Acute effects of novel selenazolidines on murine chemoprotective enzymes.

Abstract

Novel selenazolidines, designed as l-selenocysteine prodrugs and potential cancer chemopreventive agents, were examined for their ability to affect the transcription of murine hepatic chemoprotective enzymes. Compounds investigated were selenazolidine-4(R)-carboxylic acid (SCA) and six 2-substituted derivatives that cover a C log P range of -0.512 to -3.062. Their biological effects were compared with those of L-selenocystine. Gene transcripts were examined 24 h after a single dose, administered i.p. and i.g., and covered a range of chemoprotective enzymes; alpha, mu and pi class glutathione transferases (Gsts), UDP-glucuronosyltransferases (Ugts) 1a1, 1a6, 1a9, and 2b5, glutathione peroxidase 1 (Gpx), thioredoxin reductase (Tr), NAD(P)H-quinone oxidoreductase 1 (Nqo), and microsomal epoxide hydrolase (Meh). When given i.g., 2-butyl SCA (BSCA) resulted in elevations in alpha, mu and pi class Gsts, Ugt1a6, Tr, and Gpx, and 2-phenyl SCA (PhSCA) elevated GstP, Ugt1a9, Tr, Gpx (3 kb), and Meh. Other derivatives with C log P values both lower [2-(2'-hydroxy)phenyl SCA (PhOHSCA) and 2-methyl SCA (MSCA)] and higher [2-cyclohexyl SCA (ChSCA) and 2-oxo SCA (OSCA)] than BSCA and PhSCA elevated far fewer transcripts; PhOHSCA (Ugt1a1, Gpx), MSCA (Ugt1a1, Meh), ChSCA (Ugt1a1, Ugt1a9), and OSCA (Ugt1a6, Ugt1a9, GstM). When given i.p., the most pervasive transcript changes were parallel increases in Nqo and Tr transcripts which occurred with BSCA, PhSCA, MSCA, and OSCA. PhSCA also increased GstP, and PhOHSCA increased Ugt1a1 and Ugt1a6 levels. Unique among the compounds, PhSCA reduced the transcript levels of GstA, and the 1.6 kb transcript of Gpx although only when given i.p. Neither l-selenocystine nor SCA affected the level of any transcript and no compound altered the amount of Ugt2b5 mRNA. Despite chemical similarity and common ability to potentially serve as a source of l-selenocysteine, each selenazolidine compound appeared to elicit a unique pattern of mRNA responses and by either route of administration, there was no correlation between the magnitude of response of any gene and the calculated C log P values of the organoselenium compounds.

Full text

Page 1

Chemico-Biological Interactions 162 (2006) 31–42
Acute effects of novel selenazolidines on murine
chemoprotective enzymes
Wael M. El-Sayeda, Tarek Aboul-Fadla, John G. Lamba,
Jeanette C. Robertsb, Michael R. Franklina,∗
aUniversity of Utah, Department of Pharmacology and Toxicology, Salt Lake City, UT 84112, United States
bUniversity of Wisconsin, School of Pharmacy, Madison, WI 53705, United States
Received 17 March 2006; received in revised form 5 May 2006; accepted 5 May 2006
Available online 12 May 2006
Abstract
Novel selenazolidines, designed as l-selenocysteine prodrugs and potential cancer chemopreventive agents, were examined for
their ability to affect the transcription of murine hepatic chemoprotective enzymes. Compounds investigated were selenazolidine-
4(R)-carboxylicacid(SCA)andsix2-substitutedderivativesthatcoveraClogPrangeof−0.512to−3.062.Theirbiologicaleffects
werecomparedwiththoseofl-selenocystine.Genetranscriptswereexamined24hafterasingledose,administeredi.p.andi.g.,and
covered a range of chemoprotective enzymes; alpha, mu and pi class glutathione transferases (Gsts), UDP-glucuronosyltransferases
(Ugts) 1a1, 1a6, 1a9, and 2b5, glutathione peroxidase 1 (Gpx), thioredoxin reductase (Tr), NAD(P)H-quinone oxidoreductase 1
(Nqo), and microsomal epoxide hydrolase (Meh). When given i.g., 2-butyl SCA (BSCA) resulted in elevations in alpha, mu and pi
class Gsts, Ugt1a6, Tr, and Gpx, and 2-phenyl SCA (PhSCA) elevated GstP, Ugt1a9, Tr, Gpx (3kb), and Meh. Other derivatives
with ClogP values both lower [2-(2?-hydroxy)phenyl SCA (PhOHSCA) and 2-methyl SCA (MSCA)] and higher [2-cyclohexyl
SCA (ChSCA) and 2-oxo SCA (OSCA)] than BSCA and PhSCA elevated far fewer transcripts; PhOHSCA (Ugt1a1, Gpx), MSCA
(Ugt1a1, Meh), ChSCA (Ugt1a1, Ugt1a9), and OSCA (Ugt1a6, Ugt1a9, GstM). When given i.p., the most pervasive transcript
changes were parallel increases in Nqo and Tr transcripts which occurred with BSCA, PhSCA, MSCA, and OSCA. PhSCA also
increased GstP, and PhOHSCA increased Ugt1a1 and Ugt1a6 levels. Unique among the compounds, PhSCA reduced the transcript
levels of GstA, and the 1.6kb transcript of Gpx although only when given i.p. Neither l-selenocystine nor SCA affected the level of
anytranscriptandnocompoundalteredtheamountofUgt2b5mRNA.Despitechemicalsimilarityandcommonabilitytopotentially
serve as a source of l-selenocysteine, each selenazolidine compound appeared to elicit a unique pattern of mRNA responses and by
either route of administration, there was no correlation between the magnitude of response of any gene and the calculated ClogP
values of the organoselenium compounds.
© 2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Selenium;Selenocysteineprodrugs;Selenazolidines;Mouse;mRNA;Thioredoxinreductase;Glutathioneperoxidase;NAD(P)H-quinone
oxidoreductase; Glutathione transferases; UDP-glucuronosyltransferases; Microsomal epoxide hydrolase
∗Corresponding author at: University of Utah, Department of Phar-
macology and Toxicology, 30 South 2000 East, Room 201, Salt Lake
City, UT 84112, United States. Tel.: +1 801 581 7014;
fax: +1 801 585 5111.
E-mail address: mfranklin@alanine.pharm.utah.edu
(M.R. Franklin).
1. Introduction
For more than four decades selenium has been
categorized as an essential mammalian micronutrient.
Althoughseleniumandselenium-containingcompounds
provide protection against many degenerative condi-
tions, including cancer, the mechanism or mechanisms
0009-2797/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.cbi.2006.05.002

