Peptidyl 3-substituted 1-hydroxyureas as isosteric analogues of succinylhydroxamate MMP inhibitors

Article abstract of DOI:10.1016/j.ejmech.2007.07.002

To evaluate N-hydroxyurea as zinc binding group in the design of MMP inhibitors, two peptidyl 1-hydroxyureas were prepared by N-hydroxycarbamoylation of the diastereomeric dipeptides H-Leu-Phe-NHMe and H-d-Leu-Phe-NHMe. Peptidyl 1-hydroxyureas were more potent than the parent peptides, but dramatically weaker (4-5 orders of magnitude) than the isosteric (R)-succinylhydroxamate analogue, which displays IC₅₀ in the range of nM vs MMP-1, -3, -7 and sub-nM vs MMP-2, -8, and -9. The peptidyl 1-hydroxyurea 1a attained an IC₅₀ of 20 μM vs MMP-9, and substantially approaches inhibition of known N-hydroxyureas based on aminoacids or peptides against other zinc metalloenzymes and non-peptidic N-hydroxyureas against MMPs. Strong preference of the O-N1-C{double bond, long}O unit for the antiperiplanar amide bond conformation seems to be the major limit for more effective zinc chelation. Methylation of a peptidyl 1-hydroxyurea at N3, to promote the synperiplanar O-N1-C{double bond, long}O conformation required for zinc chelation and improve affinity, resulted in release of a methylimidazolidine-2,4-dione through an undesired intramolecular reaction reminiscent of the Edman peptide degradation.

Full text of DOI:10.1016/j.ejmech.2007.07.002

Available online at www.sciencedirect.com
European Journal of Medicinal Chemistry 43 (2008) 1008e1014
http://www.elsevier.com/locate/ejmech
Original article
Peptidyl 3-substituted 1-hydroxyureas as isosteric analogues of
succinylhydroxamate MMP inhibitors
Cristina Campestre a, Paolo Tortorella b, Mariangela Agamennone a, Serena Preziuso a,
Alessandro Biasone a, Elisa Nuti c, Armando Rossello c, Carlo Gallina a,
*
a Dipartimento di Scienze del Farmaco, Universita` ‘‘G. d’Annunzio’’, Via dei Vestini 31, 66013 Chieti, Italy
b Dipartimento Farmaco-Chimico, Universita` degli Studi di Bari, Via Orabona 4, 70125 Bari, Italy
c Dipartimento di Scienze Farmaceutiche, Universita` degli Studi di Pisa, Via Bonanno 6, 56126 Pisa, Italy
Received 15 April 2007; received in revised form 31 May 2007; accepted 2 July 2007
Available online 15 July 2007
Abstract
To evaluate N-hydroxyurea as zinc binding group in the design of MMP inhibitors, two peptidyl 1-hydroxyureas were prepared by N-hydrox-
ycarbamoylation of the diastereomeric dipeptides H-Leu-Phe-NHMe and H-D-Leu-Phe-NHMe. Peptidyl 1-hydroxyureas were more potent than
the parent peptides, but dramatically weaker (4e5 orders of magnitude) than the isosteric (R)-succinylhydroxamate analogue, which displays
IC50 in the range of nM vs MMP-1, -3, -7 and sub-nM vs MMP-2, -8, and -9. The peptidyl 1-hydroxyurea 1a attained an IC50 of 20 mM vs
MMP-9, and substantially approaches inhibition of known N-hydroxyureas based on aminoacids or peptides against other zinc metalloenzymes
and non-peptidic N-hydroxyureas against MMPs. Strong preference of the OeN1eC]O unit for the antiperiplanar amide bond conformation
seems to be the major limit for more effective zinc chelation. Methylation of a peptidyl 1-hydroxyurea at N3, to promote the synperiplanar
OeN1eC]O conformation required for zinc chelation and improve affinity, resulted in release of a methylimidazolidine-2,4-dione through
an undesired intramolecular reaction reminiscent of the Edman peptide degradation.
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: MMP inhibition; Peptidyl N-hydroxyureas; Succinyl hydroxamates
1. Introduction
MMP active site contains an essential zinc ion which cata-
lyzes the nucleophilic attack of a structured water molecule to
the carbonyl of the scissile peptide bond [5]. MPIs are there-
fore designed by assembling a zinc binding group (ZBG)
with a substrate like peptide or peptidomimetic chain which
is most frequently accommodated in the S0 subsites of the en-
zyme. It has been estimated [6] that more than 90% of MPIs
contain the hydroxamate function that is, by far, the most ef-
fective ZBG. Its potency is accounted for by an ideal accom-
modation [7] in the MMP catalytic site (Fig. 1a), where it not
only forms five member chelates with optimal Zneoxygen
distances, but even establishes two strong H-bonds with the
protein. Interest in non-hydroxamate MPIs, however, is re-
cently increased owing to poor clinical efficacy of hydroxa-
mate based MPIs [6,8].
Matrix metalloproteinases (MMPs) are a family of zinc-en-
dopeptidases that can degrade components of the extracellular
matrix for homeostatic regulation of the extracellular environ-
ment [1]. Overexpression and/or misregulation of these en-
zymes, however, may cause undesired breakdown of
extracellular matrix and have been implicated in pathologies
such as cardiovascular disease, cancer and arthritis [2]. Appro-
priate inhibition of MMPs has been therefore recognized as an
important therapeutic target since 1980s [3], and some MMPs
inhibitors (MPIs) are being tested in clinical trials [3c,4].
* Corresponding author. Tel.: þ39 0871 3555373; fax: þ39 0871 3555267.
E-mail address: cgallina@unich.it (C. Gallina).
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2007.07.002

C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
1009
Zn2+
(a)
(b)
(a)
(b)
2.1
O
O
2.3
O
O
1
N
3
N
1
N
3
N
HO
H
R
R
2
2
1
N
3
N
R
O
O
2.4
R
2
2
Glu198-COO
N
H
H
O
H
H
R
1
H
H
R
1
H
R
1
3.0
Fig. 3. (a) Antiperiplanar OeN1eC]O conformation in the solid state [21],
favoured when N3 is monosubstituted; (b) expected synperiplanar OeN1e
C]O conformation favoured when N3 is disubstituted; OeH hydrogen
bond represented as a dotted line.
