A new method for rapidly generating inhibitors of glyoxalase I inside tumor cells using S-(N-aryl-N-hydroxycarbamoyl)ethylsulfoxides

Article abstract of DOI:10.1021/jm980712o

The enediol analogue S-(N-p-chlorophenyl-N-hydroxycarbamoyl)glutathione is a powerful mechanism-based competitive inhibitor of the anticancer target enzyme glyoxalase I. Nevertheless, this compound exhibits limited toxicity toward tumor cells in vitro because it does not readily diffuse across cell membranes. We describe an efficient method for indirectly delivering the enzyme inhibitor into murine leukemia L1210 cells via acyl interchange between intracellular glutathione and the cell-permeable prodrug S-(N-p- chlorophenyl-N-hydroxycarbamoyl)ethylsulfoxide. The second-order rate constant for the acyl-interchange reaction in a cell-free system is 1.84 mM^{- 1} min^-1 (100 mM potassium phosphate buffer, 5% ethanol, pH 7.5, 25 °C). Incubation of L1210 cells with the sulfoxide in vitro results in a rapid increase in the intracellular concentration of the glyoxalase I inhibitor (k(app) = 1.41 ± 0.03 min^-1 (37 °C)) and inhibition of cell growth (GI₅₀ = 0.5 ± 0.1 μM). This represents an improvement in both efficiency and potency over the dialkyl ester prodrug strategy in which the inhibitor is indirectly delivered into tumor cells as the [glycyl,glutamyl] diethyl or dicyclopentyl esters. The fact that π-glutathione transferase catalyzes the acyl-interchange reaction between GSH and the sulfoxide suggests that the sulfoxide, or related compounds, might exhibit greater selective toxicity toward tumor cells that overexpress the transferase.

Full text of DOI:10.1021/jm980712o

J . Med. Chem. 1999, 42, 1823-1827
1823
A New Meth od for Ra p id ly Gen er a tin g In h ibitor s of Glyoxa la se I in sid e Tu m or
Cells Usin g S-(N-Ar yl-N-h yd r oxyca r ba m oyl)eth ylsu lfoxid es
Diana S. Hamilton,† Malcolm J . Kavarana,† Ellyn M. Sharkey,† J ulie L. Eiseman,‡ and Donald J . Creighton*,†
Department of Chemistry and Biochemistry, University of Maryland, Baltimore County, Baltimore, Maryland 21250, and
Division of Developmental Therapeutics, Greenebaum Cancer Center, and Department of Pathology, University of Maryland,
Baltimore, Maryland 21201
Received December 18, 1998
The enediol analogue S-(N-p-chlorophenyl-N-hydroxycarbamoyl)glutathione is a powerful
mechanism-based competitive inhibitor of the anticancer target enzyme glyoxalase I. Neverthe-
less, this compound exhibits limited toxicity toward tumor cells in vitro because it does not
readily diffuse across cell membranes. We describe an efficient method for indirectly delivering
the enzyme inhibitor into murine leukemia L1210 cells via acyl interchange between
intracellular glutathione and the cell-permeable prodrug S-(N-p-chlorophenyl-N-hydroxycar-
bamoyl)ethylsulfoxide. The second-order rate constant for the acyl-interchange reaction in a
cell-free system is 1.84 mM-1 min-1 (100 mM potassium phosphate buffer, 5% ethanol, pH 7.5,
25 °C). Incubation of L1210 cells with the sulfoxide in vitro results in a rapid increase in the
intracellular concentration of the glyoxalase I inhibitor (kapp ) 1.41 ( 0.03 min-1 (37 °C)) and
inhibition of cell growth (GI50 ) 0.5 ( 0.1 µM). This represents an improvement in both efficiency
and potency over the dialkyl ester prodrug strategy in which the inhibitor is indirectly delivered
into tumor cells as the [glycyl,glutamyl] diethyl or dicyclopentyl esters. The fact that
π-glutathione transferase catalyzes the acyl-interchange reaction between GSH and the
sulfoxide suggests that the sulfoxide, or related compounds, might exhibit greater selective
toxicity toward tumor cells that overexpress the transferase.