Page 2

32
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
by which this occurs continues to require investigation.
In chemoprevention against cancer, an effect related
to the maintenance or enhancement of “protective”
enzymes is commonly invoked. Protective enzymes
includetheselenoproteinsglutathioneperoxidase(GPx)
and thioredoxin reductase (TR), quinone oxidoreduc-
tase (NQO), as well as UDP-glucuronosyltransferases
(UGTs), microsomal epoxide hydrolase (mEH) and glu-
tathione transferases (GSTs). All are involved in either
sequestering reactive oxygen species and reactive elec-
trophilicmetabolitesormaintainingcellularcomponents
in their appropriate redox status.
In the study of selenium and selenium compound
efficacy in chemoprevention, naturally occurring com-
pounds have not always shown convincing and com-
prehensive activity. Inorganic selenite, in a variety
of dosing regimens but centered around a selenium
dose level of around 5ppm, was not effective against
7,12-dimethylbenz[a]anthracene (DMBA)-induced rat
mammary tumors [1], 4-(methylnitrosamino)-1-(3-
pyridyl)-1-butanone(NNK)-inducedmouselungtumors
[2,3] or azaserine-induced rat pancreatic and hepatic
neoplasms [4]. However, selenite at similar dose levels,
was effective against DMBA-induced mammary tumors
in mice [5] and rats [6–8], although only weakly so
[9]. Of the selenium-containing amino acids and their
derivatives, selenocystine at selenium dose levels of 2
and 3ppm was not effective against DMBA-induced
mammary tumors in mice [5], and only weakly effec-
tive [9], or not effective [10] in rat, whereas in mice,
and at 15ppm, it was effective for NNK-induced lung
tumors [3]. Selenomethionine at selenium dose levels
of 2–4ppm was not effective against DMBA-induced
mammary tumors in mice [5] and only weakly effec-
tive against DMBA-induced mammary tumors in rat
[9] and NNK-induced lung tumors in mice [3]. Mod-
ified seleno-amino acids, Se-methyl-, Se-propyl-, and
Se-allylselenocysteines were effective against DMBA
andN-methyl-N-nitrosourea(MNU)-inducedmammary
tumors in rat [10–12] but Se-methylselenocysteine at
similar dose levels (3ppm) was not effective against
NNK-induced lung tumors in mice [3].
In the search for more consistently efficacious com-
pounds,severalorganoseleniumcompoundsparticularly
organoselenocyanates have been created and evaluated,
although again with a variety of dosing regimens that
makes direct efficacy comparisons difficult. In general,
these compounds have been investigated at much higher
selenium dose levels (up to 40ppm) than was possible
with inorganic selenite. Benzylselenocyanate was effec-
tive against benzo[a]pyrene-induced mouse forestom-
ach tumors [13], azoxymethane-induced rat liver tumors
[14] and DMBA-induced rat mammary tumors [1]. 1,4-
Phenylenebis(methylene) selenocyanate (p-XSC) also
had wide ranging efficacy against tumors induced by
a variety of carcinogens in many tissues; in rat colon
after azoxymethane [15], in rat mammary tissue after
DMBA [16], in mouse lung after NNK [2,17–20], and
in rat tongue after 4-nitroquinoline-1-oxide [21]. It was
also effective against spontaneous familial adenoma-
touspolyposisdevelopmentinAPC(min)mice[22].The
di-glutathione conjugate of p-XSC was also effective
against azoxymethane-induced tumors in rat colon [23].
Aliphatic rather than aromatic selenocyanates have also
proved effective against DMBA-induced rat mammary
tumors [24].
Organoselenium compounds in addition to seleno-
cyanates have been investigated, mostly with rat
mammary tumor models and have met with vary-
ing degrees of efficacy. Methylphenylselenide, p-
xylylbis(methylselenide), triphenylselenonium chlo-
ride, and diphenyl selenide were investigated against
MNU-induced tumors [25], and triphenylselenon-
ium chloride and diphenyl selenide against DMBA-
induced tumors [26,27].
methoxybenzeneselenol had found it to be effective
against benzo[a]pyrene-induced mouse forestomach
tumors [13] and azoxymethane-induced rat liver tumors
[28].
Of all the organoselenium compounds developed and
evaluated, few have been examined for the changes they
elicitinchemoprotectiveenzymes,andwheresuchstud-
ies have been undertaken, most have examined activities
of only one or two, most commonly including GSTs.
Most of these changes were investigated after several
months of organoselenium compound administration,
with only a couple of studies evaluating changes after a
week,andnoneafterjustoneorafewdays.Recently,ele-
vatedGSTactivitywasreportedinmouseliver,skin,and
colon following diphenylmethylselenocyanate adminis-
tration [29,30]. In earlier studies, GSTM activity was
increased in colon mucosa of azoxymethane-treated rats
after 10 weeks of dietary exposure to the glutathione
conjugate of benzylselenocyanate [31]. With long-term
administration of p-XSC to otherwise na¨ ıve animals, no
elevation of rat liver GST (1-chloro-2,4-dinitrobenzene,
CDNB) activity was seen [21]. However, in a 1-week
study with lower p-XSC concentrations, increased GST
(CDNB) activity in liver, lung and kidney but not colon
or mammary tissue was reported [32]. Utilizing class-
selective substrates, GSTA activity was increased only
in lung, and GSTP activity in lung and colon [32].
No GST class-selective substrate activity was statis-
tically increased in liver. In a similar study in mice,
Earlystudieswith
p-