Ala161-CO
Fig. 1. Key bonding interactions of hydroxamate function (a) as found in the
˚
complex of batimastat with MMP-8 (distances in A) [7], and 3-substituted
1-hydroxyureas (b) as potential alternatives.
NH of the 1-hydroxyurea 1a, however, changes substantially
the geometry of the molecule in view of its accommodation
in the MMPs’ active site. To further evaluate and compare
the effects of stereochemistry variations in the area of the
ZBG, we synthesized also the diastereomeric 1-hydroxyurea
1b, containing the Leu residue in the D configuration.
As N-hydroxyureas bearing a hydrogen atom on N3
(Fig. 3a) adopt, in the solid state [21], a roughly antiperiplanar
OeN1eC]O conformation unproductive for zinc chelation,
it was also planned to synthesize the N3-methyl derivative
1c, where the synperiplanar OeN1eC]O conformation
(Fig. 3b) would be favoured. In addition, N-methylation of
the amide nitrogen is reported to favour pyramidalization of
nitrogen, as observed in a N3-methylated 1-hydroxyurea:MMP-
8 complex [18] and in N-methylated azapeptides [22]. Owing
to the consequent increase in flexibility, the N3-methylated 1-
hydroxyurea 3c was expected to approach its conformational
behaviour to that of analogous hydroxamates like 2a or 2c
that are potent MMP inhibitors.
The N-hydroxyurea function seems particularly suited as
hydroxamate alternative in MPIs since it incorporates the
COeNHeOH group (Fig. 1b) that is necessary to establish
the same ideal network of bonding interactions in the MMP
active site, and also presents a more favorable pharmacoki-
netic profile. N-Hydroxyurea itself, used since 1960s as
antineoplastic drug, is characterized by optimal oral bioavail-
ability (80e100%) and in vivo half-live of 3.5e4.5 h [9]. In
addition, 3-substituted 1-hydroxyureas have been reported as
5-lipoxygenase inhibitors with improved oral bioavailability
and slower in vitro metabolism with respect to analogous hy-
droxamates [10]. Several 3-substituted 1-hydroxyureas have
been also evaluated as cytostatic and antiviral agents [11], as
inhibitors of ribonuclease [12], and of enzymes that incorpo-
rate a catalytic zinc ion, including carboxypeptidase A [13],
bacterial peptide deformylase [14], carbonic anhydrase [15],
histone deacetylases [16], and MMPs [17]. We have recently
shown [18] that non-peptidic 3-substituted 1-hydroxyureas
are weak inhibitors of MMPs, binding the catalytic zinc ion
as monodentate ligands.
2. Chemistry
As an extension of our program aimed to evaluate 3-
substituted 1-hydroxyureas in the design of MPIs, we planned
to synthesize the peptidyl 1-hydroxyurea 1a (Fig. 2) as isos-
teric analogue of the potent succinylhydroxamate MMP inhib-
itor 2a [19] which is a simplified version of batimastat (2c)
[20].
The 1-hydroxyurea 1a contains the dipeptide H-Leu-Phe-
NHMe with the Leu residue in the L configuration, so that
its side chain retains the same orientation of the iso-butyl
chain of the (R)-succinyl hydroxamate 2a. Replacement of
the sp3 carbon of the succinylhydroxamate 2a with the sp2
Synthesis of the required 1-hydroxyureas 1a and 1b was
achieved through Scheme 1 that also provides dipeptides 3c
and 4c for evaluation of their potency as MPIs. Coupling of
Z-Leu-OH with H-Phe-OMe by treatment with DCC in the
presence of 1-hydroxybenzotriazole [23] gave the dipeptide
methyl ester 3a that was converted into the corresponding
N-methylamide 3b by treatment with N-methylamine in
MeOH. Removal of the Z group by hydrogenolysis gave the
free dipeptide N-methylamide 3c. Conversion into the desired
i
X-Leu-Phe-Y
X
Cbz-Leu-OH
O
R
O
O
R
O
H
N
H
N
O
CH3
O
CH3
No
Y
H
N
H
N
H
H
N
H
N
R1
N
H
O
O
R1
R
C6H5
C6H5
OCH3
3a
Cbz
ii
R1
H
No
1a
1b
1c
R
No
2a
2b
R1
H
NH-CH3
NH-CH3
NH-CH3
Cbz
3b
3c
1a
iii
i-Bu
i-Bu
i-Bu
i-Bu
i-Bu
H
H
H
iv
HO-HN-CO
CH3
H2C
S
2c
i-Bu
Scheme 1. Reagents: (i) H-Phe-OMe, 4-methylmorpholine DCC, HOBt, THF;
(ii) CH3NH2, CH3OH, 50 ꢀC; (iii) H2, Pd/C, CH3OH; (iv) p-nitrophenyl-N-hy-
droxycarbamate, CH3CN. Compound 1b was prepared accordingly through in-
termediates 4a, 4b, and 4c, starting from Cbz-D-Leu-OH.
S
Fig. 2. Peptidyl 3-substituted 1-hydroxyureas (1) and analogous succinyl hy-
droxamic acids (2).

1010
C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
No
X
Y
OCH3
5a Boc
5b Boc
ii
O
i
H
N
NH-CH3
NH-CH3
NH-CH3
Boc
X
iii
N
CH3
COOH
N
CH3
Y
5c
H
O
C6H5
iv
1c CO-NH-OH
v
H
N
O
O
vi
H
N
H3C
O
(Spontaneous)
C6H5
N
O
CH3
H3C
O
N
H
N
CH3
N
H
N
O
N
O
H
N
C6H5
O
C6H5
O
H
B
O
C6H5
6
8
7
1c
Scheme 2. Reagents: (i) H-Phe-OMe, 4-methylmorpholine DCC, HOBt, THF; (ii) CH3NH2, CH3OH, 50 ꢀC; (iii) TFA; (iv) p-nitrophenyl-N-hydroxycarbamate,
CH3CN; (v) O-benzyl-hydroxylamine$HCl, Et3N, CDI, CH2Cl2; (vi) H2, Pd/C, CH3OH.
