Division of Toxicology, Leiden/Amsterdam Center for Drug Research,
Leiden University (S.O.-M., R.A.R., G.J.M.);
Division of Organic
Chemistry, Leiden University, Leiden (R.A.R., J.H.B.);
Division of
Toxicology, TNO Nutrition and Food Research (J.H.T.M.P., P.J.B.); and
Faculty of Pharmacy, University of Toronto (R.G.T., K.S.P.)
To assess the role of GST's (glutathione S-transferases) in the
(de)toxification of their substrates, an in vivo active
inhibitor based on the structure of glutathione (GSH),
-L-glutamyl-
-(D-2-aminoadipyl)-N-2-heptylamine monoethyl ester (Et-R-Hep), was developed. To increase its effectivity, analogues esterified with alkyl chains of varying lengths and one
diesterified derivative (DiEt-R-Hep) were synthesized. The unesterified
analogue, R-Hep, was also tested. Their isoenzyme selectivity was
characterized using purified rat GST isoenzymes. Furthermore, the
extent of inhibition of the GSH conjugation of (RS)-2-bromoisovalerylurea (BIU) was evaluated in rat liver
cytosol, isolated hepatocytes, and in liver perfusions. All compounds
inhibited Alpha- (1-1 and 2-2) more effectively than Mu (3-3 and
4-4) class GSTs; Pi-(5-5) and Theta (7-7) classes were minimally
inhibited. The unesterified R-Hep was the most effective inhibitor
towards purified isoenzymes; its Ki value
towards GST 3-3 (S-BIU as substrate) was 27 µM. The mono
ethyl ester derivative, Et-R-Hep (Ki 270 µM for 3-3), was the most potent inhibitor in hepatocytes and in the
perfused liver: 50 µM inhibited the conjugation of
(S)-BIU by 50%. Longer ester chains or diesterification
did not increase the inhibitory potency; R-Hep had less inhibitory
activity. In all systems, only the (S)-enantiomer of BIU,
which is conjugated mainly by Alpha class GSTs, was inhibited,
confirming Alpha isoenzyme selective inhibition.
 |
Introduction |
GSTs1 (glutathione
S-transferases; EC 2.5.1.18) are a
family of phase II biotransformation enzymes that provide protection against a wide range of electrophilic compounds. Four major classes of
cytosolic GST can be distinguished: the Alpha, Mu, Pi, and Theta class
(2, 3) which have different but overlapping substrate specificities.
GSTs catalyze the conjugation of the endogenous tripeptide glutathione
(GSH; -L-glutamyl-L-cysteinylglycine) to compounds containing an
electrophilic center. These may be either xenobiotics, like drugs, or
endogenous compounds, such as products of lipid peroxidation (4-6).
Recent studies indicate a role of GSTs in cancer chemotherapy;
resistance to alkylating agents is frequently associated with increases
in the levels of GSH and GST (7).
In vivo inhibitors of GST are valuable tools for
characterizing the role of GSTs in drug resistance and toxicity of
xenobiotics. However, most of the existing inhibitors are either too
toxic to be used in vivo or are only effective in
vitro (7). Moreover, none of these inhibitors is specific towards
a particular GST isoenzyme. Recently, the diuretic drug ethacrynic acid
has been used as an inhibitor of GST in vivo (8). However,
because it is a substrate for GSH conjugation, it causes GSH depletion,
which by itself may have toxic consequences (7). Therefore, there is a
need for developing more selective in vivo effective GST inhibitors that would leave GSH levels unaffected.
We have recently shown that the GSH analogue Et-R-Hep
(-L-glutamyl--(D-2-aminoadipyl)-N-2-heptylamine
monoethyl ester; systematic name:
(R)-5-ethyloxycarbonyl-2--(S)-glutamylamino-N-2-heptylpentamide (fig. 1A)), is an effective
inhibitor of GST isoenzymes of the Alpha and Mu class in the rat
in vivo (9, 10). In this compound, the
L-cysteinylglycine moiety of GSH was replaced by D-Aad
(D-aminoadipic acid), to obtain a metabolically more stable analogue.
To increase its cellular uptake, the lipophilicity of the compound was
increased by introducing an ethyl ester on the 
-carboxyl group on
D-Aad, corresponding to the carboxyl group of glycine in GSH (see figs. 1 A and C). Et-R-Hep and a few other analogues
inhibited the Alpha GSTs more effectively than the Mu isoenzymes (9).
This was confirmed by experiments in which the GSH conjugation of
(R,S)-2-bromoisovalerylurea (BIU) was measured in rat liver
cytosol and hepatocytes: the conjugation of only the (S)-BIU
enantiomer, which is preferentially conjugated by the Alpha GSTs, was
inhibited. The conjugation of (R)-BIU, which is mainly
catalyzed by the Mu GSTs, was unaffected (10).
View larger version (10K):
[in this window]
[in a new window]
|
Fig. 1.
Structures of the inhibitors.
Derivaties of R-Hep (A), Glu-Oct-Hep (B) and GSH (C) are shown. *
Indicates chiral atoms of which both configurations were used as a
mixture. Thus two stereoisomers are present in the compounds in (A) and
four in the compound in (B).
|
|
We report on the effect of esterification of the R-Hep backbone (see
fig. 1) on the potency of the resulting esters as GST inhibitors in the
perfused rat liver, using (R,S)-BIU as substrate. This
project was motivated by the results of experiments in which GSH itself
was esterified, for instance, on the cellular uptake of GSH itself.