In tr od u ction
Glyoxalase I (GlxI) is receiving renewed interest as a
possible anticancer target, given its role in removing
methylglyoxal from cells and the observation that tumor
cells are exceptionally sensitive to the cytotoxic effects
of extracellular methylglyoxal.1-4 The enzyme metabo-
lizes methylglyoxal by catalyzing the conversion of the
thiohemiacetal, formed from glutathione (GSH) and
methylglyoxal, to S-D-lactoylglutathione via an enediol
proton-transfer mechanism. Glyoxalase II promotes the
subsequent hydrolysis of S-D-lactoylglutathione to D-
lactate and GSH. Years ago, Vince and Daluge hypoth-
esized that inhibitors of glyoxalase I might function as
antitumor agents by inducing elevated levels of intra-
cellular methylglyoxal.5
F igu r e 1. Enediol analogue S-(N-p-chlorophenyl-N-hydroxy-
carbamoyl)glutathione (1a , R ) H) and prodrugs 1a (Et)2 (R )
C2H5) and 1a (cyclopentyl)2 (R ) cyclopentyl). Compound 1a
is a tight-binding competitive inhibitor of human GlxI (Ki )
46 nM) and stable analogue of the enediol(ate) intermediate
that is thought to form along the reaction pathway of GlxI.9,10
into murine leukemia L1210 and B16 melanotic mela-
noma cells in culture, resulting in growth inhibition and
cell death.8 For example, the diethyl ester of one of these
enediol analogues (1a (Et)2) was found to inhibit the
growth of both L1210 and B16 cells, with GI50 values
of 7 and 15 µM, respectively (Figure 1). Compound
1a (Et)2 was also found to be much less toxic to nonpro-
liferating splenic lymphocytes, possibly reflecting re-
duced sensitivity to methylglyoxal and/or reduced chemi-
cal stability inside these cells.
Nevertheless, this prodrug strategy has at least two
limitations.8 First, the diffusion of the diethyl esters into
cells is a relatively slow process. The apparent first-
order rate constant for diffusion of 1a (Et)2 into L1210
cells in vitro is only about 0.10 min-1 at 37 °C, possibly
reflecting the fact that 1a (Et)2 is a cationic species under
physiological conditions. Second, the diethyl ester pro-
drugs undergo rapid esterase-catalyzed hydrolysis in the
In support of this prediction, Lo and Thornalley
demonstrated that the competitive inhibitor, S-p-bro-
mobenzylglutathione, is toxic to human leukemia 60
cells in vitro, once the inhibitor is delivered into the cells
as the lipophilic [glycyl,glutamyl] diethyl ester.6 This
prodrug strategy relies upon intracellular esterases to
catalyze the hydrolysis of the diethyl ester to give the
inhibitory diacid. The same laboratory reported that the
corresponding dicyclopentyl ester is about twice as
potent as the diethyl ester against a range of different
human tumor cell lines in culture.7 We subsequently
demonstrated that tight-binding enediol analogues of
the GlxI reaction could be delivered as the diethyl esters
* Corresponding author. Phone: (410) 455-2518. Fax: (410) 455-
2608.
† Department of Chemistry and Biochemistry.
‡ Greenebaum Cancer Center and Department of Pathology.
10.1021/jm980712o CCC: $18.00 © 1999 American Chemical Society
Published on Web 04/09/1999

1824 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 10
Hamilton et al.
Ta ble 1. Second-Order Rate Constants (k2) for
Sch em e 1. Acyl-Interchange Reaction between
Glutathione (GSH) and
S-(N-p-Chlorophenyl-N-hydroxycarbamoyl)ethylsulfoxide
(2a ) To Give Enediol Analogue 1a
Acyl-Interchange Reactions between Glutathione and Selected
Carbamoyl Estersa
CH3CH2XC(O)N(Y)Z + GSH f GSC(O)N(Y)Z + CH3CH2XH
substituent
carbamoyl
ester
k2
X
Y
Z
(mM-1 min-1
)
3a
2a
2b
2c
S
OH
C6H4Cl
C6H4Cl
CH3
∼3.5 × 10-7 b
1.84 ( 0.07c
0.59 ( 0.02c
0.14 ( 0.01c
S(O)
S(O)
S(O)
OH
OH
CH3
C6H4Cl
Sch em e 2. Synthetic Route to the Acylating Reagents
a
Shown in Table 1a
Conditions: potassium phosphate buffer (0.1 M), pH 7.5, 5%
ethanol, 25 °C. Calculated from the initial rate of appearance of
b
product (∼0.1% reaction) in the presence of a 15-fold excess of
glutathione, monitored by reverse-phase high-performance liquid
chromatography. c Calculated from the first-order rate of loss of
reactants under pseudo-first-order conditions (g10-fold excess of
glutathione), monitored spectrophotometrically.
a
Reagents and conditions: (i) K2CO3/Et2O; (ii) for 3a ,b, m-
chloroperbenzoic acid was used; for 3c, Oxone (2KHSO5‚KHSO4‚
K2HSO4) was used.
plasma of most inbred strains of laboratory mice,
complicating efficacy studies using murine models.
Here we show that enediol analogue 1a can be rapidly
generated inside L1210 cells via an acyl-interchange
reaction between intracellular GSH and the cell-perme-
able prodrug S-(N-p-chlorophenyl-N-hydroxycarbam-
oyl)ethylsulfoxide (2a ), Scheme 1. This prodrug strategy
has important advantages over the dialkyl ester prodrug
strategy.
