
Dimethylated 2-Oxo-3-phenylselenopropionic Acid (Phenylse-
lenopyruvate). Due to keto-enol tautomerization of 2-oxo-3-
phenylselenopropionic acid, methylation by diazomethane ap-
peared to be not only at the carboxyl position, but also at the
R-keto-enol position, as observed previously for 3-mercapto-2-
oxopropionic acid S-conjugates and R-ketoacids of cysteine
S-conjugates (20). Retention time, 10.1 min (and 10.2 min); mass
spectrum m/z (relative intensity, selenium isotope, assignment)
272 (94, 80Se, M•+), 157 (100, 80Se, C6H5Se+), 115 (40, M•+
C6H5Se), 91 (60), 77 (63, C6H5+), 59 (44, COOCH3+).
-
Diphenyldiselenide. Retention time, 11.3 min; mass spectrum
m/z (relative intensity, selenium isotope, assignment) 314 (67,
80Se, M•+), 234 (16, 80Se, M•+- Se), 157 (100, 80Se, C6H5Se+), 117
(19), 77 (66, C6H5+), 51 (39).
GC-ECD. Samples were injected onto an automated Hewlettt
Packard 5890 series II gas chromatograph equipped with a 15
mCi 63Ni electron-capture detector. GC separation was achieved
on a 25 m HP-5 column (0.32 mm i.d., 0.25 µm film thickness,
Hewlettt Packard, Amstelveen, The Netherlands). The temper-
ature of the GC oven was programmed from 160 °C (1 min) to
270 °C, at 10 °C/min, and kept at 270 °C for 10 min. The
temperatures of the injector and detector were 270 and 300 °C,
respectively. The helium column flow rate was 3.2 mL/min.
Under these conditions the retention times were: L-phenylala-
nine derivative 6.46 min and Se-phenyl-L-selenocysteine deriva-
tive 9.57 min.
LC-APCI-MS/ MS. The LC system consisted of an Inertsil
ODS-2 column (5 µm particle, 150 × 4.60 mm, Phenomenex,
Amstelveen, The Netherlands) which was eluted at a flow rate
of 1 mL/min using the following gradient: initially 2% B for 3
min followed by an increase to 92% B over 30 min; at 35 min
returned to 2% B in 1 min. Mobile phase A consisted of 1%
acetonitrile in water, and mobile phase B was 100% acetonitrile.
The eluent from the HPLC column was analyzed using an UV-
detector, set at 220 and 254 nm, and subsequently passed into
the ionization region of the mass spectrometer without flow
splitting. Mass spectrometric detection was carried out with a
Finnigan LCQ-DECA mass spectrometer (ThermoQuest, San
J ose, CA) operating in the negative ion mode using an APCI
interface probe. A potential of 5 kV was applied to the APCI
corona discharge. High purity (99.999%, Praxair) N2 was used
as drying gas, with a flow of 300 L/h). Full scanning analyses
were performed in the range of m/z 150-650. MS/MS spectra
were recorded of ions at m/z 243 for 2-oxo-3-phenylselenopro-
pionic acid (phenylselenopyruvate) and at m/z 215 for phenylse-
lenoacetic acid.
F igu r e 2. Time-dependent formation of hydrogen peroxide (A)
and pyruvate (B) upon incubation of Se-phenyl-L-selenocysteine
(0.5 mM) with L-AAO from Crotalus adamanteus (5 µg/mL).
Experiments were performed as described in the Experimental
Procedures. Results are presented as mean value ( SD for three
preparations.
Activities exceeded that of the known substrate L-
phenylalanine. The highest catalytic efficiency (kcat/Km)
for hydrogen peroxide formation was obtained for Se-
benzyl-L-selenocysteine (0.394 ( 0.099 min-1 µM-1), Se-
methyl-L-selenocysteine (0.318 ( 0.016 min-1 µM-1), Se-
phenyl-L-selenocysteine (0.307 ( 0.043 min-1 µM-1), and
CTFE-Cys (0.276 ( 0.058 min-1 µM-1). Se-Allyl-L-sele-
Resu lts
nocysteine was the SeCys conjugate with the lowest kcat
/
Km-value. S-Benzyl-L-cysteine and S-phenyl-L-cysteine
showed 5-fold higher Km-values, when compared to their
selenium analogues (Table 1).
Biotr a n sfor m a tion of SeCys Con ju ga tes by L-
AAO. Incubating SeCys conjugates and L-AAO from
Crotalus adamanteus produced both hydrogen peroxide
and pyruvate, both being linear up to 30 min (Figure 2,
panels A and B). On the basis of these results, subsequent
enzyme kinetic studies were performed using incubation
times of 20 min
The enzyme kinetic parameters of L-AAO from Crota-
lus adamanteus for hydrogen peroxide and pyruvate
formation from eight substrates are shown in Table 1.
From these results, it appeared that all substrates tested
were metabolized by oxidative deamination (Figure 1,
route A), as measured by hydrogen peroxide formation.
In contrast to oxidative deamination which was ob-
served with all substrates, formation of pyruvate was
restricted exclusively to the SeCys conjugates tested
(Table 1). None of the cysteine S-conjugates were me-
tabolized by â-elimination by L-AAO from Crotalus ada-
manteus. For all SeCys conjugates except Se-phenyl-L-
selenocysteine, catalytic efficiencies (kcat/Km) for pyruvate
formation were slightly lower than those for hydrogen
peroxide formation. For the SeCys conjugates the cata-
lytic efficiency (kcat/Km) for pyruvate formation decreased

1000 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Rooseboom et al.