Page 3

W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
33
GST (CDNB) activity, as well as GSTP and GSTM (not
GSTA) class-selective activities were elevated in lung
and liver [19]. In these latter two studies, Se-dependent-
GPxactivitywaselevatedintheratcolonbutnotliveror
lung of rat or mouse. An increase in rat colon mucosal
GPx activity by dietary p-XSC had previously been
shown [15]. Another organoselenium compound, triph-
enylselenonium chloride, did not increase rat liver GPx
activity[27].Neithertriphenylselenoniumchloride[27],
nor Se-methylselenocysteine [33] influenced rat hepatic
TR activity. Of two other protective enzyme activities
investigated, NQO activity was increased [21] and UGT
(4-nitrophenol) activity was slightly elevated [32] by p-
XSC in rat liver. In mice, neither lung nor liver UGT
activity was significantly affected by p-XSC [19].
We have pursued an l-selenocysteine prodrug
approach to delivering supranutritional levels of
bioavailable selenium. Several selenazolidines have
been created that by enzymatic or hydrolytic cleavage
are designed to slowly release l-selenocysteine [34].
Of three such prodrugs recently tested for chemopre-
ventive activity against NNK-induced lung tumors in
a mouse model, 2-oxoselenazolidine-4-carboxylic acid
at 15ppm selenium in the diet reduced the number of
lung tumors by 37.5% [3]. Chemoprotection studies are
not only costly but require an extensive time frame to
complete so there remains a compelling need for short-
term assays to prescreen newly synthesized compounds.
Towards this end, and with the mode of action of some
selenium compounds possibly linked to their ability to
affect protective enzymes, we have examined the effect
of a single dose of seven selenazolidines on the levels
of the transcription products (mRNAs) of the genes of
the protective enzymes. Since selenium-related chemo-
prevention against tumors can occur in many tissues in
animals models, including liver, we screened for these
enzyme transcript changes in liver, a readily available
and responsive tissue with a relatively high proportion
of a single cell type.
2. Materials and methods
2.1. Chemicals
SCA, MSCA, and OSCA were synthesized as pre-
viously described [34,35]. The remaining 2-substituted
selenazolidines were prepared by the condensation of
l-selenocysteine (Acros Organics, Morris Plains, NJ)
and the appropriate carbonyl compound. The struc-
tures of the target compounds were verified on the
basis of elemental and spectroscopic methods of anal-
ysis. The calculated ClogP values are shown in
Table 1.
2.2. Animal treatment and biological sample
preparation
Adult male CF-1 mice (25–35g) from Charles
River Laboratories were maintained in humidity-
and temperature-controlled environment on a 12-h
light/dark cycle with free access to food (Harlan-Teklad
Laboratory Diet 8640; Se content 0.026ppm) and
water. Compounds were administered by gavage (i.g.)
or intraperitoneal injection (i.p.) at selenium doses of
0.55mg/kg (l-selenocystine) or 1.1mg/kg (selenazo-
lidines), a dose comparable to the amount of selenium
ingested daily when the selenazolidines are included
in the diet at 15ppm as in the chemoprevention study
[3]. SCA, MSCA, OSCA, and selenocystine were
dissolved and administered in 0.2ml of water, BSCA,
ChSCA, PhSCA, and PhOHSCA in 0.2ml of corn oil.
All animal procedures were approved by the University
of Utah Animal Care and Use Committee and were
conducted in concordance with NIH guidelines for
the humane care of laboratory animals. Animals were
sacrificed 24h after the single dose, a blood sample was
collected on ice for immediate serum preparation, and
the livers were snap-cooled by perfusion in situ (via
the hepatic portal vein) with ice-cold normal saline.
Table 1
Selenazolidine l-selenocysteine prodrugs
Selenazolidine Abbreviation 2-Substituent
ClogP
Selenazolidine-4(R)-carboxylic acid
2-Methylselenazolidine-4(R)-carboxylic acid
2-Oxoselenazolidine-4(R)-carboxylic acid
2-Butylselenazolidine-4(R)-carboxylic acid
2-Phenylselenazolidine-4(R)-carboxylic acid
2-Cyclohexylselenazolidine-4(R)-carboxylic acid
2-(2?-Hydroxyphenyl)-selenazolidine-4(R)-carboxylic acid
SCAa
MSCAa
OSCAa
BSCAb
PhSCAb
ChSCAb
PhOHSCAb
H
CH3
O
C4H9
C6H5
cyclo-C6H11
2?-(OH)C6H4
−3.062
−2.543
−0.529
−0.956
−1.884
−0.512
−2.551
aAdministered to mice in water.
bAdministered to mice in corn oil.

Page 4

34
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
A 100mg sample of liver was removed, homogenized
in 2ml of TRIzol solution (Invitrogen; Carlsbad, CA)
and frozen at −80◦C for later RNA isolation. In the
study investigating any GST activity changes in animals
receiving the selenazolidines administered in corn oil,
livers, with the gall bladder carefully dissected away,
werehomogenizedin0.25Msucroseandcytosolicfrac-
tions (105,000×g for 60min) prepared by differential
centrifugation.
2.3. Hepatotoxicity evaluation
Serum alanine aminotransferase (sALT) [glutamate-
pyruvate transaminase] activity was determined using
a coupled reaction in which the serum dependent
absorbance change of NADH oxidation was monitored
at 340nm in the presence of optimized concentrations
of l-alanine, ?-ketoglutarate and purified lactic dehy-
drogenase enzyme [36].
2.4. mRNA quantification
Hepatic mRNA levels were determined by North-
ern blotting of 20?g of total RNA isolated by TRIzol
extraction. Gel electrophoresis, nucleic acid transfer to
membranes and32P probe labeling were all performed
as described previously [37]. The cDNA fragments used
as probes for northern blotting were as documented in
Table 2. The cDNA probes for the alpha and mu class
glutathionetransferaseswereeachdesignedtocaptureas
manymemberswithinaclassaspossiblebyutilizingthe
regions with highest homology within the genes of the
class, while having the greatest dissimilarity for mem-
bersofotherclasses.Hybridizedblotswerewashed(high
stringency) as described previously [38]. Washed blots
wereexposedtoautoradiographicfilmwithanintensify-
ing screen at −70◦C, and when sufficient band density
for optimal quantification had developed, the density of
themajordiscretebandorbands(twoforGpx1andGstP,
possibly representing different transcripts or enzymes)
was determined by scanning densitometry (Molecular
Analysis Software; Bio-Rad, Hercules, CA). All mRNA
bands were normalized to the same-sample cyclophilin
mRNA band. Results are expressed as fold change from
the appropriate control animals.
2.5. Glutathione transferase activity
The protein concentration of the cytosolic fraction
was determined by the method of Lowry et al. [39],
andtheglutathionetransferaseactivitytowards1-chloro-
2,4-dinitrobenzene determined spectrophotometrically
in the presence of saturating reduced glutathione as out-
lined by Habig and Jakoby [40].
2.6. Statistical analysis of data
Results are expressed as the mean±S.E.M. Fold
changes elicited by BSCA, ChSCA, PhSCA, and
PhOHSCA were compared to vehicle (corn oil)-treated
controls and SCA, MSCA, OSCA, and selenocystine
were compared to untreated animals. Treatment group
size was 3 for mice receiving SCA, MSCA, OSCA, and
selenocystine, and 4 for mice receiving corn oil. Statis-
tical analyses were performed using ANOVA, followed
by Fisher’s protected least significant difference multi-
ple range test. Differences were considered significant
at P values of <0.05.
Table 2
ProbeSequenceHomology
Ugt1a1 2-870 (rat) UGT1A1 (U20551) 90% (mouse) Ugt1a1 42-939 (NM 201645), 52% (mouse) Ugt1a6 (U09930), 52%
(mouse) Ugt1a9 (BC026561), <50% (mouse) Ugt2b5 (X06358)
Ugt1a6
Ugt1a9
Ugt2b5
Meh,
Gpx
9-765 (U09930)
1-750 (BC026561)
1-804 (X06358)
107-1531 (rat) EPHX1 (M26125)
1-670 Gpx1 (NM 008160)
93% (mouse) Ephx1 70-1440 (NM 010145), <50% (mouse) Ephx2 (BC015087)
71% Gpx2 (NM 030677), 65% Gpx3 (NM 008161), 58% Gpx4 (NM 008162),
65% Gpx5 (NM 010343)
<50% Nqor2 (NM 020282)
61% Txnrd2 (BC052758), 68% Txnrd3 (NM 153162)
97% Gsta1 (NM 00818), 76% Gsta3 (NM 010356), 66% Gsta4 (NM 010357),
<50% Gstm1 and Gstp1
87% Gstm2 (NM 008183), 91% Gstm3 (NM 010359), 66% Gstm4 (NM 026764),
86% Gstm5 (J04696), 85% Gstm6 (NM 008184), <50% Gstp1
98%Gstp2 (NM 181796)
Nqo
Tr
GstA
1-1480 Nqor1 (NM 008706)
159-539 Txnrd1 (AB027565)
1-700 Gsta2 (NM 008182)
GstM 200-784 Gstm1 (NM 010358)
GstP185-622 Gstp1a(NM 013541)
aThe GstP probe was a gift from Dr. C. Roland Wolf, University of Dundee, Scotland.

