3410 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 17
Notes
Ta ble 1. Physicochemical and Biological Properties of
S-(p-Bromobenzyl)glutathione Diesters
zyl)glutathione diesters, the propyl diester had the
lowest GC50 and TC50 values. The cyclopentyl diester,
however, was the most potent of the S-(p-bromobenzyl)-
glutathione diesters evaluated with a GC50 value of 4.23
( 0.01 µM (N ) 21) and a TC50 value of 8.86 ( 0.01 µM
(N ) 21). The antitumor activity of S-(p-bromobenzyl)-
glutathione diesters was not limited to activity against
HL-60 cells. For the National Cancer Institute panel
of leukemia, non-small lung cancer, colon cancer, CNS
cancer, melanoma, ovarian cancer, renal cancer, pros-
tate cancer, and breast cancer cell lines, S-(p-bromoben-
zyl)glutathione ethyl diester gave GC50 values in the
range 7-20 µM,11 and S-(p-bromobenzyl)glutathione
cyclopentyl diester had antitumor activity in vivo against
murine adenocarcinoma 15A.7 Incubation of HL-60 cells
with S-(p-bromobenzyl)glutathione cyclopentyl diester
lead to the delivery of diester into the cell cytosol
wherein it was hydrolyzed to S-(p-bromobenzyl)glu-
tathione cyclopentyl monoester and S-(p-bromobenzyl)-
glutathione.7
S-(p-Bromobenzyl)glutathione and S-(p-bromobenzyl)-
glutathione monoesters were inactive (GC50 and TC50
values >300 µM) although some monoester derivatives
were potent inhibitors of human glyoxalase I: the
inhibitor constant Ki value of S-(p-bromobenzyl)glu-
tathione ethyl monoester was 2.36 µM.9 In contrast,
S-(p-bromobenzyl)glutathione ethyl diester was a poor
inhibitor of human glyoxalase I and had a Ki value of
235 µM.9 This is consistent with the diesterification of
S-(p-bromobenzyl)glutathione being important for de-
livery into the cell cytosol, cf. the delivery of GSH into
the cytosol of cells by GSH diester, and also important
to confer resistance to degradation by γ-glutamyl
transpeptidase on the external surface of the plasma
membrane of cells.6
com-
pound R1,R2
yield
mp (°C) (%) GC50 (µM) N TC50 (µM) N
M+1
1
2
3
4
5
6
7
Et
532 & 534 107-109 35
8.3 ( 0.1 12 16.5 ( 0.1 27
n-Pr 560 & 562 118-120 70 7.8 ( 0.1 18 10.3 ( 0.3 18
n-Bu 588 & 590 128-130 75 18.5 ( 1.5 21 26.3 ( 1.0 21
n-Pt 616 & 618 133-135 66 22.2 ( 2.5 12 24.9 ( 1.3 12
i-Pr 560 & 562 120-122 13 19.6 ( 0.1 12 20.9 ( 1.1 12
cPt 612 & 614 136-138 71
cHx 640 & 642 121-123 60 29.2 ( 1.2 21 33.1 ( 4.7 21
4.2 ( 0.1 21 8.9 ( 0.1 21
1640 and fetal calf serum were purchased from Gibco Europe
Ltd. (Paisley, Scotland). Methanol (HPLC grade) was pur-
chased from Rathburn Chemicals Ltd. (Walkerburn, Scotland).
Silica gel 60 F234 TLC plates were purchased from BDH
Chemicals (Poole, Dorset, U.K.). S-(p-Bromobenzyl)glu-
tathione was synthesized by modification of the the method
of Vince et al.3,7 NMR spectra were recorded on a J eol ex270
MHz NMR spectrometer. FAB mass spectra were recorded
on a Kratos MS50 FAB mass spectrometer with compounds
dissolved in a glycerol matrix. S-(p-Bromobenzyl)glutathione
ethyl diester was synthesized as previously described.7
S-(p-Br om oben zyl)glu ta th ion e cyclop en tyl d iester (6).