Ta ble 1. En zym e Kin etic P a r a m eter s of L-AAO fr om Cr ota lu s a d a m a n teu s for Oxid a tive Dea m in a tion (h yd r ogen
p er oxid e for m a tion ) a n d â-Elim in a tion (p yr u va te for m a tion ) tow a r d Va r iou s Am in o Acid sa
kcat
Km
(µM)
kcat/Km
substrate
reaction
(min-1
)
(min-1 µM-1
)
Se-benzyl-L-selenocysteine
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
oxidative deamination
â-elimination
7.83 ( 0.54
13.10 ( 0.49
7.09 ( 0.27
ND
12.50 ( 0.32
57.02 ( 4.06
6.08 ( 0.47
ND
32.48 ( 0.77
47.96 ( 2.12
17.16 ( 0.56
34.17 ( 3.57
10.93 ( 0.46
ND
20 ( 4
59 ( 7
126 ( 13
ND
41 ( 6
107 ( 21
204 ( 35
ND
102 ( 6
195 ( 18
190 ( 13
608 ( 95
40 ( 5
ND
0.394 ( 0.099
0.221 ( 0.049
0.056 ( 0.004
ND
0.307 ( 0.043
0.532 ( 0.084
0.030 ( 0.007
ND
0.318 ( 0.016
0.247 ( 0.012
0.090 ( 0.003
0.056 ( 0.002
0.276 ( 0.058
ND
S-benzyl-L-cysteine
Se-phenyl-L-selenocysteine
S-phenyl-L-cysteine
Se-methyl-L-selenocysteine
Se-allyl-L-selenocysteine
CTFE-Cys
L-phenylalanine
12.85 ( 0.34
ND
125 ( 10
ND
0.103 ( 0.005
ND
a
Experiments were performed as described in the Experimental Procedures. kcat, Km, and kcat/Km values were obtained from Hanes-
Woolf plots. Results are presented as mean values ( SD for three preparations. ND not detectable.
Ta ble 2. Stoich iom etr y of Meta bolism of
Se-P h en yl-L-selen ocystein e by L-AAO fr om Cr ota lu s
a d a m a n teu sa
Ta ble 3. Effect of AOAA a n d Ca ta la se on Hyd r ogen
P er oxid e a n d P yr u va te F or m a tion fr om
Se-P h en yl-L-selen ocystein e in th e P r esen ce of L-AAO
fr om Cr ota lu s a d a m a n teu s, a n d th e Effect of Hyd r ogen
P er oxid e Gen er a ted fr om D-AAO/D-P h en yla la n in e u p on
P yr u va te F or m a tion fr om Se-P h en yl-L-selen ocystein ea
specific activity
reaction
(nmol/min/mg L-AAO)
substrate consumption
ammonia formation
pyruvate formation
626 ( 168
596 ( 46
466 ( 14
200 ( 6
hydrogen peroxide
(nmol/min/mg)
pyruvate
(nmol/min/mg)
incubation mixture
L-AAO
+ AOAA
+ catalase
hydrogen peroxide formation
182 ( 8
198 ( 10
ND
474 ( 4
466 ( 20
490 ( 8
ND
a
Experiments were performed as described in the Experimental
Procedures. Results are presented as mean value ( SD for three
preparations.
D-AAO/D-phenylalanine
+ Se-phenyl-
194 ( 10
ND
L-selenocysteine
in the order: Se-phenyl-L-selenocysteine > Se-benzyl-L-
selenocysteine ≈ Se-methyl-L-selenocysteine > Se-allyl-
L-selenocysteine.
a
Experiments were performed as described in the Experimental
Procedures. Results are presented as mean values ( SD for three
preparations.
St oich iom et r y of P r od u ct s F or m ed fr om Se-
P h en yl-L-selen ocystein e. For Se-phenyl-L-selenocys-
teine, selected as a model substrate, the stoichiometry
of the products formed upon incubation with L-AAO was
quantified (Table 2). As expected, the specific activity of
substrate consumption (626 ( 168 nmol/min/mg) equaled
the specific activity of formation of ammonia (596 ( 46
nmol/min/mg) which is formed both via the â-elimination
and oxidative deamination pathways, Figure 1. The
specific activity of pyruvate formation, 466 ( 14 nmol/
min/mg, indicated that for this SeCys conjugate, ap-
proximately 70-75% proceeds via the â-elimination
pathway. In line with this, the specific activity of
hydrogen peroxide formation, reflecting the oxidative
deamination pathway, indeed equals to approximately
30% of the total Se-phenyl-L-selenocysteine consumption.
As shown in Table 3, the addition of 1 mM AOAA did
not significantly alter the formation of pyruvate, thus
excluding contamination of commercial L-AAO by PLP-
dependent cysteine conjugate â-lyases. Neither was
hydrogen peroxide formation influenced by AOAA. When
the incubations were performed in the presence of 1%
catalase, hydrogen peroxide formation could not be
measured anymore indicating complete decomposition of
hydrogen peroxide. However, pyruvate formation was not
affected by the presence of catalase, indicating that
â-elimination of Se-phenyl-L-selenocysteine is not medi-
ated via selenoxidation/ syn-elimination by hydrogen
peroxide formed during oxidative deamination (Figure 1,
routes A and B). This conclusion was further supported
by incubations of Se-phenyl-L-selenocysteine in the pres-
ence of a hydrogen peroxide generating system consisting
of D-phenylalanine and D-amino acid oxidase (Table 3).