F igu r e 2. Time-dependent change in the intracellular con-
centration of S-(N-p-chlorophenyl-N-hydroxycarbamoyl)glu-
tathione (1a ) in L1210 cells incubated in the presence of 0.19
mM S-(N-p-chlorophenyl-N-hydroxycarbamoyl)ethylsulfoxide
(2a ) (0) and in the presence of 0.01 mM 2a (4). Error bars
represent standard deviations for quadruplicate determina-
tions. Conditions: RPMI 1640 medium containing 10% heat-
inactivated fetal calf serum, 37 °C.
Ch em istr y
The dicyclopentyl ester of enediol analogue 1a was
obtained by acid-catalyzed esterification of 1a in cyclo-
pentanol/HCl. The synthetic route to the acylating
reagents used with GSH to produce different enediol
analogues is outlined in Scheme 2. Oxidation of thioesters
3a -c to the corresponding sulfoxides 2a -c followed
published methods.11,12 N-(p-Chlorophenyl)hydroxyl-
amine was prepared by reduction of p-chloronitroben-
zene with hydrazine hydrate in the presence of 5% Rh
on carbon.13
products of the interchange reactions to be the corre-
sponding enediol analogues.
Cell P er m ea bility Stu d ies. Thus, enediol analogue
should rapidly form inside tumor cells incubated in the
presence of sulfoxide, provided that the sulfoxide can
diffuse across the cell membrane and intracellular GSH
is free to react with the sulfoxide. Indeed, incubation of
L1210 cells in the presence of 0.19 mM 2a results in
the rapid appearance of intracellular enediol analogue
1a , Figure 2. Compound 1a was isolated from cell
extracts by reverse-phase C18 column chromatography
and quantitated by interpolation from standard curves.
The maximum concentration of 1a (1.6 mM) observed
after about 20 min is close to the reported concentration
of GSH in L1210 cells (1.9 mM),14 indicating that most
of the intracellular GSH is free to react with 2a .
Moreover, the maximum intracellular concentration of
enediol analogue exceeds the extracellular concentration
of the sulfoxide by 8-fold, indicating that the rate
constant for efflux of the enediol analogue is much less
than that for influx of the sulfoxide. Similarly, incuba-
tion of the cells with 0.01 mM sulfoxide for 30 min
results in a 13-fold higher concentration of intracellular
1a , Figure 2. Thus, the enediol analogue can be con-
centrated inside tumor cells under conditions where the
Resu lts a n d Discu ssion
Ch em ica l Kin etics. The studies described in this
report were originally inspired by the observation that
intracellular sulfoxidation of herbicidal N,N-di-n-pro-
pylcarbamate thioalkyl esters promotes their catabolism
by facilitating acyl interchange with GSH.11 We rea-
soned that sulfoxides of N-aryl/alkyl-N-hydroxycarbam-
ate thioethyl esters might also undergo rapid acyl
interchange with GSH, providing an efficient method
for generating enediol analogue inhibitors of glyoxalase
I inside tumor cells. Indeed, sulfoxides 2a ,b are at least
106-fold more reactive than the thioester 3a toward
GSH, on the basis of a comparison of the second-order
rate constants for the acyl-interchange reactions, Table
1. The N-OH group inductively activates the sulfoxides
for acyl interchange, as the rate constant for 2a exceeds
that of the corresponding N-CH3 derivative 2c by 13-
fold. Ultraviolet and NMR spectroscopy confirmed the

New Method for Generating Inhibitors of Glyoxalase I
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 10 1825
Ta ble 2. Apparent First-Order Rate Constants for Delivery of
Enediol Analogue 1a into L1210 Cells via Different Prodrugs
and Comparative Toxicities of the Prodrugs to L1210 Cellsa
rate constant
GI50
TGI
LC50
b
prodrug
(min-1
)
(µM)c
(µM)c
(µM)c
2a
1.41 ( 0.03
0.5 ( 0.1 3.9 ( 0.4 13 ( 8
3.4 ( 0.1 5.3 ( 0.1 8.3 ( 1.0
1a (cyclopentyl)2 0.59 ( 0.01
1a (Et)2
0.10 ( 0.01d
7 ( 3d
27 ( 10d 54 ( 5d
a
The mean and standard deviations for the cell toxicity
parameters for each prodrug were calculated from three different
experiments carried out on different days. Conditions: RPMI 1640/
b
10% fetal calf serum, 37 °C. Derived from the slopes of plots of
initial rates of appearance of the inhibitory diacid 2 inside L1210
cells versus the extracellular concentration of prodrug. c Defined
d
in the legend to Figure 4. From ref 8.