Page 5

W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
35
3. Results
The sALT values determined 24h after a single dose
indicated that none of the investigated compounds by
either route of administration caused any acute hep-
atotoxicity (Table 3). The absence of hepatotoxicity
is important in the subsequent interpretation of any
observedchangesinmRNAsasnotbeingaconsequence
of overt liver cell damage or death.
The i.g. route of administration produced the greater
number of significant changes in mRNA levels as com-
pared to the i.p. route. This is most evident among the
microsomal enzyme mRNA responses (Ugts and Meh;
Figs. 1 and 2, respectively), where with the exception
of PhOHSCA, all statistically significant mRNA ele-
vations occurred following i.g. administration. By this
route, MSCA and ChSCA induced Ugt1a1, OSCA and
BSCAinducedUgt1a6,andOSCA,ChSCAandPhSCA
induced Ugt1a9. PhOHSCA induced Ugt1a1, but in
addition,increasedthisandtheUgt1a6transcriptfollow-
ing i.p. administration. In contrast to the inducibility of
the Ugt1a family, none of the compounds by either route
affected Ugt2b5. Significant elevations in Meh mRNA
were confined to i.g. administration, and this occurred
with MSCA and PhSCA (Fig. 2). A similar situation
existed with Gpx where elevations (seen only for the
larger (i.e. 3kb) of two transcripts) were also restricted
to i.g. administration (Fig. 3). For this gene, the ele-
vations were seen with BSCA, PhSCA and PhOHSCA
and all were very similar at approximately two-fold.
Fig. 1. Effect of selenazolidine l-selenocysteine prodrugs on hepatic mRNAs of UDP-glucuronosyltransferases. Selenazolidine compounds were
administered at 1.1mgSe/kg; l-selenocystine was administered at 0.55mgSe/kg. SCA, MSCA, OSCA, and selenocystine were compared with the
untreated mice. BSCA, PhSCA, ChSCA, and PhOHSCA were compared with mice receiving corn oil vehicle.*Significantly different from the
appropriate control, P<0.05.

Page 6

36
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
Table 3
Effect of selenazolidine l-selenocysteine prodrugs on a hepatotoxicity
marker enzyme (glutamate-pyruvate transaminase, sALT) in serum
CompoundsALT (mU/ml)
i.g.i.p.
Selenocystine
SCA
MSCA
OSCA
BSCA
PhSCA
ChSCA
PhOHSCA
43.87 ± 4.54
45.78 ± 2.24
41.02 ± 3.03
38.78 ± 5.08
27.09 ± 3.27
34.61 ± 2.58
23.45 ± 1.41
26.05 ± 3.41
54.70 ± 7.45
39.15 ± 4.25
31.70 ± 5.12
35.24 ± 3.51
24.63 ± 2.84
27.70 ± 4.54
21.89 ± 0.77
23.18 ± 2.41
Selenazolidines were administered at 1.1mgSe/kg, l-selenocystine
was administered at 0.55mgSe/kg. No value showed a statistically
significant change from its appropriate control; untreated animals for
selenocysteine, SCA, MSCA, and OSCA, corn oil-treated animals for
BSCA, ChSCA, PhSCA, and PhOHSCA.
PhSCA given i.p. produced the lone significant change
in the 1.6kb Gpx transcript, a decrease. Although the
methodology employed here does not allow definitive
assignment of the two bands on the Gpx northern blot
to either different enzymes or to alternative transcripts
of the same gene, the former is less likely since there is
only 58–71% sequence homology for other Gpx genes
Fig. 2. Effect of selenazolidine l-selenocysteine prodrugs on liver
microsomal epoxide hydrolase mRNA. Selenazolidine compounds
were administered at 1.1mgSe/kg; l-selenocystine was adminis-
tered at 0.55mgSe/kg. SCA, MSCA, OSCA, and selenocystine
were compared with the untreated animals. BSCA, PhSCA, ChSCA,
and PhOHSCA were compared with the corn oil-treated animals.
*Significantly different from the appropriate control, P<0.05.
and the northern blots were subject to high stringency
wash conditions.
The predominance of significant elevations with i.g.
administration can be similarly observed with the Gst
responses, where OSCA, BSCA and PhSCA produced
significant changes but only with PhSCA were changes
alsoseenwithi.p.administration(Fig.4).Theresponses
to PhSCA given i.p. were as robust as the changes when
given i.g.; approximately five-fold increases in both the
2.4 and 1.2kb GstP transcripts. The two bands could
either be alternative transcripts or different enzymes,
since the northern blot wash conditions would not dif-
ferentiate gene transcripts with the 98% homology that
exists between the Gstp1 probe sequence and Gstp2.
In contrast to PhSCA, BSCA (i.g.) also increased the
mRNA levels of GstA and GstM. GstA mRNA levels
were affected by one other compound, PhSCA, but for
this compound there was a reduction in the level of
the GstA transcript. Among the selenium compounds
administered in aqueous vehicle (SCA, MSCA, OSCA,
selenocystine), only OSCA which increased the GstM
mRNAwhengiveni.g.,affectedanyoftheGstenzymes.
Whether the mRNA changes had resulted in ele-
vations in GST activity during the 24h period was
evaluated for the compounds administered in corn oil
(Table 4), the compounds eliciting the largest, and all
but one of the significant, mRNA changes (Fig. 4). No
compoundshowedanelevationinGSTactivity.Theonly
statistically significant change in activity was an anoma-
lous decrease by ChSCA given i.g.
The mRNA responses of two genes that were con-
trarytothegeneraldominanceofthei.g.effectsoveri.p.
effects were with Nqo and Tr (Fig. 3). Not only were
most significant elevations seen with i.p. administration
(MSCA, OSCA, BSCA, PhSCA), but both genes were
elevated by the same compounds. Although parallel, the
Table 4
Effect of selenazolidine l-selenocysteine prodrugs on hepatic glu-
tathione transferase activity
CompoundGST (1-chloro-2,4-dinitrobenzene)
activity (nmol/mg cytosolic
protein/min)
i.g. i.p.
Corn oil (vehicle)
BSCA
PhSCA
ChSCA
PhOHSCA
4884 ± 414
5555 ± 420
5451 ± 671
3508 ± 225*
4033 ± 197
4477 ± 320
5401 ± 533
5178 ± 267
4407 ± 331
3841 ± 202
Selenazolidines were administered at 1.1mgSe/kg in corn oil vehicle.
*Significantly different (P<0.05) from same-route corn oil vehicle
control.