Anderson et al. (11) have synthesized monoesters of GSH of
varying chain lengths (methyl to butyl derivatives), which all increase
the GSH levels in organs to a similar extent. They suggest that higher
monoesters might be more effective in increasing the cellular uptake of
GSH because of their greater lipophilicity. The GSH diethyl ester was
an even more effective agent for the delivery of GSH to a cell (12,
13). To design more effective in vivo inhibitors, we
therefore synthesized ester analogues of R-Hep with varying chain
length (C2, C5,
C12) at the -carboxyl group of D-Aad or an
ethyl group on the -carboxyl group of the glutamic acid moiety to
form the diester (fig. 1A). A derivative of Et-R-Hep in
which the ethyl ester was incorporated in the D-Aad backbone structure
(i.e. the
COO-C2H5 was replaced by a
C3H7, as in Glu-Oct-Hep)
was also tested (fig. 1B). This was based on the earlier
finding (14) that the glycine-carboxyl group of GSH was not required
for effective conjugation by GSTs: glycine could be replaced by
ethylamine. Therefore, the Glu-Oct backbone might be a simple
alternative to the esterified Glu-Aad backbone, the more so because it
was expected to be at least equally lipid soluble as the latter
esterified compound.
The effect of these modifications on the isoenzyme specificity of the
compounds towards purified rat GST isoenzymes was characterized, as
well as their inhibitory potency in isolated hepatocytes and in rat
liver cytosol using (R,S)-BIU as substrate for conjugation. In addition, we have studied their effect on GSH conjugation in the
recirculating in situ rat liver perfusion.
| |
Materials and Methods |
Materials.
Racemic BIU was purchased from Brocacef (Maarsen, The Netherlands). The
(R) and (S) enantiomers of BIU were prepared as
described earlier (15). The HPLC standard (RS)-IU-G was
prepared as described elsewhere (16). D-Aad, L-methionine, and BSA
(Fraction V) were from Sigma (St. Louis, MO). CDNB was from Janssen
Chimica (Beerse, Belgium). EPNP was purchased from Eastman Kodak Co.
(Rochester, NY). The inhibitors R-Hep, Et-R-Hep, and Do-R-Hep were
prepared as described earlier (9). The other inhibitors were
synthesized by similar methods published in detail in (17).
Animals.
Male Wistar rats (CRL: (WI)WU BR; SPF quality; 280-350 g body weight;
liver weight 9 - 16.7 g) obtained from Charles River (Sulzfeld,
Germany) were used for the study. They had free access to tap water and
a commercial diet (SRM-A; Hope Farms, Woerden, The Netherlands).
Inhibition of rat GST Isoenzymes.
Rat GST isoenzymes were isolated from liver and kidney as described by
Vos et al. (18) and Mannervik et al. (19), and purity was checked as described by Bogaards et al. (20). Rat Theta enzyme 5-5 was prepared as described by Meyer et al.
(21).
The rate of enzymatic of conjugation of CDNB (1 mM) with GSH (1 mM) in
the presence of 20-50 nM GST, was spectrophotometrically monitored at
37° in 0.1 M potassium phosphate buffer pH 7.4, containing 1 mM EDTA
and corrected for nonenzymatic reaction rate (22). The reaction rate
was determined in the presence and absence of 100 and 500 µM of the
inhibitors, which were added just before CDNB. Since Do-R-Hep was
insoluble at 500 µM, measurements were only performed at 100 µM.
All inhibitors were dissolved in DMSO (final concentration < 3%
(v/v)). The assay for GST 5-5 was performed using 0.5 mM EPNP as
electrophilic substrate and 10 mM GSH (21).
Incubations with Rat Liver Cytosolic GSTs.
Rat liver cytosol (4 mg/ml) (prepared as described in (9)) was
incubated with 50 µM of the inhibitors (in DMSO) or solvent (final
concn of DMSO 1.7%v/v), 1 mM GSH and 250 µM (R)- or
(S)-BIU in 50 mM sodium phosphate buffer, pH 7.4 at 37°
for 3 hr. To an aliquot (250 µl) of the incubation mixture 500 µl
ice-cold HPLC buffer A containing 60% (v/v) methanol was added. This
mixture was centrifuged and the resulting supernatant was stored at
20° until analysis by HPLC within 7 days.
Incubations with Isolated Hepatocytes.
Hepatocytes were isolated by collagenase perfusion as described earlier
(23). The inhibitors were dissolved in methanol/chloroform = 1/4
(v/v) and added to empty plastic incubation vials such that their final
concentration in the incubations would be 50 µM; then the solvent was
evaporated under a stream of nitrogen. Incubations were performed as
described previously (9): Cells (diluted to 3 × 106 cells/ml with Hanks/Hepes buffer, pH 7.4, containing 1.5% (w/v) BSA) were added to the vials and preincubated
with 1 mM L-methionine for 30 min. The experiment was
initiated by adding a solution containing either (R)- or
(S)-BIU to the cell suspension such that the final
concentration of substrate was 250 µM in an incubation volume of 3 ml. The incubations were performed in a rotary shaker (205 rev/min)
under an atmosphere of 95% O2/5%
CO2. Samples of the incubation mixture (250 µl)
were taken at 180 min and added to 500 µl of ice-cold HPLC buffer A
containing 60% (v/v) methanol to precipitate proteins. These were
removed by centrifugation prior to HPLC analysis. The viability of the
cells was assessed before and after the experiment by Trypan Blue
exclusion and was always > 85%. The GSH content of these cells,
as determined by Ellman's reagent (24), was always between 3.3 and 4.0 µmol/ml packed hepatocytes
Recirculating in Situ Rat Liver Perfusions.
