Synthesis and activity of novel glutathione analogues containing an urethane backbone linkage

Article abstract of DOI:10.1016/S0014-827X(03)00135-6

The new GSH analogues H-Glo(-Ser-Gly-OH)-OH (5), its O-benzyl derivative 4, and H-Glo(-Asp-Gly-OH)-OH (9), characterized by the replacement of central cysteine with either serine or aspartic acid, and containing an urethanic fragment as isosteric substitu

Full text of DOI:10.1016/S0014-827X(03)00135-6

Il Farmaco 58 (2003) 787ꢀ793
/
www.elsevier.com/locate/farmac
Synthesis and activity of novel glutathione analogues containing an
urethane backbone linkage
I. Cacciatore a, A.M. Caccuri b, A. Di Stefano a, G. Luisi a, M. Nalli c, F. Pinnen a,*,
G. Ricci b, P. Sozio a
a Dipartimento di Scienze del Farmaco, Universita` degli Studi ‘G. D’Annunzio’, Via dei Vestini, I-66100 Chieti, Italy
b Dipartimento di Biologia, Universita` degli Studi ‘Tor Vergata’, Via della Ricerca Scientifica, I-00133 Rome, Italy
c Dipartimento di Studi Farmaceutici, Universita` degli Studi ‘La Sapienza’, P.le Aldo Moro 5, I-00185, Rome, Italy
Received 22 November 2002; accepted 28 December 2002
Dedicated to the memory of Professor Piero Pratesi
Abstract
The new GSH analogues Hꢀ
/
Glo(ꢀ
/
Serꢀ
/
Glyꢀ
/
OH)ꢀ
/
OH (5), its O-benzyl derivative 4, and Hꢀ
/
Glo(ꢀ
/
Aspꢀ
/
Glyꢀ
/
OH)ꢀOH (9),
/
characterized by the replacement of central cysteine with either serine or aspartic acid, and containing an urethanic fragment as
isosteric substitution of the scissile g-glutamylic junction, have been synthesized and characterized. Their ability to inhibit human
GST P1-1 (hGST P1-1) in comparison with Hꢀ
/
Glu(ꢀ
/
Serꢀ
/
Glyꢀ
/
OH)ꢀ
/
OH and Hꢀ
/
Glu(ꢀ
/
Aspꢀ
/
Glyꢀ
/
OH)ꢀOH, which are potent
/
competitive inhibitors of rat GST 3-3 and 4-4, has been evaluated. In order to further investigate the effect of the isosteric
substitution on the binding abilities of the new GSH analogues 4, 5 and 9, the previously reported cysteinyl-containing analogue Hꢀ
Glo(ꢀCysꢀGlyꢀOH)ꢀOH has been also evaluated as a co-substrate for hGSTP1-1.
# 2003 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
/
/
/
/
/
´
Keywords: g-Glutamyl junction; g-Glutamyl transpeptidase; Glutathione; Glutathione analogues; Human glutathione S-transferase; Urethanic
bond
1. Introduction
concentration in the cytosol of many tissues. Based on
sequence similarity, immunological cross-reactivity, and
subcellular distribution, cytosolic GSTs have been
divided into at least ten gene-independent classes
(Alpha, Beta, Delta, Kappa, Pi, Mu, Sigma, Theta,
Zeta and Omega), which show different but overlapping
substrate specificities, while sharing, not surprisingly, a
high specificity for the physiologic co-substrate GSH [3].
The suggestion that GSH conjugation may be concerned
with anticancer drug resistance has evoked additional
interest in this group of enzymes [4,5]. Although it is not
clear whether GST overexpression is the result or a
cause of resistance to chemotherapy, the approach of
GST inhibition with the aim of providing tumor-
directed potentiation of conventional antineoplastic
agents appears challenging. In this context the human
Pi-class GST (hGST P1-1) represents an attractive
target, since it is the predominant isoform in most
tumors [6,7].
A major pathway in the cellular (de)toxification of
both xenobiotic and endobiotic electrophiles is repre-
sented by conjugation with tripeptide glutathione (g-
Gluꢀ
/
CysꢀGly, GSH). This reaction is catalysed by the
/
isoenzyme family of glutathione S-transferases (GSTs,
E.C. 2.5.1.18) [1,2], which are mainly present in high
Abbreviations: AcOH, acetic acid; Boc, tert-butyloxycarbonyl; n-
BuOH, n-butanolo; Bzl, benzyl; CDNB, 1-chloro-2,4-dinitrobenzene;
DBU,
1,8-diazabicyclo[5.4.0]undec-7-ene;
DCC,
dicyclohexylcarbodiimide; DMF, N,N?-dimethylformamide; EDTA,
ethylenediamine tetraacetic acid; Et2O, diethyl ether; EtOAc, ethyl
acetate; Fmoc, 9-fluorenylmethoxycarbonyl; Glo,
L
-g-oxaglutamic
acid; HOBt, 1-hydroxybenzotriazole; MeOH, methanol; TEA,
triethylamine; TFA, trifluoroacetic acid; THF, tetrahydrofuran; Z,
benzyloxycarbonyl.
* Corresponding author.
E-mail address: fpinnen@unich.it (F. Pinnen).
´
0014-827X/03/$ - see front matter # 2003 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved.
doi:10.1016/S0014-827X(03)00135-6

788
I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ793
/
Structural and functional aspects concerning GSTs
are the subject of extensive studies [1,8ꢀ11]; much
interest is focused on the one part of the protein which
is essentially conserved among all the GSTs, the GSH-
binding site (G-site); here the co-substrate is accomo-
dated, adopting the proper alignment at the reaction site
prior to thiol deprotonation.
were found stable g-GT inhibitors [2,19ꢀ
/
21]. This
/
original approach has been subsequently adopted by
other Authors for the synthesis of metabolically stable
inhibitors of glyoxalase I [23] and novel g-GT resistant
GST inhibitors [24].
