Design, synthesis and activity of bisubstrate, transition-state analogues and competitive inhibitors of aspartate transcarbamylase

Article abstract of DOI:10.1016/j.ejmech.2004.01.006

Aspartate transcarbamylase initiates the de novo biosynthetic pathway for the production of the pyrimidine nucleotides, precursors of nucleic acids. This pathway is particularly active in rapidly growing cells and tissues. Thus, this enzyme has been tested as a potential target for antiproliferative drugs. In the present work, on the basis of its structural and mechanistic properties, a series of substrate analogues, including potential suicide-pseudosubstrates was synthesized and their putative inhibitory effects were tested using E. coli aspartate transcarbamylase as a model. Two of these compounds appear to be very efficient inhibitors of this enzyme.

Full text of DOI:10.1016/j.ejmech.2004.01.006

European Journal of Medicinal Chemistry 39 (2004) 333–344
www.elsevier.com/locate/ejmech
Original article
Design, synthesis and activity of bisubstrate, transition-state analogues
and competitive inhibitors of aspartate transcarbamylase
a,
a
a
a
a
Claude Grison *, Philippe Coutrot , Corinne Comoy , Laurence Balas , Stéphane Joliez ,
a
a
b
b
b
Guido Lavecchia , Patrick Oliger , Bernadette Penverne , Valérie Serre , Guy Hervé
a
INCM FR CNRS 1742, UMR CNRS-UHP Nancy I 7565, Université Henri Poincaré, institut nanceien de chimie moléculaire, laboratoire de chimie
organique biomoléculaire, faculté des sciences et techniques, BP 239, 54506 Vandoeuvre-lès-Nancy cedex, France
b
Laboratoire de biochimie des signaux régulateurs cellulaires et moléculaires, CNRS UMR 7631, Université Pierre et Marie Curie,
96, bd Raspail, 75006 Paris, France
Received 10 October 2003; received in revised form 13 January 2004; accepted 15 January 2004
Abstract
Aspartate transcarbamylase initiates the de novo biosynthetic pathway for the production of the pyrimidine nucleotides, precursors of
nucleic acids. This pathway is particularly active in rapidly growing cells and tissues. Thus, this enzyme has been tested as a potential target
for antiproliferative drugs. In the present work, on the basis of its structural and mechanistic properties, a series of substrate analogues,
including potential suicide-pseudosubstrates was synthesized and their putative inhibitory effects were tested using E. coli aspartate
transcarbamylase as a model. Two of these compounds appear to be very efficient inhibitors of this enzyme.
©
2004 Published by Elsevier SAS.
Keywords: Phosphonopeptides; Fluorinated phosphonates; Carbamylphosphate analogues; ATCase inhibitors
1
. Introduction
[7]. Numerous analogues of PALA were synthesized, but
they are often poor inhibitors of ATCase [8]. These ap-
proaches were based on structural rather than mechanistic
considerations.
In this work, we describe an investigation to the rational
design of ATCase inhibitors based on the understanding of
the mechanism of the catalytic reaction. This approach con-
sisted in the synthesis and assay towards ATCase enzymatic
activity of the bisubstrate analogues 1 and 2a and 2b [8e], a
transition-state analogue 3, and the competitive inhibitors
Aspartate transcarbamylase (ATCase, EC 2.1.3.2) cataly-
ses the first unique step of the pyrimidine pathway, that is the
carbamylation of the amino group of L-aspartate by carbam-
ylphosphate to produce N-carbamyl-L-aspartate [1]. A high
activity of this enzyme is a marker of rapidly dividing cells
[2]. As such, this enzyme is an interesting target for the
development of antiproliferative drugs and it is not surprising
that numerous approaches to the design of new antitumoral
agents were based on the search for ATCase inhibitors [3].
N-(Phosphonoacetyl)-L-aspartate (PALA), a bisubstrate ana-
logue of this enzyme, is a powerful inhibitor of its activity
4
a–b (Fig. 1).
From archae to mammals the structure of the ATCase
catalytic site and the aminoacids involved in substrate bind-
ing and catalysis are fully conserved [9]. Consequently, in
this study, the E. coli enzyme was used as a model. This
enzyme is made from the association of two trimeric catalytic
subunits with three dimeric regulatory subunits [1]. Its activ-
ity is feedback inhibited by the endproducts CTP and UTP,
and activated by ATP. These three nucleotides bind competi-
tively to the same allosteric sites on the regulatory subunits.
E. coli ATCase can be dissociated into its catalytic and
regulatory subunits. The isolated catalytic subunits are active
but, in contrast with the native enzyme, they show neither
[4]. This compound was shown to inhibit proliferative
growth of mammalian cancer cells and to exhibit consider-
able activity against certain transplantable solid tumours in
mice [5]. It has been used in clinical investigation, but a
notable decrease of effectiveness was observed, possibly due
both to difficult transportation of PALA to the enzyme active
site [6] and development of resistance by gene amplification
*
Corresponding author.
E-mail address: claude.grison@lco2.uhp-nancy.fr (C. Grison).
©
2004 Published by Elsevier SAS.
doi:10.1016/j.ejmech.2004.01.006

334
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
H
N
H
N
H
N
X
Y
NaO
P
COOH
COOH
COOH
COOH
COONa
COONa
X
OMe
(
HO)2P
O
C
O
(HO)2P
C
(NaO)2P
O
(
NaO)2P
O
C
O
O
O
O
1
2a X = H, Y = F
b X = Y = F (8e)
Fig. 1. Structure of the putative inhibitors synthesized and used in this study.
3
4a X = CH2
4b X = NH
2
sensitivity to the allosteric effectors nor homotropic coopera-
tive interactions between the catalytic sites for the binding of
the substrate aspartate.
The method allows the preparation of various phospho-
nopeptides from the easily available dialkylphosphonoal-
kanoic acids as starting materials and thus could be used for
the preparation of PALG 1 (Scheme 1). Accordingly, the
synthesis was based on the coupling between diben-
zylphosphonoacetic acid and H-Glu(OBn)-OBn using BOP
as the coupling reagent in the presence of triethylamine. The
expected product 5 was obtained after usual treatment and
purification by chromatography on silica gel with 75% yield.
It was completely debenzylated by catalytic hydrogenation
under 1 bar in the presence of palladium on charcoal 10%, in
formic acid for 36 h. The resulting PALG 1 was obtained
after separation of the catalyst on celite. This quadruple
deprotection was quantitative (Scheme 1).
2. Chemistry
The rationales for the devise of these putative inhibitors
and the procedures for their syntheses were as follows.
2.1. Synthesis of bisubstrate analogues of ATCase
2.1.1. Synthesis of N-phosphonacetyl-L-glutamate (PALG)
1
N-phosphonoacetyl-L-glutamate (PALG) was considered
as potentially able to stabilize the T conformation ofATCase.
In this enzyme, the homotropic cooperative interactions be-
tween the catalytic sites for aspartate binding are explained
by a transition from an enzyme conformation which has a
low affinity for aspartate (T state) to a conformation which
has a high affinity for this substrate (R state). The binding of
the two substrates or of PALA to the active site of a catalytic
chain induces a domain closure in which the carbamyl
phosphate-binding domain and the aspartate-binding domain
come closer [1]. This intramolecular movement is supposed
to induce the T–R transition. Consequently, in the T state, the
amino acid side chains of the catalytic site that bind to the
phosphate group and the carboxyl groups of aspartate or
PALA are more distant than in the R state. It was, thus,
postulated that PALG in which these groups are more distant
than in PALA, would bind to the T state and fail to promote
the domain closure, stabilizing, and at the same time, inhib-
iting this low affinity conformation.
2
.1.2. Synthesis of N-phosphonofluoroacetyl-
and N-phosphonodifluoroacetyl-L-aspartate (PALA(F) 2a
and PALA(FF) 2b
Because of its small size (close to the size of a hydrogen
atom) and the high energy of the carbon-fluorine bond, fluo-
rine atom has proven to be a good choice to design enzyme
inhibitors. In the case of ATCase inhibition, the introduction
of one or two fluor atoms in a-position of the phosphorus
atom could provide information on the mechanism of the
uptake of PALA in cytoplasm. It is hypothesized that PALA
enters the cells via endocytosis, but this supposition is always
debated. A best knowledge of the uptake of PALA could help
to devise new analogues. With one or two fluor atoms, the
second dissociation of the phosphonic acid moiety of 2a and
2b (Fig. 1) should occur more readily than with PALA,
because of the great electronegativity of the fluor [11]. If
PALA enters the cells by endocytosis, fluorinated PALA
would be partially protonated at the lysosomal acid pH. An
increase in the net charge of the inhibitor should decrease the
rate of diffusion across the lysosomal membrane. Conse-
In a previous paper, we described a general route to phos-
phopeptides with a N-acylphosphonate terminal group [10].
H
BOP, Et N
3
N
COOBn
COOBn
(
BnO) P CH COOH + H-Glu(OBn)-OBn
2 2
(BnO)2P
C
O
CH Cl , 25ºC, 45 mn
2
2
O
O
H
N
5
(75%)
COOH
COOH
(
HO)2P
O
C
O
H /Pd-C, HCOOH
2
5
1
bar, 25ºC, 36 h
PALG 1 (98%)
Scheme 1.