Male rats were used as liver donors. The surgery was conducted under
pentobarbital anesthesia (60 mg/kg) as described previously (25) with a
few modifications. The perfusion medium consisted of 20% (v/v) washed
bovine red blood cells (courtesy of Slachthuis Leiden bv., Leiden, The
Netherlands), 1% (w/v) BSA, 0.3% (w/v) glucose in Krebs-Henseleit
bicarbonate solution, buffered to pH 7.4. Perfusate was oxygenated with
95% oxygen: 5% carbon dioxide and kept at 37° in a 500 ml rotating
reservoir (MX International, Aurora, CO). Liver viability during the
perfusion was assessed on the basis of gross appearance, bile flow
rate, and oxygen consumption rate. The GSH content of the liver at the
end of the perfusion ranged from 2.9 to 4.8 µmol/g of liver (mean of
3.7 ± 0.6) as determined by Ellman's reagent (26)
After surgery, the livers were perfused in single pass fashion for 10 min to stabilize the liver. Thereafter, perfusion was conducted for 105 min under a constant flow (10 ml/min) in a recirculating fashion: the
venous outflow from the liver was returned to the reservoir (total
perfusate volume was 100 ml). Immediately after onset of recirculation,
the inhibitor in 1 ml of 0.1 M HCl/saline (controls received only
solvent) was added to the reservoir, to yield a final concentration of
50 µM; for Et-R-Hep, an additional concentration of 500 µM was
used. The pH of the perfusion medium was not affected by the addition
of the acidic inhibitor solution. Ten min later, 25 µmol
(RS)-BIU, dissolved in 10 ml Krebs-Henseleit bicarbonate
solution, was added to the reservoir, to result in the initial BIU
concentration of 250 µM.
Bile was collected in fractions after addition of BIU. At the end of
the experiment, perfusate was sampled and analyzed for the presence
unchanged BIU and GSH conjugates. Bile and perfusate samples were
stored at 30° until analysis.
Quantitation of the GSH Conjugates of BIU.
The GSH conjugates of BIU were analyzed according to the HPLC method
with electrochemical detection as described by Te Koppele et
al. (27) with minor modifications. The separation of the diastereomeric GSH conjugates of BIU was performed using a glass Inertsil ODS-2 column (100 × 3 mm i.d.) from Chrompack (Bergen op
Zoom, The Netherlands), fitted with a Chrompack ODS-2 guard column
(10 × 2 mm). For the analysis, a Spectroflow model 400 pump
(Kratos Analytical, Ramsey, NJ) equipped with a Promis II autoinjector
(Spark Holland, Emmen, The Netherlands) was used. The eluent used
consisted of Buffer A/methanol = 82/18 (v/v) at a flow rate of
0.45 ml/min. Buffer A contained 0.1 M NaNO3, 0.01 M KBr, 0.01 M citric acid, 0.1 mM EDTA, and 0.1 mM 1-decane sulfonic acid.
Samples of the cytosol and hepatocyte incubations were diluted five
times before analysis; bile samples were diluted 25 times. Of the
diluted sample, 25 µl was injected onto the HPLC system and the
amount of GSH conjugates present was quantitated by using a solution of
(RS)-IU-G as an external standard.
| |
Determination of Octanol/Buffer Partition Coefficients |
Partition coefficients were determined by dissolving R-Hep and
Et-R-Hep in perfusion medium, pH 7.4 (without red blood cells and BSA),
and Pt-R-Hep, Do-R-Hep, and DiEt-R-Hep in 1-octanol to a concentration
of 200 µM. One ml of this solution was intensively vortexed for 1 min
with one ml of 1-octanol (for R-Hep and Et-R-Hep) or perfusion buffer
(the other compounds). The phases were separated by centrifugation. The
amount of compound present in both phases was determined by HPLC.
Of each phase, 20 µl was injected onto a Chrompack Inertsil ODS-2
column equipped with a guard column, as described above for the
analysis of BIU. The HPLC system consisted of a Promis autoinjector
coupled to two Kratos Spectroflow 400 pumps controlled by a Kratos
Spectroflow 450 solvent programmer (Kratos Analytical instruments,
Rotterdam, The Netherlands). Detection was at 214 nm using a
Spectroflow 783 programmable absorbance detector (Separations Analytical instruments, H.I. Ambacht, The Netherlands). The mobile phase consisted of 25 mM ammonium acetate buffer, pH 4.5 and
acetonitrile. Elution was isocratic, the amount of acetonitrile varying
per inhibitor, as shown in table 1. Each
compound yielded two peaks on HPLC, corresponding to two diastereomers.
| |
Protein Binding of the Inhibitors |
The extent of protein binding of the compounds in BSA containing
perfusate (without red blood cells) was determined by
ultracentrifugation. Perfusate (with or without BSA) was spiked with
200 µM of the inhibitors (dissolved in 0.1 M HCl/saline) and
incubated for 1 hr at 37°. The perfusate (400 µl) was then
centrifuged for 10 min at 1090 × g at 37° through Amicon YMT
membrane discs fitted in a MPS-1 Micropartition system (Amicon,
Dronten, The Netherlands). The amount of compound present in the
ultrafiltrate was determined by HPLC as described above for the
determination of partition coefficients.
Statistical Analysis.
The statistical significance of the results presented in figs.
2, 3,and
4 was determined using (a) an ANOVA
(indicating significant differences at p < 0.05)
followed by (b) Dunnett's test (SAS system for Windows, version 6.12, PROC GLM) The significance levels are indicated by p 
0.05 (*).
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3.
Effect of mono- or diesterification of R-Hep
on the biliary excretion of (RS)-BIU.