F igu r e 4. Growth inhibition of L1210 cells by S-(N-p-
chlorophenyl-N-hydroxycarbamoyl)ethylsulfoxide (2a ) (RPMI
1640/10% fetal calf serum, 37 °C, 48 h). Percentage growth
(PG) was calculated from the following relationships: When
(Ftest - Ftzero) g 0, PG ) 100 × (Ftest - Ftzero)/(Fctrl - Ftzero), where
Ftzero ) cell density prior to exposure to drug, Ftest ) cell density
after 48 h exposure to drug, Fctrl ) cell density after 48 h with
no exposure to drug. When (Ftest - Ftzero) < 0, PG ) 100 × (Ftest
- Ftzero)/Ftzero. Different symbols correspond to experiments
carried out on different days. Error bars represent standard
deviations for triplicate determinations. The toxicity param-
eters were obtained by interpolation from the dose-response
curve and are defined as follows: GI50, concentration producing
50% growth inhibition in comparison to no-drug controls; TGI,
concentration producing 100% growth inhibition; LC50, con-
centration producing 50% cell killing.
F igu r e 3. Initial rates of appearance of intracellular S-(N-
p-chlorophenyl-N-hydroxycarbamoyl)glutathione (1a ) in L1210
cells as a function of the extracellular concentration of sulfox-
ide 2a . Error bars represent standard deviations for triplicate
determinations. Conditions: RPMI 1640 medium containing
10% heat-inactivated fetal calf serum, 37 °C.
noted, the minimum concentration of GSH in the L1210
cells used in this study is 1.6 mM. Therefore, the
concentration of GSH will be reduced by less than 1%
in the presence of a sulfoxide concentration correspond-
ing to the GI50 (0.5 µM), assuming that the maximum
achievable concentration of enediol analogue inside the
cells will be about 10-fold greater than that of the
extracellular sulfoxide. In principle, the toxicity of the
sulfoxide might arise from acylation of critical sulfhy-
dryl-containing proteins inside the cell. However, this
possibility seems less likely, given that the concentra-
tion of free GSH inside the cells is orders of magnitude
higher than that of protein-associated sulfhydryl groups.
Therefore, the sulfoxide is much more likely to react
with GSH than with intracellular proteins. Importantly,
the GI50 value for the sulfoxide is significantly lower
than those of the diethyl ester 1a (Et)2 and dicyclopentyl
ester 1a (cyclopentyl)2, Table 2. This can be explained
by the greater rate at which the sulfoxide enters the
cells in comparison to the dialkyl esters.
Su b st r a t e P r op er t ies w it h GSH Tr a n sfer a se.
N,N-Di-n-propyl ethylsulfoxide has been reported to be
a substrate for the GSH transferase in plant cells.11 This
prompted us to examine the kinetic properties of sul-
foxide 2a with the π isozyme of GSH transferase from
human placenta. Multidrug-resistant tumors of the
colon, lung, and breast commonly overexpress the π
isozyme of GSH transferase.16 Therefore, resistant
tumors might be exceptionally sensitive to sulfoxides
such as 2a , provided that the sulfoxides are catalytically
converted to the inhibitory enediol analogue.
extracellular concentration of sulfoxide is low. The
explanation for the decrease in concentration of enediol
analogue after 20 min is not known but could involve
several factors including enzymatic degradation and/or
export of the enediol analogue from the cells by the
GSH-conjugate export pump that is present in L1210
cells.15
The sulfoxide prodrug strategy is more efficient than
the dialkyl ester prodrug strategy at delivering enediol
analogue 1a into L1210 cells. The apparent first-order
rate constant for formation of 1a in L1210 cells incu-
bated with sulfoxide is 14-fold larger than that with the
diethyl ester prodrug 1a (Et)2 and 2-fold larger than that
with the dicyclopentyl ester 1a (cyclopentyl)2, Table 2.
These rate constants were derived from the slopes of
plots of the initial rates of appearance of 1a inside the
cells versus extracellular [prodrug], e.g., Figure 3.
Cytotoxicity Stu d ies. Sulfoxide 2a is both cytostatic
and cytotoxic to L1210 cells in tissue culture, as
reflected in the GI50 and LC50 values obtained by
interpolation from the dose-response curve of Figure
4. We believe that this most likely reflects rapid reaction
of intracellular sulfoxide with GSH to give enediol
analogue, which induces elevated levels of methylgly-
oxal in the L1210 cells by inhibiting glyoxalase I. The
toxicity of the sulfoxide is unlikely to be due to depletion
of intracellular GSH, because the total concentration of
GSH in L1210 cells is large in comparison to the value
of the growth inhibition parameter (GI50). As previously
Sulfoxide 2a proved to be a poor substrate for the
transferase from human placenta with an estimated kcat
/
Km value of 54 mM-1 min-1. This value was derived
from the first-order rates of loss of 2a (∆OD260) in the

1826 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 10
Hamilton et al.
aromatic-H); HR FAB-MS consistent with the molecular
formula of 2a (cyclopentyl)2 (C27H38N4O8SCl). Anal. 2(cyclo-
pentyl)2(HCl salt)‚3H2O (C27H44N4O11SCl2): C calcd 46.14,
found 45.51; H calcd 6.26, found 6.14; N calcd 7.97, found 7.17.