1-hydroxyurea 1a was achieved by treatment with p-nitro-
phenyl-N-hydroxycarbamate [24] in CH3CN, while the milder
acylating reagent phenyl-N-hydroxycarbamate [25], com-
monly used for N-hydroxyureas preparation, proved to be
completely ineffective for acylation of the amino group of
a-aminoacid amides. The diastereomeric dipeptide N-methyl-
amide H-D-Leu-Phe-NHMe (4c) and the 1-hydroxyurea 1b
were obtained starting from D-Leu-OMe, by following the
same synthetic sequence of Scheme 1. The two diastereomeric
succinyl hydroxamates 2a and 2b, necessary to compare their
inhibition constants with that of the 1-hydroxyureas 1a and 1b,
were also prepared following methods previously reported
[19].
Synthesis of the N-methyl-hydroxyurea 1c (Scheme 2) was
attempted starting from Boc-N-methyl-L-leucine, in a procedure
analogous to Scheme 1. When the dipeptide 5c (Scheme 2)
was subjected to N-acylation with p-nitrophenyl-N-hydroxycar-
bamate we failed to obtain the expected N-hydroxyurea 1c. To
overcome this difficulty, we tried an alternative route progress-
ing through an O-benzyl-N-hydroxyurea (6) [24], a less polar
and easier to handle intermediate, in the expectations that it
could be converted into the target compound 1c by catalytic
hydrogenolysis. When the dipeptide 5c was coupled with O-
benzylhydroxylamine in the presence of carbonyldiimidazole
in CH2Cl2 [22] at room temperature, for 24 h, the imidazoli-
dine-2,4-dione 8 was isolated in 64%yield. The structure is in
accordance with 1H and 13C NMR spectra, and stereochemistry
was assigned on the basis of the (S )-configuration of the N-
methyl-leucine residue of 5c.
The expected O-benzyl-N-hydroxyurea 6 probably under-
went a facile intramolecular reaction reminiscent of the Ed-
man peptide degradation [27]. While the original Edman
method requires strong acid catalysis (HCl in nitromethane),
the modified Edman procedure [28], applied to peptides con-
taining an N-alkylated amino terminal residue, proceeds with
release of a phenylthiohydantoin under mild basic conditions.
We believe that the presence of an additional alkyl group
on the amide nitrogen, both in N-terminal O-benzyl-N-
hydroxyureas and in phenyl-thiocarbamyl derivatives of the
modified Edman procedure, strongly increases flexibility of
the chain and pyramidalization of nitrogen. These circum-
stances probably favour bended conformations (7) required by
cyclization and increase nitrogen nucleophilicity for intramo-
lecular attack and cleavage of the LeuePhe peptide bond. Anal-
ogous effects probably intervene in the target N-hydroxyurea
1c, when formed from 5c, and favour the undesired LeuePhe
peptide bond cleavage.
3. Results and discussion
The (R)-succinyl hydroxamate 2a was selected as a typical
broad spectrum [19] peptidomimetic MMP inhibitor, in order
Table 1
In vitro inhibiting activity of 3-substituted 1-hydroxyureas (1a,b) of analogous succinyl hydroxamates (2a,b) and related dipeptides (3c,4c)
No.
MMPs IC50 (mM)a
1
2
3
7
8
9
1a
1b
2a
2b
3c
4c
a
270
>300
0.025
0.64
60
>300
0.0007
0.17
>300
>300
>300
>300
>300
>300
70
150
20
200
0.043
20
ND
>300
0.0083
4.1
>300
>300
0.000046
0.048
0.00038
0.15
NDb
ND
>300
>300
>300
>300
Errors in the range of 5e10% of the reported value (from three different assays).
Not determined.
b

C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
1011
C6H5
O
O
H
N
O
HO
N
H
H
N
H
N
N
H
N
H
COOH
O
9
10
40% Inhibition vs peptide deformilase at 1 µM [14b]
Ki 1.54 µM vs carboxypeptidase A [13]
O
O
O
H
N
H
N
O
HO
O
N
H
N
H
C6H4(p)C6H4CN(p)
C6H4-Bn(p)
NH
O
O
11
12
IC50 80 µM vs MMP-3 [17a]
KD154 µM vs MMP-12 [17b]
Fig. 4. Inhibition potencies of known N-hydroxyurea inhibitors against different zinc metalloenzymes.
to compare the target N-hydroxyurea 1a with a hydroxamate
4. Conclusions
analogue that displays nM potencies against several different
MMPs with relevant differences in the sequence and flexibility
of the active site loop [29]. As expected, the hydroxamate 2a
proved to be the most potent inhibitor (Table 1) against MMP-
2, -3, -8, and -9, as previously reported [19], and against
MMP-1 and -7. The diastereomeric (S )-succinyl hydroxamate
2b showed binding affinities at least 100-fold lower, confirm-
ing that this model of ligand attains maximum potency when
the iso-butylsuccinyl unit presents the (R)-configuration.
Results reported in Table 1 show that conversion of the sub-
strate analogue dipeptide 3a (IC50 > 300 mM) into the N-hy-
droxyurea 1a causes a measurable increase of binding
affinity. These values, however, are in the mM range, with
a dramatic decrease (4e5 orders of magnitude) in binding af-
finities with respect to the reference hydroxamate 2a. A de-
crease in binding affinity of 2e3 orders of magnitude is also
observed with respect to the (S )-hydroxamate 2b. Thus, re-
placement of the CH2eCOeNHeOH chain of the hydroxa-
mate with the NHeCOeNHeOH group causes a decrease
in potencies even higher than that observed for inversion of
configuration at the iso-butylsuccinyl unit. As previously ob-
served for a non-peptidyl MMP inhibitor [18], the dramatic
decrease in affinity of the N-hydroxyurea with respect to
closely structural related hydroxamate can be related to the ri-
gidity of eNHeCOeNHOH sequence and to the high energy
penalty required to promote the synperiplanar OeN1eC]O
amide bond conformation productive for zinc chelation. These
troubles probably represent a general limit for N-hydroxyurea
inhibitors, since the value of 20 mM attained by 1a vs MMP-9,
substantially approaches inhibition of N-hydroxyureas (Fig. 4)
based on aminoacids (9) or peptides (10) against other zinc
metalloenzymes and non-peptidic N-hydroxyureas (11 and
12) against MMPs.