C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
335
X
Y
H
N
X
Y
OH
BOP, Et3N
COOR
COOR
(
EtO)2P
C
(
EtO)2P
C
+
H-Asp(OR)-OR
R = Me, tBu
CH Cl , 25ºC
O
O
2
2
O
O
2
2
a X = H,Y= F
b X = Y = F
6a-b/6'a-b
Scheme 2.
X
H
N
Y
1
2
) Me3SiBr 2.5eq
) MeOH
COOMe 7a X = H, Y = F (98%)
7b X = Y = F (98%)
H
(HO)2P
O
C
X
Y
R = Me
COOR
N
O
(
EtO)2P
C
COOMe
H
N
X
Y
O
O
COOR R = tBu
1
2
) Me SiBr 4eq
COOH 2a : PALA(F) X = H, Y = F (85%)
b : PALA(FF) X = Y = F (85%)
3
6
(HO)2P
O
C
O
2
) MeOH
COOH
Scheme 3.
quently, PALA(F) 2a and PALA(FF) 2b could provide new
information on the mechanism of PALA uptake .
The phosphonopeptides 6a, 6a′ and 6b, 6b′ were synthe-
sized by coupling L-aspartic acid dialkylester hydrochloride
with fluoro- or difluorophophonoacetic acid [12,13] (Scheme
sodium hydroxide solution (Table 2). As expected, the effect
of fluor substitution into PALA was very pronounced. It
could be calculated (Fig. 2) that, at pH 5, 39% of 7b was a
tetraanion, whereas 99% of PALA was trianionic. As a con-
sequence, PALA, 7a and 7b should have a different behav-
iour in vitro towards the target enzyme in the hypothesis of a
membrane uptake by endocytosis. PALA (F) 2a, and PALA
(FF) 2b should have still more difficulties to cross the lyso-
somal membrane.
1
2). In the presence of BOP as coupling reagent, the desired
peptides were obtained in good yields (Table 1). The reaction
occurred at room temperature with two equivalents of tri-
ethylamine in dichloromethane and was complete after stir-
ring for 45 min when X = H, Y = F and 2 h when X= Y = F.
The formed HMPT was easily removed by chromatography
on silica gel. The chemioselective deprotection of the dieth-
ylphosphonic ester of 6a and 6b was then achieved with a
small excess of bromotrimethylsilane (2.5 equivalents) fol-
lowed by methanolysis of the disilylated phosphonate. The
resulting crude products were washed with ether and af-
forded the compounds 7a and 7b [14]. Simultaneous removal
of the diethylphosphonic ester and tert-butyl carboxylic ester
groups was carried out in a one-step sequence involving the
reaction between 6′a or 6′b and a large excess of bromotrim-
ethylsilane (four equivalents, 6 h in dichloromethane) fol-
lowed by methanolysis. After vacuum evaporation, the phos-
phonic acids were obtained as a brown pasty solid, which
was purified by washing with acetone. In these conditions,
2
.2. Synthesis of the transition-state analogue PALA(P), 3
A potential mechanism-based inhibitor was designed. If
we consider the enzymatic ATCase reaction and the mecha-
nism of carboxamide bond formation, the reactive centre is a
tetrahedral site. As a consequence, PALA is not a real transi-
tion state mimic as previously claimed, but a bisubstrate
analogue. Thus, we postulated that incorporation of a phos-
phorus tetrahedral atom, instead of the carbonyl carbon atom,
would provide a compound which would resemble more
closely the transition state and could improve notably the
affinity for the catalytic site of ATCase (Fig. 3).
2
.2.1. Synthesis of PALA(P) 3
In a previous paper, we described the one-pot synthesis of
6′a and 6′b afforded directly the entirely deprotected target
phosphonophosphonamide 8 [15]; we wish to report herein
the complete deprotection of 8 to give the target compound 3.
The choice of the protective groups in 8 was directed by the
fragility of the P–C–P–N chain. Actually, it is well known
that unsymmetrical alkylidene diphosphorylated derivatives
are unstable in the presence of nucleophiles since these later
attack the most electrophilic phosphorus atom, leading to a
P–C cleavage. In other hand, it is also well documented that
P–N linkage are acid sensitive [16].
molecules PALA(F) 2a and PALA(FF) 2b (Scheme 3). If the
phosphonic acids are suspected to being contaminated with
siloxanes they may be purified by treatment with cyclohexy-
lamine. Cyclohexylammonium salts were isolated by pre-
cipitation from acetone.
2
The pKa values of 7a, 7b and PALA were evaluated with
an automatic titrator (Tacussel TT Processor 2) using a 0.1 N
1
The best deprotection conditions were the tandem reac-
tion silylation/hydrolysis with careful control of pH. The use
of a large excess of triethylamine (13 equivalents) during the
N-Phosphonodifluoroacetyl-L-aspartate 2b has been described by Lin-
dell and Turner [8e] but ATCase-inhibitory properties has been evaluated
with mung bean ATCase.

336
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
4
3
3
2
2
1
1
0
5
0
5
0
5
0
5
0
2
X
H
H
F
Y
H
F
pK
a
PALA
7.70
6.20
5.20
7
7
a
b
F
PALA
PALA(F) 2a
PALA(FF) 2b
Fig. 2. Degree of ionisation of the fluorinated derivatives of PALA.
COO
NH2
COO
COO
COO
COO
(
O) P O C NH +
( O)2P
O
O
C
O
NH
( O) P O
+
2
2
2
O
O
H N
2
COO
H
O
H N
C
NH
2
O
transition state of the reaction
catalysed byATCase
O
NH
COO
EtO
P
NH
COO
NH
COOMe
(
O)2P
O
P
(
O)2P
C
O
(
EtO)2P
O
O
O
O
COO
COO
COOMe
PALA
transition-state mimic for ATCase
inhibition PALA(P) 3
8
Fig. 3. Structure of the transition state analogues PALA(P).
silylation was a good solution in response to the need for
smooth phosphonic ester-deprotection in the presence of
acid-sensitive P–N bond (Scheme 4). Addition of triethy-
lamine as scavenger of the in situ generated bromhydric acid
efficiently avoided the P–N linkage degradation. The silyla-
tion was slowed down and needed 3 days to complete the
reaction. The treatment of the resulting trimethylsiloxyphos-
phoryl moieties with sodium hydroxide in methanol led to
the expected phosphonophosphonamide 3 in interesting
atom linked to the nitrogen atom (PN) resonated as a ddt,
while the other phosphorus atom resonated as dt only. This
result was confirmed by the detection of a correlation be-
tween the CH proton of aspartate and the phosphorus atom
PN in a bidimensional P/ H-NMR experiment. Positive
electron spray mass spectrum was in good agreement with
complete deprotected structure.
3
1
1
2.2.2. Stability of PALA(P) 3
yield (69%).
The lability of the P–C–P chain required a study of the
stability of the transition-state analogue PALA(P) under con-
ditions where the inhibition of ATCase was measured.
2
Observation of a triplet (2.17 ppm, J = 18 Hz) in
HP
1
1
13
H-NMR, a triplet (32.0 ppm, J = 114 Hz) in C-NMR
CP
3
1
revealed the presence of the P–C–P linkage. The AM system
The stability of this compound was studied in P-NMR
2
–5
(
two doublets, 22.6 ppm and 14.1 ppm, JPP = 9 Hz) showed
under the followed conditions: 8.10 mole of phosphono-
the non-equivalence of the two phosphorus atoms and was in
accordance with the P–C–P–N chain. Multiplicities observed
in non H-decouplated P-NMR showed the connexion of
the amino acid residue on the P–C–P moiety: the phosphorus
phosphonamide 3 (the sensitivity of the NMR technique
needed to concentrate the phosphorylated compound eight
times as compared to the conditions of the inhibition assay)
was added to an aqueous solution of tris-acetate (20 ml,
1
31
H
H
EtO
NaO
N
COOMe
COOMe
1) Me SiBr (5eq), Et N (13eq)
3
3
N
COONa
COONa
(
EtO)2P
O
P
(NaO)2P
O
P
ClCH CH Cl
2
2
O
O
2) NaOH 1M / MeOH
Scheme 4.
8
PALA(P) 3 (69%)

