Synthesis and evaluation of new ω-borono-α-amino acids as rat liver arginase inhibitors

Article abstract of DOI:10.1016/j.bmc.2005.01.053

Recent studies have demonstrated that arginase plays important roles in pathologies such as asthma or erectile dysfunctions. We have synthesized new ω-borono-α-amino acids that are analogues of the previously known arginase inhibitors S-(2-boronoethyl)-l-cysteine (BEC) and 2-amino-6- boronohexanoic acid (ABH) and evaluated them as inhibitors of purified rat liver arginase (RLA). In addition to the distance between the B(OH)₂ and the α-amino acid functions, the position of the sulfur atom in the side chain also appears as a key determinant for the interaction with the active site of RLA. Furthermore, substitution of the alkyl side chain of BEC by methyl groups and conformational restriction of ABH by incorporation of its side chain in a phenyl ring led to inactive compounds. These results suggest that subtle interactions govern the affinity of inhibitors for the active site of RLA.

Full text of DOI:10.1016/j.bmc.2005.01.053

Bioorganic & Medicinal Chemistry 13 (2005) 2373–2379
Synthesis and evaluation of new x-borono-a-amino acids
as rat liver arginase inhibitors
Olivier Busnel,a Franc¸ois Carreaux,a Bertrand Carboni,a Stephanie Pethe,b
Sandrine Vadon-Le Goff,b Daniel Mansuyb and Jean-Luc Boucherb,*
a
` ` ´
Laboratoire de Synthese et Electrosynthese Organiques, Institut de Chimie, Universite de Rennes 1,
Campus de Beaulieu, 35042 Rennes Cedex, France
Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, UMR 8601 CNRS, Universite Rene Descartes,
`
45 Rue des Saints Peres, 75270 Paris Cedex 06, France
b
´
´
Received 19 October 2004; revised 20 January 2005; accepted 26 January 2005
Abstract—Recent studies have demonstrated that arginase plays important roles in pathologies such as asthma or erectile dysfunc-
tions. We have synthesized new x-borono-a-amino acids that are analogues of the previously known arginase inhibitors S-(2-boro-
noethyl)-L-cysteine (BEC) and 2-amino-6-boronohexanoic acid (ABH) and evaluated them as inhibitors of purified rat liver
arginase (RLA). In addition to the distance between the B(OH)2 and the a-amino acid functions, the position of the sulfur atom
in the side chain also appears as a key determinant for the interaction with the active site of RLA. Furthermore, substitution of
the alkyl side chain of BEC by methyl groups and conformational restriction of ABH by incorporation of its side chain in a phenyl
ring led to inactive compounds. These results suggest that subtle interactions govern the affinity of inhibitors for the active site of
RLA.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
NOSs, is a potent competitive inhibitor of arginases with
Ki value of 10–40 lM.7,8 This resulted in the hypothesis
that under various physiological situations, arginases
may be essential in the regulation of NOSs activities by
modulating local L-arginine concentration. Recent stud-
ies support this hypothesis and in cells producing NO,
localized NOHA and L-arginine concentrations were
found to modulate arginase activity and inhibition of
arginases enhanced the formation of NO.9–11 The inhibi-
tion of arginases by selective and potent inhibitors thus
became the focus of potential therapies for treating sev-
eral NO-dependent smooth muscle disorders, including
asthma and erectile dysfunctions.12–15
In mammalian cells, L-arginine is metabolized by two
major pathways. Arginases catalyze its hydrolysis to L-
ornithine and urea, whereas NO synthases (NOSs) cata-
lyze its oxidation to L-citrulline and nitric oxide, NO.1
The general importance of arginases lies in their roles
in controlling nitrogen excretion and cellular levels of
L-arginine and L-ornithine involved in protein synthesis,
as well as production of creatine, proline, and polyam-
ines.2 NO is an important biological molecule involved
in vasodilation, neurotransmission, and cytotoxicity.3–5
Since NOSs and arginases can be found in similar tissues
and cells, and because their expression may be regulated
in response to the same stimuli (cytokines, endotoxines),
both enzymes are believed to participate in the regulation
of NO biosynthesis by competing for the common sub-
strate, L-arginine.6 Conversely, Nx-hydroxy-L-arginine
(NOHA), an intermediate in the reaction catalyzed by
Arginine hydrolysis by arginases is achieved by a
metal-activated water molecule that bridges a (MnII)2
cluster at their active site.16 The hydrolysis is postulated
to proceed through a tetrahedral intermediate resulting
from the nucleophilic attack of metal-bridging hydr-
oxide ion at the guanidinium carbon of L-arginine.17
This proposal resulted in the synthesis of the first boro-
nic analogues of L-arginine, 2-amino-6-boronohexanoic
acid (ABH),18, and S-(2-boronoethyl)-L-cysteine (BEC)
that act as high-affinity and slow-binding inhibitors of
arginase (Scheme 1).14,19
Keywords: Rat liver arginase; Inhibitors; a-Amino acids; Boronic
acids.
*
Corresponding author. Tel.: +33 1 42 86 21 91; fax: +33 1 42 86 83
87; e-mail: boucher@univ-paris5.fr
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.01.053

2374
O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
X
Y
HO2C
H2N
HO2C
H2N
X
B(OH)2
ABH
BEC
X = -(CH2)4-
X = -CH2-S-(CH2)2-
6: 4-BPA X = B(OH)2, Y = H
7: 3-BPA X = H, Y = B(OH)2
1: BMC
2: BMHC X = -(CH2)2-S-CH2-
X = -CH2-S-CH2-
3: BEHC X = -(CH2)2-S-(CH2)2-
4: 2MBEC X = -CH2-S-CH2-CH(CH3)-
5: BEPA X = -C(CH3)2-S-CH2-CH2-
Scheme 1. Structure of ABH, BEC, and x-borono-a-amino acids 1–7.
In order to define better relationships between the
chemical structure of x-borono-a-amino acids and their
recognition by RLA, we have synthesized five new a-
amino acids, compounds 1–5, that are BEC analogues.