Both peptidyl (1a, 1b, and 10) and non-peptidyl [17,18]
3-substituted 1-hydroxyurea inhibitors attain inhibition
constants that are 4e5 order of magnitude higher than that
of analogous hydroxamates. Potency improvements could
probably be achieved by use of peptidomimetic backbones
that favour zinc chelation by the NHeCOeNHOH in its
planar conformation. The energy penalty for promotion of
the synperiplanar OeN1eC]O amide bond conformation,
however, seems to be the major limit for more effective zinc
chelation. Experimental measurements for zinc complexation
constants in solution could be useful to disclose N-hydroxy-
urea attitude for zinc chelation.
5. Experimental protocols
5.1. Chemistry
Melting points were determined on a Buchi B-540 appara-
¨
tus and are uncorrected. Optical rotations were measured with
a PerkineElmer 241 digital polarimeter at 20 ꢀC; concentra-
1
tions are expressed as g/100 mL. H and 13C NMR spectra
were recorded on a Varian VXR 300 instrument; chemical
shifts are expressed in d units, using TMS as an internal stan-
dard. Microanalyses indicated by the symbols of the elements
are within ꢁ0.4% of the theoretical values and were performed
on a Carlo Erba mod. 1106 analyzer. All solvents and reagents
were obtained from commercial sources and used without
further purification.
5.1.1. N-(Benzyloxycarbonyl)-L-leucyl-L-phenylalanine
methyl ester (3a)
To a solution of L-phenylalanine methyl ester hydrochloride
Methylation of the N3 atom was attempted to favour pyra-
midalization of N3 and flexibility of the chain with improved
resemblance to hydroxamate and increased affinity for MMPs.
This modification of the N-hydroxyurea group and its O-
benzyl derivative 6, however, strongly increases tendency to
cyclization and appears to prevent any access to N3-methylated
peptidyl N-hydroxyureas.
(3.11 g,
14.4 mmol),
4-methylmorpholine
(1.6 mL,
14.4 mmol) and N-(benzyloxycarbonyl)-L-leucine (3.82 g,
14.4 mmol) in anhydrous THF (35 mL), a solution of DCC
(3.56 g, 17.3 mmol) and HOBt (3.28 g, 28.8 mmol) in anhy-
drous THF (30 mL) was added dropwise at 0 ꢀC. The reaction
mixture was stirred for 1 h at 0 ꢀC, and for additional 20 h at
room temperature. After removal of N,N0-dicyclohexylurea by

1012
C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
filtration, the mixture was diluted with EtOAc (150 mL) and
washed with 2 N HCl (2 ꢂ 70 mL), NaHCO3 (2 ꢂ 70 mL)
and brine. The organic layer was dried (Na2SO4) and concen-
trated to give a residue that was triturated with hexane to pro-
vide ester 3a as a white solid (5.96 g, 97%). Mp 80.0e81.5 ꢀC;
[a]2D0 ꢃ22.8ꢀ (c 1.0, CH3OH); 1H NMR (CDCl3) d 0.91 (d, 6H,
J ¼ 6.3 Hz), 1.42e1.63 (m, 3H), 3.09 (dd, 1H, J ¼ 13.7 and
5.7 Hz), 3.12 (dd, 1H, J ¼ 13.7 and 5.7 Hz), 3.72 (s, 3H),
4.12e4.18 (m, 1H), 4.84 (dd, 1H, J ¼ 13.7 and 5.7 Hz),
5.04e5.17 (m, 2H), 6.41 (d, 1H, J ¼ 7.0 Hz), 7.08 (d, 1H,
J ¼ 6.9 Hz), 7.19e7.29 (m, 5H), 7.31e7.35 (m, 5H).
NMR (DMSO-d6) d 20.1, 21.4, 22.4, 23.8, 36.1, 40.0, 49.8,
52.1, 124.6, 126.4, 127.5, 136.1, 159.1, 169.5, 170.6. Anal.
C17H26N4O4 (C, H, N).
5.1.5. N-(Benzyloxycarbonyl)-D-leucyl-L-phenylalanine
methyl ester (4a)
Prepared as described for 3a, starting from D-leucine. Crys-
tallization from EtOAc/n-hexane gave 16.0 g (98%) of 4a as
a white solid: mp 114.1e114.5 ꢀC; [a]D20 þ50.1ꢀ (c 1.0,
CHCl3); 1H NMR (CDCl3) d 0.87 (d, 6H, J ¼ 6.3 Hz),
1.39e1.59 (m, 3H), 3.04 (dd, 1H, J ¼ 13.8 and 6.9 Hz), 3.14
(dd, 1H, J ¼ 15.0 and 5.4 Hz), 3.69 (s, 3H), 4.16e4.24 (m,
1H), 4.83e4.89 (m, 1H), 5.04e5.12 (m, 2H), 5.25 (d, 1H,
J ¼ 8.1 Hz), 6.60 (d, 1H, J ¼ 7.5 Hz), 7.18e7.25 (m, 5H),
7.29e7.35 (m, 5H); 13C NMR (CDCl3) d 22.1, 23.1, 24.9,
38.1, 41.8, 52.6, 53.2, 53.7, 67.3, 127.4, 128.3, 128.4, 128.7,
128.7, 129.5, 136.0, 136.4, 156.3, 172.0, 172.1.