C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
337
%) 1
00
(
9
0
0
0
0
0
0
0
0
0
0
8
7
6
5
4
3
2
1
conditions 1: 20°C, without substrates of ATCase
conditions 2: 37°C, without substrates of ATCase
conditions 3: 37°C, with substrates of ATCase
Sé3rie1
Sé3rie’2
1
2
3
conditions
Fig. 4. Degradation of PALA(P3) .
–
3
4
2
0.10 M, pH = 8.0). Different experiments were realized at
0 °C or 37 °C with or without the two substrates of ATCase,
2.3. Synthesis of the competitive inhibitors CP(C), 4a
and CP(N), 4b
carbamylphosphate and aspartate. After stirring for 10 min,
an analytical sample was cut off reaction solution and analy-
sed rapidly in P-NMR in sweep-off mode. Results are
reported in Fig. 4.
3
1
We propose here a new concept of inhibition based on the
synthesis of a chemically reactive substrate analogue of AT-
Case able to react with the other substrate of the enzyme to
generate in situ an inhibitor of ATCase. The principle is the
use of the enzyme as a catalyst for the synthesis of its
self-inhibitor.
3
1
The P-NMR spectrum displayed two compounds, or
four if carbamylphosphate was present (Scheme 5):
•
•
carbamylphosphate (–4.6 ppm),
phosphate (0.0 ppm) produced by degradation of car-
bamylphosphate,
This new approach is illustrated by the synthesis of two
mimics of carbamylphosphate, CP(C) (4a) and CP(N) (4b)
where the carboxylic acid methyl ester 4a or the carbamic
acid methyl ester 4b could act as an electrophile towards
aspartate to afford respectively, in situ, PALA or a new PALA
analogue.
•
•
signal of 3 (17.3 ppm, 12.3 ppm) that decreased in
intensity when temperature increased,
singulet at 13.7 ppm that increased to the detriment of 3.
This signal was assigned to the tetrasodium salt 3′ de-
rived from the methylene bisphosphonic acid. This
formed side-product evidenced the sensitivity of the
phosphorus atom in the diphosphorylated compound 3
to nucleophilic attack of water.
2.3.1. Synthesis of methoxycarbonylmethylphosphonic acid
disodium salt 4a
As carbamylphosphate, the natural substrate of ATCase,
PALA(P) was thermally unstable. Nevertheless, the decom-
position was partial and easily quantified. Twenty-five per-
cent of carbamylphosphate and 43% of the transition state
analogues were degraded after 10 min at 37 °C. It was
important to notice that the effect of the tetrasodium salt 3′
derived from the methylene bisphosphonic acid on the AT-
Case was known and did not interfere with the measure of the
ATCase activity (see paragraph 3).
Methyl diethylphosphonoacetate is commercially avail-
able and can be used without purification as starting material.
Chemioselective deprotection of the diethylphosphonic
ester was achieved with a small excess of bromotrimethylsi-
lane (2.5 equivalents) followed by a methanolysis to keep the
methoxycarbonyl group. Washing of the resulting brown
solid with methanol afforded quantitatively the target com-
pound 4a (Scheme 6).
(
O)2P O C NH2
( O) P O
2
O
O
O
-
4.6ppm
0
.0ppm
H
N
NaO
COONa
COONa
NH2
COONa
COONa
(
NaO)2P
O
P
(
NaO)2P
P(ONa)2
O
+
O
O
3
3
'
+
17.3ppm + 12.3ppm
+
13.7ppm
Scheme 5.

338
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
1
) Me SiBr (2.5eq), ClCH CH Cl
3
2
2
OMe
OMe
2
5ºC, 5 h
(NaO)2P
O
C
O
(
EtO)2P
O
C
O
2) MeONa/MeOH, 25ºC , 15 min
CP(C) 4a (99%)
Scheme 6.
Et O
thylchloroformiate led to the formation of the stable interme-
diate 10′. Two equivalents of NaH were essential for a quan-
titative formation of 10′. Subsequent acidification of the
reaction mixture with a saturated HCl ether solution at
pH = 7, centrifugation and removal of the solvents led to pure
compound 10 in good yield (79%) (Scheme 8).
2
(
EtO)2P C1
+
NH3
(EtO)2P NH2 + NH4Cl
O
-
35ºC, 1h
O
9
(87%)
Scheme 7.
2
.3.2. Synthesis of methoxycarbonylamidophosphonic acid
This simple procedure constitutes a good phosphonocar-
bamate 10 synthesis, superior to known methods.
disodium salt 4b
Several methods are reported in literature for the synthesis
of the methyl (diethylphosphono)carbamate precursor 10
After demonstrating that this process gave an attractive
access to 10 we investigated the chemioselective deprotec-
tion of the diethylphosphonyl group. The treatment of 10
with 8 equivalents of bromotrimethylsilane and a large ex-
cess of triethylamine afforded the methyl (ditrimethylsi-
lylphosphono)carbamate. The complete silylation was slow
(Scheme 8).
The decomposition of dialkylphosphonourea into dialky-
lphosphonoisocyanate under acidic conditions, followed by
alcoholysis has been described for the preparation of carbam-
ate 10 [17].
(56 h). After basic methanolysis, compound 4b was isolated.
A different approach was reported by Kirsanov and Mare-
nets [18] who built the urethanphosphoric ester by thermal
decomposition of trichlorophosphazocarbonic ester. These
later can also be transformed into urethanphosphoric ester by
solvolysis. The two ways use the phosphazocarbonic inter-
mediate, an explosive reagent.
The crude product was purified by washing with CH Cl ,
2
2
AcOEt and Et O and then filtered. The solid residue was then
2
triturated with the mixture MeOH/Me CO to eliminate
2
NaBr. These operationally simple treatments led to the pure
compound 4b in 50% yield (Scheme 9).
These preparations are rather laborious, and we report
here a simple and efficient synthesis of this useful derivative
3. Enzymology
10, starting from versatile reagents.
In a first step, chloride ion of diethylchlorophosphate was
easily displaced by ammoniac. The reaction was fast (1 h at
35 °C) and no by-product was formed. After centrifugation
and removal of the solvent, diethylamidophosphate 9 was
isolated without difficulty. The crude product was purified by
distillation (Scheme 7).
On the basis of the above-mentioned working hypotheses
the influence of the different compounds obtained on the
activity of E. coli, ATCase was examined under various
conditions as indicated in Experimental, in the presence of
5 mM carbamylphosphate, 20 mM [ C]-L-aspartate or
otherwise indicated.
–
1
4
The second step is the methoxycarbonylation of 9. We
31
have examined this reaction in some detail by P-NMR
spectroscopy. Deprotonation of 9, in THF at 60 °C, with two
equivalents of sodium hydride produced the anion 9′ which
was stable at this temperature. The formation of 9′ was
complete after 10 min. Addition of one equivalent of me-
3.1. Influence of the bisubstrate analogues PALG, PALA(F)
and PALA(FF) on ATCase activity
The response of theATCase activity to increasing concen-
trations of PALG, PALA(F) and PALA(FF) is shown in
Na
EtO)2P NH2 NaH (2eq) THF (EtO)2P NHNa
CH OCOC1
3
HC1
(
(EtO)2P
O
N COOMe
(EtO)2P NH C OMe
O
10ºC, 1 h
Et2O
O
60ºC, 10 min
O
O
9
(9 ppm)
9' (23 ppm)
10' (5 ppm)
10 (-4 ppm) (79%)
Scheme 8.
1
) Me SiBr (8eq), Et N,
3
3
(
EtO) P NH C OMe
(NaO) P NH C OMe
2
2
C1CH CH C1
2
2
O
O
O
O
2)MeONa / MeOH
1
0
CP(N) 4b (50%)
Scheme 9.

C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
339
Fig. 6. Influence of the transition state analogues PALA(P) on ATCase
activity. The ATCase activity was measured as indicated in “Experimental”
in the presence of increasing concentrations of PALA(P). (•-•): in absence of
ATP; (h-h): in the presence of 1 mM ATP.
importance of the stereo-electronic effects on the loose bind-
ing at the enzyme active site. PALA(FF) was previously
reported to be an efficient inhibitor of mung bean ATCase
[8e]. This difference is most probably due to the fact that
plant ATCases have a much simpler organization than the
homologous enzymes from E. coli and mammals, since they
are simple catalytic trimers in which the catalytic site is more
readily accessible.
The slight activation which is observed in the presence of
low concentrations of PALA is known and expected. It re-
sults from the influence of this analogue on the T–R equilib-
rium which is not entirely shifted towards R in the presence
of 5 mM aspartate.
Fig. 5. Influence of the different substrate analogues onATCase activity. The
ATCase activity was measured as indicated in “Experimental” in the absence
or presence of increasing concentrations of the different substrate analogues
and PALA as a control. (A) Effect of the inhibitors in the presence of 5 mM
aspartate. (C-C): PALG; (h-h): PALA(F); (D-D): PALA(FF); (•-•): PALA.
3
.2. Influence of the transition-state analogue PALA(P) on
ATCase activity
(B) Effect of PALG on ATCase and its isolated catalytic subunits in the
presence of 20 mM aspartate. (h-h): ATCase; (k-k): isolated catalytic
subunits.
The influence of the putative transition-state analogue
PALA(P) on ATCase activity was measured, using the same
methodology. It can be seen in Fig. 6 that, unexpectedly, this
compound has no inhibitory effect on the native enzyme, but
that at concentrations higher than 0.2 mM, it behaves as a
activator whatever the aspartate concentration. Such an acti-
vation was not observed in the case of the isolated catalytic
subunits, a result which suggested that this stimulation could
result from a shift in the T–R equilibrium towards the R
conformation. However, the fact that neither the aspartate
concentration nor the presence of 1 mM ATP (Fig. 6) de-
creases the extent of this activation rules out this explanation.
Fig. 5A in comparison with the influence of PALA under the
same conditions. It appears that PALG did not provoke inhi-
bition for concentrations lower than 1 mM. In order to test the
significance of the 20% inhibition observed at 5 mM, the
influence of this compound was measured in the presence of
20 mM aspartate, using both the native ATCase and the
isolated catalytic subunits.
Fig. 5B shows that this increase in aspartate concentration
does not alter the level of inhibition by PALG. In addition,
the native enzyme and its isolated catalytic subunits exhibit
the same sensitivity to this analog. Taken together, these
observations indicate that the small inhibition observed is not
due to the stabilization of the T conformation of the enzyme.
At a concentration of 5 mM, the inhibition by PALA(F)
raises to about 45%, at the same concentration PALA(FF)
does not provoke inhibition. This difference points to the
3.3. Influence of the competitive inhibitors CP(C) 4a
and CP(N) 4b on the activity of ATCase
In the presence of 20 mM aspartate and 10 mM carbam-
ylphosphate, CP(C) and CP(N) appear to be very potent