Page 7

W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
37
Fig. 3. Effect of selenazolidine l-selenocysteine prodrugs on hepatic mRNAs of glutathione peroxidase, NAD(P)H-quinone oxidoreductase and
thioredoxin reductase. Selenazolidine compounds were administered at 1.1mgSe/kg; l-selenocystine was administered at 0.55mgSe/kg. SCA,
MSCA, OSCA, and selenocystine were compared with the untreated animals. BSCA, PhSCA, ChSCA, and PhOHSCA were compared with the
corn oil-treated animals.*Significantly different from the appropriate control, P<0.05.
magnitude of the increase was greater for Tr than for
Nqo. In addition to the effects seen with i.p. administra-
tion, Tr mRNA but not Nqo mRNA was also elevated by
the i.g. administration of BSCA and PhSCA.
4. Discussion
The ratio of maximum tolerable dose (MTD):dose
producing 50% reduction (ED50) of total tumor num-
ber has frequently been low for inorganic selenium
compounds and naturally occurring selenium con-
taining amino acids, and this has fueled searches for
compounds with more favorable properties. Dominant
among them have been modified selenium amino
acids [12], organoselenides [25], and organoseleno-
cyanates among which p-XSC has been most studied.
This compound showed efficacy against a variety of
chemically induced tumors in several organs of rat and
mouse [2,15,16,21]. Surprisingly, given the expense of
mounting tumor prevention studies, and their duration,
few if any of the compounds developed have been
analyzed for their acute effects at the level of gene
expression in animals. Gene expression has most often
been deduced from changes in tissue enzyme activities
determined following chronic administration, usually
in the diet. Enzyme activities were not universally
determined in the present single dose study because
with other inducing agents of protective enzymes in
vivo, the maximal changes in hepatic mRNA levels of
the genes occurred around the 18-h period [41] and it
was anticipated that any translation-dependent changes
would not be sufficiently robust until sometime beyond
the 24-h period at which the animals were sacrificed.
This assumption was borne out by limited studies

Page 8

38
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
Fig. 4. Effect of selenazolidine l-selenocysteine prodrugs on mRNAs of glutathione transferases in liver. Selenazolidine compounds were adminis-
teredat1.1mgSe/kg;l-selenocystinewasadministeredat0.55mgSe/kg.SCA,MSCA,OSCA,andselenocystinewerecomparedwiththeuntreated
animals. BSCA, PhSCA, ChSCA, and PhOHSCA were compared with the corn oil-treated animals.*Significantly different from the appropriate
control, P<0.05.
examining GST activities for compounds producing the
greatestmRNAelevationswherenoincreasesinactivity
were observed (Table 4). Acute gene expression studies
have the potential to yield information that can serve as
an early indication of efficacy, safety and toxicity and
can provide clues as to the mechanism or mechanisms
involved in tumor reduction. Also, acute transcription
effects more likely reflect properties of the compounds
themselves, rather than secondary effects arising from
their ability to maintain or enhance selenoenzymes (e.g.
GPx and Tr) which though their catalytic activities on
cellular constituents could also alter gene expression.
Selenium-related studies where mRNA levels have
been investigated in animals have been largely confined
tostudieswithselenoproteinmRNAs;GpxandTr.Early
studies indicated that selenium deficiency for >40 days
depressed hepatic Se-GPx mRNA levels [42,43] but the
elevated levels in a selenium sufficient diet were due
to cytosolic mRNA stabilization, not elevated transcrip-
tion [42]. While decreases in GPx mRNA closely par-
alleled the degree of selenium deficiency [44] and were
related to “nonsense mediated degradation” [45], sup-
plying selenium at four-fold selenium-adequate levels
did not increase levels above those seen at the selenium-
adequate level [44]. Selenium deficiency also decreased
Gpx1 mRNA levels in the intestines of mice [23]. In
studies monitoring Tr mRNA, selenium deficiency for
28 days decreased this transcript in rat liver [46]. Sele-
nium in the form of selenite did not alter Tr transcription
[47]orTrmRNAlevels[48]inHepG2cellsbutwasable
to stabilize Tr mRNA induced by sulforaphane. In sev-
eralhumancancercelllines,seleniteincreasedTrmRNA

Page 9

W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
39
in part, or in total, by increasing mRNA stability [49].
In rats, the direction of the effects of sodium selenite
(i.p. for 15 days) on hepatic Tr (and Gpx) mRNAs var-
ied with dose, with low doses elevating, and high doses
depressing [50].
In studies related to organoselenium compound
effects on the mRNA levels of other chemoprotective
enzymes, selenocysteine Se-conjugates induced mul-
tiple GSTA forms (2, 3, and 5) and GSTP but not
GSTM in primary rat hepatocytes and H35 Reuber rat
hepatoma cells [51] and p-XSC induced GSTs of all
classes in DMBA-induced rat mammary adenocarcino-
mas[52].Interestingly,seleniumdeficiencyalsoinduced
rat hepatic GSTA mRNA [53]. In these studies, as in the
presentstudies,elevatedmRNAlevelsdonotdistinguish
between elevated transcription rates and mRNA stabi-
lizationandrepresentsalimitationtotheinterpretationof
the possible mechanism of action of selenocompounds.
Theselenazolidinesthatarethesubjectofthisinvesti-
gation were designed as l-selenocysteine prodrugs with
different cell permeability (ClogP values ranging from
−3.06 to −0.51). Based on reasoning expressed earlier,
the acute effect of these compounds on the transcription
of a comprehensive range of chemoprotective enzymes
only two of which are selenoproteins, was undertaken.
Some of the selenazolidines will require enzymatic
releaseofselenocysteine(SCA,OSCA),whiletheothers
are anticipated to release selenocysteine by spontaneous
hydrolysis, albeit at likely differing rates. The variety
of gene transcription responses seen could be an indi-
cation that the responses are not solely the result of the
common ability of the compounds to provide cells with
selenocysteine. Responses to a common entity, likely
selenocysteine, may be present, as is suggested by the
elevation of Nqo and Tr by four of the seven SCA com-
pounds given i.p. (MSCA, OSCA, BSCA, and PhSCA)
but responses of other genes that were not in common
forthesefourcompoundssuggestthatselectothergenes
can respond to the parent compounds. However, caution
isnecessaryinsuchaninterpretationastherearecaveats
inherent in a single-dose, single-time point study. Thus,
the absence of a response of a gene or genes (e.g. Nqo
and Tr by SCA, ChSCA, and PhOHSCA) could be the
result of inadequate generation, or inability to provide
sustained concentrations of an entity such as selenocys-
teinefromthesecompoundswithinthetimeframeofthe
investigation. These considerations also relate to the fre-
quent differences in response to a single compound that
areseenbetweenthetworoutesofadministration.Given
i.g.,thecompoundsareexposedonafirstpasstoanenvi-
ronment, enzymes and transporters that do not exist in
the i.p. route. The mix of parent compound, metabolites
or breakdown products as well as the concentrations of
each to which the liver is exposed may therefore vary as
a consequence.
With the common response of Nqo and Tr to many
of the compounds, it is tempting to speculate that such
induction might arise via the antioxidant/electrophile
response element known to be present in the promoter
regionsofthesegenes[54,55],eventhoughselenitedoes
notactinthismanner[45,47].Evidenceagainstis(i)that
selenocystine does not elicit this induction and through
thetransientoxidativestressaccompanyingthedepletion
ofglutathioneasitisreducedtoselenocysteine,mightbe
expectedtodoso;(ii)GstAtranscriptioncanalsobereg-
ulated by an antioxidant/electrophile response element
[56–58]andwasnotup-regulatedwhenbothNqoandTr
mRNAs were elevated. The frequency with which there
was an absence of GstA induction when Nqo was ele-
vated (MSCA, OSCA, BSCA and PhSCA i.p.) or vice
versa (BSCA i.g.) adds substantial evidence to the idea
that interaction at the ARE/EpRE is not the mechanism
through which these genes are responding to the selena-
zolidines.
Based on the observations of the present study, and
within the limitations inherent in a single-time point
single-dosescreen,itisapparentthatinsteadofaseriesof
compounds (l-selenocysteine prodrugs) with the same
spectrum of properties, but differing in extent, each of
theselenazolidinesappearscapableofactivatingitsown
spectrum of protective enzymes, or for one compound,
SCA, none at all. The inducing properties appear to dif-
ferevenwithincloselyrelatedgenefamilies.Amongthe
single Ugt inducers, MSCA can induce Ugt1a1, BSCA
Ugt 1a6, and PhSCA Ugt 1a9. For compounds inducing
two Ugts, PhOHSCA can induce Ugt1a1 and Ugt 1a6,
ChSCAUgt1a1andUgt1a9,andOSCAUgt1a6andUgt
1a9. Similarly among the Gsts, some compounds induce
none (SCA, MSCA, ChSCA, PhOHSCA), OSCA can
induce GstM, PhSCA can induce GstP and BSCA can
induce all three Gst classes examined, alpha, mu, and pi.
For the Gst changes, the experimental approach using
class-specific rather than enzyme-specific probes was
designedtocaptureanychanges,shouldtheyexist.Such
an approach does have limitations in that elevations in
onlyonememberoftheclassmaybeobscuredordiluted
out to non-significance by other members that are not
changed, or elevations in one may be cancelled out by
decreases in another.
In addition to the absence of a signature mRNA
response common to all the selenazolidines, there was
also no pattern of response that delineated i.p. or
i.g. administration of the organoselenium compounds.
As indicated earlier, route of administration might be