5.1.2. N-(Benzyloxycarbonyl)-L-leucyl-L-phenylalanine
methylamide (3b)
To a solution of methyl ester 3a (6.12 g, 14.4 mmol) in
CH3OH (40 mL), a solution (33% in EtOH) of methylamine
(17.0 mL, 143.5 mmol) was added. After 5 h at 50 ꢀC, the vol-
atiles were evaporated. Crystallization from EtOAc/n-hexane
gave 5.49 g (90%) of 3b as a white solid. Mp 180.7e
5.1.6. N-(Benzyloxycarbonyl)-D-leucyl-L-phenylalanine
methylamide (4b)
181.0 ꢀC; [a]D20 ꢃ33.0ꢀ (c 1.0, CH3OH); H NMR (CDCl3)
1
d 0.89 (d, 3H, J ¼ 6.0 Hz), 0.90 (d, 3H, J ¼ 6.0 Hz), 1.34e
1.59 (m, 3H), 2.69 (d, 3H, J ¼ 5.2 Hz), 3.00e3.10 (m, 2H),
4.00e4.18 (m, 1H), 4.57 (dd, 2H, J ¼ 14.5 and 8.0 Hz),
5.00e5.11 (m, 2H), 5.30 (d, 1H, J ¼ 5.2 Hz), 5.93 (s, 1H),
6.81 (d, 1H, J ¼ 9.0 Hz), 7.15e7.41 (m, 10H).
Prepared as described for 3b, starting from methyl ester 4a.
Crystallization from CH3OH/Et2O gave 14.3 g (92%) of 4b as
a white solid. Mp 182.9e183.9 ꢀC; [a]D20 ꢃ3.0ꢀ (c 1.0,
1
CH3OH); H NMR (CDCl3) d 0.81 (d, 3H, J ¼ 6.0 Hz), 0.83
(d, 3H, J ¼ 6.0 Hz), 1.26e1.52 (m, 3H), 2.70 (d, 3H,
J ¼ 5.4 Hz), 3.07 (dd, 1H, J ¼ 14.1 and 7.2 Hz), 3.17 (dd,
1H, J ¼ 14.7 and 7.2 Hz), 3.89e3.96 (m, 1H), 4.65e4.70
(m, 1H), 5.04e5.12 (m, 2H), 5.18 (d, 1H, J ¼ 6.3 Hz), 6.27
(d, 1H, J ¼ 9.3 Hz), 6.44 (br s, 1H), 7.16e7.35 (m, 10H);
13C NMR (CDCl3) d 22.6, 22.8, 24.7, 26.5, 37.9, 41.1, 51.6,
54.5, 67.5, 127.3, 128.2, 128.6, 128.8, 128.9, 129.4, 136.1,
136.9, 156.7, 171.1, 172.4.
5.1.3. L-Leucyl-L-phenylalanine methylamide (3c)
A solution of the methylamide 3b (3.00 g, 7.0 mmol) in
CH3OH (100 mL) was stirred under hydrogen, for 5 h, at
room pressure and temperature, in the presence of 5% Pd/C
(300 mg). The reaction mixture was filtered and evaporated
under reduced pressure to give the crude amine. Trituration
with Et2O/hexane 1:1 gave 2.02 g (98%) of 3c as a white solid.
1
Mp 118.4e120.3 ꢀC; [a]D20 þ4.1ꢀ (c 1.0, CH3OH); H NMR
5.1.7. D-Leucyl-L-phenylalanine methylamide (4c)
(CDCl3)
d
0.86 (d, 3H, J ¼ 9.3 Hz), 0.88 (d, 3H,
Prepared as described for 3c, starting from methylamide 4b.
Crystallization from CH2Cl2/n-hexane afforded 1.86 g (98%)
of 4c as a white solid. Mp 130.9e131.5 ꢀC; [a]D20 þ16.7ꢀ (c
J ¼ 9.3 Hz), 1.08e1.49 (m, 3H), 1.47 (br s, 2H), 2.70 (d,
3H, J ¼ 4.8 Hz), 3.01 (dd, 1H, J ¼ 13.5 and 7.5 Hz), 3.13
(dd, 1H, J ¼ 13.8 and 6.9 Hz), 3.28e3.36 (m, 1H), 4.50e
4.56 (m, 1H), 6.19 (s, 1H), 7.18e7.30 (m, 5H), 7.84 (d, 1H,
J ¼ 8.4 Hz). Anal. C16H25N3O2 (C, H, N).
1
1.0, CHCl3); H NMR (CDCl3) d 0.88 (d, 3H, J ¼ 9.0 Hz),
0.90 (d, 3H, J ¼ 9.0 Hz), 1.20e1.63 (m, 3H), 1.95 (s, 2H),
2.71 (d, 3H, J ¼ 5.1 Hz), 3.10 (dd, 1H, J ¼ 13.5 and 7.8 Hz),
3.13 (dd, 1H, J ¼ 13.5 and 7.2 Hz), 3.28e3.36 (m, 1H),
4.50e4.56 (m, 1H), 6.17 (s, 1H), 7.20e7.30 (m, 5H), 7.77
(d, 1H, J ¼ 7.5 Hz); 13C NMR (CDCl3) d 21.7, 23.5, 25.0,
26.4, 38.1, 44.1, 53.3, 54.8, 127.1, 128.8, 129.5, 147.2,
171.8, 176.0. Anal. C16H25N3O2 (C, H, N).