As an extension of our research in this field, we
become interested in developing peptidic GSH analo-
gues stabilized against g-GT degradation, to be tested as
potential hGST P1-1 inhibitors. The GSH-related pseu-
dopeptides considered in this study are characterized by
the replacement of central cysteine with serine ed
aspartic acid, respectively, and contain an urethanic
fragment as isosteric substitution of the scissile glu-
tamylic junction, which is expected to confer the desired
metabolic stability; thus, from a formal point of view,
Thus several GSH analogues and derivatives, inclu-
ding a variety of g-glutamyl modifications, have been
synthesized in an attempt to design specific GST
inhibitors and/or as probes for the G-site in GSTs
[12,13]. However, GST inhibitors retaining the GSH
isopeptidic structure are expected to be ineffective in
vivo, as they are still processed by g-glutamyl transpep-
tidase (g-GT), a cell-surface enzyme that initiates the
breakdown of GSH and its S-substituted derivatives at
the level of the g-glutamyl junction [14,15]. Interestingly,
with very few notable exceptions [16] the majority of the
chemical variations reported so far does not involve the
GSH g-glutamylic junction, in spite of the relevant
consequences this approach is expected to introduce in
terms of resistance to g-GT.
the designed analogues contain the
L
-g-oxaglutamic acid
(Glo) replacing the Glu residue. The replacement of
cysteine thiol with more electronegative groups such as
carboxylates or hydroxyls (Ser or Asp) is a well-
documented strategy to potent GST competitive inhibi-
tors at the GSH binding site [25,26]. On the other hand,
urethane bonds appear to be promising peptide bond
surrogates due to structure planarity, hydrogen-bonding
characteristics and defined conformational features.
Unlike the majority of peptide bond surrogates, which
mimic the two atom sequence of conventional peptide
unit, the urethane bond consists of an extended tria-
As a prosecution of our studies on the structural
modification of GSH [17ꢀ22], we started a research
/
program aimed at studying GSH analogues characte-
rized by two fundamental properties: (i) close resem-
blance with the natural substrate as to overall peptidic
structure, molecular size, and functional group arrange-
ment; and (ii) altered structure and reactivity of the g-
tomic (Oꢀ
/
CꢀN) planar system, made of an ester and an
/
amide portion, suitable to introduce unique conforma-
tional and electronic properties at the backbone level
[27,28]. We report here synthesis, characterization and
biological activity of novel GSH-related compounds, the
glutamylic COꢀNH junction obtained through chemical
/
modification or isosteric replacement of the g-CH2
bound to the amide carbonyl group. The first property
should play a favorable role during the molecular
recognition step, thus rendering these models suitable
tools for the study of the reaction mechanism of the
enzymatic system involved. The chemical alteration of
the glutamylic g-CH2 represents on the other hand a
versatile strategy for introducing relevant consequences
at both the conformational and biochemical level. The
two oxy-analogues Hꢀ
and its O-benzyl derivative Hꢀ
OH]ꢀOH (4), and HꢀGlo(ꢀAspꢀ
/
Glo(ꢀ
/
Serꢀ
Glo[ꢀ
Glyꢀ
/
Glyꢀ
Ser(Bzl)ꢀ
OH)ꢀOH (9)
/
OH)ꢀ
/OH (5)
/
/
/
Glyꢀ
/
/
/
/
/
/
/
(Schemes 1 and 2). The analogue 4, featuring a PhCH2
appendix at the central residue, appears as a suitable
model to explore the hydrophobic requirements of the
adduct-binding site.
peculiar properties of the COꢀNH group such as
/
polarity, H bond capacity, double bond character and
metabolic stability can be in fact profoundly altered by
changing the nature of the group directly bound to the
amide carbonyl. These modified compounds may be-
have as substrates, often with quite different kinetics
constants, or inhibitors, often of competitive nature if
the chemical modifications do not alter dramatically the
spatial structure.
This chemical modification has been recently intro-
duced by us with the synthesis of backbone-modified
GSH analogues, obtained through replacement of the
Glu g-carbon atom adjacent to the isopeptidic bond
with an aza (NH) and an oxa (O) unit [20,21]. The
resulting pseudopeptides, characterized respectively by
the presence of ureic (NHCONH) and urethanic
(OCONH) junctions as g-glutamylic bond surrogates,
2. Experimental
2.1. Peptide synthesis
Melting points were determined on a Buchi B-450
¨
apparatus and are uncorrected. TLC was performed on
Merck 60 F254 silica gel plates developed with the
following solvent system: (a) CHCl3:MeOH (98:2); (b)
CHCl3:MeOH (96:4); (c) CHCl3:Et2O (1:1); (d) n-
BuOH:AcOH:H2O (4:5:1); (e) n-BuOH:AcOH:H2O
(6:4:4), (f) CHCl3:MeOH (99:1). Column chromatho-
graphy was carried out using Merck 60 silica gel (230ꢀ
400 mesh). Optical rotations were taken at 20 8C with a
PerkinꢀElmer 241 polarimeter. IR spectra were re-
corded employing a PerkinꢀElmer 983 FTIR-1600
/
/
/

I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ
/793
789
Scheme 1. Reagents and conditions. (a) HCl×
dioxane, 80 8C, 10 h; (d) TFA, r.t., 5 h; (e) 1 N aq. NH3, r.t., 30 min; (f) H2, 10% Pdꢀ
/
Hꢀ
/
OBut , HOBt, NMM, DCCl, THF, 0 8C, 4 h, then 5 8C, 16 h; (b) DBU, CH2Cl2, r.t., 20 min; (c)
C, AcOH, r.t., 6 h.