S-(p-Bromobenzyl)glutathione cyclopentyl diester was pre-
pared and purified by methods similar to those described for
the corresponding ethyl diester except the acid-catalyzed
esterification of S-(p-bromobenzyl)glutathione was performed
in cyclopentanol for 7 days at room temperature. It was
characterized by melting point, FAB mass spectrometry (Table
1
1), H and 13C NMR spectroscopy, TLC, and elemental CHN
analysis: 1H NMR (DMSO-d6) δ 8.54 (t, 1H, J ) 5.86 Hz, glycyl
NH), 8.23 (d, 1H, J ) 7.81 Hz, cysteinyl NH), 7.49 (d, 2H, J )
8.43 Hz, benzyl m-H), 7.28 (d, 2H, J ) 8.43 Hz, benzyl ï-H),
5.06 (m, 2H, cyclopentyl 1-H), 4.54 (m, 1H, cysteinyl 2-H), 3.76
(s, 2H, benzyl 1-H and d, 2H, J ) 5.86 Hz, glycyl 2-H), 3.41 (t,
1H, J ) 5.35 Hz, γ-glutamyl 1-H), 2.74 (q, 1H, J ) 4.94 Hz
and -12.03 Hz, cysteinyl 3A-H), 2.52 (q, 1H, J ) 4.94 Hz and
-12.03 Hz, cysteinyl 3B-H), 2.25 (t, 2H, J ) 8.12 Hz,
γ-glutamyl 4-H), 1.6-1.9 (m, 10H, γ-glutamyl 3-H and cyclo-
pentyl-2,3,4,5-H); 13C NMR (DMSO-d6) δ 174.5 (γ-glutamyl
C-5), 172.3 (γ-glutamyl C-1), 171.1 (cysteinyl C-1), 169.4 (glycyl
C-1), 138.1 (phenyl C-4), 131.4 (phenyl C-3,5), 131.3 (phenyl
C-2,6), 120.1 (phenyl C-1), 77.4 (cyclopentyl(γ-glutamyl) C-1),
77.2 (cyclopentyl(glycyl) C-1), 53.5 (cysteinyl C-2), 51.8 (γ-
glutamyl C-2), 41.3 (glycyl C-2), 34.5 (cysteinyl C-3), 33.2
(benzyl C-1), 32.4 (cyclopentyl(γ-glutamyl) C-2,5), 32.2 (cyclo-
pentyl(glycyl) C-2,5), 31.7 (γ-glutamyl C-4), 29.8 (γ-glutamyl
C-3), 23.4 (cyclopentyl(γ-glutamyl) C-3,4 & cyclopentyl(glycyl)
C-3,4). The TLC Rf value on silica gel was 0.90 (mobile
phase: chloroform, methanol, acetic acid, 8:1:1). Anal.
(C27H40N3O7SBr) C, H, N, for S-(p-bromobenzyl)glutathione
cyclopentyl diester‚H2O.
Precedents for maximal pharmacological response of
the n-propyl and cyclopentyl diesters were in a series
of n-alkyl esters of ibuprofen,10 and the delivery of
cysteine into rat lung tissue by cysteine esters.8 This
ester structure-activity effect is probably due to a
balance of factors influential on the ability of S-(p-
bromobenzyl)glutathione diesters to deliver S-(p-bro-
mobenzyl)glutathione into cells: (i) the ability of the
ester groups to be resistant to cleavage by plasma and
extracellular plasma membrane-bound nonspecific es-
terases, and (ii) the susceptibility of the ester groups to
hydrolysis by nonspecific esterase in the cytosol of the
target cells.
Other S-(p-bromobenzyl)glutathione diesters were prepared,
purified, and characterized similarly.
Evalu ation of An tileu kem ia Activity of S-(p-Br om oben -
zyl)glu ta th ion e Diester s in Vitr o. HL-60 cells were incu-
bated at 37 °C in RPMI 1640 media containing 10% fetal calf
serum under an atmosphere of 5% CO2 in air, 100% humidity.7
Cells were seeded at an initial density of 5 × 104/mL and
incubated with 0.5-300 µM S-(p-bromobenzyl)glutathione
diester for 2 days.7 A stock solution of S-(p-bromobenzyl)-
glutathione diester was prepared in dimethyl sulfoxide and
diluted into the growth medium such that the final concentra-
tion of dimethyl sulfoxide did not exceed 5 mM, a concentration
which did not induce differentiation or toxicity in HL-60 cells.
Cell viability was judged by the ability of cells to exclude
trypan blue. GC50 and TC50 values were determined by logistic
regression of viable cell number and cytotoxicity data on
diester concentration, respectively. Nonlinear regression was
performed using the ENZFITTER program (Biosoft, Cam-
bridge, U.K.).
Con clu sion
S-(p-Bromobenzyl)glutathione diesters, glyoxalase I
inhibitor prodrugs, have antitumor activity in vitro
where a limited ester structure-activity study sug-
gested that n-propyl and cyclopentyl diesters have
potent antitumor activity. Their prospective mechanism
of action is different from current clinical antitumor
agents,7,11 and they are good candidates for further
pharmacological evaluation.
Exp er im en ta l Section
Trypan blue, reduced glutathione, Dowex 1 (chloride form),
and dimethylsulfoxide were purchased from Sigma Chemical
Co. Ltd. (Poole, Dorset, U.K.). p-Bromobenzyl bromide, hy-
drogen chloride gas, trifluoroacetic acid, dimethyl sulfoxide-
d6, and alcohols used were purchased from Aldrich Chemical
Co. Ltd. (Poole, Dorset, U.K.). Tissue culture medium RPMI
Ack n ow led gm en t. M.J .L. was on study leave from
the Department of Biochemistry, Usamanu Danfodiyo