While hydrogen peroxide formation equaled that ob-
served in incubations of Se-phenyl-L-selenocysteine in the
presence of L-AAO, not a trace of pyruvate formation was
noticed.
No products were observed when Se-phenyl-L-seleno-
cysteine or L-phenylalanine was incubated in the pres-
ence of D-AAO, indicating absolute stereospecificity of this
enzyme, as described previously (21).
Id en tifica tion of Selen iu m -Con ta in in g P r od u cts
by Ma ss Sp ectr om etr y. GC-MS analysis of methylated
extracts of incubations of Se-phenyl-L-selenocysteine with
L-AAO showed five selenium-containing products as
indicated by the characteristic selenium-isotope patterns.
The first selenium-containing product, at a retention time
of 5.4 min (peak I, Figure 3) showed a molecular ion at
m/z 172 (80Se isotope) and a fragment at m/z 157,
apparently due to loss of a methyl-radical (Figure 4A).
Comparison with a synthetical reference revealed that
this product was methylphenylselenide, formed by meth-
ylation of phenylselenol by diazomethane. A second
selenium-containing product had a retention time of 8.2
min (peak II, Figure 3) and a molecular ion at m/z 230
(80Se isotope) (Figure 4B). This mass is consistent with
the molecular weight of methylated phenylselenoacetic
acid, the product resulting from oxidative decarboxylation

Selenocysteine Se-Conjugates and Amino Acid Oxidases
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1001
(peak IIIA and IIIB, Figure 3) occurring at retention
times 10.1 and 10.2 min had comparable mass spectra
with a molecular ion at m/z 272 (80Se isotope) (Figure 4C).
These mass spectra most likely correspond to cis- and
trans-isomers of dimethylated 2-oxo-3-phenylselenopro-
pionic acid (phenylselenopyruvate), which is the product
resulting from oxidative deamination. Due to keto-enol
tautomerization of 2-oxo-3-phenylselenopropionic acid,
methylation by diazomethane occurs both at the carboxyl
position and at the enol-oxygen position, as previously
observed in case of 3-mercapto-2-oxo-propionic acid S-
conjugates, the R-ketoacids formed from cysteine S-
conjugates (20). The identity of 2-oxo-3-phenylselenopro-
pionic acid was also confirmed by LC-APCI-MS analyses
(negative ion mode) of an underivatized reaction mixture.
A selenium-containing product eluted at 4.5 min with an
ion at m/z 243 [M - H]- (highest mass and80Se isotope),
which is corresponding to the molecular weight of 2-oxo-
3-phenylselenopropionic acid. MS/MS analysis of the ion
at m/z 243 showed a product ion at m/z 157 corresponding
to the phenylseleno-moiety (data not shown). The fourth
product identified by GC-MS analysis was diphenyl
diselenide (Figure 3, peak IV, and Figure 4D), according
to its retention time (11.3 min) and mass spectrum (m/z
314 and 80Se isotope) identical with commercially avail-
able reference compound. Diphenyl diselenide is expected
to result from autoxidation of phenylselenol formed
during the â-elimination reaction.
F igu r e 3. Total ion current (TIC) chromatogram of an incuba-
tion of Se-phenyl-L-selenocysteine with L-AAO. Samples were
acidified, extracted and derivatized with ethereal diazomethane
as described in “Experimental Procedures”. Five selenium-
containing metabolites were detected as indicated by their
characteristic selenium-isotope patterns. Peak I represents
methylphenylselenide, II methylated phenylselenoacetic acid,
III dimethylated 2-oxo-3-phenylselenopropionic acid (cis- and
trans-isomers) and IV diphenyldiselenide. Corresponding mass
spectra are depicted in Figure 4.
of 2-oxo-3-phenylselenopropionic acid by hydrogen per-
oxide. In incubations in the presence of catalase and
HRP/HPA, this product was not detectable, indicative for
the involvement of hydrogen peroxide. The identity of
phenylselenoacetic acid as a product was confirmed by
LC-APCI-MS analyses (negative ion mode) of an un-
derivatized sample showing a selenium-containing prod-
uct with a retention time of 3.3 min with an ion at m/z
215 [M - H]- (highest mass and 80Se isotope) consistent
to the molecular weight of phenylselenoacetic acid. MS/
MS analysis of the ion at m/z 215 showed a product ion
at m/z 171 that is likely produced by decarboxylation
(data not shown). Two other selenium-containing peaks
GC-MS analysis of methylated extracts of incubations
of Se-methyl-L-selenocysteine and Se-allyl-L-selenocys-
teine with L-AAO showed only the presence of the
corresponding dimethylated R-keto acids. In these incu-
bations, the corresponding selenols and diselenides could
not be detected as products. In incubations performed in
F igu r e 4. Mass spectra of an incubation of Se-phenyl-L-selenocysteine with L-AAO. Samples were acidified, extracted and derivatized
with ethereal diazomethane as described in the Experimental Procedures. (A) Peak I represent methylphenylselenide (5.4 min, m/z
172), (B) peak II methylated phenylselenoacetic acid (8.2 min, m/z 230), (C) peak IIIA dimethylated 2-oxo-3-phenylselenopropionic
acid (phenylselenopyruvate), and (D) peak IV diphenyldiselenide. Due to keto-enol tautomerization of 2-oxo-3-phenylselenopropionic
acid, methylation by diazomethane occurred not only at the carboxyl position, but also at the R-keto-enol position, as observed
previously for 3-mercapto-2-oxopropionic acid S-conjugates, R-ketoacids of cysteine S-conjugates (20). Fragmentation patterns and
experimentals are described in the Experimental Procedures.