340
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
compounds exhibit the same efficiency. TheATCase reaction
proceeds through an ordered mechanism in which carbam-
ylphosphate binds first and induces a conformational change
which allows the subsequent binding of the second substrate
aspartate. A consequence of this process is that the bi-
substrate analogue PALA binds to the catalytic site of this
enzyme in competition with carbamylphosphate. However
the interaction between the two carboxyl groups of this
compound with positively charged residues of the catalytic
site contributes energetically to its affinity for this site. In
contrast, CP(C) and CP(N) are analogues or carbamylphos-
phate alone.The fact that under similar conditions these two
inhibitors and PALA provoke inhibitions which are of the
same order of magnitude suggest that CP(C) and CP(N)
would bind more tightly to the catalytic site of ATCase than
carbamylphosphate. Further investigation is needed to deter-
mine whether these two new anti-metabolites are acting
strictly as very potent inhibitors in pure competition with
carbamylphosphate or as suicide substrates.
Fig. 7. Influence of the carbamylphosphate analogues CP(C) and CP(N) on
the ATCase activity. The ATCase activity was measured as indicated in
“
Experimental” in the presence of either CP(C): (•-•) or CP(N): (h-h).
5
5
5
. Experimental
.1. Chemistry
.1.1. General methods
Melting points were determined on an Electrothermal
IA9100 digital apparatus. Infrared spectra were obtained
–
1
using a Nicolet 205 spectrometer and are given in cm .
NMR spectra were recorded using a BRUKER AC250 spec-
1
13
trometer. For H and C-NMR data, chemical shifts are
reported in parts per million (d, ppm) downfield from CHCl
3
3
1
as an internal standard while P-NMR are reported with
1
9
8
5% H PO as an external standard. F-NMR are reported
3 4
from CFCl as standard. NMR coupling constants (J values)
3
are listed en hertz (Hz) and spin multiplicities are reported as
singulet (s), doublet (d), triplet (t), multiplet (m) and broad
(br). Mass spectra were obtained with a TRIO 5000 spec-
trometer. Organic solvents were purified according to the
methods described by W.L.F. Armagero and D.D. Perrin
[19]. All no aqueous reactions were performed in oven-dried
glassware under nitrogen atmosphere. n-Butyllithium and
s-butyllithium were purchased from Aldrich and were ti-
trated in tetrahydrofuran for n-BuLi and benzene for s-BuLi
according to the Watson and Eastham procedure [20].
Fig. 8. Kinetic of the ATCase reaction in the presence of CP(C) or CP(N).
The ATCase activity was measured as indicated in “Experimental” in the
presence of either CP(C): (k-k) or CP(N): (h-h) at a concentration of
0.15 mM.
inhibitors ofATCase (Fig. 7).At a concentration of 0.05 mM,
the inhibition is about 50% and raises to 80% at 0.33 mM.
In a preliminary attempt to detect a putative inactivation of
the enzyme by these compounds the kinetics of the ATCase
reaction were followed over 20 min in the presence of these
two inhibitors at a concentration of 0.15 mM. The result of
this experiment is shown in Fig. 8. No decrease of the rate of
the reaction is observed over that period of time.
3
1
Advancement of reactions was followed by P-NMR spec-
troscopy.
5
.1.2. Preparation of N-(phosphonoacetyl)-L-glutamic
acid 1
5
.1.2.1. Dibenzyl methylphosphonate. To a suspension of
freshly washed sodium hydride (NaH) (55% suspension in
mineral oil, 1.90 g, 79.2 mmol) in toluene (10.0 ml) was
added dropwise a solution of dibenzyl H-phosphonate
4
. Conclusion
Among the different substrate analogues synthesized and
tested in this study, only CP(C) and CP(N) appear to be
strong inhibitors of the activity of ATCase, and these two
(18.88 g, 72.0 mmol) in toluene (10.0 ml), under N atmo-
sphere, at room temperature and using an ice bath checked
2