The distance between the two functions has been modi-
fied by introducing the boronic acid at the end of an al-
kyl side chain containing 2, 3, or 4 C-atoms. The
distance between the B- and S-atoms of BEC has been
reduced, and one or two methyl groups have been intro-
duced in the alkyl side chain of BEC. We also synthe-
sized two conformationally restricted ABH analogues,
racemic 4- and 3-boronophenylalanine, 6 and 7, that
contain a phenyl ring in the alkyl side chain (Scheme
1). These BEC and ABH analogues have been evaluated
as inhibitors of purified rat liver arginase (RLA) and we
used the recent crystallographic data to discuss our
results.
nate with 2-(bromomethyl)-4,4,5,5-tetramethyl-1,3,2-
dioxaborolane21 in THF led to the expected product in
a 87% yield. In the case of L-homocysteine derivative,
the thiolate anion was generated in the presence of the
electrophile to avoid the formation of tert-butyl(2-oxo-
tetrahydro-3-thienyl)carbamate resulting from an intra-
molecular cyclization. Under those conditions, the
desired S-alkylated adduct was obtained in 90% yield.
Removal of the protective groups of the a-amino acid
function was carried out using treatments with aqueous
solutions of NaOH and HCl. Boronic acids of S-(boro-
nomethyl)-L-cysteine, BMC 1 and S-(boronomethyl)-L-
homocysteine, BMHC 2, were obtained from boronic
esters by reacting phenylboronic acid in a two-phase
ether/water system (Scheme 2).22
2.2. Synthesis of S-(boronoethyl)-D/L-homocysteine,
BEHC 3, S-(2-methyl-2-boronoethyl)-L-cysteine,
2MBEC 4, and S-(boronoethyl)-D-penicillamine, BEPA 5
2. Results
The strategy previously described to prepare BEC al-
lowed us to get a simple access to the higher homologue
BEHC 3, and to compound 5, BEPA, that contains a
gem-dimethyl group adjacent to the a-CH of L-cyst-
eine.14,23 Addition of D/L-homocysteine and D-penicill-
amine to di(n-butyl)-vinylboronate24 in the presence of
AIBN led to the formation of BEHC and BEPA in
55% and 35% yields, respectively (Scheme 3). Com-
pound 4 that contains a methyl group adjacent to the
B-atom was similarly obtained in 42% yield by reacting
L-cysteine and di(n-butyl)-2-propeneboronate.24
2.1. Synthesis of S-(boronomethyl)-L-cysteine, BMC 1
and S-(boronomethyl)-L-homocysteine, BMHC 2
Our strategy to prepare BEC analogues 1 and 2 is based
on the nucleophilic attack of orthogonally protected L-
cysteine and L-homocysteine on bromomethylboronic
ester (Scheme 2). Methyl esters of Boc-L-cysteine and
Boc-L-homocysteine were synthesized from L-cystine
and L-homocystine, respectively, following a described
procedure.20 The S-alkylation of methyl Boc-L-cystei-
CO2H
CO2CH3
S
NH2
CO2H
S
NHBoc
a
b
S
n
S
n
H2N
n
BocHN
n
CO2CH3
1a: n = 1
2a: n = 2
O
CO2H
CO2CH3
SH
CO2CH3
d
c
B(OH)2
HCl
S
n
S
n
B
H2N
O
BocHN
BocHN
n
1b: n = 1
2b: n = 2
1c: n = 1
2c: n = 2
1: n = 1
2: n = 2
Scheme 2. Synthesis of BMC, 1 (n = 1) and BMHC, 2 (n = 2). Reagents and conditions: (a) (i) SOCI2, MeOH, reflux, 12 h; (ii) NaOH, dioxane/H2O
(2:1) (Boc)2O, rt, 12 h; (b) Ph3P, NaOAc, MeOH/H2O/AcOH, reflux; (c) 2-(bromomethyl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane, NaH, THF, rt,
2 h; (d) (i) 1 Maq NaOH, 50 ꢁC, 2 h, (ii) 2 Maq HCI, Et O, PhB(OH)2 rt, 9 h.
3

O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
2375
R
HO2C
H2N
HO2C
H2N
MeOH-H2O (1:1)
80˚C, 6-8h.
O-C4H9
O-C4H9
+
B(OH)2
B
X
SH
X
S
AIBN
R
3: BEHC X = -CH2 CH2
R = H
R = CH3
R = H
-
-
4: 2MBEC X = -CH2
-
5: BEPA
X = -C(CH3)2
-
Scheme 3. Synthesis of BEHC, 3, 2MBEC, 4, and BEPA, 5.
Table 1. Effects of BEC and x-borono-a-amino acids 1–7 on the
activity of purified RLA. Half-inhibitory concentrations (IC50) values
were measured at pH 7.4 and pH 9.0 following the hydrolysis of [14C]L-
arginine to [14C] urea as indicated in Experimental
2.3. Synthesis of 4-(dihydroxyboryl)-D/L-phenylalanine,
4-BPA 6 and 3-(dihydroxyboryl)-D/L-phenylalanine,
3-BPA 7
Compounds
IC50
The synthesis of 3-BPA, 7, was based on the Horner–
Emmons–Wittig reaction of carbonyl compounds with
N-acylphosphonoglycine trimethyl ester following the
methodology previously described for the synthesis of
4-BPA, 6 (Scheme 4).25 Condensation of protected 3-
formylphenylboronic acid with N-(benzyloxycarbonyl)-
phosphonoglycine trimethyl ester26 using DBU in
CH2Cl2 was successfully achieved at ambient tempera-
ture to provide the intermediate a-amino acrylate in a
55% yield and predominantly in the form of the Z-iso-
mer. Amino acrylate was hydrogenated using 10% Pd–
C to afford the protected boronophenylalanine with
concomitant debenzylation. Finally, successive treat-
ments with NaOH and HCl in the presence of phenyl-
boronic acid provided 3-BPA 7 in an overall yield of
35% from 3-formylphenylboronic acid.22
pH 7.4a
pH 9.0a
lM
7.5 1.0 mM
BEC
1 BMC
5 1 lM20
3.8 0.5 mM
4
b
2 BMHC
3 BEHC
4 2MBEC
5 BEPA
6 4-BPA
7 3-BPA
350 100(l4M0%>)1 mM
>10 mM(35%)
b
b
>10 mM(15%)
>10 mM (10%)b
9 1 mM
>3 mM(15%)
b
b
>3 mM(10%)
b
6.0 1.0 m(2M0%>)10 mM
b
>10 mM(30%)
b
>10 mM(10%)
a Mean values SD from three independent experiments.