5.1.4. N-[(Hydroxyamino)carbonyl]-L-leucyl-L-
phenylalanine methylamide (1a)
To a solution of L-leucyl-L-phenylalanine methylamide 3c
(716 mg, 2.46 mmol) in CH3CN (30 mL), p-nitrophenyl-N-
hydroxycarbamate (730 mg, 3.68 mmol) was added portion-
wise under N2. After 24 h under stirring, the reaction mixture
was concentrated in vacuo. Purification by silica gel column
chromatography (MeOH/EtOAc 1:9), followed by crystalliza-
tion from EtOAc afforded 712 mg (83%) of the title compound
1a as a white solid. Mp 145.2e145.5 ꢀC; [a]D20 ꢃ35.7ꢀ (c 1.0,
5.1.8. N-[(Hydroxyamino)carbonyl]-D-leucyl-L-
phenylalanine methylamide (1b)
Prepared as described for 1a, starting from N-methylamide
4c. Yield 75%, white solid; mp 195.0e196.0 ꢀC; [a]D20 ꢃ10.4ꢀ
(c 1.0, CH3OH). 1H NMR (DMSO-d6) d 0.70 (t, 6H,
J ¼ 6.3 Hz), 1.04e1.19 (m, 3H), 2.61 (d, 3H, J ¼ 4.8 Hz),
2.68 (dd, 1H, J ¼ 13.5 and 9.0 Hz), 3.05 (dd, 1H, J ¼ 14.1
and 7.5 Hz), 4.04e4.11 (m, 1H), 4.32e4.40 (m, 1H), 6.37
(d, 1H, J ¼ 7.2 Hz), 7.15e7.22 (m, 5H), 7.77 (d, 1H,
J ¼ 4.8 Hz), 8.44 (d, 1H, J ¼ 8.7 Hz), 8.53 (s, 1H), 8.70
(s, 1H); 13C NMR (DMSO-d6) d 22.9, 23.3, 24.5, 26.3, 37.9,
1
CH3OH); H NMR (DMSO-d6) d 0.80 (t, 6H, J ¼ 6.3 Hz),
1.21e1.46 (m, 3H), 2.52 (d, 3H, J ¼ 4.8 Hz), 2.77 (dd, 1H,
J ¼ 13.5 and 9.0 Hz), 2.92 (dd, 1H, J ¼ 14.1 and 6.0 Hz),
4.08e4.16 (m, 1H), 4.35e4.46 (m, 1H), 6.57 (d, 1H,
J ¼ 8.7 Hz), 7.15e7.25 (m, 5H), 7.90 (d, 1H, J ¼ 4.5 Hz),
8.07 (d, 1H, J ¼ 8.7 Hz), 8.50 (s, 1H), 8.71 (s, 1H); 13C

C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
1013
1
42.0, 52.0, 54.9, 126.8, 128.6, 129.8, 138.8, 161.6, 172.0,
176.0. Anal. C17H26N4O4 (C, H, N).
oil. [a]2D0 ꢃ15.4ꢀ (c 1.0, CHCl3); H NMR (CDCl3) d 0.89 (d,
3H, J ¼ 6.6 Hz), 0.91 (d, 3H, J ¼ 6.6 Hz), 1.55e1.80 (m, 3H),
2.92 (d, 3H, J ¼ 1.3 Hz), 3.74e3.78 (m, 1H), 5.14 (s, 2H),
7.26e7.51 (m, 5H); 13C NMR (CDCl3) d 22.9, 23.2, 24.3,
28.6, 38.1, 58.4, 79.4, 128.7, 129.5, 130.2, 133.8, 153.3,
167.7. Anal. C15H20N2O3 (C, H, N).
5.1.9. N-(tert-Butoxycarbonyl)-N-methyl-L-leucyl-L-
phenylalanine methyl ester (5a)
Prepared as described for 3a, starting from N-tert-butoxy-
carbonyl-N-methyl-L-leucine and L-phenylalanine methyl
ester. Purification by silica gel chromatography (CHCl3/
5.2. MMP inhibition assays [30]
1
hexane 2:8) gave 2.27 g (68%) of 5a as colourless liquid. H
NMR (CDCl3) d 0.86 (d, 3H, J ¼ 6.5 Hz), 0.90 (d, 3H,
J ¼ 6.5 Hz), 1.42 (s, 9H), 1.50e1.80 (m, 3H), 2.51 (s, 3H),
3.14 (d, 1H, J ¼ 5.1 Hz), 3.18 (d, 1H, J ¼ 6.0 Hz), 3.72 (s,
3H), 4.56e4.65 (m, 1H), 4.81e4.85 (m, 1H), 6.38 (d, 1H,
J ¼ 7.0 Hz), 7.06e7.29 (m, 5H); 13C NMR (CDCl3) d 21.7,
23.4, 24.7, 28.5, 29.6, 36.4, 38.2, 53.4, 56.2, 57.3, 81.0,
127.3, 128.8, 129.3, 136.2, 156.7, 171.2, 172.0.
Recombinant human progelatinase A (pro-MMP-2) and B
(pro-MMP-9) from transfected mouse myeloma cells, were
supplied by Prof. Gillian Murphy (Department of Oncology,
University of Cambridge, UK); pro-MMP-1, pro-MMP-3,
pro-MMP-7 and pro-MMP-8 were purchased from Calbio-
chem. Proenzymes were activated immediately prior to use
with p-aminophenylmercuric acetate (APMA 2 mM for 1 h
at 37 ꢀC for MMP-2, MMP-1 and MMP-8, 1 mM for 1 h at
37 ꢀC for MMP-9 and MMP-7). Pro-MMP-3 was activated
with trypsin 5 mg/mL for 15 min at 37 ꢀC followed by soybean
trypsin inhibitor (SBTI) 62 mg/mL.
For assay measurements, the inhibitor stock solutions
(DMSO, 100 mM) were further diluted, at seven different con-
centrations (0.01 nMe300 mM) for each MMP in the fluori-
metric assay buffer (FAB: Tris 50 mM, pH ¼ 7.5, NaCl
150 mM, CaCl2 10 mM, Brij 35 0.05% and DMSO 1%). Ac-
tivated enzyme (final concentration 2.9 nM for MMP-2,
2.7 nM for MMP-9, 2.4 nM for MMP-7, 1 nM for MMP-3,
1.5 nM for MMP-8 and 0.20 nM for MMP-1) and inhibitor so-
lutions were incubated in the assay buffer for 4 h at 25 ꢀC. Af-
ter the addition of 200 mM solution of the fluorogenic substrate
Mca-Arg-Pro-Lys-Pro-Val-Glu-Nva-Trp-Arg-Lys(Dnp)-NH2
(Sigma) for MMP-3 and Mca-Lys-Pro-Leu-Gly-Leu-Dpa-Ala-
Arg-NH2 (Bachem) for all the other enzymes in DMSO (final
concentration 2 mM), the hydrolysis was monitored every 15 s
for 20 min recording the increase in fluorescence (lex ¼ 325 nm,
lem ¼ 400 nm) using a Molecular Device SpectraMax Gemini
XS plate reader. The assays were performed in triplicate in
a total volume of 200 mL per well in 96-well microtitre plates
(Corning, black, NBS). Control wells lack inhibitor. The MMP
inhibition activity was expressed in relative fluorescent units
(RFU). Percent of inhibition was calculated from control reac-
tions without the inhibitor. IC50 was determined using the for-
mula: Vi/Vo ¼ 1/(1 þ [I]/IC50), where Vi is the initial velocity
of substrate cleavage in the presence of the inhibitor at
concentration [I] and Vo is the initial velocity in the absence
of the inhibitor. Results were analyzed using SoftMax Pro
software and GraFit software [31].