C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
341
temperature reaction. After stirring for 1 h, end of gas libera-
tion was observed then iodomethane (10.22 g, 72.0 mmol) in
toluene (10.0 ml) was added dropwise. Reaction mixture was
stirring overnight at room temperature. After hydrolysis and
extraction with dichloromethane (CH Cl ), combined or-
the expected phosphonopeptide 5 was obtained as an oil in
75% yield: R (hexane/acetone: 1/1) = 0.75; IR (film) m 1010,
f
1
1250, 1680, 1743, 3200; H-NMR (CDCl ) d 1.90–2.40 (m,
4H, CH CH, CH CO), 2.89 (d, 2H, J = 21 Hz, PCH2),
3
2
2
2
HP
4.60–4.70 (m, 1H, CH), 4.90–5.15 (m, 8H, OCH Ph), 7.20–
2
2
2
3
1
13
ganic layers were dried over magnesium sulfate (MgSO4),
filtered and solvents were removed in vacuum to afford the
expected crude compound in 70% yield: IR (film) m 1050,
7.40 (m, 20H, Ph); P-NMR (CDCl ) d 20.6 (s, 1P); C-
3
1
NMR (CDCl ) d 28.5 (d, J = 175 Hz, PCH ), 29.0 (s,
3
CP
2
CHCH ), 31.5 (s, CH CO), 54.0 (s, CH), 69.2 (s, OCH ),
2
2
2
1
2
1
250; H-NMR (CDCl ) d 1.49 (d, 3H, J = 17.5 Hz,
127.7 (s, CPh), 127.8 (s, CPh), 128.2 (s, CPh), 136.0 (s,
Carom), 163.8 (s, CONH), 171.0 (s, COO), 172.1 (s, COO).
3
HP
PCH ), 4.90–5.20 (m, 4H, OCH ), 7.35–7.45 (m, 10H, Ph);
3
2
3
1
13
P-NMR (CDCl ) d 29.3 (s, 1P); C-NMR (CDCl ) d 11.0
3
3
1
2
CP 3 CP
(
1
d, J = 144 Hz, PCH ), 66.4 (d, J = 6 Hz, OCH2),
3
5
.1.2.4. N-(Phosphonoacetyl)-L-glutamic acid 1. A solution
27.3–128.0 (m, CHarom), 135.7 (d, J = 2 Hz, Carom).
CP
of phosphonopeptide 5 (0.82 g, 1.3 mmol) in formic acid
(50 ml) and 10% palladium on charcoal catalyst (0.10 g) was
stirred at room temperature under hydrogen atmosphere for
5
.1.2.2. (Dibenzylphosphono)acetic acid. n-Butyllithium (n-
BuLi) (17.5 ml of a 1.6 M solution in hexane, 28.0 mmol)
was added to tetrahydrofuran (THF) (28 ml) at –70 °C, under
nitrogen. Then a solution of dibenzyl methylphosphonate
3
1
36 h. Reaction advancement was followed by P-NMR
spectroscopy. Catalyst was removed by filtration using celite
and evaporation of formic acid under vacuum provided the
expected N-(phosphonoacetyl)-L-glutamic acid 1 as a solid
in quantitative yield: IR (KBr) m 1080, 1150, 1600, 1720,
(7.73 g, 28.0 mmol) in THF (6 ml) was added dropwise, at
–70 °C.After mechanic stirring for 30 min at –70 °C, reaction
solution was poured with stirring into a Dewar containing a
saturated dry ice/diethyl ether (Et O, 200 ml) solution. A few
minutes later, the mixture was poured into a beaker and
allowed to warm to room temperature (2 h) with stirring.
After hydrolysis with water (30 ml), organic layer was
washed with a saturated aqueous sodium hydrogencarbonate
1
2200–3700; H-NMR (D O) d 1.70–2.20 (m, 2H, CH CH),
2
2
2
3
2.38 (t, 2H, JHH = 7 Hz, CH COOH), 2.79 (d, 2H,
JHP = 20 Hz, CH P), 4.10–4.30 (m, 1H, CH); P-NMR
(D O) d 14.1 (s, 1P); C-NMR (D O) d 30.3 (s, CHCH ),
2
2
31
2
1
3
2
2
2
1
35.2 (s, CH CO), 39.1 (d, J = 168 Hz, PCH ), 57.8 (s,
2
CP
2
solution (NaHCO ). Combined aqueous layers were then
CH), 175.6 (s, COO), 182.3 (s, CONH). Anal. Calcd. for
C H NO P: C, 31.24; H, 4.49; N, 5.20. Found: C, 30.85; H,
3
washed twice with Et O, acidified to pH 1 with a 6 M HCl
2
7
12
8
aqueous solution, saturated with NaCl and extracted with
CH Cl . Organic layer was dried with MgSO and solvents
were removed under vacuum to give (dibenzylphospho-
4.70; N, 5.50; P, 11. 90.
2
2
4
5.1.3. Preparation of N-[(phosphono)fluoroacetyl)]-L-
aspartic acid (2a) and N-[(phosphono) difluoroacetyl]-
L-aspartic acid 2b
no)acetic acid as an oil in 60% yield: IR (film) m 1025, 1215,
1
1
720, 2500–3200; H-NMR (CDCl ) d 3.01 (d, 2H,
3
2
3
JHP = 22 Hz, PCH ), 5.08 (d, 4H, J = 8.5 Hz, POCH Ph),
.20–7.39 (m, 10H, Ph), 10.00 (br s, 1H, COOH); P-NMR
2
HP
2
31
5
.1.3.1. Peptidic coupling reaction.
5.1.3.1.1. N-[(Diethylphosphono)fluoroacetyl]-L-aspar-
7
13
(
CDCl ) d 22.0 (s, 1P); C-NMR (CDCl ) d 34.3 (d,
3 3
2
1
tic acid dimethyl diester 6a. Phosphonopeptide 6a was
obtained according to the coupling procedure described for
derivative 5 starting from L-aspartic acid dimethyl diester
JCP = 136 Hz, PCH ), 68.5 (d, J = 6 Hz, OCH ), 127.9 (s,
2
CP
2
2
CPh), 128.4 (s, CPh), 135.5 (s, CPh), 167.7 (d, J = 6 Hz,
CP
Carom).
hydrochloride (H N-L-Asp(OMe)-OMe, Cl) (0.79 g,
3
5
.1.2.3. N-[(Dibenzylphosphono)acetyl]-L-glutamic acid
4.0 mmol) and (diethylphosphono)fluoroacetic acid (0.86 g,
4.0 mmol). Reaction mixture was stirred for 45 min at room
temperature. A flash chromatography purification (eluent:
hexane/ethyl acetate) afforded the two diastereoisomers of
dibenzyl diester 5. Triethylamine (Et N) (0.40 g, 4.0 mmol)
was added to a solution of L-glutamic acid dibenzyl diester
tosylate (H N-L-Glu(OBn)-OBn, TsO) (2.00 g, 4.0 mmol) in
3
3
CH Cl (40 ml). Then, (dibenzylphosphono)acetic acid
phosphonopeptide 6a as an oil in 82% yield: IR (Film) m
2
2
1
(
1.28 g, 4.0 mmol) in CH Cl2 (20 ml), benzotriazol-1-
1150, 1250, 1682, 1743, 3200; H-NMR (CDCl
3
) d 1.36 (t,
2
3
2
yloxytris(dimethylamino)phosphonium
hexafluorophos-
6H, JHH = 7 Hz, OCH
2
CH
3
), 2.97 (dd, 1H, JH–H = 20 Hz,
3
2
phate (BOP) (1.77 g, 4.0 mmol) in CH Cl (20 ml) and Et N
JH–H = 5 Hz, CH COO), 2.99 (dd, 1H, J = 20 Hz,
2
2
3
2
H–H
3
(
0.40 g, 4.0 mmol) in CH Cl (10 ml) were added succes-
JHH = 5 Hz, CH COO), 3.71 (s, 3H, COOCH ), 3.78 (s, 3H,
2 3
2
2
sively. After stirring for 45 min at room temperature, CH Cl
COOCH ), 4.20–4.32 (m, 4H, CH O), 4.93 (dt, 1H,
2
2
3 2
3
3
2
3
2
2
2
2
(50 ml) was added. Organic layer was washed at first three
JHH = 8 Hz, J = 5 Hz, CH), 5.23 (dd, 1H, J = 46 Hz,
HH HF
3
times with a 10 N sulfuric acid solution, with a saturated
aqueous NaCl solution, with a saturated aqueous NaHCO3
solution then with a saturated aqueous NaCl solution again
J
= 11 Hz, PCHF), 7.43 (d, 1H, JHH= 8 Hz, NH);
HP
1
2
P-NMR (CDCl ) d 8.1 (d, 1P, J = 70 Hz), 8.4 (d, 1P,
JPF = 70 Hz); F-NMR (CDCl ) d 174.6 (dd, 1F,
JFP = 70 Hz, J = 46 Hz), -174.3 (dd, 1F, J = 70 Hz,
3
PF
19
3
2
2
FH FP
and dried with MgSO .After evaporation of solvent and flash
4
chromatography purification (eluent: hexane/ethyl acetate),
JFH = 46 Hz).