Page 10

40
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
anticipated to affect both the concentration and compo-
sition of selenocompounds (parent, hydrolysis products,
and metabolites) reaching the liver and the transcrip-
tional response may be different for each component.
Moreextensivestudiesthanthoseundertakenherewould
be needed to resolve the reason(s) for route differences
of this single dose study, but while of possible interest
for academic reasons, such differences are likely to be
mootforchemopreventionstudieswherethecompounds
are likely to be included in the diet and continuously
ingested. The bolus administration of the present study
was viewed as a method to screen for, and accurately
compare, the acute effects of a defined per animal dose
of the compounds and escape any variability associated
with ad libitum food ingestion.
Insummary,wehavedemonstratedthatamongseven
selenazolidines, all of which share the property of being
l-selenocysteine prodrugs, six out of seven were able,
when given as a single dose, to induce their own unique
spectrum of mRNAs of chemoprotective enzymes. The
seventh compound (SCA), the compound without a sub-
stitution at the 2-position, failed to significantly increase
any mRNAs. We have also observed that the response
observedcanvarywiththeroutebywhichthecompound
isadministered,afeaturelikelylinkedtovariationsinthe
concentration, duration, and mix of selenocompounds
reaching the liver. The extensive range of enzyme genes
affectedbyselenazolidinesholdspromisefortheirpossi-
ble use as chemoprotectants against toxicants with toxic
effects in addition to carcinogenicity.
Acknowledgement
This project was supported by a USPHS Grant No.
GM 058913.
References
[1] J. Nayini, K. El-Bayoumy, S. Sugie, L.A. Cohen, B.S. Reddy,
Chemoprevention of experimental mammary carcinogenesis by
the synthetic organoselenium compound, benzylselenocyanate,
in rats, Carcinogenesis 10 (1989) 509–512.
[2] K. El-Bayoumy, P. Upadhyaya, D.H. Desai, S. Amin, S.S. Hecht,
Inhibition of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone
tumorigenicity in mouse lung by the synthetic organose-
lenium compound, 1,4-phenylenebis(methylene)selenocyanate,
Carcinogenesis 6 (1993) 1111–1113.
[3] L. Li, Y. Xie, W.M. El-Sayed, J.G. Szakacs, M.R. Franklin, J.C.
Roberts, Chemopreventive activity of selenocysteine prodrugs
against tobacco-derived nitrosamine (NNK) induced lung tumors
in the A/J mouse, J. Biochem. Mol. Toxicol. 19 (2005) 396–405.
[4] T.J. Curphey, E.T. Kuhlmann, B.D. Roebuck, D.S. Longnecker,
Inhibition of pancreatic and liver carcinogenesis in rats by
retinoid- and selenium-supplemented diets, Pancreas 3 (1988)
36–40.
[5] H.W. Lane, P. Teer, J. Dukes, J. Johnson, M.T. White, The
effect of four chemical forms of selenium on mammary
tumor incidence in BALB/c female mice treated with 7-12-
dimethylbenz[a]anthracene, Cancer Lett. 50 (1990) 39–44.
[6] C. Ip, Factors influencing the anticarcinogenic efficacy of sele-
nium in dimethylbenz[a]anthracene-induced mammary tumori-
genesis in rats, Cancer Res. 41 (1981) 2683–2686.
[7] C. Ip, Prophylaxis of mammary neoplasia by selenium supple-
mentation in the initiation and promotion phases of chemical
carcinogenesis, Cancer Res. 41 (1981) 4386–4390.
[8] C.Ip,D.Sinha,Anticarcinogeniceffectofseleniuminratstreated
withdimethylbenz[a]anthraceneandfeddifferentlevelsandtypes
of fat, Carcinogenesis 2 (1981) 435–438.
[9] C. Ip, G. White, Mammary cancer chemoprevention by inorganic
and organic selenium: single agent treatment or in combination
with Vitamin E and their effects on in vitro immune functions,
Carcinogenesis 8 (1987) 1763–1766.
[10] C. Ip, C. Hayes, R.M. Budnick, H.E. Ganther, Chemical form
of selenium, critical metabolites, and cancer prevention, Cancer
Res. 51 (1991) 595–600.
[11] C. Ip, H.E. Ganther, Comparison of selenium and sulfur analogs
in cancer prevention, Carcinogenesis 13 (1992) 1167–1170.
[12] C. Ip, Z. Zhu, H.J. Thompson, D. Lisk, H.E. Ganther, Chemo-
prevention of mammary cancer with Se-allylselenocysteine and
other selenoamino acids in the rat, Anticancer Res. 19 (1999)
2875–2880.
[13] K.El-Bayoumy,Effectsoforganoseleniumcompoundsoninduc-
tion of mouse forestomach tumors by benzo(a)pyrene, Cancer
Res. 45 (1985) 3631–3635.
[14] S. Sugie, B.S. Reddy, K. El-Bayoumy, T. Tanaka, Inhibition by
dietary benzylselenocyanate of hepatocarcinogenesis induced by
azoxymethane in Fischer 344 rats, Jpn. J. Cancer Res. 80 (1989)
952–957.
[15] B.S. Reddy, A. Rivenson, N. Kulkarni, P. Upadhyaya, K. El-
Bayoumy, Chemoprevention of colon carcinogenesis by the syn-
thetic organoselenium compound, 1,4-phenylenebis(methylene)
selenocyanate, Cancer Res. 52 (1992) 5635–5640.
[16] K.El-Bayoumy,Y.H.
Meschter, L.A. Cohen, B.S. Reddy, Inhibition of 7,12-
dimethylbenz(a)anthracene-induced tumors and DNA adduct
formation in the mammary glands of female Sprague–Dawley
rats bythesyntheticorganoselenium
phenylenebis(methylene)selenocyanate, Cancer Res. 52 (1992)
2402–2407.
[17] K. El-Bayoumy, P. Upadhyaya, D.H. Desai, S. Amin,
D. Hoffmann, E.L. Wynder, Effects of 1,4-phenylenebis
(methylene)selenocyanate, phenethyl isothiocyanate, indole-3-
carbinol, and d-limonene individually and in combination
on the tumorigenicity of the tobacco-specific nitrosamine
4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in A/J mouse
lung, Anticancer Res. 16 (1996) 2709–2712.
[18] B. Prokopczyk, S. Amin, D.H. Desai, C. Kurtzke, P.
Upadhyaya, K. El-Bayoumy, Effects of 1,4-phenylenebis
(methylene)selenocyanateand
(methylnitrosamino)-1-(3-pyridyl)-1-butanone-inducedtumorig-
enesis in A/J mouse lung, Carcinogenesis 18 (1997) 1855–1857.
[19] B. Prokopczyk, J.G. Rosa, D. Desai, S. Amin, O.S. Sohn,
E.S. Fiala, K. El-Bayoumy, Chemoprevention of lung tumori-
genesis induced by a mixture of benzo(a)pyrene and 4-
(methylnitrosamino)-1-(3-pyridyl)-1-butanone by the organose-
lenium compound 1,4-phenylenebis(methylene)selenocyanate,
Cancer Lett. 161 (2000) 35–46.
Chae, P.Upadhyaya, C.
compound, 1,4-
selenomethionine on 4-