1002 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Rooseboom et al.
F igu r e 5. Activity profiles of â-elimination and oxidative deamination with SeCys conjugates in FPLC-fractions from rat renal
cytosol applied to a Mono Q anion-exchange column. (A) â-Elimination of Se-phenyl-L-selenocysteine with FPLC-fractions (closed
symbols), and in the presence of AOAA (open symbols). (B) â-Elimination of Se-(4-methylbenzyl)-L-selenocysteine with FPLC-fractions
at various KMB concentrations. (C) Oxidative deamination of Se-phenyl-L-selenocysteine with FPLC-fractions. Experiments were
performed as described in the Experimental Procedures.
Ta ble 4. Sp ecific Activity for Oxid a tive Dea m in a tion (h yd r ogen p er oxid e for m a tion ) a n d â-Elim in a tion (p yr u va te
for m a tion ) tow a r d Va r iou s Am in o Acid s in F P LC-F r a ction 12a
oxidative deamination
â-elimination
Substrate
pH 8.6
pH 7.2
pH 8.6
pH 7.2
Se-benzyl-L-selenocysteine
S-benzyl-L-cysteine
Se-phenyl-L-selenocysteine
S-phenyl-L-cysteine
CTFE-Cys
L-phenylalanine
â-chloro-L-alanine
4.11 ( 0.21
0.69 ( 0.02
1.62 ( 0.07
0.49 ( 0.02
0.47 ( 0.03
0.59 ( 0.01
0.32 ( 0.02
1.63 ( 0.17
4.97 ( 1.07
ND
1.13 ( 0.20
ND
ND
ND
ND
3.92 ( 0.04
ND
1.17 ( 0.21
ND
ND
ND
ND
0.06 ( (2 × 10-3
0.32 ( 0.03
)
0.02 ( (8 × 10-5
0.03 ( (1 × 10-3
ND
)
)
0.03 ( (1 × 10-3
)
a
Experiments were performed as described in the Experimental Procedures. Pyruvate determination was performed in the presence
of 1 mM AOAA. Specific activities are presented in nanomoles per minute per milligram of protein. Results are presented as mean values
( SD for three preparations. ND, not detectable.
the presence of catalase, to prevent oxidation of methyl-
and allylselenol, the corresponding selenols were also not
detectable.
in the presence of AOAA (Figure 5A). Furthermore, the
highest oxidative deamination, measured as hydrogen
peroxide formation, was observed in fraction 12, while
in fraction 10 no significant hydrogen peroxide was
detected (Figure 5C). The oxidative deamination and
â-elimination of several amino acids was determined in
the presence of AOAA in fraction 12 at pH 7.2 and 8.6
(Table 4). These results show that all amino acids were
oxidatively deaminated and that the specific activity is
lower at pH 7.2 than at pH 8.6. Furthermore, the specific
activity of oxidative deamination was much higher for
SeCys conjugates than for the corresponding sulfur
analogues. Formation of pyruvate from SeCys conjugates
by â-elimination in the presence of AOAA was again only
observed for the SeCys conjugates. The specific activity
of â-elimination for Se-benzyl-L-selenocysteine was higher
than for Se-phenyl-L-selenocysteine at both pH 7.2 and
pH 8.6.