/
spectrophotometer. 1H and 13C NMR spectra were
determined on a Varian VXR 300 MHz instrument (d
expressed in ppm). NMR resonances of analogues 4, 5
and 9 were assigned with standard 2D NMR techniques
(HETCOR). Elemental analyses for the new compounds
2.1.1. Fmocꢀ
/
Ser(Bzl)ꢀ
/
Glyꢀ
/
OBut (1)
OH (2 g, 4.8 mmol) was dissolved in
FmocꢀSer(Bzl)ꢀ
/
/
THF (8 ml) and HOBt (0.65 g, 4.8 mmol) was added
with stirring. The solution was cooled to 0 8C and an
iceꢀ
/
cold solution containing HCl×
/
Hꢀ
/
Glyꢀ
OBut (0.8 g,
/
4.8 mmol) in THF (8 ml) was added, followed by
portionwise addition of a solution of DCC (0.99 g, 4.8
mmol) in THF (1 ml). After 4 h at 0 8C and 16 h at 5 8C,
the reaction mixture was filtered and the resulting
are within the 9
Ser(Bzl)ꢀOH and Zꢀ
from Novabiochem. HCl×
from Sigma Aldrich.
/
0.4% of the theoretical values. Fmocꢀ
Asp(OBut)ꢀ
OH were obtained
HꢀGlyꢀ
OBut was purchased
/
/
/
/
/
/
/
Scheme 2. Reagents and conditions. (a) HCl×
/
Hꢀ
/
Glyꢀ
/
OBut , HOBt, NMM, DCCl, THF, 0 8C, 4 h, then 5 8C, 16 h; (b) H2, 10% Pdꢀ
C, MeOH, r.t., 2
/
h; (c) dioxane, 80 8C, 10 h; (d) TFA, r.t., 5 h; (e) 1 N aq. NH3, r.t., 30 min.

790
I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ793
/
solution was evaporated under vacuum. The residue was
taken up in EtOAc and the organic layer washed with 1
N KHSO4, saturated acqueous NaHCO3 and H2O. The
residue obtained after drying and evaporation was
cromatographed on silica gel using CHCl3:MeOH
(99:1) as eluant to give dipeptide t-butyl ester 1 as a
white solid, further purified by crystallization from
Ser(Bzl)ꢀ
further purification.
/
Glyꢀ
/
OH]ꢀOH, which was used without
/
The trifluoroacetate (0.94 g, 1.89 mmol) was dissolved
in 1 N aqueous NH3 (14.6 ml) at room temperature.
After 30 min the aqueous solution was concentrated and
subjected to column chromatography on Sephadex LH-
20 using H2O as eluant to afford tripeptide acid 4 as a
EtOAc (2.4 g, 94%). m.p. 128ꢀ
[a]546 3.08 (cꢁ1, CHCl3). IR (KBr) 3405, 3250,
1725, 1655, 1535 cmꢃ1. H NMR (CDCl3) dꢁ
9H, 3ꢄCH3), 3.6 (m, 1H, Ser b-CHB), 3.9 (m, 3H, Gly
/
9 8C. Rf (a)ꢁ
/
0.7.
foam (0.67 g, 92%). Rf (d)ꢁ
H2O). IR (KBr) 3255 br, 1710, 1670ꢀ
cmꢃ1. 1H NMR (D2O) dꢁ
3.7 (m, 2H, Ser b-CH2), 3.75
/
0.85. [a]Dꢁ
/
ꢃ
/
3.08 (cꢁ1,
/
ꢁ
/
ꢂ
/
/
/1550, 1395, 1250
1
/1.5 (s,
/
/
(app s, 2H, Gly CH2), 3.95 (m, 1H, Glo a-CH), 4.35 (m,
1H, Ser a-CH), 4.4 (m, 2H, Glo CH2), 4.45 (m, 2H,
CH2 and Ser a-CH), 4.2 (m, 1H, Fmoc CH), 4.4 (m, 3H,
Fmoc OCH2 and Ser b-CHA), 4.6 (m, 2H, PhCH2O), 5.7
PhCH2O), 7.3 (m, 5H, ArH). 13C NMR (D2O) dꢁ
46.00
/
(br d, 1H, Ser NH), 6.95 (br t, 1H, Gly NH), 7.2ꢀ
/
7.4 (m,
(Gly Ca), 56.80 (Glo Ca), 57.91 (Ser Ca), 66.65 (Glo Cb),
71.85 (Ser Cb), 75.84 (PhCH2O), 131.21, 131.25, 131.57
and 139.87 (aromatics), 159.79 (Glo OCO), 173.88,
174.30 and 178.66 (CO).
9H, ArH), 7.6 (2d overlapped, 2H, ArH) and 7.8 (2d
overlapped, 2H, ArH).