Livers were perfused in situ with (RS)-BIU (25 µmol)
and solvent, R-Hep, Et-R-Hep or DiEt-R-Hep. The cumulative biliary
excretion of the GSH conjugates of (S)-BIU,
(R)-BIU and (S)-BI after 90 min is shown
as mean ± SD of three experiments. * indicates values significantly different from control (p 0.05)
|
|
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of monoesters of varying chain length
on the biliary excretion of BIU GSH conjugates in the liver
perfusion.
25 µmol (RS)-BIU recirculated the perfused liver in
the presence of 50 µM Et-R-Hep, Pt-R-Hep or Do-R-Hep. The cumulative
excretion of the GSH conjugates of (S)-BIU,
(R)-BIU and (S)-BI in 90 min is shown as
means ± SD (N = 3). * indicates values
significantly different from control (p 0.05)
|
|
The statistical significance of the results presented in fig.
5 was determined using Shapiro-Wilk's
test which indicated that the data followed the normal distribution and
ANOVA (which indicated overall significant differences at
p < 0.05) followed by Tukey's multiple comparison set
at = 0.05 (SAS system for Windows, version 6.12, PROC GLM).
Significant differences are indicated at p = 0.05 (*).
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 5.
Inhibition of (S)-BIU conjugation in rat
liver cytosol and hepatocytes.
Rat liver cytosol or isolated rat hepatocytes were incubated with
(S)-BIU (250 µM) and 50 µM inhibitor. Means ± SD of three experiments are shown. Columns with the same letter are not
statistically significantly different.
|
|
| |
Results |
Inhibition of Rat GST Isoenzymes.
To characterize their isoenzyme selectivity, the inhibitors were tested
with purified rat GST isoenzymes (table
2). The analogues were most effective
towards the Alpha class isoenzymes, 1-1 and 2-2; the Mu class
isoenzymes 3-3 and 4-4 were inhibited to a lesser extent. Although
R-Hep was the most potent inhibitor, towards both Alpha and Mu class
isoenzymes (confirming earlier results (9)), the esterified analogues
were quite active as well: except for the diethylester DiEt-R-Hep, they
were almost equally inhibitory towards 1-1 and 2-2 as R-Hep. However,
towards the Mu isoenzymes 4-4 and, especially, 3-3 they were
considerably less active than R-Hep: at 500 µM they had only 50% of
the inhibitory potency of R-Hep (table 2).
Introduction of an ethyl ester at the -carboxyl group of D-Aad
(Et-R-Hep) resulted in a slightly decreased potency towards the Alpha
isoenzymes (71% to 68% inhibition for 1-1, and 100% to 62%
inhibition for 2-2 at 100 µM inhibitor) but a much more pronounced
decrease towards the Mu enzymes (from 51% to 4% inhibition for 3-3,
and 67% to 28% for 4-4). Introduction of a second ethyl ester group
at the -carboxyl group of the glutamic acid moiety (producing
DiEt-R-Hep) yielded an even less effective inhibitor, giving no more
than 37% inhibition of Alpha GSTs and only 5 and 23% inhibition of Mu
GSTs 3-3 and 4-4, respectively.
Glu-Oct-Hep, in which the
COOC2H5 moiety of the D-Aad
group of Et-R-Hep was replaced by a
C3H7 straight alkyl chain
(fig. 1), was equally potent as Et-R-Hep towards Alpha class GSTs and more effective towards GSTs of the Mu class.
None of the compounds inhibited GSTs 5-5 and 7-7 strongly. At 500 µM, DiEt-R-Hep inhibited the rate of conjugation of CDNB by GST 7-7
by 29%. Et-R-Hep, Pt-R-Hep, and Glu-Oct-Hep inhibited isoenzyme 5-5
by 20-27% at that concentration; since in the 5-5 assay the GSH
concentration was 10 mM (in stead of the 1 mM GSH concentration in the
assays for the other isoenzymes) the degree of inhibition of GST 5-5
may actually be underestimated as compared to the other isoenzymes.
The conjugation of the two BIU enantiomers was studied with pure GST
3-3. The Km values for the
(R)- and (S)-BIU enantiomers were 1.5 and 5.2 mM,
respectively, whereas the Km for GSH was 50 µM towards both substrates. Both R-Hep and Et-R-Hep inhibited the
enzyme in a competitive fashion with both GSH and the acceptor substrates. With GSH as variable substrate the
Ki of R-Hep was 44 ± 4 µM
(R-BIU; mean ± SD) and 27 ± 2 µM (S-BIU), whereas the Ki of Et-R-Hep was 967 ± 163 µM (R-BIU) and 273 ± 19 µM (S-BIU).
Inhibition of BIU GSH Conjugation by Rat Liver Cytosol and Isolated
Rat Hepatocytes.
The GSH conjugation of (R)-BIU was not inhibited by any of
the compounds, neither in hepatocyte incubations nor in rat liver cytosol. In contrast, all the analogues inhibited (S)-BIU
conjugation (fig. 5). R-Hep and the esterified analogues, Pt-R-Hep and
Do-R-Hep, were the most potent compounds in cytosol, causing some
65-70% inhibition. The mono- and diethyl ester compounds Et-R-Hep and DiEt-R-Hep were somewhat less effective (45-55% inhibition).
In hepatocytes, the mono- and diester analogues were equally potent,
but the unesterified, less lipophilic R-Hep, was less inhibitory (24%
versus 40-45% inhibition by the esterified compounds). Glu-Oct-Hep was by far the least effective towards BIU, in both cytosol
and hepatocytes.
Rat Liver Perfusions.