b % Inhibition at the concentration indicated.
inactive compounds. Interestingly, BMHC 2, which dif-
fers from BEC by the position of the S-atom in the side
chain and the distance between the S-atom and the
boronic acid function, was 70-fold less potent than
BEC (IC50 = 350 50 lM, Table 1). Finally, the two
conformationally restricted ABH analogues, 6 and 7,
were almost inactive at both pH values (IC50 > 5 mM,
Table 1).
2.4. Effects of compounds 1–7 on rat liver arginase
activity
The ability for compounds 1–7 to inhibit [14C]urea form-
ation from [Guanido-14C]L-arginine in the presence
of purified RLA was investigated following usual proto-
cols27,28 and compared to the inhibitory effects of BEC.
Experiments were performed at physiological pH 7.4
and at pH 9.0 where RLA displays maximal activity.2
As previously described,14 BEC led to a potent concen-
tration-dependent inhibition of purified RLA activity
with half-maximal inhibition (IC50) obtained in the pres-
ence of 5 1 lMBEC ( Table 1). The shorter and the
longer analogues BMC 1 and BEHC 3, were almost
inactive at both pH (IC50 > 3 mM). Introduction of a
methyl group in the a-position from the B-atom, in com-
pound 4, or of a gem-dimethyl group in close proximity
to the a-amino acid function, in compound 5, led to
2.5. Discussion
New amino acids 1–7 were easily obtained following
previously described protocols and were tested as rat li-
ver arginase inhibitors. The comparison of the inhibi-
tory effects of BEC and its analogues 1 and 3 fully
identified the distance between the a-amino acid func-
tion and the terminal OH group of the side chain as a
key determinant in their binding at the active site of
RLA. In the case of ABH analogues, such a key
role for the distance between the two functions was
similarly observed and one CH2-group shorter or
longer analogues of ABH were much less active.29
CO2CH3
NH-Z
CO2H
NH2
CO2CH3
CHO
NH-Z
a
b
c
X
X
B(OH)2
X
O
O
X = B
7a
7b
7
Scheme 4. Synthesis of 3-BPA 7. Reagents and conditions: (a) N-(benzyloxycarbonyl)phosphonoglycine trimethyl ester, DBU/CH2CI2, rt, 3 h (E/Z =
30:70); (b) 10% Pd–CH2/MeOH–H2Cl2 (2:1), rt, 5 atm, 21 h; (c) (i) 1 Maq NaOH, 50 ꢁC, 2 h, (ii) 2 Maq HCI, Et O, PhB(OH)2, rt, 9 h.
2

2376
O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
Dehydro-ABH, which only differs from ABH by intro-
duction of a double-bond in the alkyl side chain, dis-
played lower inhibitory effects than ABH.29 This could
explain that introduction of the alkyl-chain of ABH in a
phenyl ring, in compounds 6 and 7, resulted in a complete
loss of interaction with RLA. The determinants in the
binding of these new x-borono-a-amino acids to RLA
thus seem very similar to those required for the binding
of another group of potent RLA inhibitors, Nx-hydroxy-
guanidino-a-amino acids, NOHA, and nor-NOHA.30,31
250 or 300 MHz Bruker spectrometers operating in the
Fourier transform mode. 13C NMR spectra were ob-
tained with broadband proton decoupling. Chemical
shifts are expressed as d values in parts per million rela-
tive to tetramethylsilane, except for 11B NMR where
chemical shifts are relative to BF3ÆOEt2. Coupling con-
stants (J) are expressed in hertz. Optical rotations were
measured using 10 cm path-length cell at 20 ꢁC, and
the concentration is expressed in grams per deciliter.
HRMS were obtained with a Varian MAT 311 mass
´
spectrometer at Centre Regional de Mesures Physiques
Very unexpected results resided in the poor inhibitory
effects of BMHC compared to those of BEC. BMHC
only differs from BEC by the position of its S-atom in
the alkyl side chain. Shortening the distance between
the S- and B-atoms by one CH2-group resulted in a
70-fold decrease of the inhibitory effects. Introduction
of one methyl group in the a-position to the B-atom,
in 2MBEC 4, or of a gem-dimethyl group, in BEPA 5,
completely abolished the inhibitory effects of BEC. Pos-
sible explanations could be found from the binding of
BEC as observed in the X-ray crystal structure of the
RLA–BEC complex.14 This structure reveals that the
S-atom of BEC is very close to His141 and His126 resi-
dues in the active site (pdb access 1HQ5).14 Shortening
the distance between the S-atom and the B(OH)2 func-
tion could introduce steric and/or electronic interactions
between the S-atom of BMHC and one (or both) of
these His-residues. Similarly, introduction of a methyl
group close to the B-atom could introduce steric con-
straints with one of these His-residues and reduce the
affinity. The a-amino acid function of BEC establishes
direct or water-mediated H-bonds with residues
Asn130, Ser137, Asn139, Asp181, Asp183, and Glu186
of the protein.14 Alteration of this network of H-bonds
and restriction of the possible conformations by intro-
duction of an hydrophobic bulky gem-dimethyl group
could explain the low affinity of BEPA. Although the
binding mode of BEC is close to that of ABH, subtle dif-
ferences appear between the two inhibitors and can ex-
plain their 20-fold difference in affinity. Our results
demonstrate for the first time that further modifications
of the structure of BEC, by introduction of a methyl
group close to the B-atom or by shortening the distance
between the B- and S-atoms, strongly compromise the
affinity for the active site of RLA. X-Ray studies are
underway to more precisely explain these subtle
differences.