342
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
1
3
5.1.3.1.2. N-[(Diethylphosphono)difluoroacetyl]-L-aspar-
H-NMR (CDCl ) d 1.25 (t, 6H, J = 7 Hz, OCH CH ),
3 HH 2 3
tic acid dimethyl diester 6b. Phosphonopeptide 6a was
obtained according to the coupling procedure described for
derivative 5 starting from L-aspartic acid dimethyl diester
1.48 (s, 18H, (CH )3C), 2.90–3.00 (m, 2H, CH CO), 4.32–
3
2
4.38 (m, 4H, CH O), 4.65–4.75 (m, 1H, CH), 7.60 (d, 1H,
2
3
2
2
31
JHH= 8 Hz, NH); P-NMR (CDCl ) d 0.7 (t, 1P,
JPF = 96 Hz); F-NMR (CDCl ) d 117.7 (d, 1F,
JFP = 96 Hz), -117.5 (d, 1F, J = 96 Hz); C-NMR
(CDCl ) d 14.0 (d, JCP = 6 Hz, OCH CH ), 27.6 (s,
3 2 3
3
19
hydrochloride (H N-L-Asp(OMe)-OMe, Cl) (0.79 g,
3
3
2
13
4.0 mmol) and (diethylphosphono)difluoroacetic acid
FP
3
(0.93 g, 4.0 mmol). Reaction mixture was stirred for 2 h at
room temperature. A flash chromatography purification (elu-
ent: hexane/ethyl acetate) afforded the expected phospho-
nopeptide 6b as an oil in 79% yield: IR (Film) m 1250, 1710,
(CH )3C), 27.8 (s, (CH )3C), 36.7 (s, CH CO), 49.3 (s, CH),
3 3 2
2
60.2 (d, J = 5 Hz, OCH CH ), 81.8 (s, (CH )3C), 134.0–
CP
2
3
3
137.4 (m, CF ), 168.4 (s, COO), 169.5 (s, COO), 170.6 (s,
2
1
3
1
740, 3200; H-NMR (CDCl ) d 1.39 (t, 6H, J = 7 Hz,
CONH).
3
HH
2
3
OCH CH ), 2.90 (dd, 1H, J
CH COO), 3.12 (dd, 1H, J
= 20 Hz, J
= 20 Hz, J
= 5 Hz,
= 5 Hz,
H–H
2
3
H–H
H–H
H–H
2
3
5.1.3.2. Protective group removal.
2
5
.1.3.2.1. N-[(Phosphono)fluoroacetyl]-L-aspartic acid
CH COO), 3.71 (s, 3H, COOCH ), 3.78 (s, 3H, COOCH ),
2
3
3
3
dimethyl diester 7a. Bromotrimethylsilane (3.83 g,
25.0 mmol) was added to a solution of diethylphosphonopep-
tide dimethyl diester 6a (3.60 g, 10.0 mmol) in dichloroet-
hane (5 ml) under nitrogen atmosphere. Reaction mixture
was stirred for 3 h at room temperature. After evaporation of
solvent, excess of methanol was added. Then, removal under
4
.29–4.40 (m, 4H, CH O), 4.87 (dt, 1H, J
JHH = 5 Hz, CH), 7.62 (d, 1H, J = 8 Hz, NH); P-NMR
CDCl ) d 0.6 (t, 1P, J = 96 Hz); F-NMR (CDCl ) d
13.3 (d, 1F, J = 96 Hz).
= 8 Hz,
2
HH
3
3
31
HH
2
3 PF 3
19
(
1
2
FP
5.1.3.1.3. N-[(Diethylphosphono)fluoroacetyl]-L-aspar-
tic acid di-tert-butyl diester 6′a. Phosphonopeptide 6′a was
obtained according to the coupling procedure described for
derivative 5 starting from L-aspartic acid di-tert-butyl diester
vacuum of solvent, the expected derivative 7a was obtained
1
in quantitative yield: H-NMR (CDCl ) d 2.87–3.10 (m, 2H,
3
CH CO), 3.71 (s, 3H, COOCH ), 3.78 (s, 3H, COOCH ),
2
3
3
hydrochloride (H N-L-Asp(OtBu)-OtBu, Cl) (1.13 g,
3
2
4
.84–5.10 (m, 1H, CH), 5.40 (dd, 1H, JHF = 46 Hz,
JHP = 20 Hz, CHF), 7.43 (d, 1H, J = 8 Hz, NH), 11.20 (br
s, 2H, OH); P-NMR (CDCl ) d 6.0 (br s, 1P).
4
4
.0 mmol) and (diethylphosphono)fluoroacetic acid (0.86 g,
.0 mmol). Reaction mixture was stirred for 45 min at room
2
3
HH
3
1
3
temperature. A flash chromatography purification (eluent:
hexane/ethyl acetate) afforded the two diastereoisomers of
phosphonopeptide 6′a as an oil in 82% yield: Rf
5
.1.3.2.2. N-[(phosphono)difluoroacetyl]-L-aspartic acid
dimethyl diester 7b. Compound 7b was obtained according
to the coupling procedure described for derivative 7a starting
from bromotrimethylsilane (3.83 g, 25.0 mmol) and dieth-
ylphosphonopeptide dimethyl diester 6b (3.75 g,
(
1
acetone/hexane: 1/1) = 0.70; IR (Film) m 1150, 1267, 1695,
1
3
3 HH
737, 3200; H-NMR (CDCl ) d 1.19 (t, 6H, J = 7 Hz,
3
OCH CH ), 1.29 (t, 6H, J = 7 Hz, OCH CH ), 1.38 (s,
2
3
HH
2
3
1
0.0 mmol). Reaction mixture was stirred for 3 h at room
1
8H, (CH )3C), 2.50–2.80 (m, 2H, CH COO), 4.10–4.20
3
2
temperature. The expected derivative 7b was obtained in
(
m, 4H, CH O), 4.60–4.70 (m, 1H, CH), 5.23 (dd, 1H,
2
1
quantitative yield: H-NMR (CDCl ) d 2.91–3.09 (m, 2H,
2
3
2
2
2
2
3
J
= 46 Hz, J = 11 Hz, PCHF), 7.34 (br s, 1H, NH);
HF
HP
CH CO), 3.71 (s, 3H, COOCH ), 3.78 (s, 3H, COOCH ),
1
2
2
3
3
P-NMR (CDCl ) d 8.1 (d, 1P, J = 70 Hz), 8.8 (d, 1P,
JPF = 70 Hz); F-NMR (CDCl ) d 174.6 (dd, 1F,
JFP = 70 Hz, J = 46 Hz), -174.3 (dd, 1F, J = 70 Hz,
3
PF
3
4
.84–5.10 (m, 1H, CH), 7.62 (d, 1H, J = 8 Hz, NH), 10.45
br s, 2H, OH); P-NMR (CDCl ) d 3.1 (br s, 1P,
19
HH
3
31
(
2
2
3
FH
13
FP
2
JPF = 95 Hz).
.1.3.2.3. N-[(Phosphono)fluoroacetyl]-L-aspartic acid
a. Bromotrimethylsilane (1.23 g, 8.0 mmol) was added to a
3
JFH = 46 Hz); C-NMR (CDCl ) d 16.2 (d, J = 6 Hz,
3
CP
5
OCH CH ), 27.7 (s, (CH )3C), 27.8 (s, (CH )3C), 37.1 (s,
2
3
3
3
2
2
CH CO), 48.6 (s, CH), 48.8 (s, CH), 64.0 (d, J = 5 Hz,
2
CP
solution of diethylphosphonopeptide di-tert-butyl diester 6′a
0.88 g, 2.0 mmol) in dichloroethane (1 ml) under nitrogen
OCH CH ), 81.7 (s, (CH )3C), 82.5 (s, (CH )3C), 86.2 (dd,
2
3
3
3
(
1
2
1
JCP = 158 Hz, JCF = 206 Hz, PCHF), 163.6 (d,
JCP = 17 Hz, CONH), 168.6 (s, COO), 168.7 (s, COO),
atmosphere. After stirring for 6 h at room temperature,
methanol (0.200 g, 8.0 mmol) was added, and then the
reaction solution was allowed to stir for 2 h more. Evapora-
tion under vacuum of solvents afforded the expected deriva-
169.5 (s, COO), 169.6 (s, COO).
5.1.3.1.4. N-[(Diethylphosphono)difluoroacetyl]-L-as-
partic acid di-tert-butyl diester 6′b. Phosphonopeptide 6′b
was obtained according to the coupling procedure described
for derivative 5 starting from L-aspartic acid di-tert-butyl
tive 2a as an oil in 85% yield: IR (Film) m 1090, 1235, 1615,
1
1750, 2200–3700; H-NMR (CD CN) d 1.64 (m, 2H,
3
2
CH CO), 3.65–3.75 (m, 1H, CH), 4.16 (dd, 1H, JH–
= 46 Hz, PCHF), 6.23 (d, 1H, J
NH); P-NMR (CD CN) d 6.1 (br s, 1P); F-NMR
2
2
3
diester dibenzenesulfimide (H N-L-Asp(OtBu)-OtBu,
P
= 12 Hz, J
= 7 Hz,
3
H–F
H–H
19
3
1
(
PhSO ) N) (2.17 g, 4.0 mmol) and (diethylphosphono)dif-
2
2
3
luoroacetic acid (0.93 g, 4.0 mmol). Reaction mixture was
stirred for 2 h at room temperature. A flash chromatography
purification (eluent: hexane/ethyl acetate) afforded the ex-
pected phosphonopeptide 6′b as an oil in 90% yield: Rf
(CD CN) d 179.6–180.5 (m, 1F).
3
Purification of 2a was obtained by derivation to its tri(cy-
clohexyammonium) salt. To a solution of phosphonoacid 2a
(0.41 g, 1.5 mmol) in ethanol (2 ml) wad added dropwise
successively cyclohexylamine (0.45 g, 4.5 mmol) and dry
(acetone/hexane: 1/1) = 0.68; IR (Film) m 1250, 1740, 3200;