Biotr a n sfor m a tion of SeCys Con ju ga tes by D-
AAO. Experiments with Se-(n-propyl)-D-selenocysteine
and Se-(4-methylphenyl)-D-selenocysteine showed that
D-amino acid oxidase from porcine kidney catalyzed both
the oxidative deamination and â-elimination of these
substrates (Table 5). In case of Se-(n-propyl)-D-seleno-
cysteine, activity of oxidative deamination and â-elimina-
tion were comparable, while for Se-(4-methylphenyl)-D-
selenocysteine â-elimination was the major pathway
accounting for 75% of the total biotransformation by
D-AAO. The corresponding L-enantiomers were metabo-
lized by L-AAO from Crotalus adamanteus via oxidative
deamination and â-elimination (Table 5). Both enzymes
were highly stereoselective since the L-SeCys conjugates
F r a ction a tion of Ra t Kid n ey Cytosol by An ion -
Exch a n ge F P LC. Rat kidney cytosol was fractionated
using Mono Q anion-exchange FPLC, and 2 mL fractions
were collected. The fractions were screened for â-elimina-
tion activity, measured as pyruvate formation, with Se-
phenyl-L-selenocysteine as a substrate in the presence
of 0.5 mM ΚΜΒ, a known cofactor of PLP-dependent
cysteine conjugate â-lyase enzymes (Figure 5A). The
highest â-elimination activity was observed in fraction
10, corresponding to rat kidney glutamine transaminase
K/cysteine conjugate â-lyase enzyme (7). In presence of
AOAA, a well-known inhibitor of PLP-dependent en-
zymes, however, the â-elimination activity was highest
in fraction 12 (Figure 5A). The â-elimination activity in
fraction 10 was completely blocked by AOAA, indicating
full dependence of PLP-dependent enzymes in this frac-
tion. â-Elimination activity in fraction 12 was only
partially inhibited by AOAA. The same pattern was
observed using different concentrations of ΚΜΒ, a cofac-
tor of PLP-dependent cysteine conjugate â-lyase enzymes,
and Se-(4-methylbenzyl)-L-selenocysteine as a substrate
(Figure 5B). In presence of relatively high ΚΜΒ concen-
trations (>0.2 mM), the highest â-elimination activity is
present in fraction 10 and the activity profile is similar
to Figure 5A. However, at lower ΚΜΒ concentrations
(0.1-0.2 mM), the highest â-elimination activity was
present in fractions 11 and 12. In the absence of ΚΜΒ,
fraction 12 displayed the highest â-elimination activity
and the activity profile was similar to the one obtained

Selenocysteine Se-Conjugates and Amino Acid Oxidases
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1003
Ta ble 5. Sp ecific Activity for Oxid a tive Dea m in a tion (h yd r ogen p er oxid e for m a tion ) a n d â-Elim in a tion (p yr u va te
for m a tion ) of L-SeCys Con ju ga tes by L-AAO fr om Cr ota lu s a d a m a n teu s a n d D-SeCys Con ju ga tes by D-AAO fr om P or cin e
Kid n eya
hydrogen peroxide
(nmol/min/mg)
pyruvate
(nmol/min/mg)
substrate
Se-(n-propyl)-L-selenocysteine
Se-(n-propyl)-D-selenocysteine
Se-(4-methylphenyl)-L-selenocysteine
Se-(4-methylphenyl)-D-selenocysteine
453 ( 26
2.0 ( 0.16
530 ( 16
2.0 ( 0.24
384 ( 74
1.8 ( 0.08
551 ( 32
6.0 ( 0.51
a
Experiments were performed as described in the Experimental Procedures. Results are presented as mean values ( SD for three
preparations.
were exclusively metabolized by L-AAO and the D-SeCys
conjugates were exclusively metabolized by D-AAO. Spe-
cific activities of L-AAO and D-AAO, however, should not
be compared quantitatively because of the different
source of the enzymes.
To investigate whether mammalian amino acid oxidase
is also able to catalyze â-elimination reactions of SeCys
conjugates, rat kidney cytosol was fractionated by anion-
exchange chromatography. When SeCys conjugates were
incubated in the presence of AOAA, to inhibit PLP-
dependent enzymes, one of the fractions displayed both
pyruvate formation and hydrogen peroxide formation
(Figure 5). In presence of phenylalanine and cysteine
S-conjugates, this fraction only displayed hydrogen per-
oxide formation. These results strongly suggest that
amino acid oxidase is involved in the non-PLP-dependent
â-elimination of SeCys conjugates in rat renal cytosol.
Rat renal cytosolic L-AAO, also called R-hydroxy acid
oxidase or B isoform, is a flavin-dependent peroxisomal
oxidase which can metabolize both L-R-hydroxy acids and
L-amino acids to R-keto acids (25). The oxidative deami-
nation was higher at pH 8.6 than at pH 7.2 for all
substrates (Table 4), which is consistent with the pH-
optimum for rat renal L-AAO, known to range between
pH 8.0 and 10 depending on the amino acid used (26).
Similar to L-AAO from Crotalus adamanteus, specific
activities of oxidative deamination in rat renal cytosol
were higher for the SeCys conjugates than for the
corresponding sulfur analogues. â-Elimination by cyto-
solic rat renal L-AAO was only obtained for the SeCys
conjugates.
Discu ssion
The present study shows that both pyruvate and
hydrogen peroxide is formed upon incubations of purified
L-AAO from Crotalus adamanteus and purified D-AAO
from porcine kidney with L-SeCys conjugates and D-
SeCys conjugates, respectively. Pyruvate formation ap-
pears to proceed via a direct â-elimination implicating
formation of selenols as products (Figure 1, route C),
instead of being mediated by selenoxidation by hydrogen
peroxide (Figure 1, route B). With Se-phenyl-L-seleno-
cysteine, formation of phenylselenol was indeed con-
firmed by analysis of selenium-containing products by
GC-MS (Figure 3). For the chemopreventive SeCys
conjugates Se-methyl-L-selenocysteine and Se-allyl-L-
selenocysteine, however, formation of the anticipated
selenols could not be demonstrated yet due to their high
volatility and/or their oxidation by hydrogen peroxide.
Addition of catalase to prevent this oxidation was unsuc-
cessful. It might therefore be possible that in the presence
of catalase and hydrogen peroxide methyl- and allyl-
selenols are rapidly oxidized similar as has been de-
scribed for small alcohols, such as methanol and allyl
alcohol (35).