2.1.2. Hꢀ
/
Ser(Bzl)ꢀ
/
Glyꢀ
OBut (2)
/
To a solution of the above reported t-butyl ester 1
(2.3 g, 4.35 mmol) in CH2Cl2 (30 ml) DBU (0.69 g, 4.36
mmol) was added at room temperature. After 20 min the
solution was evaporated to dryness and the residue
chromatographed on silica gel using CHCl3:MeOH
(95:5) as eluant to yield pure N-deprotected dipeptide
2.1.5. Hꢀ
/
Glo(ꢀ
/
Serꢀ
/
Glyꢀ
/
OH)ꢀOH (5)
/
A solution of the O-benzyl derivative 4 (0.65 g, 1.7
mmol) in glacial AcOH (42 ml) was hydrogenated in the
presence of 10% Pd on activated charcoal (0.25 g). After
6 h the catalyst was filtered off and the filtrate was
evaporated under vacuum. The residue was chromatho-
graped on Sephadex LH-20 using H2O:MeOH (2:1) as
eluant to yield title compound 5 as a white solid (0.35 g,
ester 2 as an oil (1.32 g, 98%). Rf (b)ꢁ
38.08 (cꢁ1, CHCl3). IR (CHCl3) 3370, 1735, 1670
cmꢃ1. H NMR (CDCl3) dꢁ
1.5 (s, 9H, 3ꢄCH3), 1.8
(s, 2H, Ser NH2), 3.6ꢀ3.75 (m, 3H, Ser a-CH and b-
CH2), 3.9 (ABq, Jꢁ10 Hz, 2H, Gly CH2), 4.55 (s, 2H,
PhCH2O), 7.3 (m, 5H, ArH), 7.9 (br t, 1H, Gly NH).
/
0.3. [a]Dꢁ
/
ꢃ
/
/
1
/
/
70%). Rf (e)ꢁ
3235 br, 1670ꢀ
/
0.25; [a]Dꢁ
/
ꢃ
/
7.08 (cꢁ
/
1, H2O). IR (KBr)
.
/
/
1555, 1400, 1255 cmꢃ1
/
2.1.6. Zꢀ
/
Asp(OBut)ꢀ
/
Glyꢀ
OBut (6)
This compound was prepared by following the
/
2.1.3. Bocꢀ
/
Glo[ꢀ
/
Ser(Bzl)ꢀ
/
Glyꢀ
/
OBut]ꢀ
/
OBut (3)
The above described dipeptide t-butyl ester 2 (1.3 g,
procedure reported for the N-protected dipeptide t-
Glyꢀ
OBut (1.5 g,
butyl ester 1, starting from HCl×
/
Hꢀ
/
/
4.25 mmol) was dissolved in dioxane (10 ml) and a
OBut [21] (1.8 g, 4.25
8.8 mmol) and Zꢀ OH (3.0 g, 8.8 mmol).
/
Asp(OBut)ꢀ
The resulting residue was purified by column chromato-
graphy (CHCl3: MeOH 99:1 as eluant) to give the
/
solution of Bocꢀ
/
Glo(ONp)ꢀ
/
mmol) in dioxane (10 ml) was added. After 10 h at
80 8C the reaction mixture was evaporated under
reduced pressure and the residue taken up in CHCl3.
The organic layer was washed with 0.5 N HCl, saturated
acqueous Na2CO3 and H2O. The residue obtained after
drying and evaporation was cromatographed on silica
gel using CHCl3:MeOH (99:1) as eluant to give tripep-
dipeptide t-butyl ester 6 as an oil (3.38 g, 88%). Rf (f)ꢁ
0.50. [a]Dꢁ 15.58 (cꢁ1, CHCl3). IR (KBr) 3410,
3360, 2980, 1725, 1655, 1530 cmꢃ1. H NMR (CDCl3)
dꢁ1.5 (s, 18H, 6ꢄCH3), 2.65 (m, 1H, Asp b-CHB), 2.9
(m, 1H, Asp b-CHA), 3.9 (m, 2H, Gly CH2), 4.6 (m, 1H,
Asp a-CH), 5.15 (m, 2H, PhCH2O), 6.0 (d, Jꢁ7.92 Hz,
1H, Asp NH), 6.95 (br t, 1H, Gly NH), 7.2ꢀ7.4 (m, 5H,
ArH).
/
/
ꢂ
/
/
1
/
/
/
tide t-butyl ester 3 as a foam (1.97 g, 78%). Rf (c)ꢁ
[a]Dꢁ 248 (cꢁ1), CHCl3). IR (CHCl3): 3425, 1735ꢀ
1700, 1680, 1500 cmꢃ1. H NMR (CDCl3) dꢁ
27H, 9ꢄCH3), 3.95 (d, 2H, Gly CH2), 4.3ꢀ4.5 (m, 6H,
Ser and Glo a-CH and b-CH2), 4.6 (ABq, Jꢁ12.0 Hz,
2H, PhCH2O), 5.4 (d, Jꢁ4.8 Hz, 1H, Glo NH), 5.7 (br
d, 1H, Ser NH), 7.0 (t, Jꢁ4.5 Hz, 1H, Gly NH), 7.4 (m,
5H, ArH).
/0.6;
/
/
ꢂ
/
/
/
1
/
1.45 (s,
/
/
2.1.7. Hꢀ
/
Asp(OBut)ꢀ
/
Glyꢀ
OBut (7)
/
/
A solution of the N-benzyloxycarbonyl derivative 6
(3.2 g, 7.3 mmol) in MeOH (360 ml) was hydrogenated
in the presence of 10% Pd on activated charcoal (2.1 g).
After 2 h the catalyst was removed by filtration and the
solution evaporated under reduced pressure to afford
the pure N-deprotected dipeptide ester 7 as an oil (2.14
/
/
2.1.4. Hꢀ
/
Glo[ꢀ
/
Ser(Bzl)ꢀ
/
Glyꢀ
/
OH]ꢀOH (4)
/
The preceding foam 3 (1.55 g, 2.6 mmol) was treated
with TFA (40 ml) at room temperature. After 5 h the
solvent was removed and the residue repeatedly evapo-
g, 97%). Rf (f)ꢁ
(neat): 3380, 3005, 2980, 1725, 1675, 1520, 1455 cmꢃ1
1H NMR (CDCl3) dꢁ
1.5 (s, 18H, 6ꢄCH3), 1.8 (s, 2H,
Asp NH2), 2.51 (m, 1H, Asp b-CHB), 2.82 (m, 1H, Asp
/
0.25. [a]Dꢁ
/
ꢃ
/
14.58 (cꢁ1, CHCl3). IR
/
.