C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
343
acetone (50 ml). The solution was cooled at –18 °C and
5.1.5. Preparation of methoxycarbonylmethylene-
phosphonic acid disodium salt 4a
precipitation of a white solid was observed. After removing
of liquid phase, solid residue was washed twice with diethyl
ether and then dried by desiccation: IR (KBr) m 1090, 1235,
5
.1.5.1. Diethylphosphonoacetic acid methyl ester. Methyl
3
1
bromoacetate (12.0 g, 78.4 mmol) was warmed to 80 °C and
triethylphosphite (10.0 g, 60.2 mmol) was added dropwise
swiftly. Increase of reaction temperature until 130 °C al-
lowed to distil bromoethane. Stirring of reaction mixture was
extended for 30 min. Distillation under reduced pressure
afforded the expected diethylphosphonoacetic acid methyl
1
620, 1750, 2200–3700; P-NMR (D O) d 2.6 (t, 1P,
2
2
19
2
JPF = 76 Hz); F-NMR (D O) d -114.0 (d, 1F, J = 76 Hz).
2
FP
Anal. Calcd. for C H N O FP: C, 50.52; H, 8.48; N, 9.82;
2
4 48 4 8
F, 3.33; P, 5.43 Found: C, 50.92; H, 8.73; N, 9.50; P, 5.70.
5
.1.3.2.4. -[(Phosphono)difluoroacetyl]-L-aspartic
N
acid 2b. Compound 2b was obtained according to the proce-
dure described for derivative 2a starting from bromotrimeth-
ylsilane (1.23 g, 8.0 mmol) and diethylphosphonopeptide
di-tert-butyl diester 6′b (0.92 g, 2.0 mmol) and subsequent
addition of methanol (0.200 g, 8.0 mmol). Compound 2b was
afforded as an oil in 85% yield: IR (Film) m 1090, 1235, 1690,
ester as an oil in 88% yield: Eb
= 128 °C; IR (Film) m
1
0 mmHg
1
1
052, 1120, 1742; H-NMR (CDCl ) d 1.35 (t, 6H,
3
3
2
JHP = 7 Hz, CH CH O), 2.98 (d, 2H, J = 22 Hz, PCH ),
3
2
HP
2
31
3
.75 (s, 3H, OCH ), 4.10–4.20 (m, 4H, CH CH O); P-
3 3 2
NMR (CDCl ) d 17.2 (s, 1P).
3
1
2
1
H
H
750, 2200–3700; H-NMR (CD OD) d 2.52 (dd, 1H, J
3
H–
5.1.5.2. Methoxycarbonylmethylenephosphonic acid diso-
dium salt 4a. Bromotrimethylsilane (9.11 g, 59.5 mmol) was
added to a solution of diethylphosphonoacetic acid methyl
ester (5.0 g, 23.8 mmol) in dichloroethane (12 ml). After
stirring for 5 h, a 1 N MeONa in MeOH solution (60.0 ml,
59.5 mmol) was added. Reaction mixture was stirred for
15 min. After evaporation of solvents, residue was washed
with MeOH. After evaporation, solid residue was dissolved
3
2
= 17 Hz, J
= 6 Hz, CH COO), 2.54 (dd, 1H, J
= 6 Hz, CH COO), 4.49 (t, 1H, J = 6 Hz,
2 HH
H–H
2
H–
3
3
= 17 Hz, J
H–H
3
1
2
19
CH); P-NMR (CD OD) d 3.1 (t, 1P, J = 95 Hz); F-
NMR (CD OD) d 116.9 (d, 2F, J = 95 Hz).
3
PF
2
3
FP
Tetra(cyclohexyammonium) salt of phosphonoacid 2b
was prepared according to the method carried out for 2a
starting from compound 2b (0.44 g, 1.5 mmol) and cyclo-
hexylamine (0.59 g, 6.0 mmol): IR (KBr) m 1096, 1230, 1620,
in water and lyophilised to afforded the expected salt 4a as a
3
1
1
1
740, 2200–3700;
P-NMR (D O)
d
2.6 (t, 1P,
white solid in 99% yield: (KBr) m 1702; H-NMR (D O) d
2
2
2
19
2
2
JPF = 76 Hz); F-NMR (D O) d 114.1 (d, 1F, J = 76 Hz),
113.7 (d, 1F, JFP
2.67 (d, 2H, J = 20 Hz, PCH ), 3.69 (s, 3H, OCH );
2
FP
HP
2
3
2
31
1
13
2 2
-
=
76 Hz). Anal. Calcd. for
P-NMR (D O) d 10.5 (s, 1P); C-NMR (D O) d 40.6 (d,
C H N O F P: C, 52.39; H, 8.79; N, 10.18; F, 5.52; P,
JCP = 109 Hz, PCH ), 54.9 (s, OCH ), 177.7 (s, COO).
30
60
5
8
2
2
3
4.50 Found: C, 52.55; H, 9.05; N, 9.74. P, 4.85.
5
.1.6. Preparation of methoxycarbonylamidophosphonic
acid disodium salt 4b
5.1.4. Preparation of N-[(phosphonomethyl)phosphinato]-
L-aspartic acid 3
A solution of N-{[(diethylphosphono)methyl]ethoxy-
phosphinyl}-L-aspartic acid dimethyl diester 8 (1.25 g,
5
.1.6.1. Diethyl amidophosphate 9. A solution of diethyl
chlorophosphate (40.0 g, 231.6 mmol) and triethylamine
23.43 g, 231.6 mmol) in Et O (200.0 ml) was cooled to
(
–
2
3
4
.1 mmol) in CH Cl (16 ml) and triethylamine (5.82 ml,
1.9 mmol) was cooled to 0–5 °C then bromotrimethylsilane
2 2
35 °C. Gas ammoniac was added at –35 °C and with effi-
cient stir for 1 h. Ammonium chloride was removed by
centrifugation then evaporation of solvent followed by a
distillation under reduced pressure gave the expected phos-
phoramide 9 as a white amorphous solid in 87% yield:
(
2.05 ml, 15.51 mmol) was added dropwise. Reaction mix-
ture was stirred at room temperature for 3.5 days. A solution
M sodium hydroxide in methanol (31.0 ml) was then added
1
and stirring was extended for 1 h at room temperature. After
evaporation of solvents, residue was washed three times with
a mixture CH Cl /EtOH (1/1). Three steps of work-up were
1
Eb0.02 mbar = 120 °C; IR (KBr) m 1033, 1231; H-NMR
3
(CDCl ) d 1.34 (t, 6H, J = 7 Hz, CH ), 2.80 (br s, 2H,
3
HP
3
31
2
2
NH2), 4.05–4.15 (m, 4H, CH ); P-NMR (CDCl ) d 6.8 (s,
2
3
realized successively as follow: residue in suspension in
THF, or MeOH or EtOH was stirred, respectively, 1.5 h at
room temperature, 4 h at 65 °C and 2 h at 65 °C. After last
filtration, the expected compound 3 was obtained as a solid in
13
3
3 CP 3
1
P); C-NMR (CDCl ) d 16.0 (d, J = 7 Hz, CH ), 62.1 (d,
2
JCP = 5 Hz, CH2).
.1.6.2. Methyl N-(diethylphosphono)carbamate 10. A sus-
5
69% yield: IR (KBr) m 1096, 1447, 1596, 1672; MS (positive
pension of freshly washed sodium hydride (NaH) (55% sus-
pension in mineral oil, 2.07 g, 43.12 mmol) in THF (15 ml)
was stirred at 65 °C. Phosphoramide 9 (3.0 g, 19.6 mmol) in
THF (10 ml) was then added dropwise. After stirring for
10 min, reaction temperature was allowed to cool to 0 °C.
Methyl chloroformiate (1.85 g, 19.6 mmol) in THF (10 ml)
was added dropwise at 0 °C. After stirring for 1 h, a saturated
HCl diethyl ether solution was added until pH = 7 of reaction
mixture. After centrifugation of NaCl and evaporation of
+
+
+
1
electrospray) m/z 356 (M+H –2Na ), 424.1 (M+Na ); H-
2
NMR (D O) d 2.7 (t, 2H, J = 18 Hz, PCH P), 2.38 (dd,
2
HP
2
2
3
1
H, J
= 14 Hz, J
= 5 Hz, CH CO) 2.45 (dd, 1H,
H–H
H–H 2
2
3
JH–H = 14 Hz, J
CHN); P-NMR (D O) d 22.6 (d, 1P, J = 9 Hz, PN), 14.1
= 9 Hz, CH CO), 3.65–3.70 (m, 1H,
H–H
2
3
1
2
2
PP
2
13
(
d, 1P, JPP = 9 Hz, PO); C-NMR (D O) d 32.0 (t,
JCP = 114 Hz, PCH P), 44.4 (s, CH CO), 54.9 (s, CHN),
2
1
2
2
180.7 (s, COONa), 182.8 (s, COONa).