As yet little is known about the substrate specificity
of cytosolic rat renal L-AAO. The enzyme is frequently
called long-chain R-hydroxy acid oxidase, because good
substrates for this enzyme are those containing benzyl-
or alkyl-groups of 2-5 carbons (25). Consistent to this
structure-activity relationship, Se-benzyl-L-selenocys-
teine appeared to be an excellent substrate for cytosolic
rat renal L-AAO, showing the highest activities of both
oxidative deamination and â-elimination (Table 4). There-
fore, the relatively poor inhibition (21%) of â-elimination
activity in rat renal cytosol by AOAA, as described
previously (6), may be explained by the high activity of
L-AAO for this substrate in combination with a relatively
low activity of cysteine conjugate â-lyase/glutamine trans-
aminase K (7). In case of Se-phenyl-L-selenocysteine,
â-elimination in rat renal cytosol appeared to be mediated
almost completely (95%) by PLP-dependent enzymes (6).
This SeCys conjugate indeed showed a very high activity
when incubated with PLP-dependent cysteine conjugate
â-lyase/glutamine transaminase K (7), whereas in the
present study a relatively low activity is observed with
rat renal L-AAO (Table 4).
In contrast to the SeCys conjugates, in case of the
corresponding sulfur analogues no pyruvate formation
was observed. A similar result was observed previously
when SeCys conjugates and the corresponding S-conju-
gates were incubated with highly purified PLP-dependent
rat renal cysteine conjugate â-lyase/glutamine transami-
nase K (7). Possible explanations for the higher suscep-
tibility of SeCys conjugates for â-elimination may be the
weaker bond strength of the C-Se bond (234 kJ /mol)
versus the C-S bond (272 kJ /mol) (22) and/or a more
facile â-proton abstraction of the selenocysteine moiety
(23). Interestingly, CFTE-Cys, which is one of the best
substrates known for PLP-dependent cysteine conjugate
â-lyases, was also not â-eliminated at all by L-AAO,
indicating that different factors determine substrate
selectivity when compared to PLP-dependent enzymes.
Although no pyruvate formation was detected, all cys-
teine S-conjugates tested were actively metabolized by
oxidative deamination reactions, as reflected by hydrogen
peroxide formation. Consistent with previous studies
with different SeCys conjugates and cysteine S-conju-
gates (11, 24), the cysteine S-conjugates showed up to
10-fold lower deamination activities when compared to
corresponding SeCys conjugates (Table 1).
The present study indicates that in addition to PLP-
dependent cysteine conjugate â-lyases and flavin-con-
taining monooxygenases, AAOs may contribute to the
bioactivation of SeCys conjugates, as summarized in
Figure 6. Both â-elimination and oxidative deamination
pathways may be relevant for chemopreventive activity,

1004 Chem. Res. Toxicol., Vol. 14, No. 8, 2001
Rooseboom et al.
teine conjugate â-lyases, such as glutamine transaminase
K, AAOs may be involved in the bioactivation of SeCys
conjugates. The relative contribution of these enzymes
in the chemopreventive activity of SeCys conjugates
remains to be established, however. In case of the
involvement of reactive metabolites of SeCys conjugates
in the chemopreventive activity, tissue distribution of
these enzymes may determine the organ selectivity of
SeCys conjugates as chemopreventive agents.
Ack n ow led gm en t. We thank Dr. Herman Zappey
for his kind assistance with mass spectrometry experi-
ments.
Refer en ces
(1) Ip, C. (1998) Lessons from basic research in selenium and cancer
prevention. J . Nutr. 128, 1845-1854.
F igu r e 6. Identified enzymes active in the bioactivation of
SeCys conjugates. Reaction A: selenoxidation by flavin-contain-
ing monooxygenases followed by a spontaneous syn-elimination
of the selenoxide. Reaction B: â-elimination by PLP-dependent
cysteine conjugate â-lyases and amino acid oxidases. Reaction
C: oxidative deamination by amino acid oxidases. Reaction D:
transamination by PLP-dependent cysteine conjugate â-lyases.
(2) Ganther, H. E. (1999) Selenium metabolism, selenoproteins and
mechanisms of cancer prevention: complexities with thioredoxin
reductase. Carcinogenesis 20, 1657-1666.
(3) Ip, C., Zhu, Z., Thompson, H. J ., Lisk, D., and Ganther, H. E.
(1999) Chemoprevention of mammary cancer with Se-allylsele-
nocysteine and other selenoamino acids in the rat. Anticancer Res.
19, 2875-2880.
(4) Andreadou, I., Menge, W. M. P. B., Commandeur, J . N. M.,
Worthington, E. A., and Vermeulen, N. P. E. (1996) Synthesis of
novel Se-substituted selenocysteine derivatives as potential kid-
ney selective prodrugs of biologically active selenol compounds:
Evaluation of kinetics of â-elimination reactions in rat renal
cytosol. J . Med. Chem. 39, 2040-2046.
(5) Rooseboom, M., Vermeulen, N. P. E., Andreadou, I., and Com-
mandeur, J . N. M. (2000) Evaluation of kinetics of â-elimination
reactions of selenocysteine Se-conjugates in human renal cytosol.
Possible implications for the use as kidney selective prodrugs. J .
Pharmacol. Exp. Ther. 294, 762-769.
(6) Andreadou, I., Van De Water, B., Commandeur, J . N. M.,
Nagelkerke, F. J ., and Vermeulen, N. P. E. (1996) Comparative
cytotoxicity of 14 novel selenocysteine Se-conjugates in rat renal
proximal tubular Cells. Toxicol. Appl. Pharmacol. 141, 278-287.