/
/
rated with ether to give 0.94 g (73%) of TFA×
/
Hꢀ
/
Glo[ꢀ
/

I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ
/793
791
b-CHA), 3.68 (m, 1H, Asp a-CH), 3.9 (m, 2H, Gly CH2),
7.83 (br t, 1H, Gly NH).
Glyꢀ
/
OH)ꢀ
/
OH (9) and Hꢀ
/
Glu(ꢀ
/
Aspꢀ
/
Glyꢀ
/
OH)ꢀ
/
OH.
9.6 mMꢃ1 cmꢃ1),
Activity was recorded at 340 nm (oꢁ
/
with a Kontron double-beam Uvikon 940 spectrophoto-
meter and normalized for the rate of reaction obtained
in the absence of inhibitor.
2.1.8. Bocꢀ
To a solution of the foregoing dipeptide t-butyl ester
7 (2.0 g, 6.6 mmol) in DMF (20 ml) a solution of Bocꢀ
Glo(ONp)ꢀ
OBut (2.81 g, 6.6 mmol) in DMF (20 ml)
/
Glo[ꢀ
/
Asp(OBut)ꢀ
/
Glyꢀ
/
OBut]ꢀ
/
OBut (8)
/
/
was added. After 10 h at 80 8C the reaction mixture was
evaporated under vacuum and the residue partitioned
between EtOAc and H2O. The organic layer was washed
with 1 N KHSO4, saturated acqueous Na2CO3 and
H2O. The residue obtained after drying and evaporation
was cromatographed on silica gel using CHCl3:MeOH
(99:1) as eluant to give tripeptide t-butyl ester 8 as a
3. Results and discussion
3.1. Chemical synthesis
The GSH analogues 4, 5 and 9 were synthesized in
good overall yields using solution procedures by step-
wise elongation of the peptide chain in the C-to-N
direction. The common adopted synthetic strategy,
which provides for the introduction of the urethane
foam (2.1 g, 54%). Rf (f)ꢁ
/
0.55. [a]Dꢁ
/
ꢂ
/
12.88 (cꢁ1,
/
CHCl3). IR (neat): 3345, 2980, 2930, 1730, 1520, 1370
1
cmꢃ1. H NMR (CDCl3) dꢁ
2.63 (m, 1H, Asp b-CHB), 2.90 (m, 1H, Asp b-CHA),
3.89 (m, 2H, Gly CH2), 4.30ꢀ4.54 (m, 4H, Asp a-CH,
Glo a-CH and Glo b-CH2), 5.35 (d, Jꢁ6.6 Hz, 1H, Glo
NH), 5.96 (d, Jꢁ7.91 Hz, 1H, Asp NH), 6.96 (t, Jꢁ
6.18 Hz, 1H, Gly NH).
/
1.43 (s, 36H, 12ꢄ
/
CH3),
OCONH linkage in place of the g-glutamyl CH2ꢀ
/
COꢀ
/
NH fragment, is based on the utilization of an O-
functionalized serine residue as building block of the
modified g-glutamyl moiety. Thus, the active carbonate
/
/
/
/
Bocꢀ
/
Glo(ONp)ꢀ
/
OBut was synthesized starting from
Bocꢀ
/
Serꢀ
/
OBut and p-nitrophenyl chloroformate, as
described elsewhere [21], to be condensed with the
amino group of the protected dipeptides.
2.1.9. Hꢀ
/
Glo[ꢀ
/
Asp(OH)ꢀ
/
Glyꢀ
/
OH]ꢀOH (9)
/
The above described tripeptide 8 (2.0 g, 3.4 mmol)
was dissolved in TFA (10 ml). After 5 h at room
temperature the solvent was removed and the residue
The route towards the g-glutamyl modified oxy-
analogues 4 and 5 is outlined in Scheme 1. Dipeptide
amide Hꢀ
conventional DCC/HOBt condensation of Fmocꢀ
Ser(Bzl)ꢀOH with HꢀGlyꢀ
OBut and subsequent re-
/
Ser(Bzl)ꢀ
/
Glyꢀ
OBut (2) was prepared by
/
repeatedly evaporated with ether to give TFA×
AspꢀGlyꢀOH)ꢀOH in quantitative yield, which was
used as such.
/
Hꢀ
/
Glo(ꢀ
/
/
/
/
/
/
/
/
moval of the Fmoc protecting group with DBU at r.t.
[18]. The N-deprotected dipeptide 2 was then coupled
The trifluoroacetate (1.9 g, 4.3 mmol) was dissolved in
1 N aqueous NH3 (100 ml) at room temperature. After
30 min the aqueous solution was concentrated and
subjected to column chromatography on Sephadex LH-
20 using H2O:MeOH (2:1) as eluant to afford 9 as a
with Bocꢀ
to give the fully protected tripeptide Bocꢀ
Ser(Bzl)ꢀGlyꢀ
OBut]ꢀOBut (3) in 80% yield. Subse-
/
Glo(ONp)ꢀ
OBut in dioxane at 80 8C for 10 h
/
/
Glo[ꢀ
/
/
/
/
quent removal in a single step of the tert-butyl and
tert-butyloxycarbonyl protecting groups by using TFA,
followed by treatment of the resulting trifluoroacetate
white solid (1.15 g, 83%). Rf (d)ꢁ
(cꢁ1, H2O). IR (neat): 3135 br, 2360, 1705, 1650, 1400
cmꢃ1
/
0.45. [a]Dꢁ
/
ꢂ19.48
/
/
.