Page 11

W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
41
[20] A. Das, D. Desai, B. Pittman, S. Amin, K. El-Bayoumy, Com-
parison of the chemopreventive efficacies of 1,4-phenylenebis
(methylene)selenocyanate and selenium-enriched yeast on
4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone induced lung
tumorigenesis in A/J mouse, Nutr. Cancer 46 (2003) 179–
185.
[21] T. Tanaka, H. Makita, K. Kawabata, H. Mori, K. El-Bayoumy,
1,4-Phenylenebis(methylene) selenocyanate exerts exceptional
chemopreventive activity in rat tongue carcinogenesis, Cancer
Res. 57 (1997) 3644–3648.
[22] C.V. Rao, I. Cooma, J.G. Rodriguez, B. Simi, K. El-
Bayoumy, B.S. Reddy, Chemoprevention of familial adenoma-
tous polyposis development in the APC(min) mouse model by
1,4-phenylene bis(methylene)selenocyanate, Carcinogenesis 21
(2000) 617–621.
[23] C.V. Rao, C.Q. Wang, B. Simi, J.G. Rodriguez, I. Cooma,
K. El-Bayoumy, B.S. Reddy, Chemoprevention of colon
cancerbya glutathione
(methylene)selenocyanate, a novel organoselenium compound
with low toxicity, Cancer Res. 61 (2001) 3647–3652.
[24] C.Ip,S.Vadhanavikit,H.E.Ganther,Cancerchemopreventionby
aliphatic selenocyanates: effect of chain length on inhibition of
mammary tumors and DMBA adducts, Carcinogenesis 16 (1995)
35–38.
[25] C. Ip, D.J. Lisk, H.E. Ganther, Activities of structurally-
related lipophilic selenium compounds as cancer chemopreven-
tive agents, Anticancer Res. 18 (1998) 4019–4025.
[26] C. Ip, D.J. Lisk, H.E. Ganther, H.J. Thompson, Triphenylseleno-
nium and diphenylselenide in cancer chemoprevention: compar-
ative studies of anticarcinogenic efficacy, tissue selenium levels
and excretion profile, Anticancer Res. 17 (1997) 3195–3199.
[27] C. Ip, H.J. Thompson, Z. Zhu, H.E. Ganther, In vitro and in vivo
studies of methylseleninic acid: evidence that a monomethylated
selenium metabolite is critical for cancer chemoprevention, Can-
cer Res. 60 (2000) 2882–2886.
[28] T. Tanaka, B.S. Reddy, K. El-Bayoumy, Inhibition by dietary
organoselenium, p-methoxybenzene-selenol, of hepatocarcino-
genesis induced by azoxymethane in rats, Jpn. J. Cancer Res.
76 (1985) 462–467.
[29] R.K. Das, S. Bhattacharya, Anti-tumour promoting activity of
diphenylmethyl selenocyanate against two-stage mouse skin car-
cinogenesis, Asian Pacific J. Cancer Prev. 6 (2005) 181–188.
[30] S. Ghosh, R.K. Das, A. Sengupta, S. Bhattacharya, Inhibition
of azoxymethane-induced aberrant crypt foci in rat by diphenyl-
methyl selenocyanate through downregulation of COX-2 and
modulation of glutathione transferase and lipid peroxidation,
Biol. Trace Elem. Res. 105 (2005) 171–185.
[31] T. Kawamori, K. El-Bayoumy, B.Y. Ji, J.G. Rodriguez, C.V.
Rao, B.S. Reddy, Evaluation of benzyl selenocyanate glutathione
conjugate for potential chemopreventive properties in colon car-
cinogenesis, Int. J. Oncol. 13 (1998) 29–34.
[32] O.S. Sohn, E.S. Fiala, P. Upadhyaya, Y.H. Chae, K. El-Bayoumy,
Comparative effects of phenylenebis(methylene)selenocyanate
isomers on xenobiotic metabolizing enzymes in organs of female
CD rats, Carcinogenesis 20 (1999) 615–621.
[33] H.E. Ganther, C. Ip, Thioredoxin reductase activity in rat liver is
not affected by supranutritional levels of monomethylated sele-
nium in vivo and is inhibited only by high levels of selenium in
vitro, J. Nutr. 131 (2001) 301–304.
[34] Y.Xie,M.D.Short,P.B.Cassidy,J.C.Roberts,Selenazolidinesas
novel organoselenium delivery agents, Bioorg. Med. Chem. Lett.
11 (2001) 2911–2915.
conjugate of1,4-phenylenebis
[35] M.D. Short, Y. Xie, L. Li, P.D. Cassidy, J.C. Roberts, Character-
istics of selenazolidine prodrugs of selenocysteine: toxicity and
glutathione peroxidase induction in V79 cells, J. Med. Chem. 46
(2003) 3308–3313.
[36] F.Wroblewski,J.S.LaDue,Serumglutamicpyruvictransaminase
in cardiac with hepatic disease, Proc. Soc. Exp. Biol. Med. 91
(1956) 569–571.
[37] H.T. Le, J.G. Lamb, M.R. Franklin, Drug metabolizing enzyme
induction by benzoquinolines, acridine, and quinacrine; tricyclic
aromatic molecules containing a single heterocyclic nitrogen, J.
Biochem. Toxicol. 11 (1996) 297–303.
[38] W.M. El-Sayed, T. Aboul-Fadl, J.G. Lamb, J.C. Roberts, M.R.
Franklin, Effect of selenium-containing compounds on hep-
atic chemoprotective enzymes in mice, Toxicology 220 (2006)
179–188.