5.1.10. N-(tert-Butoxycarbonyl)-N-methyl-L-leucyl-L-
phenylalanine methylamide (5b)
Prepared as described for 3b, starting from methyl ester 5a.
Purification by silica gel column chromatography (2% iso-prop-
anol in CHCl3) provided 1.75 g (88%) of 5b as a white solid. Mp
98.8e100.5 ꢀC; [a]D20 ꢃ57.4ꢀ (c 1.0, CH3OH); 1H NMR
(CDCl3) d 0.88 (d, 3H, J ¼ 8.1 Hz), 0.91 (d, 3H, J ¼ 8.1 Hz),
1.46 (s, 9H), 1.53e1.63 (m, 3H), 2.47 (s, 3H), 2.73 (d, 3H,
J ¼ 4.8 Hz), 3.04 (d, 2H, J ¼ 6.9 Hz), 4.21e4.58 (m, 2H),
5.93 (s, 1H), 6.47 (br s, 1H), 7.15e7.31 (m, 5H); 13C NMR
(CD3OD) d 20.7, 22.4, 24.7, 25.1, 27.5, 29.2, 36.7, 37.9, 54.5,
56.4, 80.0, 126.6, 128.4, 129.0, 137.1, 155.6, 172.2, 172.7.
5.1.11. N-Methyl-L-leucyl-L-phenylalanine methylamide (5c)
Compound 5b (1.54 g, 3.8 mmol) was dissolved in TFA
(20 mL). After 4 h at room temperature, the reaction mixture
was concentrated, diluted with EtOAc (200 mL), and washed
with NaHCO3 (3 ꢂ 100 mL) and brine. Evaporation of the sol-
vents, followed by crystallization from EtOAc/n-hexane gave
894 mg (77%) of 5c as a white solid. Mp 134.3e135.3 ꢀC;
[a]2D0 ꢃ1.99ꢀ (c 1.0, CH3OH); H NMR (CDCl3) d 0.84 (d,
1
3H, J ¼ 6.6 Hz), 0.86 (d, 3H, J ¼ 6.6 Hz), 1.09e1.55 (m, 3H),
2.27 (s, 3H), 2.73 (d, 3H, J ¼ 5.2 Hz), 2.90e3.18 (m, 3H),
4.53e4.61 (m, 1H), 6.10 (br s, 1H), 7.19e7.30 (m, 5H), 7.66
(d, 1H, J ¼ 7.5 Hz); 13C NMR (CDCl3) d 22.2, 23.3, 25.2,
26.4, 35.6, 38.0, 42.8, 54.3, 63.6, 127.0, 128.8, 129.4, 137.3,
171.9, 175.6.
5.1.12. (5S )-3-(Benzyloxy)-5-isobutyl-1-
methylimidazolidine-2,4-dione (8)
To a solution of O-benzylhydroxylamine hydrochloride
(159 mg, 1.0 mmol) and Et3N (112 mL, 0.8 mmol) in anhydrous
CH2Cl2 (7 mL), 1,10-carbonyldiimidazole (133 mg, 0.8 mmol)
was added [26]. After 1 h under stirring, compound 5c
(205 mg, 0.7 mmol) was added, and the resulting mixture was
stirred for 24 h. The reaction mixture was then diluted with
CH2Cl2 (10 mL) and washed with 1 N HCl (2 ꢂ 10 mL),
NaHCO3 (2 ꢂ 10 mL) and brine. The organic layer was concen-
trated to dryness and purified by column chromatography
(EtOAc/n-hexane 2:8), to provide 8 (82 mg, 64%) as a yellow
Acknowledgments
Financial support by Italian MIUR is gratefully
acknowledged.
References
[1] J.F. Woessner, H. Nagase (Eds.), Matrix Metalloproteinases and TIMPS,
Oxford University Press, New York, NY, USA; 2000.

1014
C. Campestre et al. / European Journal of Medicinal Chemistry 43 (2008) 1008e1014
[2] (a) M.L. Lindsey, Heart Fail. Rev. 9 (2004) 7e19;
(b) L.M. Coussens, B. Fingleton, M.L. Matrisian, Science 295 (2002)
2387e2392;
[17] (a) M.R. Michaelides, J.F. Dellaria, J. Gong, J.H. Holms, J.J. Bouska,
J. Stacey, C.K. Wada, H.R. Heyman, M.L. Curtin, Y. Guo,
C.L. Goodfellow, I.B. Elmore, D.H. Albert, T.J. Magoc, P.A. Marcotte,
D.W. Morgan, S.K. Davidsen, Bioorg. Med. Chem. Lett. 11 (2001)
1553e1556;
(b) C. Mannino, M. Nievo, F. Machetti, A. Papakyriakou, V. Calderone,
M. Fragaid, A. Guarna, Bioorg. Med. Chem. 14 (2006) 7392e7403.
[18] C. Campestre, M. Agamennone, P. Tortorella, S. Preziuso, A. Biasone,
E. Gavuzzo, G. Pochetti, F. Mazza, O. Hiller, H. Tschesche,
V. Consalvi, C. Gallina, Bioorg. Med. Chem. Lett. 16 (2006)
20e24.
(c) C. Overall, O. Kleifeld, Br. J. Cancer 94 (2006) 941e946.
[3] (a) M. Whittaker, C.D. Floyd, P. Brown, A.J.H. Gearing, Chem. Rev. 99
(1999) 2735e2376;
(b) J.W. Skiles, N.C. Gonnella, A.Y. Jeng, Curr. Med. Chem. 11 (2004)
2911e2977;
(c) H. Matter, M. Schudok, Curr. Opin. Drug Discov. Devel. 7 (2004)
513e535;
(d) B.G. Rao, Curr. Pharm. Des. 11 (2005) 295e322.