presence and absence of known amounts of trans-
ferase: conditions of 0.09 mM potassium phosphate
buffer (pH 7.5), 5% ethanol, 0.1 mM GSH (saturating
concentration), 0.005 mM 2a , 25 °C. This represents a
rate enhancement of about 30-fold, on the basis of a
comparison with the second-order rate constant for the
nonenzymic acyl-interchange reaction (Table 1). Thus,
transferase activity is unlikely to have a major influence
on the steady-state concentration of enediol analogue
derived from 2a in tumor cells containing high concen-
trations of GSH (∼2 mM) and low levels of transferase
activity. Nevertheless, this observation emphasizes the
potential importance of using the transferase to gener-
ate enediol inhibitors of glyoxalase I inside tumor cells.
Ser u m Sta bility. Finally, sulfoxide 2a exhibits rea-
sonable stability in serum samples obtained from DBA/2
mice, routinely used to evaluate the in vivo efficacy of
drugs against L1210 cells. The first-order rate constant
for loss of sulfoxide in serum samples from these mice
is 0.06 ( 0.01 min-1 (T1/2 ) 13 ( 2 min), n ) 3. This
compares with a half-life of less that 30 s for 2(Et)2, due
to the high levels of esterase activity in plasma.8 The
stability of the sulfoxide, together with its rapid rate of
diffusion into L1210 cells, should optimize the chances
of delivering toxic levels of drug to L1210 cells implanted
in tumor-bearing mice.
N-Hyd r oxy-N-p-ch lor op h en ylca r ba m a te Th ioeth yl Es-
ter (3a ). A solution of thioethyl chloroformate (2.46 g, 20
mmol) in diethyl ether (10 mL) was added dropwise to a ice-
cold stirring mixture of N-p-chlorophenylhydroxylamine (2.83
g, 20 mmol), K2CO3 (1.36 g, 10 mmol), diethyl ether (40 mL),
and water (2 mL) over a period of 10 min. The reaction mixture
was allowed to come to room temperature and followed to
completion by TLC (∼180 min). The ether layer was removed,
washed once with 5% aqueous HCl and water, and dried over
anhydrous MgSO4. The solvent was removed in vacuo and the
residue recrystallized from benzene/petroleum ether to give
the final product as transparent needles: yield 83%; mp 119
°C dec; IR (KBr) 1600, 1490, 1400, 1340, 1070, 820 cm-1; 300
MHz 1H NMR (CDCl3, TMS) δ 1.32 (t, J ) 7.5 Hz, methyl-
H3), 2.90 (q, J ) 7.5 Hz, methylene-H2), 6.68 (bs, OH), 7.34 (d,
J ) 8.7 Hz, p-chlorophenyl ring meta-H2), 7.52 (d, J ) 8.7 Hz,
p-chlorophenyl ring ortho-H2). Anal. (C9H10NO2SCl): C; H
calcd 4.32, found 4.29; N calcd 6.05, found 6.00.
N-Hyd r oxy-N-m eth ylca r ba m a te Th ioeth yl Ester (3b).
This compound was prepared by the same general method used
in the preparation of 3a : yield 70%; IR (KBr) 3220, 2920, 1620,
1
1400, 1265, 1200, 1080, 900 cm-1; 300 MHz H NMR (CDCl3,
TMS) δ 1.71 (t, J ) 7.5 Hz, methyl-H3), 3.29 (q, J ) 7.5 Hz,
methylene-H2), 3.70 (s, methyl-H3), 7.08 (bs, 1H), HR FAB-
MS consistent with C4H9NO2S. Anal. (C4H9NO2S): C; H calcd
6.67, found 6.72; N calcd 10.37, found 10.27.
N-Meth yl-N-p-ch lor op h en ylca r ba m a te Th ioeth yl Es-
ter (3c). This compound was prepared by the same general
method used in the preparation of 3a : yield 82%, mp 60-61
°C; IR (KBr) 2970, 1650, 1580, 1480, 1400, 1345, 1260, 1085,
Con clu sion s
Acyl interchange between intracellular GSH and
substituted (N-hydroxycarbamoyl)alkylsulfoxides is a
new prodrug strategy for indirectly delivering mecha-
nism-based competitive inhibitors of glyoxalase I into
tumor cells. In principle, this method could be expanded
to include any S-conjugate of GSH. Sulfoxide prodrugs
offer advantages over dialkyl ester prodrugs in terms
of increased rates of intracellular delivery, a corre-
sponding increase in potency, and the potential to
undergo GSH transferase-catalyzed conversion to the
inhibitory enediol analogues.