(7) Commandeur, J . N. M., Andreadou, I., Rooseboom, M., Out M.,
de Leur, L. J ., Groot, E., and Vermeulen, N. P. E. (2000)
Bioactivation of selenocysteine Se-conjugates by a highly purified
rat renal cysteine conjugate beta-lyase/glutamine transaminase
K. J . Pharmacol. Exp. Ther. 294, 753-761.
because both selenols and hydrogen peroxide are known
to possess antitumor activity (1, 27, 28). Furthermore,
hydrogen peroxide may oxidize the selenols formed to
selenenic acids and seleninic acids. Seleninic acids were
recently shown to be chemopreventive both in vivo as in
vitro (29). As mode of action of selenium compounds,
interference with the Zn-homeostasis has been proposed
recently (2, 30-32). One of the enzymes strongly regu-
lated by the Zn-homeostasis is caspase-3, which plays an
important role in apoptosis (33). Release of inhibitory zinc
by complexing to selenols was proposed as a mechanism
leading to apoptosis (2). Recently, it was shown that zinc-
release from metallothionein by benzeneselenol was
dramatically enhanced by addition of tert-butylhydrop-
eroxide (31). It is therefore tempting to speculate that
parallel formation of selenols and hydrogen peroxide by
amino acid oxidase, as observed in the present study, may
be involved in chemopreventive activity of SeCys conju-
gates, however, this remains to be established.
(8) Rooseboom, M., Commandeur, J . N. M., Floor, G. C., Rettie, A.
E., and Vermeulen, N. P. E. (2000) Selenoxidation by flavin-
containing monooxygenases as a novel pathway for â-elimination
of selenocysteine Se-conjugates. Chem. Res. Toxicol. 14, 127-134.
Recently, three SeCys conjugates were shown to be
very active in the chemoprevention against methyl-
nitrosourea-induced mammary tumors in rats (3). The
SeCys conjugates tested, however, were racemic mixtures
of L- and D-SeCys conjugates. In the present study it was
shown that D-AAO is also able to catalyze the â-elimina-
tion as well as the oxidative deamination of D-SeCys
conjugates. Therefore, it may be anticipated that D-AAOs
may also be involved in case racemic mixtures of SeCys
conjugates are administered. In this respect, it is inter-
esting to note that recently a novel cancer gene therapy
paradigm was exploited based on the expression of D-AAO
in tumor cells in which the peroxisomal targeting se-
quence was removed (34). Exposure of these cells to
D-alanine caused cytotoxicity mediated by oxidative stress
at concentrations that were nontoxic to parental cells.
Using D-isomers of SeCys conjugates in this therapy
would not only lead to oxidative stress due to hydrogen
peroxide, but also to the generation of pharmacologically
active selenols.
(9) Cini, C., and De Maro, C. (1978) Carboxymethyl-selenopyruvic
acid as the product of the oxidative deamination of carboxymethyl-
selenocysteine. Ital. J . Biochem. 27, 104-110.
(10) De Marco, C., Rinaldi, M., Dessi, M. R., and Dernini, S. (1976)
Oxidation of Se-carboxymethyl-selenocysteine by L-amino acid
oxidase and by D-aspartate oxidase. Mol. Cell Biochem. 12, 89-
92.
(11) Coccia, R., Blarzino, C., Foppoli, C., and Cini, C. (1988) Oxidative
deamination of Se-(1-carboxyethyl)-, Se-(1-carboxypropyl)- and
Se-(2-carboxyethyl)-selenocysteine by snake venom L-amino acid
oxidase. Physiol. Chem. Phys., Med. NMR 20, 115-122.
(12) Walsh, C. T., Schonbrunn, A., and Abeles, R. H. (1971) Studies
on the mechanism of action of D-amino acid oxidase. Evidence
for removal of substrate R-hydrogen as a proton. J . Biol. Chem.
246, 6855-6866.
(13) Walsh, C. T., Krodel, E., Massey, V., and Abeles, R. H. (1973)
Studies on the elimination reaction of D-amino acid oxidase with
R-amino-â-chlorobutyrate. Further evidence for abstraction of
substrate R-hydrogen as a proton. J . Biol. Chem. 248, 1946-1955.
(14) Dang, T.-Y., Cheung, Y.-F., and Walsh, C. (1976) Reactions of
â-fluoroalanine and â-bromoalanine with D-amino acid oxidase.
Biochem. Biophys. Res. Commun. 72, 960-968.
(15) Miura, R., Shiga, K., Miyake, Y., Watari, H., and Yamano, T.
(1980) Studies on the reaction of D-amino acid oxidase with beta-
cyano-D-alanine. Observation of an intermediary stable charge-
transfer complex. J . Biochem. (Tokyo) 87, 1469-1481.
In conclusion, the present study indicates that, next
to flavin-containing monooxygenase enzymes and cys-

Selenocysteine Se-Conjugates and Amino Acid Oxidases
Chem. Res. Toxicol., Vol. 14, No. 8, 2001 1005
(16) Commandeur, J . N. M., Brakenhof, J . P. G., De Kanter, F. J . J .,
and Vermeulen, N. P. E. (1988) Nephrotoxicity of mercapturic
acids of three stucturally related 2,2-difluoroethylenes in the rat.
Biochem. Pharmacol. 37, 4495-4504.
(17) Stijntjes, G. J ., Te Koppele, J . M., and Vermeulen, N. P. E. (1992)
High-performance liquid chromatography-fluorescence assay of
pyruvic acid to determine cysteine conjugate â-lyase activity:
application to S-1,2-dichlorovinyl-L-cysteine and S-2benzothi-
azolyl-L-cysteine. Anal. Biochem. 206, 334-343.