[4] G. Abbenante, D.P. Fairlie, J. Med. Chem. 1 (2005) 71e104.
[5] B. Lovejoy, A.M. Hassel, M.A. Luther, D. Weigl, S.R. Jordan, Biochem-
istry 33 (1994) 8207e8217.
[19] (a) J.P. Dickens, D.K. Donald, G. Kneen, W.R. McKey, Eur. Pat. Appl.
EP0214639, 1987;
(b) D.A. Levy, F. Lapierre, W. Liang, W. Ye, C.W. Lange, X. Li,
D. Grobelny, M. Casabonne, D. Tyrrell, K. Holme, A. Nadzan,
R.E. Galardy, J. Med. Chem. 41 (1998) 199e233.
[6] D.T. Puerta, M.O. Griffin, J.A. Lewis, D. Romero-Perez, R. Garcia,
F.J. Villareal, S.M. Cohen, J. Biol. Inorg. Chem. 11 (2006) 131e138.
[7] F. Grams, M. Crimmin, L. Hinnes, P. Huxley, M. Pieper, H. Tschesche,
W. Bode, Biochemistry 34 (1995) 14012e14020.
[20] C. Campion, A.H. Davidson, A.E.H. Gearing, M.J. Crimmin, PCT Int.
Appl. WO9005719, British Bio-Technology Ltd, U.K., 1990.
[21] B.B. Nielsen, K. Frydenvang, I.K. Larsen, Acta Crystallogr. C49 (1993)
1018e1022.
[8] E. Breuer, J. Frant, R. Reuven, Expert Opin. Ther. Pat. 15 (2005)
253e269.
[9] R.C. Donehower, in: B. Chabner (Ed.), Pharmacological Principles of
Cancer Treatment, W.B. Saunders Co., Philadelphia, 1982, pp. 269e275.
[10] A.O. Stewart, P.A. Bhatia, J.G. Martin, J.B. Summers, K.E. Rodriques,
M.B. Martin, J.H. Holms, J.L. Moore, R.A. Craig, T. Kolasa,
J.D. Ratajczyk, H. Mazdiyasni, F.A.J. Kerdesky, S.L. DeNinno,
R.G. Maki, J.B. Bouska, P.R. Young, C. Lanni, R.L. Bell, G.W. Carter,
C.D.W. Brooks, J. Med. Chem. 40 (1997) 1955e1968.
[11] N. Opacic, M. Barbaric, B. Zorc, M. Cetina, A. Nagl, D. Frkovic,
M. Kralj, K. Pavelic, J. Balzarini, G. Andrei, R. Snoeck, E. De Clercq,
S. Raic-Malic, M. Mintas, J. Med. Chem. 48 (2005) 475e482.
[12] J.J. Higgin, G.I. Yakovlev, V.A. Mitchevich, A.A. Makarov, R.T. Raines,
Bioorg. Med. Chem. Lett. 13 (2003) 409e412.
[22] H.J. Lee, J.W. Song, Y.S. Choi, H.M. Park, K.B. Lee, J. Am. Chem. Soc.
124 (2002) 11881e11893 and literature therein.
[23] B.M. Schiavi, D.J. Richard, M.M. Joullie, J. Org. Chem. 67 (2002)
620e624.
[24] P. Beier, J. Mindl, V. Sterba, J. Hanusek, Org. Biomol. Chem. 2 (2004)
562e569.
[25] A. Defoin, A. Brouillard-Poichet, J. Streith, Helv. Chim. Acta 74 (1991)
103e109.
[26] H.A. Staab, W. Benz, Angew. Chem. 73 (1961) 657.
[27] J.P. Greenstein, M. Winitz, Chemistry of the Aminoacids, vol. 2, Wiley
and Sons Inc., New York, 1961, pp 1566e1568.
[28] P. Rydberg, B. Luning, C.A. Wachtmeister, L. Eriksson, M. Tornqvist,
Chem. Res. Toxicol. 15 (2002) 570e581.
[13] S.J. Chung, D.H. Kim, Bioorg. Med. Chem. 9 (2001) 185e189.
[14] (a) A. Thorarensen, M.R. Douglas Jr., D.C. Rohrer, A.F. Vosters,
A.W. Yem, V.D. Marshall, J.C. Lynn, M.J. Bohanon, P.K. Tomich,
G.E. Zurenko, M.T. Sweeney, R.M. Jensen, J.W. Nielsen, E.P. Seest,
L.A. Dolak, Bioorg. Med. Chem. Lett. 11 (2001) 1355e1358;
(b) H.K. Smith, R.P. Beckett, J.M. Clements, S. Doel, S.P. East,
S.B. Launchbury, L.M. Pratt, Z.M. Spavold, W. Thomas, R.S. Todd,
M. Whittaker, Bioorg. Med. Chem. Lett. 12 (2002) 3595e3599.
[15] C. Temperini, A. Innocenti, A.T. Scozzafava, C. Supuran, Bioorg. Med.
Chem. Lett. 16 (2006) 4316e4320.
[29] (a) G.E. Terp, G. Cruciani, I.T. Christensen, F.S. Jørgensen, J. Med.
Chem. 45 (2002) 2675e2684;
(b) K. Maskos, Biochimie 87 (2005) 249e263;
(c) P. Cuniasse, L. Devel, A. Makaritis, F. Beau, D. Georgiadis,
M. Matziari, A. Yiotakis, V. Dive, Biochimie 87 (2005) 393e402.
[30] (a) G. Murphy, T. Crabbe, Methods Enzymol. 248 (1995) 470e484;
(b) T. Tuccinardi, A. Martinelli, E. Nuti, P. Carelli, F. Balzano,
G. Uccello-Barretta, G. Murphy, A. Rossello, Bioorg. Med. Chem. 14
(2006) 4260e4276.
[16] S. Hanessian, V. Vinci, L. Auzzas, M. Marzi, G. Giannini, Bioorg. Med.
Chem. Lett. 16 (2006) 4784e4787.
[31] (a) SoftMax Pro 4.7.1 by Molecular Device;
(b) GraFit version 4 by Erithecus Software.