1
1010, 840 cm-1; 300 MHz H NMR (CDCl3, TMS) δ 1.24 (t, J
) 7.2 Hz, methyl-H3), 2.86 (q, J ) 7.2 Hz, methylene-H2), 3.31
(s, methyl-H3), 7.19 (d, J ) 8.4 Hz, 2H), 7.38 (d, J ) 8.8 Hz,
2H); HR FAB-MS consistent with C10H12NOSCl. Anal. (C10H12
-
NOSCl): C; H calcd 5.23, found 5.28; N calcd 6.10, found 6.01.
S-(N-H yd r oxy-N-p -ch lor op h en ylca r b a m oyl)et h ylsu l-
foxid e (2a ). A solution of m-chloroperbenzoic acid (2.12 g, 12.3
mmol) in diethyl ether (60 mL) was added dropwise to a ice-
cold stirring solution of 3a (2.85 g, 12.3 mmol) dissolved in
diethyl ether (30 mL) over a period of 15 min. The reaction
mixture was allowed to come to room temperature and stirred
for an additional 30 min. The precipitate was removed from
the reaction mixture by filtration, thoroughly washed with
diethyl ether, and allowed to dry. The final product is a white
powder: yield 96%; mp 143-144 °C dec; IR (KBr) 1700, 1490,
1410, 1350, 1260, 1090, 1000, 800 cm-1; 300 MHz 1H NMR
(DMSO-d6, TMS) δ 1.20 (t, J ) 7.5 Hz, methyl-H3), 7.54 (d, J
) 9.2 Hz, 2H), 7.71 (d, J ) 9.2 Hz, 2H), 11.69 (bs, OH); HR
FAB-MS consistent with C9H10NO3SCl. Anal. (C9H10NO3S-
Cl): C; H calcd 4.07, found 4.12; N calcd 5.66, found 5.57.
Exp er im en ta l Section
Synthetic methods are outlined in Scheme 2. NMR spectra
were taken on a GE QE-300 NMR spectrometer. Mass spectral
data were obtained at the Midwest Center for Mass Spectrom-
etry, University of Nebraska-Lincoln. Elemental analyses
were obtained at Atlantic Microlabs, Inc., Norcross, GA, and
are within (0.4% of the calculated values unless otherwise
indicated. Glutathione transferase was purchased from Sigma
Chemical Co., and excess glutathione was removed by filtration
through a Centricon filter.
S -(p -Ch lor op h e n yl-N -h yd r oxyca r b a m oyl)glu t a t h i-
on e Dicyclop en tyl Ester (2a (cyclop en tyl)2). This com-
pound was prepared by acid-catalyzed esterification of 2a ,
prepared as previously described,8 in cyclopentanol/HCl (2.3
N) (2 days, room temperature). The orange-colored oil was
dissolved in a minimum of 40% ethanol in water and decol-
orized by stirring with activated charcoal overnight. The
charcoal was removed by filtration through Celite and the
filtrate brought to dryness in vacuo to give the HCl salt of the
dicyclopentyl ester as a colorless oil. Yield > 90%. The product
was estimated to be greater than 98% pure by reverse-phase
C18 HPLC, using methanol/water (55:45) containing 0.25%
acetic acid as an eluting solvent: 300 MHz 1H NMR (D2O,
DSS) δ 1.5-2.0 (16 H, m, cyclopentyl-CH2), 2.21 (2H, m, Glu-
CâH2), 2.54 (2H, m, Glu-Cγ H2), 3.20 (1H, q, Cys-CâHa, J )
7.8, 14.5 Hz), 3.42 (1H, q, Cys-CâHb, J ) 5.1, 14.5 Hz), 4.01
(2H, s, Gly-CH2), 4.08 (1H, t, Glu-CRH, J ) 6.0 Hz), 4.67 (1H,
m, Cys-CRH), 5.18 (2H, m, cyclopentyl-CH), 7.3-7.5 (4H, m,
S-(N-Hydr oxy-N-m eth ylcar bam oyl)eth ylsu lfoxide (2b).
This compound was prepared by the same general method used
to prepare 2a : yield 86%; mp 110-112 °C; IR (KBr) 2700,
1710, 1380, 1190, 1010, 970, 840 cm-1; 300 MHz 1H NMR
(DMSO-d6, TMS) δ 1.16 (t, J ) 7.5 Hz, methyl-H3), 2.80-3.05
(m, methylene-H2), 3.23 (s, methyl-H3), 10.71 (s, 1H); HR FAB-
MS consistent with C4H9NO3S. Anal. (C4H9NO3S): C; H calcd
5.96, found 6.03; N calcd 9.27, found 9.21.
S-(N-Meth yl-N-p-ch lor op h en ylca r ba m oyl)eth ylsu lfox-
id e (2c). To an ice-cold stirring solution of 3c (0.5 g, 2.2 mmol)
in methanol (15 mL) was added dropwise over 10 min a
solution of Oxone (Aldrich; 0.87 g of 2KHSO5‚KHSO4‚K2SO4,
containing 2.8 mmol of KHSO5) in water (25 mL). The
resulting slurry was stirred at room temperature for 1.5 h.