The effectiveness of the inhibitors in the intact liver was assessed in
a recirculating liver perfusion model, again using (RS)-BIU
as substrate. (R) and (S)-BIU undergo
stereoinversion when conjugated by GSTs to form (S)-IU-G and
(R)-IU-G, respectively. In the perfused liver, these are
excreted either into bile or perfusate (25, 26). (S)-BIU can
also undergo amidase catalyzed hydrolysis to
(S)-2-bromoisovaleric acid ((S)-BI) which can
then be conjugated to its GSH conjugate (R)-I-G and excreted
in bile.
Two concentrations of Et-R-Hep as added to the perfusion medium were
tested to determine the extent of inhibition of BIU conjugation in the
perfused liver. Based on earlier in vivo studies with
Et-R-Hep (10), we estimated that 500 µM would be required for
sufficient inhibition. However, a 50 µM concentration produced an
almost equally strong inhibition: the cumulative excretion of
(R)-IU-G was inhibited by approximately 50% (fig.
2A). As in cytosol and hepatocyte experiments, in liver
perfusions also, Et-R-Hep had no effect on (R)-BIU
conjugation (fig. 2B). None of the other inhibitors had any
effect on the conjugation of (R)-BIU (figs. 3 and 4).
The effect of esterification of the two carboxylate groups was assessed
by comparing the effect of the unesterified compound (R-Hep), the
monoethyl ester (Et-R-hep), and the diethyl ester (DiEt-R-Hep) (fig.
3). The unesterified R-Hep caused a statistically nonsignificant
decrease of 22% of the biliary excretion of the (S)-BIU
conjugate. Surprisingly, the diethylester showed only a 13%
statistically nonsignificant decrease.
The extent of inhibition of (S)-BIU conjugation by monoester
analogues of varying chain lengths was approximately equal: a 50-60%
inhibition was observed. Et-R-Hep was only slightly more effective than
Pt-R-Hep and Do-R-Hep (fig. 4). Again, (R)-BIU conjugation
was not affected. Glu-Oct-Hep was not evaluated in the rat liver
perfusion because it was only slightly effective towards BIU in cytosol
and hepatocytes.
The compounds had no statistically significant effect on the excretion
of (R)-I-G, the minor metabolite of (S)-BIU
(figs. 2, 3, 4).
After 90 min 42-63% of the dose administered was recovered in bile in
the form of GSH conjugates and 12-21% in the perfusate, mainly as
unconjugated BIU.
Octanol-Buffer Partition Coefficients and Protein Binding
Experiments.
Octanol-buffer partition coefficients and the extent of protein binding
of the compounds used in the perfusion experiments were determined
(table 3). As could be expected, the
unesterified R-Hep was highly water soluble, compared with the
monoesterified compounds, which had higher log p values with
increasing chain lengths. DiEt-R-Hep was of intermediate lipophilicity
between Et-R-Hep and Pt-R-Hep.
R-Hep and its mono- and diethyl ester analogues were bound to protein
to similar extents. Introduction of a longer alkyl chain, as in
Pt-R-Hep, resulted in considerably higher protein binding. Protein
binding of Do-R-Hep could not be measured because it formed aggregates
that could not pass through the ultrafiltration membrane even in the
absence of BSA. These aggregates had a diameter of 1000-5000 nm, as
determined by photon correlation spectroscopy, using a Malvern 4700 C
system (Malvern Instruments, Malvern, Worcs., UK), at 27° and a 90°
angle between laser and detector.
The extent of protein binding of all compounds was stereoselective, the
diastereomer eluting from the HPLC column first being bound to protein
much more strongly than the second. This effect was most pronounced for
Pt-R-Hep.
| |
Discussion |
These results indicate that against the GST isoenzymes themselves,
the unesterified R-Hep is the most effective inhibitor, while in
cellular systems esterified derivatives are more effective because they
penetrated the cells more readily.
Esterification presumably decreases the affinity of the inhibitor for
binding to the active site of GSTs. This is confirmed by the
Ki values of R-Hep and Et-R-Hep as
determined for isoenzyme 3-3 towards (S)- and
(R)-BIU: the values for R-Hep were one order of magnitude
lower than those for Et-R-Hep. The DiEt-R-Hep diester would be expected
to have an even lower affinity because the -carbonyl group of the
-Glu moiety (which is esterified in DiEt-R-Hep) engages into an
interaction with groups in the G-site of the enzyme (14). In the
monoester the ethyl group presumably decreases affinity because the
carboxyl group of the glycine moiety also interacts with the G-site
(14). That may be the reason why Glu-Oct-Hep, the analogue of Et-R-Hep
in which the ester function was replaced by a carbon-carbon bond (see
fig. 1B), was approximately as potent as the monoester
compounds toward GST 1-1 and 2-2. Yet, Et-R-Hep is still a potent
inhibitor, especially of the Alpha family members 1-1 and 2-2; this
seems in contrast to the results of Morgan et al.
(unpublished data quoted in (30)) who reported that monoester derivatives of their GSH analogues were very poor GST inhibitors.
Very few detailed studies are available on the kinetic mechanism of the
reaction catalyzed by the GSTs. The most extensive studies have been
performed with GST 3-3, which was also used in the present study.
Jakoby et al. (31) proposed a hybrid mechanism: a
combination of ordered sequential and ping-pong mechanism. Later work
indicated that the mechanism was most likely random sequential (32,
33). In intact cells, where the intracellular GSH concentration is
high, the mechanism is probably ordered, with GSH adding first (34).
Our results are in agreement with a random-order model: competitive
inhibition patterns towards both GSH and acceptor substrate indicates
that both substrates and inhibitor compete for the free enzyme. If the
mechanism were ordered, with GSH adding first, competition would be
observed only towards GSH as the variable substrate (35).