de lꢀOuest. Microanalysis were performed at the Central
Laboratory for Analysis, CNRS (Solaize, France). 4-
Dihydroxyboryl-D/L-phenylalanine hydrochloride, 4-
BPA 6, was prepared as previously described.25
3.2. N,N0-Bis[(tert-butyloxy)carbonyl]-L-cystine dimethyl
ester 1a
Thionyl chloride (3.8 mL, 52 mmol) was added dropwise
at 0 ꢁC to 80 mL of methanol. L-Cystine (5 g,
20.8 mmol) was then added and the reaction mixture
was refluxed overnight. The volatile reagents and sol-
vent were evaporated in vacuo and a white solid
(6.9 g, 97%) was obtained. L-Cystine dimethyl ester
dihydrochloride was used directly in the next step with-
1
out further purification. H NMR (200 MHz, D2O) d
3.29–3.33 (m, 4H), 3.78 (s, 6H), 4.49 (m, 2H).
To a stirred solution of L-cystine dimethyl ester dihydro-
chloride (6.9 g, 20.2 mmol) in 1 MNaOH (40.4 mL) and
dioxane/water (2:1, 60 mL), di-tert-butyl pyrocarbonate
(10.1 g, 46.3 mmol) was added in one portion at 0 ꢁC.
The ice was removed, and the solution was stirred for
24 h at room temperature. After concentration of the
reaction mixture, ethyl acetate (60 mL) was added.
The pH of the mixture was adjusted to 3.0 with aqueous
2 MKHSO 4. The aqueous layer was extracted with
ethyl acetate (2 · 30 mL) and the combined organic lay-
ers were washed with brine (30 mL) and water (30 mL),
and dried over MgSO4. Evaporation of ethyl acetate in
vacuo and recrystallization from ethyl acetate/heptane
gave N,N0-bis[(tert-butyloxy)carbonyl]-L-cystine di-
methyl ester 1a (7.3 g, 77%). Mp 96–98 ꢁC; lit.20 mp
20
96–97 ꢁC; ½aꢀ +73.5 (c 1.08, CHCl3); 1H NM R
589
(200 MHz, CDCl3) d 1.49 (s, 18H), 3.20 (d, J = 5.3 Hz,
4H), 3.81 (s, 6H), 4.64 (m, 2H), 5.40 (br d, 2H). Anal.
Calcd for C18H32N2O8S2: C, 46.14; H, 6.88; N, 5.98.
Found: C, 46.27; H, 6.93; N, 5.95.
3. Experimental
3.3. N-[(tert-Butyloxy)carbonyl]-L-cysteine methyl ester
1b
3.1. General procedures
N,N0-Bis[(tert-butyloxy)carbonyl]-L-cystine dimethyl es-
ter 1a (2.5 g, 5.34 mmol), triphenylphosphine (1.48 g,
5.49 mmol), and sodium acetate (0.17 g, 2.11 mmol)
were suspended in a mixture of methanol (20 mL), water
(10 mL), and glacial acetic acid (0.17 mL) and heated
under reflux for 24 h. The mixture was diluted with
260 mL CH2Cl2, washed with 100 mL of water, 50 mL
of saturated sodium chloride solution, and dried over
MgSO4. The solvent was removed in vacuo and the resi-
due purified by column chromatography on silica gel
The reagents were purchased from Acros and Aldrich
and were used without further purification unless other-
wise specified. Tetrahydrofuran was distilled from deep
blue solutions of sodium/benzophenone ketyl prior to
use. Methylene chloride was shaken with concentrated
H2SO4, dried over K2CO3, and distilled. Merck silica
gel was used for column chromatography. All melting
points were determined on a Kofler apparatus and are
1
uncorrected. H NMR spectra were recorded on 200,

O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
2377
(ethyl acetate/heptane 2:8) to give 1b (2.3 g, 91%) as
cystine (1.0 g, 3.7 mmol) was added and the reaction
mixture was refluxed overnight. After evaporation of
the reaction mixture in vacuo, a white solid (1.3 g,
95%) was obtained. L-Homocystine dimethyl ester dihy-
drochloride was used directly in the next step without
20
589
1
a colorless oil. ½aꢀ +97.7 (c 2.12, CHCl3); H NM R
(200 MHz, CDCl3) d 1.50 (s, 9H), 2.90 (dd, J = 4.3,
8.9 Hz, 2H), 3.80 (s, 3H), 4.65 (m, 1H), 5.46 (br s,
1H); 13C NMR (50 MHz, CDCl3) d 27.7, 28.7, 53.1,
55.2, 80.1, 155.5, 171.2. Anal. Calcd for C9H17NO4S:
C, 45.94; H, 7.28; N, 5.95. Found: C, 46.10; H, 7.30;
N, 6.02.