The mixture was diluted to 80 mL with water and extracted
three times with methylene chloride (70 mL). The combined
organic layers were washed once with water and once with
brine and dried over anhydrous Na2SO4, and the solvent was
removed in vacuo to give a clear oil as crude product. The

New Method for Generating Inhibitors of Glyoxalase I
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 10 1827
product was purified by flash chromatography on a silica gel
column (CH2Cl2:methanol (40:1)), the solvent was removed
from the peak fractions, and the product was recrystallized
from CH2Cl2/petroleum ether to give the final product as
colorless cubic crystals: yield 43%; mp 113-115 °C; IR (KBr)
1680, 1480, 1350, 1245, 1050, 1000, 840 cm-1; 300 MHz 1H
NMR (CDCl3, TMS) δ 1.23 (t, J ) 7.2 Hz, methyl-H3), 2.98-
3.18 (m, methylene-H2), 3.42 (s, methyl-H3), 7.24 (d, J ) 8.4
Hz, 2H), 7.45 (d, J ) 8.4 Hz, 2H). Anal. (C10H12NO2SCl): C
calcd, 48.88, found 48.62; H calcd 4.89, found 4.86; N calcd
5.70, found 5.62.
Cell P er m ea bility Stu d ies. To suspensions of L1210 cells
(∼1.5 × 106 cells/mL) in RPMI 1640 medium, containing 10%
heat-inactivated fetal calf serum, gentamycin (10 µg/mL), and
L-glutamate, at 37 °C, were added fixed concentrations of 2a .
Aliquots (1 mL) were removed from the cell suspension as a
function of time, overlaid onto 0.4 mL of silicone oil (SF-1250
silicone fluids; General Electric Co., Silicone Products Division,
Waterford, NY) contained in an Eppendorf centrifuge tube, and
centrifuged at 11000g (15 min) at 25 °C. The supernatant and
silicone oil were decanted from the cell pellet, and residual oil
was removed from the inside of the centrifuge tube with a
cotton swab. To the pellet was added 1 mL of 70% ethanol in
water, and the suspension was sonicated for 2 min. The
denatured protein was sedimented by centrifugation (11000g,
15 min), and the clear supernatant was removed. The super-
natant was brought to dryness under a stream of nitrogen,
and the resulting residue was fractionated by reverse-phase
HPLC (Waters µBondapak C18, 0.78 × 30 cm), using a mobile
phase composed of methanol:water (1:1) containing 0.25%
acetic acid at a flow rate of 2 mL/min. The integrated intensity
of the peak corresponding to S-(N-p-chlorophenyl-N-hydroxy-
carbamoyl)glutathione (1a ) was determined at λ ) 260 nm.
Concentrations were extrapolated from standard curves that
were linear in the range 0.1-4.0 nmol (r ) 0.997).
Cytotoxicity Stu dies. Murine lymphocytic leukemia (L1210,
G050141) cells were obtained from the NCI, DCT, Tumor
Repository (Frederick, MD) and were maintained in RPMI
1640 medium containing L-glutamate (Gibco BRL, Gaithers-
burg, MD), supplemented with 10% heat-inactivated fetal calf
serum and gentamycin (10 µg/mL), under 37 °C humidified
air containing 5% CO2. The L1210 cells have a doubling time
of approximately 14 h. Cells in logarithmic growth were
introduced into 96-well tissue culture plates at a density of
5000 cells/well (0.15 mL) in the absence and presence of at
least five different concentrations of drug, spanning the IC50
concentration (in triplicate). Fresh drug was added to the
media every 12 h, to compensate for spontaneous hydrolysis
of the drug (T1/2 ) 4.5 h). After a 48-h incubation period, cell
densities were determined with the use of a Coulter Counter
(Model ZBI, Coulter Electronics). Cell viability was determined
by the trypan blue exclusion method.17
Sta bility of Su lfoxid e 2a in Mou se Ser u m . To 0.5-mL
portions of serum from DBA/2 mice, at 37 °C, was added
sulfoxide 2a to an initial concentration of approximately 0.8
mM. As a function of time, 0.1-mL aliquots of the incubation
mixture were transferred to separate microfuge tubes. The
samples were immediately deproteinized by the addition of
70% ethanol (0.9 mL), and the protein precipitate sedimented
by centrifugation at 13000g. The supernatants were then
was determined at 259 nm. The rate constant for decomposi-
tion was calculated from the first-order loss of 2a as a func-
tion of time.
Ack n ow led gm en t. This work was supported by a
grant from the National Institutes of Health (CA 59612).
Refer en ces
(1) Thornalley, P. J . Advances in glyoxalase research. Glyoxalase
expression in malignancy, anti-proliferative effects of methyl-
glyoxal, glyoxalase
I inhibitor diesters and S-D-lactoylglu-
tathione, and methylglyoxal-modified protein binding and en-
docytosis by the advanced glycation end product receptor. Crit.