Esterification at the -carboxyl group of D-Aad, regardless of the
alkyl chain length strongly decreased the inhibitory potency towards Mu
class GST's, and to a lesser extent towards Alpha GST's. This
difference presumably is related to differences in the G-site between
the two isoenzyme families.
The preferential inhibition of Alpha as compared to Mu class GST's may
explain the results with the racemic substrate BIU. (S)-BIU
is conjugated more effectively by Alpha class GST and (R)-BIU by Mu class GST (36). Indeed, the predominantly
Alpha GST substrate (S)-BIU was inhibited by all compounds,
but (R)-BIU was not inhibited by any of the derivatives.
R-Hep was again the most potent inhibitor; the monoesters were slightly
less effective, and the diethylester was much less potent.
Surprisingly, Glu-Oct-Hep, which was as effective as the monoesters
with isoenzymes, was much less effective towards BIU in rat liver
cytosol. This most likely is because of the different substrates used:
CDNB was the electrophilic substrate in the isoenzyme studies, whereas
BIU was the substrate used in cytosol.
Inhibition of GST in Hepatocytes and Rat Liver Perfusion.
The inhibitory potency of the compounds in intact cells is a result of
three factors: the extent of cell penetration (dependent largely on
lipophilicity of the compound), the rate of hydrolysis of the ester
(yielding the most effective inhibitor, R-Hep), and the intrinsic
inhibitory potency of the compounds (esters versus R-Hep)
towards the various GSTs.
The lipophilicity of a compound is reflected by its octanol/buffer
partition coefficient. Mono- and diesterification of the charged, polar
carboxylic acid groups in R-Hep yielded more lipophilic compounds;
especially the compounds with long alkyl side chains, Pt-R-Hep and
Do-R-Hep. Thus they would be expected to penetrate into the cells most
efficiently.
Furthermore, usually only the fraction of the compound unbound to
protein (BSA) is available for uptake into cells. The more lipophilic
compounds are more tightly bound to BSA. Although the extent of protein
binding could not be measured for Do-R-Hep, we estimate that it is even
more strongly protein bound, because of the presence of an even longer
alkyl side chain. Because there is a complex interaction between the
factors lipophilicity and protein binding, a simple correlation between
the magnitude of these parameters and the extent of cell penetration
cannot be made.
Only (S)-BIU conjugation was inhibited in both isolated
hepatocytes and the perfused liver. In isolated hepatocytes, R-Hep was
much less effective than the esterified compounds, most likely because
of its lower lipophilicity, resulting in a slower uptake into the cell.
The monoester compounds were roughly equally effective in hepatocytes;
once they pass the cell membrane, they may be hydrolyzed by esterases,
to yield the same unesterified residue R-Hep, which is the most
effective inhibitor of Alpha class GST, or the esterified compounds as
such may inhibit GST (table 2).
Previously, the conjugation of BIU has been studied in detail in a
single-pass perfused rat liver system (29). Owing to limited
availability of the inhibitors, we have characterized the effect of the
inhibitors on the conjugation of (RS)-BIU in a recirculating
rat liver perfusion system to assess their relative potency towards GSH
conjugation in the intact liver. In accordance with the results in
hepatocytes and cytosol, (S)-BIU but not (R)-BIU conjugation was inhibited by all inhibitors, indicating isoenzyme selective inhibition of Alpha GSTs. For Et-R-Hep, we had already observed this in the rat in vivo (9).
The three monoesters were equipotent in the inhibition of
(S)-BIU conjugation; apparently, difference in
lipophilicity, protein binding, rate of hydrolysis, and their intrinsic
potency compensate so that the end result is equipotency. All of them
were more active than unesterified R-Hep, presumably owing to the low
lipophilicity of R-Hep. The same has been demonstrated for GSH itself;
monoethyl GSH is transported into the cell much more efficiently than
unesterified GSH (12). It has been reported, however, that for some GSH
conjugates and GSH analogues monoesterification does not enhance
cellular uptake (30, 37).
Diesterification of GSH, certain GSH conjugates, and GSH analogues (12,
30, 37) yields compounds that penetrate the cell even better than the
monoesters. Surprisingly, however, the diester of Et-R-Hep was less
effective in liver perfusions, although it was as effective as the
monoesters in hepatocytes.
In conclusion, Et-R-Hep is the most effective GST inhibitor in isolated
hepatocytes and liver perfusion; derivatives with longer alkyl chains
are slightly less potent. Towards purified isoenzymes or in rat liver
cytosol however, unesterified R-Hep is the more potent compound; it may
be the active form of the inhibitors inside the cell. These compounds
are useful tools to study the role of GSTs in biotransformation,
toxicity, and in physiological processes in vivo.
The authors thank Irma Meijerman and Marty Blom for their generous
supply of isolated hepatocytes and Mathieu Ouwerkerk for his assistance
with the statistical analysis of the results.
The work was supported by a grant from the Netherlands
Organization for Scientific Research (NWO; reg. no. 900-21-142).
| 1.
|
B. Mannervik,
Y. C. Awasthi,
P. G. Board,
J. D. Hayes,
C. Di Ilio,
B. Ketterer,
I. Listowski,
R. Morgenstern,
M. Muramatsu,
W. R. Pearson,
C. B. Pickett,
K. Sato,
M. Widersten, and
C. R. Wolf:
Nomenclature for human glutathione transferases.
Biochem. J.
282,
305-308 (1992).
|
| 2.
|
D. J. Meyer,
B. Coles,
S. E. Pemble,
K. S. Gilmore,
G. M. Fraser, and
B. Ketterer:
Theta, a new class of glutathione transferases purified from rat and man.