with aqueous ammonia, afforded H-Glo[ꢀ
GlyꢀOH]ꢀOH (4) in 70% yields. The O-benzyl deriva-
tive 4 was then converted by catalytic hydrogenolysis
with 10% PdꢀC in AcOH into the GSH oxy-analogue
HꢀGlo(ꢀSerꢀGlyꢀOH)ꢀOH (5), without any side re-
/
Ser(Bzl)ꢀ
/
/
/
2.2. Biological assays
/
Inhibition experiments in vitro concerning com-
pounds 4, 5 and 9 were performed on human isoenzyme
GST P1-1 (EC 2.5.1.18), using CDNB as acceptor
substrate to characterize GSH conjugation. Human
placenta GST P1-1 was expressed in Escherichia coli
and purified as previously described [29]. GSTP1-1
activity was measured at 25 8C in 1 ml of 0.1 M K-
phosphate buffer pH 6.5 containing 1 mM GSH {or 1
/
/
/
/
/
actions and in good yields.
Accordingly, synthesis of compound 9 was straight-
forward (see Scheme 2), commencing with the DCC/
HOBt mediated acylation of Zꢀ
/
Asp(OBut)ꢀ
OH with
/
HꢀGlyꢀ
/
/
OBut to give the protected dipeptide 6. The
amino functionality was recovered by catalytic hydro-
genolysis and coupled to activated tert-butyloxycarbo-
mM Hꢀ
/
Glo(ꢀ
/
Cysꢀ
/
Glyꢀ
/
OH)ꢀ
/
OH [21]}, 1 mM CDNB
nyl-
reported for dipeptide 2, to give acceptable amounts of
the protected tripeptide BocꢀGlo[ꢀ Glyꢀ
Asp(OBut)ꢀ
OBut]ꢀOBut (8). In analogy with the oxy-compound
3, complete deprotection was easily performed by
acidolysis with TFA, followed by addition of aqueous
ammonia to ensure the conversion into the free amino
L
-g-oxaglutamic acid under the same conditions
and 0.1 mM EDTA. GSTP1-1 inhibition experiments
were performed in the presence of variable amounts of
/
/
/
/
GSH analogues, ranging from 0.01 to 8 mM for Hꢀ
Glo[ꢀSer(Bzl)ꢀGlyꢀOH]ꢀOH (4), HꢀGlo(ꢀSerꢀGlyꢀ
OH)ꢀOH (5), and HꢀGlu(ꢀSerꢀGlyꢀOH)ꢀOH
(GOH), and from 0.01 to 2 mM for HꢀGlo(ꢀAspꢀ
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/

792
I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ793
/
tripeptide acid Hꢀ
/
Glo(ꢀ
/
Aspꢀ
/
Glyꢀ
/
OH)ꢀOH (9) in
/
good yields.
The final peptides were purified to apparent homo-
geneity by gel-permeation chromatography and fully
1
characterized by H and 13C NMR spectroscopy (see
Table 1).
3.2. Biological studies
The new GSH analogues Hꢀ
OH (5), its O-benzyl derivative 4, and Hꢀ
GlyꢀOH)ꢀOH (9) were tested for their ability to inhibit
human GST P1-1 (hGST P1-1) in comparison with Hꢀ
Glu(ꢀSerꢀGlyꢀOH)ꢀOH (GOH) and HꢀGlu(ꢀAspꢀ
GlyꢀOH)ꢀOH, which are potent competitive inhibitors
/
Glo(ꢀ
/
Serꢀ
/
Glyꢀ
/
OH)ꢀ
/
/
Glo(ꢀ
/
Aspꢀ
/
/
/
/
/
/
/
/
/
/
/
/
/
of rat GST 3-3 and 4-4, as previously reported [25,26].
Fig. 1 shows the effect on the enzymatic activity of the
three GSH oxy-analogues 4, 5, and GOH (panel A), and
of the two aspartylꢀ
OH)ꢀOH (panel B). Compounds 4, 5 and 9 are not
inhibitors of hGST P1-1 while, in the same assay
conditions, GOH and HꢀGlu(ꢀAspꢀGlyꢀOH)ꢀOH
show IC50 values of 0.7990.04 and 0.0590.01 mM,
respectively. Assuming for these compounds a compe-
titive inhibition towards GSH {Kmꢁ0.22 mM [30]}, the
/
peptides 9 and Hꢀ
/
Glu(ꢀ
/
Aspꢀ
/
Glyꢀ
/
Fig. 1. Inhibition of human placenta GSTP1-1 by GSH analogues.
GST activity was measured at 25 8C in the presence of 1 mM GSH and
/
1 mM CDNB, as described under Section 2. Panel A: Hꢀ
GlyꢀOH)ꢀOH (GOH) (k), HꢀGlo[ꢀSer(Bzl)ꢀGlyꢀOH]ꢀ
(j), HꢀGlo(ꢀSerꢀGlyꢀOH)ꢀOH (5) (I). The solid line is the best
fit of the experimental data to a hyperbolic binding equation which
fulfils an IC50 0.7990.04 mM. Assuming a competitive inhibition vs.
GSH, a Kiꢁ142 mM can be calculated for the inhibitor GOH. Panel B:
HꢀGlu(ꢀAspꢀGlyꢀOH)ꢀOH (k), HꢀGlo(ꢀAspꢀGlyꢀOH)ꢀOH
(9) (j). The solid line is the best fit of the experimental data to a
hyperbolic binding equation which fulfils an IC50 0.0590.01 mM.