Rev. Oncol. Hematol. 1994, 20, 99-128.
(2) Vander J agt, D. L. The Glyoxalase system. In Coenzymes and
Cofactors: Glutathione; Dolphin, D., Poulson, R., Avramovic, O.,
Eds.; J ohn Wiley and Sons: 1989; Vol. 3 (part A), pp 597-641.
(3) Creighton, D. J .; Pourmotabbed, T. Glutathione-dependent al-
dehyde oxidation reactions. In Molecular Structure and Energet-
ics: Principles of Enzyme Activity; Liebman, J . F., Greenberg,
A., Eds.; VCH Publishers: New York, 1988, Vol. 9, pp 353-386.
(4) Creighton, D. J .; Murthy, N. S. R. K. Stereochemistry of enzyme-
catalyzed reactions at carbon. In The Enzymes, 3rd ed.; Sigman,
D. S., Boyer, P. D., Eds.; Academic Press: New York, 1990; Vol.
19, pp 323-421.
(5) Vince, R.; Daluge, S. A possible approach to anticancer agents.
J . Med. Chem. 1971, 14, 35-37.
(6) Lo, T. W. C.; Thornalley, P. J . Inhibition of proliferation of
human leukaemia 60 cells by diethyl esters of glyoxalase
inhibitors in vitro. Biochem. Pharmacol. 1992, 44, 2357-2363.
(7) Thornalley, P. J .; Ladan, M. J .; Ridgeway, S. J . S.; Kang, Y.
Antitumor activity of S-(p-bromobenzyl)glutathione diesters in
vitro: a structure activity study. J . Med. Chem. 1996, 39, 3409-
3411.
(8) Kavarana, M. K.; Kovaleva, E. G.; Creighton, D. J .; Wollman,
M. B.; Eiseman, J . L. Mechanism-based competitive inhibitors
of glyoxalase I: intracellular delivery, in vitro antitumor activi-
ties, and stabilities in human serum and mouse serum. J . Med.
Chem. 1999, 42, 221-228.
(9) Hamilton, D. S.; Creighton, D. J . Inhibition of glyoxalase I by
the enediol mimic S-(N-hydroxy-N-methylcarbamoyl)glutathi-
one: the possible basis of a tumor-selective anticancer strategy.
J . Biol. Chem. 1992, 267, 24933-24936.
(10) Murthy, N. S. R. K.; Bakeris, T.; Kavarana, M. J .; Hamilton, D.
S.; Lan, Y.; Creighton, D. J . S-(N-Aryl-N-hydroxycarbamoyl)-
glutathione derivatives are tight-binding inhibitors of glyoxalase
I and slow substrates for glyoxalase II. J . Med. Chem. 1994, 37,
2161-2166.
(11) Casida, J . E.; Gray, R. A.; Tilles, H. Thiocarbamate sulfoxides:
potent, selective, and biodegradable herbicides. Science 1974,
184, 573-574.
(12) Trost, B. M.; Curran, D. P. Chemoselective oxidation of sulfides
to sulfones with potassium hydrogen persulfate. Tetrahedron
Lett. 1981, 22, 1287-1290.
(13) Entwistle, I. D.; Gilkerson, T.; J ohnstone, R. A. W.; Telford, R.
P. Rapid catalytic transfer reduction of aromatic nitro com-
pounds to hydroxylamines. Tetrahedron 1978, 34, 213-215.
(14) Reed, D. J .; Brodie, A. E.; Meredith, M. J . Cellular heterogeneity
in the status and functions of cysteine and glutathione. In
Functions of Glutathione: Biochemical, Physiological, Toxico-
logical, and Clinical Aspects; Larsson, A., Orrenius, S., Holmgren,
A., Mannervik, B., Eds.; Raven Press: New York, 1983; pp 39-
49.
(15) Ishikawa, T.; Ali-Osman, F. Glutathione-associated cis-diam-
minedichloroplatinum(II) metabolism and ATP-dependent efflux
from leukemia cells. J . Biol. Chem. 1993, 268, 20116-20125.
(16) Montali, J . A.; Wheatley, J . B.; Schmidt, D. E., J r. Expression
of GST in normal and malignant tissues. Cell. Pharmacol. 1995,
2, 241-247.
(17) Kaltenbach, J . P.; Kaltenbach, M. H.; Lyons W. B. Nigrosin for
differentiating live and dead ascites cells. Exp. Cell Res. 1958,
15, 112-117.
fractionated by reverse-phase HPLC (Waters µBondapak C18
,
0.78 × 30 cm), using a mobile phase composed of methanol:
water (1:1) containing 0.25% acetic acid at a flow rate of 2 mL/
min. The integrated intensity of the peak corresponding to 2a
J M980712O