1
further purification. H NMR (200 MHz, D2O) d 2.30
(m, 2H), 2.79 (t, J = 6.9 Hz, 2H), 3.76 (s, 3H), 4.22 (t,
J = 6.5 Hz, 1H). L-Homocystine dimethyl ester dihydro-
chloride (1.3 g, 3.52 mmol) was dissolved in dioxane/
water (2:1, 12 mL). Sodium carbonate (0.8 g,
7.54 mmol) in water (4 mL) was added to the stirred
solution and the mixture was cooled to 0–5 ꢁC. Di-
tert-butyl pyrocarbonate (1.9 g, 8.7 mmol) was added
in one portion and the mixture was stirred at room tem-
perature for 18 h. The mixture was concentrated to
about 3 mL in vacuo. The precipitate was filtered off
and resuspended in hot ethanol (12 mL). Insoluble par-
ticles were filtered off and the filtrate was kept at 0–5 ꢁC
for 18 h. The resulting colorless crystals were collected
3.4. N-[(tert-Butyloxy)carbonyl)]-S-[(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)methyl]-L-cysteine methyl ester
1c
To a suspension of NaH (0.15 g, 3.8 mmol) in THF
(10 mL) at 0 ꢁC, N-[(tert-butyloxy)carbonyl]-L-cysteine
methyl ester 1b (0.8 g, 3.40 mmol) in THF (5 mL) was
added dropwise under N2. The mixture was warmed to
room temperature, stirred 30 min and 2-(bromo-
methyl)-4,4,5,5-tetramethyl-1,3,2-dioxaborolane21 (0.75 g,
3.4 mmol) in THF (5 mL) was added dropwise. After
stirring for 12 h at room temperature, the mixture was
quenched by the addition of H2O (20 mL), THF was
evaporated and the residue extracted with ethyl acetate
(3 · 20 mL). The organic phase was washed with brine
(20 mL), dried over MgSO4, and evaporated. Flash col-
umn chromatography on silica gel (ethyl acetate/hep-
by filtration and dried in vacuo (1.43 g, 82%). Mp
20
589
100–102 ꢁC; ½aꢀ +26.7 (c 0.52, CHCl3); 1H NM R
(200 MHz, CDCl3) d 1.48 (s, 18H), 1.88–2.15 (m, 4H),
2.75 (t, J = 7.6 Hz, 4H), 3.80 (s, 6H), 4.43 (m, 2H),
5.15 (br s, 2H); 13C NMR (50 MHz, CDCl3) d 28.7,
33.0, 34.9, 52.8, 52.9, 80.1, 156.5, 173.0. Anal. Calcd
for C20H36N2O8S2: C, 48.37; H, 7.31; N, 5.64. Found:
C, 48.52; H, 7.35; N, 5.74.
tane 2:8) gave pure 1c (1.1 g, 87%) as a colorless oil.
20
589
½aꢀ +14.7 (c 1.15, CHCl3); 1H NMR (200 MHz,
CDCl3) d 1.31 (s, 12H), 1.48 (s, 9H), 2.03 (m, 2H),
3.02 (m, 2H), 3.80 (s, 3H), 4.56 (m, 1H), 5.51 (br s,
1H); 13C NMR (50 MHz, CDCl3) d 23.8, 27.3,
35.7, 51.6, 51.9, 79.0, 81.1, 157.3, 170.8. Anal. Calcd
for C16H30BNO6S: C, 51.21; H, 8.06; N, 3.73. Found:
C, 51.25; H, 8.09; N, 3.74.
3.7. N-[(tert-Butyloxy)carbonyl]-L-homocysteine methyl
ester 2b
N,N0-bis[(tert-butyloxy)carbonyl]-L-homocystine dimethyl
ester 2a (1.26 g, 2.54 mmol), triphenylphosphine (0.7 g,
2.67 mmol), and sodium acetate (82 mg, 1 mmol) were
suspended in a mixture of 9.8 mL of methanol, 4.7 mL
of water, and 81 lL of glacial acetic acid and heated
under reflux for 30 min. The mixture was diluted with
120 mL CH2Cl2, washed with 50 mL of water, 25 mL
of brine, and dried over MgSO4. The solvent was
removed in vacuo yielding 2.0 g of a clear oil, which
was purified by column chromatography on silica gel
3.5. S-[(Dihydroxyboryl)methyl]-L-cysteine 1
A mixture of 1c (0.7 g, 1.87 mmol) and 1 Maq NaOH
(16 mL) was stirred at 45–50 ꢁC for 2 h and the pH of
the mixture was adjusted to 1 with aqueous 3 MHCl.
After being stirred at ambient temperature for 12 h, phen-
ylboric acid (1.87 mmol) and ether (16 mL) were added.
After 3 h at room temperature, the ether layer was
removed and replaced by fresh ether (16 mL). This
treatment was repeated two-fold, and the aqueous solu-
tion was concentrated and dried under high vacuum.
After silica gel chromatography (CHCl3/MeOH/NH3/
(ethyl acetate/heptane 3:7) to give 2b (1.2 g, 95%) as a
20
D
colorless oil. ½aꢀ +19.4 (c 1.5, CHCl3); 1H NM R
(200 MHz, CDCl3) d 1.65 (s, 9H), 2.00 (m, 1H), 2.20
(m, 1H), 2.78 (t, J = 7.9 Hz, 2H), 3.80 (s, 3H), 4.67 (m,
1H), 5.40 (br s, 1H); 13C NMR (50 MHz, CDCl3) d
21.0, 28.6, 37.5, 52.6, 52.8, 80.4, 155.8, 173.2. Anal.
Calcd for C10H19NO4S: C, 48.17; H, 7.68; N, 5.62.
Found: C, 48.30; H, 7.75; N, 5.71.
H2O 6:6:0.5:1.5) compound 1 was obtained (181 mg,
20
D
49%) as an amorphous solid. ½aꢀ ꢁ7.2 (c 0.8,
20
365
1
H2O); ½aꢀ ꢁ13.7 (c 0.8, H2O); H NMR (200 MHz,
D2O + DCl) d 1.94 (d, J = 14.1 Hz, 1H), 2.02 (d,
J = 14.1 Hz, 1H), 2.97 (dd, J = 14.6, 7.6 Hz, 1H), 3.10
(dd, J = 14.6, 4.5 Hz, 1H), 4.22 (dd, J = 7.6, 4.5 Hz,
1H); 13C NMR (50 MHz, D2O) d 39.9, 51.9, 170.7; 11B
NMR (96 MHz, D2O) d 30.1. HRMS: m/z: calcd for
C18H21N3O8SB with matrix 3-nitrobenzyl alcohol:
450.1142. Found 450.1144 [M+H]+.
3.8. N-[(tert-Butyloxy)carbonyl)]-S-[(4,4,5,5-tetramethyl-
1,3,2-dioxaborolan-2-yl)methyl]-L-homocysteine methyl
ester 2c
To a suspension of NaH (96 mg, 2.40 mmol) in THF
(4 mL), 2-(bromomethyl)-4,4,5,5-tetramethyl-1,3,2-dioxa-
borolane (0.48 g, 2.16 mmol) in THF (3.5 mL) was
added under N2. N-[(tert-butyloxy)carbonyl]-L-homo-
cysteine methyl ester 2b (0.54 g, 2.16 mmol) in THF
(4 mL) was added dropwise at 0 ꢁC. The mixture was
warmed to room temperature, stirred for 2 h, and
quenched by the addition of H2O (15 mL). THF was
evaporated and the residue extracted with ethyl acetate
3.6. N,N0-Bis[(tert-butyloxy)carbonyl]-L-homocystine
dimethyl ester 2a
Thionyl chloride (0.7 mL, 9.31 mmol) was added drop-
wise with stirring to 15 mL of cold methanol. L-Homo-

2378
O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
(3 · 15 mL). The organic phase was washed with brine
(15 mL), dried, and evaporated. Flash column chroma-
tography on silica gel (ethyl acetate/heptane 2:8) gave
3.07–3.23 (m, 2H), 3.98 (m, 1H); 13C NMR (50 MHz,
D2O + DCl) d 14.6, 20.6 (br s), 31.5, 35.6, 52.8, 170.8;
11B NMR (96 MHz, D2O) d 31.3; HRMS: m/z: calcd
for C20H25O8N3SB with matrix 3-nitrobenzyl alcohol:
478.1343. Found: 478.1462 [M+H+].
20
pure 2c (0.76 g, 90%) as a colorless oil. ½aꢀ +7.6 (c
D
7.2, CHCl3); 1H NMR (200 MHz, CDCl3) d 1.29 (s,
12H), 1.45 (s, 9H), 1.97 (s, 2H), 2.15 (m, 2H), 2.60 (t,
J = 7.6 Hz, 2H), 3.75 (s, 3H), 4.35 (m, 1H), 5.20 (br s,
1H); 13C NMR (50 MHz, CDCl3) d 23.6, 27.3, 29.0,
31.0, 51.4, 51.9, 81.9, 83.0, 154.3, 171.9; 11B NM R
(96 MHz, D2O) d 31.8. Anal. Calcd for C17H32BNO6S:
C, 52.45; H, 8.28; N, 3.60. Found: C, 52.48; H, 8.29;
N, 3.63.
3.12. S-[2-(Dihydroxyboryl)ethyl]-D-penicillamine 5
Reaction of di-(n-butyl)-vinylboronate (0.37 g, 2.0 mmol)
and D-penicillamine (0.340 g, 2.1 mmol) following the
protocol described for compound 3 provided S-[2-
(dihydroxyboryl)ethyl]-penicillamine 5 as an hygro-
1
scopic white solid (165 mg, 35%). H NMR (300 MHz,
3.9. S-[(Dihydroxyboryl)methyl]-L-homocysteine 2
D2O + DCl) d 0.95 (t, J = 7.8 Hz, 2H), 1.25 (s, 3H),
1.44 (s, 3H), 3.62 (m, 2 H), 3.98 (s, 1H); 13C NM R
(50 MHz, D2O + DCl) d 23.2, 23.9, 26.8, 46.7, 53.6,
169.5; 11B NMR (96 MHz, D2O) d 31.4; HRMS: m/z:
calcd for C21H27O8N3SB with matrix 3-nitrobenzyl alco-
hol: 492.1612. Found 492.1610 [M+H+].
A mixture of 2c (0.60 g, 1.56 mmol) and 1 Maq NaOH
(15 mL) was stirred at 45–50 ꢁC for 2 h and the pH of
the mixture was adjusted to 1 with aqueous 3 MHCl.
After being stirred at ambient temperature for 12 h, phen-
ylboric acid (1.56 mmol) and ether (15 mL) were added.
After 3 h at room temperature, the ether layer was
removed and replaced by fresh ether (15 mL). This
treatment was repeated two-fold and the aqueous phase
was concentrated and dried under high vacuum. After
silica gel chromatography (CHCl3/MeOH/NH3/H2O
3.13. Methyl (E/Z)-2-(benzyloxycarbonylamino)-3-[3-
(5,5-dimethyl-1,3,2-dioxaboran-2-yl)phenyl] acrylate 7a
To a stirred solution of N-(benzyloxycarbonyl)phos-
phonoglycine ester (1.38 g, 4.16 mmol) in anhydrous
CH2Cl2 (5 mL) was added DBU (0.57 mL, 3.78 mmol).
After 10 min at room temperature, a solution of 2-(3-
6:6:0.5:1.5), compound 2 was obtained (100 mg, 30%)
20
as an hygroscopic colorless solid. ½aꢀ ꢁ11.7 (c 0.48,
D
H2O); 1H NMR (200 MHz, D2O + DCl) d 1.85 (s,
2H), 2.02 (m, 2H), 2.51 (t, J = 7.6 Hz, 2H), 3.72 (dd,
J = 7.1, 5.5 Hz, 1H); 13C NMR (50 MHz, D2O) d 29.1,
54.2, 54.6, 175.2; 11B NMR (96 MHz, D2O) d 29.5;
HRMS: m/z: calcd for C19H23N3O8SB with matrix 3-
nitrobenzyl alcohol: 464.1299. Found 464.1295 [M+H]+.