Assuming a competitive inhibition vs. GSH, a Kiꢁ9 mM can be
calculated for the inhibitor HꢀGlu(ꢀAspꢀGlyꢀOH)ꢀOH. Data
/
Glu(ꢀ
/
Serꢀ
/
/
/
/
/
/
/
/
/
/
/
/
/
OH (4)
/
/
/
/
/
/
/
ꢁ/
/
/
/
IC50 values correspond to Ki of 142 and 9 mM,
respectively. The present data indicate that GOH and
/
/
/
/
/
/
/
/
/
/
Hꢀ
/
Glu(ꢀ
/
Aspꢀ
/
Glyꢀ
/
OH)ꢀOH can inhibit even the hu-
/
ꢁ/
/
man GST P1-1 isoform, although ten times less than rat
liver GST 3-3 and 4-4.
In order to further investigate the effect of the
isosteric substitution on the binding abilities of the
new GSH analogues 4, 5 and 9, we tested the previously
/
/
/
/
/
/
shown in the figure represent the mean of three different experiments;
the standard error for each point does not exceed 5%.
synthesized cysteinyl-containing pseudopeptide Hꢀ
/
Glo(ꢀCysꢀGlyꢀOH)ꢀOH [21] as a co-substrate for
/
/
/
/
Table 1
1H and 13C NMR data (in D2O at 25 8C) for compounds 5 and 9
hGSTP1-1. The reactivity of this GSH analogue is quite
low, as the isosteric substitution resulted in a fivefold
decrease of catalytic efficiency compared to that ob-
tained with GSH.
Residue
H*
/
Glo(ꢀ
/
Serꢀ
/
Glyꢀ
/
Hꢀ/Glo-
OH)ꢀ/OH (5)
(ꢀAspꢀ
/
/
Glyꢀ
/
OH)ꢀ
/
OH (9)
dH
dC
dH
dC
Glo
Ca
4.0
4.45
57.53
3.85
53.98
63.80
173.57 b
157.01
4. Conclusions
Cb
66.63
174.83 a
159.91
4.2ꢀ4.45
/
COꢀ
/O
Results from the inhibition studies on hGST P1-1
show that the tested compounds 4, 5 and 9 are devoid of
activity, although they meet some requirements for good
in vivo GST inhibitors. In fact all the pseudopeptides
have been designed to be g-GT resistant, through
isosteric replacement of the g-glutamylic isopeptidic
bond with an urethane moiety [21]. Furthermore they
are modified at the central residue with the introduction
OꢀCOꢀN
/
/
Ser or Asp
Ca
Cb
4.2
3.75
59.81
64.27
170.58
4.3
52.76 b
2.45 (HB), 2.6 (HA) 38.13
COꢀ/N
171.29 b
177.00
COꢀ/O
Gly
Ca
3.7
46.20
174.91 a
3.6
43.12
176.04
CO
of L-amino acids featuring electronegative groups at the
Cb atom; the exploitment of this last approach has
previously led to potent GST 3-3 and 4-4 inhibitors. In
a
Assignments may be interchanged.
Assignments may be interchanged.
b

I. Cacciatore et al. / Il Farmaco 58 (2003) 787ꢀ
/793
793
this regard it is interesting to note that also the
isoenzyme of the Pi-class is inhibited by the two reported
References
[1] M.C.J. Wilce, M.W. Parker, Biochim. Biophys. Acta 1205 (1994)
1ꢀ18.
[2] A.E. Salinas, M.G. Wong, Curr. Med. Chem. 6 (1999) 279ꢀ
[3] D. Sheehan, G. Meade, V.M. Foley, C.A. Dowd, Biochem. J. 360
(2001) 1ꢀ16.
[4] D.J. Waxman, Cancer Res. 50 (1990) 6449ꢀ
[5] S. Kuzmich, K.D. Tew, Med. Res. Rev. 11 (1991) 185ꢀ
[6] W.H.M. Peters, F.M. Nagengast, T. Wobbes, Carcinogenesis 10
(1989) 2371ꢀ2374.
[7] S.-I. Hamada, M. Kamada, H. Furumoto, T. Hirao, T. Aono,
Gynaecol. Oncol. 52 (1994) 313ꢀ319.
[8] A.E.P. Adang, J. Brussee, A. van der Gen, G.J. Mulder, Biochem.
J. 269 (1990) 47ꢀ54.
inhibitors Hꢀ
/
Glu(ꢀ
/
Serꢀ
/
Glyꢀ
/
OH)ꢀ
/
OH (GOH) and Hꢀ
/
/
Glu(ꢀAspꢀGlyꢀ
/
/
/
OH)ꢀ
/OH, however with at least one
/
309.
order of magnitude less affinity. These data are con-
sistent with slight variations in the G-site structure
among these isoenzymic classes, which result in different
hydrogen-bonding co-substrate interactions involving
protein residues other than the conserved ones.
These results suggest that substitution of the oxygen
for the g-CH2 bound to the g-Glu amide carbonyl is
responsible for a non productive binding of the GSH
analogues to the G-site, confirming that the g-glutamyl
moiety is the major binding determinant in protein-
ligand multiple interaction pattern. Accordingly, the
cysteinyl-containing isostere shows a poor propensity to
act as a co-substrate in the GST-mediated conjugation
with CDNB.
The lack of inhibitory activity observed for the new
compounds may be explained in terms of both con-
formational and electronic consequences of the isosteric
replacement. Several conformational investigations on
this peptide bond surrogate are centered on N-terminal
protected tert-butoxycarbonyl- and benzyloxycarbonyl-
amino acids and peptides [27,28,31]. A notable excep-
tion is represented by the work of Parkinson et al., who
reported in detail the conformational properties of
urethane linkages when inserted into the peptide back-
bone [32]. They evidenced the tendency of this moiety to
/
/
6454.