[39] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein
measurement with the Folin phenol reagent, J. Biol. Chem. 193
(1951) 265–275.
[40] W.H. Habig, W.B. Jakoby, Glutathione S-transferases (rat and
human), Meth. Enzymol. 77 (1981) 218–231.
[41] J.G.Lamb,M.R.Franklin,Earlyeventsintheinductionofrathep-
atic UDP-glucuronosyltransferases, glutathione S-transferase,
and microsomal epoxide hydrolase by 1,7-phenanthroline: com-
parison with oltipraz, tert-butyl-4-hydroxyanisole, and tert-
butylhydroquinone, Drug Metab. Dispos. 28 (2000) 1018–
1023.
[42] M.J. Christensen, K.W. Burgener, Dietary selenium stabilizes
glutathione peroxidase mRNA in rat liver, J. Nutr. 122 (1992)
1620–1626.
[43] K.E.Hill,P.R.Lyons,R.F.Burk,Differentialregulationofratliver
selenoproteinmRNAsinseleniumdeficiency,Biochem.Biophys.
Res. Commun. 185 (1992) 260–263.
[44] G. Bermano, F. Nicol, J.A. Dyer, R.A. Sunde, G.J. Beckett, J.R.
Arthur, J.E. Hesketh, Tissue-specific regulation of selenoenzyme
gene expression during selenium deficiency in rats, Biochem. J.
311 (1995) 425–430.
[45] S. Weiss-Sachdev, R.A. Sunde, Selenium regulation of transcript
abundanceandtranslationalefficiencyofglutathioneperoxidase-
1 and -4 in rat liver, Biochem. J. 357 (2001) 851–858.
[46] K.B. Hadley, R.A. Sunde, Selenium regulation of thioredoxin
reductase activity and mRNA levels in rat liver, J. Nutr. Biochem.
12 (2001) 693–702.
[47] K.J.Hintze,K.A.Wald,H.Zeng,E.H.Jeffery,J.W.Finley,Thiore-
doxin reductase in human hepatoma cells is transcriptionally
regulated by sulforaphane and other electrophiles via an antioxi-
dant response element, J. Nutr. 133 (2003) 2721–2727.
[48] J. Zhang, V. Svehlikova, Y. Bao, A.F. Howie, G.J. Beckett, G.
Williamson, Synergy between sulforaphane and selenium in the
induction of thioredoxin reductase 1 requires both transcriptional
andtranslationalmodulation,Carcinogenesis24(2003)497–503.
[49] A.Gallegos,M.Berggren,J.R.Gasdaska,G.Powis,Mechanisms
of the regulation of thioredoxin reductase activity in cancer cells
by the chemopreventive agent selenium, Cancer Res. 57 (1997)
4965–4970.
[50] L.Gan,Q.Liu,H.B.Xu,Y.S.Zhu,X.L.Yang,Effectsofselenium
overexposure on glutathione peroxidase and thioredoxin reduc-
tase gene expressions and activities, Biol. Trace Elem. Res. 89
(2002) 165–175.
[51] P.A.’t. Hoen, M. Rooseboom, M.K. Bijsterbosch, T.J. van
Berkel, J.N. Vermeulen, Commandeur, Induction of glutathione
S-transferase mRNA levels by chemopreventive selenocysteine
Se-conjugates, Biochem. Pharmacol. 63 (2002) 1843–1849.

Page 12

42
W.M. El-Sayed et al. / Chemico-Biological Interactions 162 (2006) 31–42
[52] K. El-Bayoumy, B.A. Narayanan, D.H. Desai, N.K. Narayanan,
B. Pittman, S.G. Amin, J. Schwartz, D.W. Nixon, Elucidation of
molecular targets of mammary cancer chemoprevention in the
rat by organoselenium compounds using cDNA microarray, Car-
cinogenesis 24 (2003) 1505–1514.
[53] M.Chang,J.R.Burgess,R.W.Scholz,C.C.Reddy,Theinduction
of specific rat liver glutathione S-transferase subunits under inad-
equate selenium nutrition causes an increase in prostaglandin F2
alpha formation, J. Biol. Chem. 265 (1990) 5418–5423.
[54] L.V. Favreau, C.B. Pickett, Transcriptional regulation of the
rat NAD(P)H:quinone reductase gene. Characterization of a
DNA-protein interaction at the antioxidant responsive element
and induction by 12-O-tetradecanoylphorbol 13-acetate, J. Biol.
Chem. 268 (1993) 19875–19881.
[55] W.H. Watson, J. Pohl, W.R. Montfort, O. Stuchlik, M.S. Reed, G.
Powis, D.P. Jones, Redox potential of human thioredoxin 1 and
identification of a second dithiol/disulfide motif, J. Biol. Chem.
278 (2003) 33408–33415.
[56] V. Daniel, R. Sharon, A. Bensimon, Regulatory elements
controlling the basal and drug-inducible expression of glu-
tathione S-transferase Ya subunit gene, DNA 8 (1989) 399–
408.
[57] R.S. Friling, A. Bensimon, Y. Tichauer, V. Daniel, Xenobiotic-
inducible expression of murine glutathione S-transferase Ya
subunit gene is controlled by an electrophile-responsive ele-
ment, Proc. Natl. Acad. Sci. U.S.A. 87 (1990) 6258–
6262.
[58] T.H. Rushmore, C.B. Pickett, Transcriptional regulation of the
rat glutathione S-transferase Ya subunit gene. Characteriza-
tion of a xenobiotic-responsive element controlling inducible
expression by phenolic antioxidants, J. Biol. Chem. 265 (1990)
14648–14653.