Biochem. J.
274,
409-414 (1991).
|
| 3.
|
Y. C. Awasthi,
R. Sharma, and
S. S. Singhal:
Human glutathione S-transferases.
Int. J. Biochem.
26,
295-308 (1994)[Medline].
|
| 4.
|
J. N. M. Commandeur,
G. J. Stijntjes, and
N. P. E. Vermeulen:
Enzymes and transport systems involved in the formation and disposition of glutathione S-conjugates.
Pharmacol. Rev.
47,
271-330 (1995)[Medline].
|
| 5.
|
B. Ketterer and
G. J. Mulder:
Glutathione conjugation. In
"Conjugation reactions in drug metabolism: An integrated Approach" (G. J. Mulder, ed.), pp. 307-364. Taylor and Francis, London, UK, 1990.
|
| 6.
|
D. J. Waxman:
Glutathione S-transferases: Role in alkylating agent resistance and possible target for modulation chemotherapy: a review.
Cancer Res.
50,
6449-6454 (1990)[Free Full Text].
|
| 7.
|
G. J. Mulder and
S. Ouwerkerk-Mahadevan:
Modulation of glutathione conjugation in vivo: how to decrease glutathione conjugation in vivo or in intact cellular systems in vitro.
Chem. Biol. Interact.
105,
17-34 (1997)[Medline].
|
| 8.
|
P. J. O'Dwyer,
F. LaCreta,
S. Nash,
P. W. Tinsley,
R. Schilder,
M. L. Clapper,
K. D. Tew,
L. Panting,
S. Litwin,
R. L. Comis, and
R. F. Ozols:
Phase I study of thiotepa in combination with the glutathione transferase inhibitor ethacrynic acis.
Cancer Res.
51,
6059-6065 (1991)[Abstract/Free Full Text].
|
| 9.
|
S. Ouwerkerk-Mahadevan,
J. H. van Boom,
M. C. Dreef-Tromp,
J. H. T. M. Ploemen,
D. J. Meyer, and
G. J. Mulder:
Glutathione analogues as novel inhibitors of rat and human glutathione S-transferase isoenzymes, as well as of glutathione conjugation in isolated rat hepatocytes and in rat in vivo.
Biochem. J.
308,
283-290 (1995).
|
| 10.
|
S. Ouwerkerk-Mahadevan,
J. H. van Boom, and
G. J. Mulder:
Isoenzyme-selective inhibition of glutathione conjugation in vivo: selective inhibition of the conjugation of S-2- bromoisovalerylurea in the rat.
J. Pharmacol. Exp. Ther.
276,
923-928 (1996)[Abstract/Free Full Text].
|
| 11.
|
M. E. Anderson,
F. Powrie,
R. M. Puri, and
A. Meister:
Glutathione monoethyl ester: preparation, uptake by tissues and conversion to glutathione.
Arch. Biochem. Biophys.
239,
538-548 (1985)[Medline].
|
| 12.
|
E. J. Levy,
M. E. Anderson, and
A. Meister:
Transport of glutathione diethyl ester into human cells.
Proc. Natl. Acad. Sci. USA
90,
9171-9175 (1993)[Abstract/Free Full Text].
|
| 13.
|
M. E. Anderson,
E. J. Levy, and
A. Meister:
Preparation and use of glutathione monoesters.
Meth. Enzymol.
234,
492-499 (1994)[Medline].
|
| 14.
|
A. E. P. Adang,
J. Brussee,
A. van der Gen, and
G. J. Mulder:
The glutathione binding site in glutathione S-transferases: Investigation of the cysteinyl, glycyl and  -glutamul domains.
Biochem. J.
269,
47-54 (1990)[Medline].
|
| 15.
|
J. M. Te Koppele,
S. W. Esajas,
J. Brussee,
A. van der Gen, and
G. J. Mulder:
Stereo-selective glutathione conjugation of -bromoisovaleric acid enantiomers in the rat in vivo and in rat liver subcellular fractions.
Biochem. Pharmacol.
37,
29-35 (1988)[Medline].
|
| 16.
|
J. M. Te Koppele,
E. J. van der Mark,
J. C. Olde Boerrigter,
J. Brussee,
A. van der Gen,
J. van der Greef, and
G. J. Mulder:
-Bromoisovaleryl urea as model substrate for studies on pharmacokinetics of glutathione conjugation in the rat. I. (Bio-)synthesis, analysis and identification of diastereomeric glutathione conjugates and mercapturates.
J. Pharmacol. Exp. Ther.
239,
898-904 (1986)[Abstract/Free Full Text].
|
| 17.
| S. Ouwerkerk-Mahadevan: Inhibition of glutathione conjugation
in vitro and in rat in vivo. Thesis, Leiden
University, Germany (1997)
|
| 18.
|
R. M. E. Vos,
M. C. Snoek,
W. H. J. van Berker,
F. Müller, and
P. J. van Bladeren:
Differential induction of rat hepatic glutathione S-transferase isoenzymes by hexachlorobenzene and benzyl isothiocyanate: comparison with induction of phenobarbital and 3-methylcholanthrene.
Biochem. Pharmacol.
37,
1077-1082 (1988)[Medline].
|
| 19.
|
B. Mannervik and
C. Guthenberg:
Glutathione transferase (human placenta).
Meth. Enzymol.
77,
231-235 (1981)[Medline].
|
| 20.
|
J. J. P. Bogaards,
B. van Ommen, and
P. J. van Bladeren:
An improved method for the separation and quantification of glutathione S-transferase subunits in rat tissue using high-pressure liquid chromatography.