/217.
/
/
/
[9] G. Ricci, A.M. Caccuri, M. Lo Bello, N. Rosato, G. Mei, M.
Nicotra, E. Chiessi, A.P. Mazzetti, G. Federici, J. Biol. Chem. 271
(1996) 16187ꢀ
[10] A.J. Oakley, M. Lo Bello, A. Battistoni, G. Ricci, J. Rossjohn,
H.O. Villar, M.W. Parker, J. Mol. Biol. 274 (1997) 84ꢀ100.
[11] A. Gustafsson, M. Etahadieh, P. Jemth, B. Mannervick, Bio-
chemistry 38 (1999) 16268ꢀ16275.
[12] A.E.P. Adang, D.J. Meyer, J. Brussee, A. van der Gen, B.
Ketterer, G.J. Mulder, Biochem. J. 264 (1989) 759ꢀ764.
/
16192.
/
/
/
[13] M.H. Lyttle, M.D. Hocker, H.C. Hui, C.G. Caldwell, D.T.
˚
Aaron, A. Engqvist-Goldstein, J.E. Flatgaard, K.E. Bauer, J.
Med. Chem. 37 (1994) 189ꢀ
[14] O.W. Griffith, A. Meister, Proc. Natl. Acad. Sci. USA 76 (1979)
268ꢀ272.
[15] S.S. Tate, A. Meister, Mol. Cell. Biochem. 39 (1981) 357ꢀ
[16] A.E.P. Adang, J. Brussee, A. van der Gen, G.J. Mulder, J. Biol.
Chem. 266 (1991) 830ꢀ836.
[17] G. Luisi, A. Calcagni, F. Pinnen, Tetrahedron Lett. 34 (1993)
2391ꢀ2392.
[18] A. Calcagni, G. Lucente, G. Luisi, F. Pinnen, D. Rossi, Amino
Acids 17 (1999) 257ꢀ265.
[19] G. Lucente, G. Luisi, F. Pinnen, Farmaco 53 (1998) 721ꢀ
[20] A. Calcagni, S. Dupre`, G. Lucente, G. Luisi, F. Pinnen, D. Rossi,
Int. J. Pept. Protein Res. 46 (1995) 434ꢀ439.
[21] A. Calcagni, S. Dupre`, G. Lucente, G. Luisi, F. Pinnen, D. Rossi,
A. Spirito, Arch. Pharm. Pharm. Med. Chem. 329 (1996) 498ꢀ
/194.
/
/368.
/
/
/
adopt a trans ꢀ
ester and amide portions of the extended planar system,
respectively; in this particular arrangement the OꢀCO
/trans conformation at the level of the
/735.
/
/
fragment features the pseudopeptide oxygen in place of
the peptide Ca atom and its lone pairs at the position of
the Cb group. The lack of activity in the tested
compounds should be ascribed to alterations in enzyme
interactions connected with the specific conformational
properties of the urethane bond. However, the likely
occurrence of alternative conformations at the modified
g-Glu linkage is not the only explanation for inactivity;
the electronic properties as well as the different H-
bonding patterns of the urethane junction may be as
well of significance in modulating the interactions in the
protein active sites.
/
502.
[22] E. Morera, M. Nalli, F. Pinnen, D. Rossi, G. Lucente, Bioorg.
Med. Chem. Lett. 10 (2000) 1585ꢀ
[23] R. Vince, J. Brownell, L.B. Akella, Bioorg. Med. Chem. Lett. 9
(1999) 853ꢀ856.
[24] D. Burg, D.V. Filippov, R. Hermanns, G.A. van der Marel, J.H.
van Boom, G.J. Mulder, Bioorg. Med. Chem. 10 (2002) 195ꢀ205.
[25] W.-J. Chen, C.C. Boehlert, K. Rider, R.N. Armstrong, Biochem.
Biophys. Res. Comm. 128 (1985) 233ꢀ240.
[26] G.F. Graminski, Y. Kubo, R.N. Armstrong, Biochemistry 28
(1989) 3562ꢀ3568.
[27] E. Benedetti, C. Pedone, C. Toniolo, G. Nemethy, M.S. Pottle,
H.A. Scheraga, Int. J. Pept. Protein Res. 16 (1980) 156ꢀ172.
[28] A.C. Chernovitz, T.B. Freedman, L.A. Nafie, Biopolymers 26
(1987) 1879ꢀ1900.
/
1588.
/
/
/
/
/
/
[29] G. Ricci, G. Del Boccio, A. Pennelli, M. Lo Bello, R. Petruzzelli,
A.M. Caccuri, D. Barra, G. Federici, J. Biol. Chem. 266 (1991)
Acknowledgements
21409ꢀ21415.
/
[30] A.M. Caccuri, G. Antonini, P. Ascenzi, M. Nicotra, M. Nucce-
telli, A.P. Mazzetti, G. Federici, M. Lo Bello, G. Ricci, J. Biol.
The Istituto di Chimica Biomolecolare del CNR,
Sezione di Roma, and financial support from the
Ministero dell’Universita` e della Ricerca Scientifica e
Tecnologica (MURST, 60%) (Italy) are gratefully ac-
knowledged.
Chem. 274 (1999) 19276ꢀ
[31] E. Benedetti, C. Pedone, C. Toniolo, M. Dudek, G. Nemethy,
H.A. Scheraga, Int. J. Pept. Protein Res. 21 (1983) 163ꢀ181.
[32] G.N. Parkinson, Y. Wu, P. Fan, J. Kohn, J. Baum, H.M.
Berman, Biopolymers 34 (1994) 403ꢀ414.
/19280.
/
/