(25) Hamilton, G. A. (1985) Peroxisomal oxidases and suggestions for
the mechanism of action of insulin and other hormones. Adv.
Enzymol. 57, 86-178.
(26) Paik, W. K., and Kim, S. (1965) pH-substrate relationships of
L-amino acid oxidases from snake venom and rat kidney. Biochim.
Biophys. Acta 96, 66-74.
(27) Parnham, M. J ., and Graf, E. (1991) Pharmacology of synthetic
organic selenium compounds. Prog. Drug Res. 36, 9-47.
(28) Suhr, S. M., and Kim, D. S. (1996) Identification of the snake
venom substance that induces apoptosis. Biochem. Biophys. Res.
Commun. 224, 134-139.
(18) Hyslop, P. A., and Sklar, L. A. (1984) A quantitative fluorimetric
assay for the determination of oxygen production by polymorpho-
nuclear leukocytes: its use in the simultaneous fluorimetric assay
of cellular activation processes. Anal. Biochem. 141, 280-286.
(19) Wang, J ., Huang, Z.-H., Gage, D. A., and Watson, J . T. (1994)
Analysis of amino acid by gas chromatography-flame ionization
detection and gas chromatography-mass spectrometry: simul-
taneous derivatization of functional groups by an aqueous-phase
chloroformate-mediated reaction. J . Chromatogr. A. 663, 71-78.
(20) Commandeur, J . N. M., King, L. J ., Koymans, L., and Vermeulen,
N. P. E. (1996) Bioactivation of S-(2,2-dihalo-1,1-difluoroethyl)-
L-cysteines and S-(trihalovinyl)-L-cysteines by cysteine S-conju-
gate â-lyase: indications for formation of both thionoacylating
species and thiiranes as reactive intermadiates. Chem. Res.
Toxicol. 9, 1092-1102.
(21) Curti, B., Rorchi, S., and Simonetta, M. P. (1992) D- an L-amino
acid oxidases. In Chemistry and Biochemistry of Flavoenzymes
(Mu¨ller, F., Ed.) pp 69-94, CRC Press, Boca Raton, FL.
(22) Krief, A. (1987) Synthesis of selenium and tellurium ylidesand
carbanions: application to organic synthesis. In The chemistry
of organic selenium and tellurium compounds (Patai, S., and
Rappoport, Z., Eds.) Vol. 1, pp 675-757, J ohn Wiley & Sons,
Chichester, U.K.
(29) Ip, C., Thompson, H. J ., Zhu, Z., and Ganther, H. E. (2000) In
vitro and in vivo studies of methylseleninic acid: Evidence that
a
monomethylated selenium metabolite is critical for cancer
chemoprevention. Cancer Res. 60, 2882-2886.
(30) J acob, C., Maret, W., and Vallee, B. L. (1998) Control of zinc
transfer between thionein, metallothionein, and zinc proteins.
Proc. Natl. Acad. Sci. U.S.A. 95, 3489-3494.
(31) J acob, C., Maret, W., and Vallee, B. L. (1999) Selenium redox
biochemistry of zinc-sulfur coordination sites in proteins and
enzymes. Proc. Natl. Acad. Sci. U.S.A. 96, 1910-1914.
(32) Maret, W., J acob, C., Vallee, B. L., and Fischer, E. H. (1999)
Inhibitory sites in enzymes: Zinc removal and reactivation by
thionein. Proc. Natl. Acad. Sci. U.S.A. 96, 1936-1940.
(33) Perry, D. K., Smyth, M. J ., Stennicke, H. R., Salvesen, G. S.,
Duriez, P., Poirier, G. G., and Hannun, Y. A. (1997) Zinc is a
potent inhibitor of the apoptotic protease, caspase-3. A novel
target for zinc in the inhibition of apoptosis. J . Biol. Chem. 272,
18530-18533.
(34) Stegman, L. D., Zheng, H., Neal, E. R., Ben-Yoseph, O., Pollegioni,
L., Pilone, M. S., and Ross, B. D. (1998) Induction of cytotoxic
oxidative stress by D-alanine in brain tumor cells expressing
Rhodotorula gracilis D-amino acid oxidase: a cancer gene therapy.
Hum. Gene Ther. 9, 185-193.
(23) Miles, E. W. (1986) Pyridixal phophate enzymes catalyzing
â-elimination and â-replacement reactions. In Coenzymes and
cofactors. Vitamin B6 pyridoxal phosphate: chemical, biocehmical
and medical aspects. Part B (Dolphin, D., Poulson, R., and
Avramovic, O., Eds.) Vol. 1, pp 253-310, J ohn Wiley & Sons, New
York.
(24) Foppoli, C., Coccia, R., and Blarzino, C. (1986) Oxidative deami-
nation of S-(1-carboxyethyl)-L-cysteine and S-(1-carboxypropyl)-
L-cysteine by L-amino acid oxidase. It. J . Biochem. 35, 385-390.
(35) DeMaster, E. G., Dahlseid, T., and Redfern, B. (1994) Comparative
oxidation of 2-propyn-1-ol with other low molecular weight
unsaturated and saturated primairy alcohols by bovine liver
catalase in vitro. Chem. Res. Toxicol. 7, 414-419.
TX000265R