J. Chromatogr.
474,
435-440 (1989)[Medline].
|
| 21.
|
D. J. Meyer,
B. Coles,
S. E. Pemble,
K. S. Gilmore,
G. M. Fraser, and
B. Ketterer:
Theta, a new class of GST purified from rat and man.
Biochem J.
274,
409-414 (1991).
|
| 22.
|
A. G. Clark,
J. W. Smith, and
T. W. Speir:
Cross specificity in some vertebrate and insect glutathione transferases with methyl parathion, 1-chloro-2.4-dinitrobenzene and S-crotonyl-N-acetylcystemic as substrates.
Biochem. J.
135,
385-392 (1973)[Medline].
|
| 23.
|
J. F. Nagelkerke,
K. P. Barto, and
T. J. C. van Berkel:
In vivo and in vitro uptake and degradation of acetylated low density lipoprotein by rat liver endothelial Kupffer and parenchymal cells.
J. Biol. Chem.
258,
12221-12227 (1983)[Abstract/Free Full Text].
|
| 24.
|
P. Dogterom,
G. J. Mulder, and
J. F. Nagelkerke:
Lipid peroxidation and independent protein thiol modifications in isolated rat hepatocytes.
Chem. Biol. Interact.
71,
291-306 (1989)[Medline].
|
| 25.
|
K. S. Pang and
J. R. Gilette:
Kinetics of metabolite formation and elimination in the perfused rat liver preparation: differences between the elimination of preformed acetaminophen formed form phenacetin.
J. Pharmacol. Exp. Ther.
207,
178-194 (1978)[Abstract/Free Full Text].
|
| 26.
|
Y. Zhao,
C. A. W. Snel,
G. J. Mulder, and
K. S. Pang:
Localization of glutathione conjugating activities towards bromosulfophthalein in perfused rat liver.
Drug Metab. Dispos.
21,
1070-1078 (1993)[Abstract].
|
| 27.
|
J. M. Te Koppele,
E. J. van der Mark, and
G. J. Mulder:
Liquid chromatographic determination of a -bromoisovalerylurea in rat and urine by electrochemically generated bromine.
J. Chromatogr.
427,
67-77 (1988)[Medline].
|
| 28.
|
M. Polhuijs,
D. K. F. Meijer, and
G. J. Mulder:
The fate of diastereomeric glutathione conjugates of -bromoisovalerylurea in blood in the rat in vivo and in the perfused liver.
J. Pharmacol. Exp. Ther.
256,
458-461 (1991)[Abstract/Free Full Text].
|
| 29.
|
M. Polhuijs,
I. Gasinka,
W. F. Cherry,
G. J. Mulder, and
K. S. Pang:
Stereoselectivity in glutathione conjugation and amidase catalyzed hydrolysis of the 2-bromoisoval erylurea enantiomers in the single-pass perfused rat liver.
J. Pharmacol. Exp. Ther.
265,
1406-1412 (1993)[Abstract/Free Full Text].
|
| 30.
|
A. S. Morgan,
P. J. Ciaccio,
K. D. Tew, and
L. M. Kauvar:
Isozyme specific glutathione S-transferase inhibitors potentiate drug sensivity in cultured human tumour cell lines.
Cancer Chemother. Pharmacol.
37,
363-370 (1996)[Medline].
|
| 31.
| M. J. Pabst, W. H. Habig, and W. B. Jakoby:
Glutathione S-transferase: A novel kinetic mechanism in
which the major reaction pathway depends on substrate concentration.
J. Biol. Chem. 249, 7140-7150.
|
| 32.
|
I. Jakobson,
M. Warholm, and
B. Mannervik:
Multiple inhibition of glutathione S- transferase A from rat liver by glutathione derivatives: kinetic analysis supporting a steady state random sequential mechanism.
Biochem. J.
177,
861-868 (1979)[Medline].
|
| 33.
| V. L. Schramm, R. McCluskey, F.A. S. Emig, and G. Litwack: Kinetic studies and active site binding properties of
glutathione S-transferase using spin-labeled glutathione, a product
analogue. J. Biol. Chem. 259, 714-722.
|
| 34.
| W.-J. Chen, C. C. Boehlert, K. Rider, and R. N. Armstrong: Synthesis and characterization of the oxygen and desthio
analogues of glutathione as dead-end inhibitors of glutathione
S-transferase. Biochem. Biophys. Res. Commun.
128, 230-240.
|
| 35.
|
T. Palmer:
Kinetics of multi-substrate enzyme-catalyzed reactions, In
"Understanding Enzymes" (A. Wiseman, ed.), 3rd ed., pp. 166-180. Ellis Horwood Ltd, Chichester UK, 1991.
|
| 36.
|
J. M. Te Koppele,
B. Coles,
B. Ketterer, and
G. J. Mulder:
Stereoselectivity of rat liver glutathione transferase isoenzymes for -bromoisovaleric acid and -bromo-isovalerylurea enantiomers.
Biochem. J.
252,
137-142 (1988)[Medline].
|
| 37.
|
T. W. C. Lo and
P. J. Thornalley:
Inhibition of proliferation of human leukaemia 60 cells by diethyl esters of glyoxalase inhibitors in vitro.
Biochem. Pharmacol.
44,
2357-2363 (1992)[Medline].
|
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Abstract
Introduction
Abstract
) or 500 µM (
) Et-R-Hep.
Values are mean ± SD of three experiments. Note that
(S)-BIU is converted to the (R)-IUG
conjugate because of stereoinversion during conjugation; the same
applies to the other enantiomers.