formylphenyl)-5,5-dimethyl-1,3,2-dioxaborane
(0.8 g,
3.65 mmol) in anhydrous CH2Cl2 (6 mL) was added
dropwise to the mixture and stirring was continued
for 3 h. Then, the solution was diluted with CH2Cl2
(75 mL), washed with 0.5 Maq H 2SO4 (20 mL) and
then with H2O (4 · 40 mL), dried (MgSO4) and concen-
trated in vacuo. The crude product was purified by sil-
ica gel chromatography (ethyl acetate/heptane 3:7) to
give a E/Z mixture of 7a (0.97 g, 55%, E/Z = 30/70) as
3.10. S-[2-(Dihydroxyboryl)ethyl]-D/L-homocysteine 3
1
Di-(n-butyl)-vinylboronate24 (0.37 g, 2.0 mmol) and D/
L-homocysteine (0.345 g, 2.0 mmol) were mixed in
MeOH (5 mL) and water (3 mL) under an inert atmo-
sphere. AIBN (40 mg) was added and the mixture was
heated at 80 ꢁC for 1 h. Twenty milligram of AIBN were
further added and the mixture was heated for 6 h at
80 ꢁC. The solvents were evaporated and the ammonium
salt of (dihydroxyboryl)ethyl-D/L-homocysteine 3 was
obtained after silica gel chromatography (CHCl3/
MeOH/NH3/H2O 1:4:0.5:1) as an hygroscopic colorless
a colorless solid. H NMR (200 MHz, CDCl3) d 1.06
(s, 6H), 3.65 (s, 0.9H), 3.71 (s, 1.1H), 3.79 (s, 2.9H),
3.85 (s, 2.1H), 5.17 (s, 1.4H), 5.22 (s, 0.6H), 6.43 (br
s, 0.7H), 7.05 (br s, 0.3H), 7.33–8.00 (m, 10H); 13C
NMR (50 MHz, CDCl3) d 22.3, 22.4, 32.3, 53.0, 53.8,
67.8, 68.1, 72.7, 124.4, 128.3, 128.6, 128.8, 128.9,
129.0, 131.9, 132.7, 133.2, 135.4, 136.3, 136.4, 154.3,
154.5, 166.2, 166.8. Anal. Calcd for C23H26BNO6: C,
65.27; H, 6.19; N, 3.31. Found: C, 65.21; H, 6.37; N,
3.46.
1
solid (245 mg, 55%); H NMR (200 MHz, D2O + DCl)
d 1.12 (t, J = 7.9 Hz, 2H), 2.15 (m, 2H), 2.69 (m, 4H),
3.82 (t, J = 5.6 Hz, 1H); 13C NMR (50 MHz, D2O) d
15.2 (br s), 26.4, 26.5, 29.6, 51.9, 171.7; 11B NM R
3.14. Methyl 2-amino-3-[3-(5,5-dimethyl-1,3,2-dioxabo-
ran-2-yl)phenyl]propionate 7b
(96 MHz, D2O)
C20H25O8N3SB with matrix 3-nitrobenzyl alcohol:
d
30.2; HRMS: m/z: calcd for
To a stirred solution of (E/Z)-7a (0.67 g, 1.58 mmol) in
anhydrous MeOH/CH2Cl2 (2:1, 18 mL) was added
10% Pd–C (35 mg). After 21 h under 5 atm of hydrogen,
the mixture was passed through a silica plug and the fil-
trate was concentrated in vacuo to give 7b (0.45 g, 98%)
as a colorless oil. The product was used in the next step
without further purification. 1H NMR (200 MHz,
CDCl3) d 1.05 (s, 6H), 2.96 (dd, J = 7.8, 13.6 Hz, 1H),
3.20 (dd, J = 5.0, 13.6 Hz, 1H), 3.76 (s, 3H), 3.79 (s,
4H), 3.94 (m, 1H), 7.28–7.35 (m, 2H), 7.62–7.78 (m,
2H); 13C NMR (50 MHz, CDCl3) d 22.3, 32.3, 40.6,
52.6, 56.0, 72.7, 128.3, 132.1, 133.0, 135.1, 136.0,
174.6; 11B NMR (96 MHz, CDCl3) d 29.1.
478.1343. Found 478.1462 [M+H+].
3.11. S-[2-Methyl-2-(dihydroxyboryl)ethyl]-L-cysteine 4
Reaction of di-(n-butyl)-2-propeneboronate24 (0.435 g,
2.2 mmol) and L-cysteine (0.270 g, 2.2 mmol) in MeOH
(4 mL) and water (3 mL) following the protocol previ-
ously described for compound 3 yielded S-[2-methyl-2-
(dihydroxyboryl)ethyl]-L-cysteine 4 as a white powder
1
(210 mg, 42%). H NMR (250 MHz, D2O) d 1.10 (d,
J = 7.2 Hz, 3H), 1.36–1.48 (m, 1H), 2.66–2.77 (m, 2H),

O. Busnel et al. / Bioorg. Med. Chem. 13 (2005) 2373–2379
2379
3. Alderton, W. K.; Cooper, C. E.; Knowles, R. G. Biochem.
J. 2001, 357, 593.
3.15. 3-Dihydroxyboryl-D/L-phenylalanine 7
4. Pfeiffer, S.; Mayer, B.; Hemmens, B. Angew. Chem., Int.
Ed. 1999, 38, 1714.
5. Moncada, S.; Higgs, E. A. FASEB J. 1995, 9, 1319.
6. Boucher, J.-L.; Moali, C.; Tenu, J.-P. Cell Mol. Life Sci.
1999, 55, 1015.
7. Daghigh, F.; Fukuto, J. M.; Ash, D. E. Biochem. Biophys.
Res. Commun. 1994, 202, 174.
8. Boucher, J.-L.; Custot, J.; Vadon, S.; Delaforge, M.;
Lepoivre, M.; Tenu, J.-P.; Yapo, A.; Mansuy, D. Biochem.
Biophys. Res. Commun. 1994, 203, 1614.
9. Wei, L. H.; Jacobs, A. T.; Morris, S. M.; Ignarro, L. J.
Am. J. Physiol. 2000, 279, C248.
10. Hecker, M.; Nematollahi, H.; Hey, C.; Busse, R.; Racke,
K. FEBS Lett. 1995, 359, 251.
11. Buga, G. M.; Wei, L. H.; Bauer, P. M.; Fukuto, J. M.;
Ignarro, L. J. Am. J. Physiol. 1998, 275, R1256.
12. Zimmermann, N.; King, N. E.; Laporte, J.; Yang, M.;
Mishra, A.; Pope, S. M.; Muntel, E. E.; Witte, D. P.; Pegg,
A. A.; Foster, P. S.; Hamid, Q.; Rothenberg, M. E. J. Clin.
Invest. 2003, 111, 1863.