344
C. Grison et al. / European Journal of Medicinal Chemistry 39 (2004) 333–344
solvents, the expected derivative 10 was obtained as an oil in
M. Soroka, Synthesis (1982) 219–221; (l) G. Zanotti, H.L. Monaco,
J. Foote, J. Am. Chem. Soc. 106 (1984) 7900–7904; (m) H. Ke,
W.N. Lipscomb, Y. Cho, R.B. Honzatko, J. Mol. Biol. 204 (1988)
725–747; (n) J. Gloede, H. Gross, P. Henklein, H. Niedrich, S. Tan-
neberger, B. Tschiersch, Pharmazie 43 (1988) 434; (o) P. Henklein,
J.Z. Gloede, Z. Chem. 29 (1) (1989) 19–520; (p) A.D. Morris,
A.A. Cordi, Synth. Commun. 27 (1997) 1259–1266.
1
7
9% yield: IR (KBr) m 1742; H-NMR (CDCl ) d 1.34 (t, 6H,
3
3
JHP = 7 Hz, CH ), 3.71 (s, 3H, OCH ), 4.15–4.25 (m, 4H,
3
3
31
CH ), 6.90 (br s, 1H, NH); P-NMR (CDCl ) d 4.0 (s, 1P);
2
3
13
3
C-NMR (CDCl ) d 15.9 (d, J = 7 Hz, CH ), 52.8 (s,
3
2
CP
3
2
OCH ), 63.7 (d, J = 5 Hz, CH ), 154.6 (d, J = 5 Hz,
3
CP
2
CP
[5] (a) E.A. Swyryd, S.S. Seaver, G.R. Stark, J. Biol. Chem. 246 (1974)
6945–6950; (b) R.K. Johnson, T. Inouye, A. Goldin, G.R. Stark,
Cancer Res 36 (1976) 2720–2725.
COO).
[
6] (a) T.D. Kempe, E.A. Swyryd, M. Bruits, G.R. Stark, Cell 9 (1976)
41–550; (b) H.N. Jayaram, D.A. Cooney, D.T. Vistica, S. Kariya,
5
.1.6.3. Methoxycarbonylamidophosphonic acid disodium
5
salt 4b. Compound 4b was obtained according to the proce-
dure described for derivative 4a starting from bromotrimeth-
ylsilane (9.25 g, 60.4 mmol) and compound 10 (1.50 g,
R.K. Jonhson, Cancer Treat. Rep. 63 (8) (1979) 1291–1302; (c)
T.W. Kensler, G. Mutter, J.G. Hankerson, L.J. Reck, C. Harley,
N. Han, B. Ardalan, R.L. Cysyk, R.K. Johnson, H.N. Jayaram,
D.A. Cooney, Cancer Res 41 (1981) 894–904; (d) J.C. White,
L.H. Hines, Cancer Res 44 (1984) 507–513; (e) J.L. Grem, S.A. King,
P.J. O’Dwyer, B. Leyland-Jones, Cancer Res 48 (1988) 4441–4454;
(f) A. Sharma, N.L. Straubinger, R.M. Straubinger, Pharm. Res. 10
7.1 mmol). Reaction mixture was stirred for 56 h at room
temperature. Work-up was carried out a solution 1 M MeONa
in MeOH (60.6 ml). The crude compound was washed suc-
cessively with CH Cl , AcOEt and Et O, and filtered. Then
trituration in a mixture MeOH/Me CO afforded the expected
compound 4b as a white solid in 50% yield: (KBr) m 1618,
(
1993) 1434–1441.
7] (a) G.M. Wahl, R.A. Padgett, G.R. Stark, J. Biol. Chem. 254 (1979)
679–8689; (b) E. Giulotto, I. Saito, G. Stark, EMBO J. 5 (1986)
2115–2121.
2
2
2
[
2
8
1
31
1
638; H-NMR (D O) d 4.98 (s, 3H, OCH ); P-NMR
[8] (a) M.F. Roberts, S.J. Opella, M.H. Schaffer, H.M. Philipps,
G.R. Stark, J. Mol. Biol. 251 (19) (1976) 5976–5985; (b) J.J. Good-
son, C.J. Wharton, R.J. Wrigglesworth, J. Chem. Soc. Perkin Trans. 1
2
3
1
3
(D O) d 9.3 (s, 1P); C-NMR (D O) d 51.8 (s, OCH ), 184.0
2
2
3
(s, COO).
(
1980) 2721–2727; (c) P. Kafarski, B. Lejczak, P. Mastalerz, D. Dus,
C. Radzikowski, J. Med. Chem. 28 (1985) 1555–1558; (d) G.K. Far-
rington, A. Kumar, F.C. Wedler, J. Med. Chem. 28 (1985) 1668–1673;
5
5
.2. Enzymology
(
5
e) S.D. Lindell, R.M. Turner, Tetrahedron Lett. 31 (1990) 5381–
384; (f) N. Laing, W.W.C. Chan, D.W. Hutchison, B. Oberg, Bio-
.2.1. ATCase preparation and assay
Purified ATCase was prepared and assayed as previously
chemistry 260 (2) (1990) 206–208; (g) M. Ben-Bari, G. Dewynter,
C. Aymard, T. Jei, J.L. Montero, Phosphorus Sulfur Silicon 105
(1995) 129–144; (h) P. Coutrot, P. Oliger, C. Grison, S. Joliez, New J.
Chem. 23 (1999) 981–987; (i) P. Oliger, M. Hebrant, C. Grison,
P. Coutrot, C. Tondre, Langmuir 17 (2001) 6426–6432; (j) P. Oliger,
M. Schmutz, M. Hebrant, C. Grison, P. Coutrot, C. Tondre, Langmuir
17 (2001) 3893–3897.
described [21] and the isolated catalytic subunits were pre-
pared by the method of Gerhart and Holoubek [22]. The
enzymatic assay was performed at 37 °C, in the presence of
5
2
0 mM Tris–HCl pH = 8, 5 mM carbamylphosphate and
0 mM aspartate, or otherwise indicated.
[
9] (a) C. Purcarea, G. Hervé, M. Ladjimi, R. Cunin, J. Bact. 179 (1997)
4
143–4157; (b) M. Nagy, M. Le Gouar, S. Potier, J.L. Souciet,
G. Hervé, J. Biol. Chem. 264 (1989) 8366–8374; (c) J.P. Simmer,
R.E. Kelly, J.L. Scully, D.R. Grayson, D.R. Rinker, A.G. Bergh,
D.E. Evans, Proc. Natl. Acad. Sci. USA 86 (1989) 4382–4389; (d)
L. Jin, B. Stec, W.N. Lipscomb, E.R. Kantrowitz, Proteins 37 (1999)
References
[
1] (a) T.W. Kensler, D.A. Cooney, Adv. Pharmacol. Chemother 18
1981) 273–352; (b) G. Hervé, in: G. Hervé (Ed.), Allosteric
7
29–742; (e) J.A. Endrizzi, P.T. Beerninck, T. Alber,
(
H.K. Schachman, Proc. Natl. Acad. Sci. USA 97 (2000) 5077–5082;
Enzymes, C.R.C., Boca Raton, 1989, pp. 61–79; (c) M. Le Maire,
R. Chabaud, G. Hervé, Un modèle d’étude: l’aspartate transcarbamy-
lase, Collection de Biochimie, Masson, 1990; (d) W.N. Lipscomb,
Adv. Enzymol 68 (1994) 67–152.
(
(
f) S. Van Boxstael, R. Cunin, S. Khan, D. Maes, J. Mol. Biol. 326
2003) 203–216.
[
[
[
[
10] P. Coutrot, C. Grison, C. Charbonnier-Gérardin, Tetrahedron 48
1992) 9841–9868.
(
[2] (a) S. Madani, J. Baillon, J. Fries, O. Belhadj, A. Bettaieb, M. Ben
Hamida, G. Hervé, Eur. J. Cancer Clin. Oncol. 23 (1987) 1485–1490;
11] G.M. Blackburn, D. Brown, S.J. Martin, M.J. Paratt, J. Chem. Soc.
Perkin Trans. 1 (1987) 181–186.
12] P. Coutrot, C. Grison, R. Sauvêtre, J. Organomet. Chem. 332 (1987)
(
b) G. Hervé, X.G. Xi, in: P. Sebastien (Ed.), “Mécanismes des
pneumopathies professionnelles”, vol. 203, Editions de l’INSERM,
991, pp. 195–201.
3] (a) N. Laing, W.W. Chan, D.W. Hutchinson, B. Oberg, FEBS Lett 260
1990) 206–208; (b) P.L. Dutta, W.O. Foye, J. Pharm. Sci. 79 (1990)
47–452.
4] (a) K.D. Collins, G.R. Stark, J. Biol. Chem. 246 (1971) 6599–6605;
b) E.A. Swyryd, S.S. Seaver, G.R. Stark, J. Biol. Chem. 246 (1974)
945–6950; (c) R.K. Johnson, T. Inouye, A. Goldin, G.R. Stark,
1
–8.
1
13] G.M. Blackburn, D. Brown, S.J. Martin, J. Chem. Res. (S) (1985)
92–93.
[14] (a) C.E. McKenna, M.T. Higa, N.H. Cheung, M.C. McKenna, Tetra-
hedron Lett 2 (1977) 155–158; (b) C.E. McKenna, J. Schmidhauser, J.
Chem. Soc. Chem. Commun. (1979) 739.
[
[
(
4
(
6
[15] C. Grison, F. Charbonnier, P. Coutrot, Tetrahedron Lett 35 (1994)
5425–5428.
Cancer Res 36 (1976) 2720–2725; (d) K.K. Tsuboi, H.N. Edmunds,
L.K. Kwong, Cancer Res 37 (1977) 3080–3087; (e) R.K. Johnson,
E.A. Swyryd, G.R. Stark, Cancer Res 38 (1978) 371–378; (f)
H.N. Jayaram, D.A. Cooney, Cancer Treat. Rep. 63 (1979) 1095–
[16] J.I.G. Cadogan, Organophosphorus Reagents in Organic Synthesis,
Academic Press Inc., London, 1979 (and references cited therein).
[17] L.A.Antokhina, Z.Y. Latypov, K.V. Nirondrov, Zh. Obshch. Khim. 53
(1983) 1258–1260.
1
(
108; (g) D. Magnuson, J. Kuhn, Drugs Future 6 (1981) 1259–1266;
h) B. Ardalan, R.I. Glazer, T.W. Kensler, H.N. Jayaram, T.V. Pham,
J.S. MacDonald, D.A. Cooney, Biochem. Pharm. 30 (1981) 2045–
049; (i) B.B. Natale, A. Yagoda, D.P. Kelsen, R.J. Gralla, R.C. Wat-
[18] A.V. Kirsanov, M.S. Marenets, Zh. Obshch. Khim. 29 (1959) 2256.
[19] W.L.F. Armagero, D.D. Perrin, In Purification of Laboratory Chemi-
cals, fourth ed, Butterworth-Heinemann, 1998.
[20] S.C. Watson, J.F. Eastham, J. Organomet. Chem. 9 (1967) 165–168.
[21] B. Perbal, G. Hervé, J. Mol. Biol. 70 (1972) 511–529.
[22] J.C. Gerhart, H. Holoubek, J. Biol. Chem. 242 (1967) 2886–2892.
2
son, Cancer Treat. Rep. 66 (1982) 2091; (j) J.L. Montero, J.L. Imbach,
Eur. J. Med. Chem. Chim. Ther. 17 (1982) 97–2099; (k) P. Kafarski,