A mixture of 7b (0.45 g, 1.55 mmol) and 1 Maq NaOH
(6 mL) was stirred at 45–50 ꢁC for 2 h and then, the pH
was adjusted to 1–2 by addition of 3 Maq HCl to the
cooled (0 ꢁC) solution. After stirring at ambient temper-
ature for 12 h, phenylboric acid (1.55 mmol) and ether
(14 mL) were added. After 3 h at room temperature,
the ether layer was removed and replaced by fresh ether
(14 mL). This treatment was repeated two-fold and the
resulting aqueous phase was washed with ether. The
aqueous solution was concentrated and dried under high
vacuum. After silica gel chromatography (CHCl3/
MeOH/NH3/H2O 6:6:0.5:1.5), compound 7 was ob-
tained (257 mg, 73%) as a colorless solid. Mp 128–
1
130 ꢁC; H NMR (200 MHz, D2O + DCl) d 3.01 (dd,
J = 7.6, 14.6 Hz, 1H), 3.17 (dd, J = 5.5, 14.6 Hz, 1H),
3.87 (dd, J = 5.5, 7.6 Hz, 1H), 7.23–7.40 (m, 2H),
7.45–7.62 (m, 2H); 13C NMR (50 MHz, D2O + DCl) d
36.7, 56.3, 128.9, 132.0, 133.0, 134.7, 174.3; 11B NM R
(96 MHz, D2O + DCl) d 29.2. HRMS: m/z: calcd for
C23H23N3O8B with matrix 3-nitrobenzyl alcohol:
480.1578. Found 480.1584 [M+H]+.
13. Cox, J. D.; Kim, N. N.; Traish, A. M.; Christianson, D.
W. Nat. Struct. Biol. 1999, 6, 1043.
14. Kim, N. N.; Cox, J. D.; Baggio, R. F.; Emig, F. A.;
Mistry, S. K.; Harper, S. L.; Speicher, D. W.; Morris, S.
M.; Ash, D. E.; Traish, A.; Christianson, D. W. Biochem-
istry 2001, 40, 2678.
3.16. Biological assays
15. Meurs, H.; McKay, S.; Maarsingh, H.; Hamer, M. A. M.;
Macic, L.; Molendijk, N.; Zaagsma, J. Br. J. Pharmacol.
2002, 136, 391.
16. Reczkowski, R. S.; Ash, D. E. J. Am. Chem. Soc. 1992,
114, 10992.
17. Kanyo, Z. F.; Scolnick, L. R.; Ash, D. E.; Christianson,
D. W. Nature 1996, 383, 554.
18. Baggio, R. F.; Elbaum, D.; Kanyo, Z. F.; Carroll, P. J.;
Cavalli, R. C.; Ash, D. E.; Christianson, D. W. J. Am.
Chem. Soc. 1997, 119, 8107.
19. Colleluori, D. M.; Ash, D. E. Biochemistry 2001, 40,
9356.
20. Ottenheijm, H. C. J.; Liskamp, R. M. J.; Van Nispen, S. P.
J. M.; Boots, H. A.; Tijhuis, M. W. J. Org. Chem. 1981,
46, 3273.
21. Michnick, T. J.; Matteson, D. S. Synlett 1991, 631.
22. Coutts, S. J.; Adams, J.; Krolikowski, D.; Snow, R. J.
Tetrahedron Lett. 1994, 35, 5109.
23. Matteson, D. S.; Soloway, A. H.; Tomlinson, D. W.;
Campbell, J. D.; Nixon, G. A. J. Med. Chem. 1964, 7, 640.
24. Matteson, D. S. J. Am. Chem. Soc. 1960, 82, 4228.
25. Park, K. C.; Yoshino, K.; Tomiyasu, H. Synthesis 1999,
2041.
Rat liver arginase was purified from male Sprague–
Dawley rats using an Amicon-Green dye ligand media
as previously described.16 Assays quantitated the
[14C]urea produced from [Guanido-14C]L-arginine fol-
lowing a usual protocol.27,28 A typical assay was per-
formed in a total volume of 100 lL. Increasing
concentrations of the tested compounds and RLA were
preincubated for 10 min at room temperature in 0.2 M
Tris buffer pH 7.4 or 9.0 and the assays were initiated
by the addition of L-arginine (final concentration
10 mM) and 106 cpm [Guanido-14C]L-arginine. The
reactions were stopped after 10 min at 37 ꢁC by the
addition of 150 lL of a stop-buffer containing 0.25 M
acetic acid and 7 Murea. [ 14C]Urea was separated from
unreacted [Guanido-14C]L-arginine by mixing with
250 lL of a 1:1 v/v slurry of Dowex (H+ form) in
stop-buffer and centrifugation of the mixture. [14C]Urea
was quantitated by mixing 200 lL aliquots of the super-
natants to 2 mL of Pico-Fluor 40 scintillation cocktail
and counting in a Packard 2100 scintillation counter.
26. Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1984, 53.
27. Ruegg, U. T.; Russell, A. S. Anal. Biochem. 1980, 102, 206.
28. Custot, J.; Boucher, J.-L.; Vadon, S.; Guedes-Moali, C.;
Dijols, S.; Delaforge, M.; Mansuy, D. J. Biol. Inorg.
Chem. 1996, 1, 73.
29. Collet, S.; Carreaux, F.; Boucher, J.-L.; Pethe, S.; Lepoi-
vre, M.; Danion-Bougot, R.; Danion, D. J. Chem. Soc.,
Perkin Trans. 1 2000, 177.
¨
Acknowledgements
The authors thank M. E. Jaouen (UMR 8601 CNRS,
Paris) for her help in the preparation and purification
of RLA, V. Fagot and L. Casse for technical assistance.
30. Custot, J.; Moali, C.; Brollo, M.; Boucher, J.-L.; Dela-
forge, M.; Mansuy, D.; Tenu, J.-P.; Zimmermann, J.-L. J.
Am. Chem. Soc. 1997, 119, 4086.
References and notes
1. Wu, G.; Morris, S. M. Biochem. J. 1998, 336, 1.
2. Jenkinson, C. P.; Grody, W. W.; Cederbaum, S. D. Comp.
Biochem. Physiol. 1996, 114B, 107.
31. Moali, C.; Brollo, M.; Custot, J.; Sari, M.-A.; Boucher,
J.-L.; Stuehr, D. J.; Mansuy, D. Biochemistry 2000, 39,
8208.