Carbamoylphosphonate MMP inhibitors. Part 4: The influence of chirality and geometrical isomerism on the potency and selectivity of inhibition

Article abstract of DOI:10.1016/j.tetasy.2004.07.002

Matrix metalloproteinases (MMPs) are a family of over twenty zinc-dependent enzymes that hydrolyze connective tissue and are involved in a variety of diseases, which are associated with undesired tissue breakdown. Previously we described the synthesis of

Full text of DOI:10.1016/j.tetasy.2004.07.002

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 2415–2420
Carbamoylphosphonate MMP inhibitors. Part 4: The influence
of chirality and geometrical isomerism on the potency and
selectivity of inhibition
a
b
b,
*
Eli Breuer,a, Yiffat Katz, Rivka Hadar and Reuven Reich
*
aThe Department of Medicinal Chemistry, The School of Pharmacy, The Hebrew University of Jerusalem,
PO Box 12065, Jerusalem, 91120 Israel
bThe Department of Pharmacology, The School of Pharmacy, The Hebrew University of Jerusalem,
PO Box 12065, Jerusalem, 91120 Israel
Received 24 May 2004; accepted 2 July 2004
Abstract—Matrix metalloproteinases (MMPs) are a family of over twenty zinc-dependent enzymes that hydrolyze connective tissue
and are involved in a variety of diseases, which are associated with undesired tissue breakdown. Previously we described the syn-
thesis of a series of achiral alkyl and cycloalkylcarbamoylphosphonic acids and their biological evaluation. Herein we report the
effect of chirality and geometrical isomerism on the potency and selectivity of inhibition. The inhibitory potencies of pairs of enan-
tiomeric and stereoisomeric alkyl and cycloalkylcarbamoylphosphonic acids were evaluated on recombinant MMP-1, MMP-2,
MMP-3, MMP-8, and MMP-9 enzymes. The results show that the enantiomers and stereoisomers studied differ considerably in their
inhibitory potencies and selectivities on the enzyme subtypes studied. Such a result is consistent with the assumption that the car-
bamoylphosphonates interact with a chiral environment such as an enzyme.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
complexes of considerable stability, and thus confirm
that they are capable of serving as zinc-selective binding
groups.8 Following our previous results, we posed the
question whether there is any stereoselectivity in the
interaction between the carbamoylphosphonate inhibi-
tors and the MMP enzymes. An affirmative answer to
this question would confirm that the inhibitors indeed
interact with a chiral environment, such as an enzyme,
and not by another kind of mechanism, for example,
by simple chelation of a cation crucial to the enzymatic
catalysis. In our quest for an answer to this question, we
synthesized pairs of enantiomeric and stereoisomeric
alkyl and cycloalkylcarbamoylphosphonic acids, and
evaluated their inhibitory potencies on five recombinant
MMP enzyme subtypes.
Matrix metalloproteinases (MMPs), are a family of
structurally related zinc enzymes that mediate the break-
down of connective tissue and are therefore targets for
therapeutic inhibitors in inflammatory, malignant, and
degenerative diseases associated with excessive enzymatic
activity.1 Such conditions include cancer,2 tumor metas-
tasis,3 arthritis,4 and cardiovascular5 diseases, as well as
wound healing.6 The Zn2+ ion present at the active sites
of these enzymes is crucial for enzymatic activity, there-
fore virtually all attempts to develop inhibitors have in-
volved so-called zinc-binding groups (ZBG). We
reported recently that alkyl and cycloalkylcarbamoyl-
phosphonates can act as MMP inhibitors, and that sev-
eral of them show considerable potency and selectivity
toward MMP-2, in vitro, as well as considerable activity
in a cancer metastasis model in vivo.7 We have also re-
cently reported that carbamoylphosphonates form zinc
Herein we report the synthesis of two pairs of enantio-
meric carbamoylphosphonates, as well as two pairs of
geometrical isomers, and the results of studies of their
effects on five different MMP subtypes. Our results show
that the enantiomers and geometric isomers studied dif-
fer considerably in their inhibitory potencies and selec-
tivities on the enzyme subtypes studied.
*
Corresponding authors. Tel.: +972-2-6758704; fax: +972-2-
6758934; e-mail: breuer@cc.huji.ac.il
0957-4166/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.07.002

2416
E. Breuer et al. / Tetrahedron: Asymmetry 15 (2004) 2415–2420
2. Results and discussion
tures and purity. The homochiral compounds were also
characterized by their optical rotation.
The N-alkylcarbamoylphosphonates were synthesized
using the approach previously described, based on the
reaction of triethyl phosphonothiolformate9 1 with com-
mercially available amines, followed by bromotrimeth-
ylsilane mediated dealkylation to the corresponding N-
(cyclo)alkylcarbamoylphosphonic acids 3 (Scheme 1).7
The final products were isolated as crystalline free acids.
The synthesis of N-(S)-(+)-2-butylcarbamoylphosphonic
acid is given as a representative procedure. All other
products were prepared using the same method.
Inhibition constants (IC50 values) of all compounds
were determined on five recombinant MMP subtypes,
namely MMP-1, MMP-2, MMP-3, MMP-8, and
MMP-9.11 The IC50 values obtained from the new com-
pounds, and of some previously reported ones for com-
parison, are displayed in Table 1.
It is noteworthy that among the MMPs, gelatinases
(MMP-2 and MMP-9) are especially important in con-
nection with the processes of tumor growth, invasion,
and metastasis.12 Thus, gelatinase inhibitors have been
studied extensively as new types of anticancer drugs.
Furthermore, special attention has been paid to the dis-
covery of MMP-2 (and/or MMP-9) inhibitors. Achiev-
ing highly selective inhibition of MMP-2, the enzyme
involved and mainly responsible for the breaching of
basement membranes, would especially be advanta-
geous, as selectivity might help to evade the undesirable
side effects as a result of nonspecific inhibition of other
MMPs. Therefore, the selectivity of an inhibitor is an
asset.
O
C
O
OCH2CH3
OCH2CH3
OCH2CH3
OCH2CH3
RNH2
C
R
P
EtS
P
NH
2
O
1
O
1. TMSBr
2. MeOH
O
OH
OH
C
P
R
NH
3
Herein, we examine two pairs of enantiomeric car-
bamoylphosphonates, one derived from optically active,
commercially available (R)- and (S)-2-butylamines and
the other from (R)- and (S)-1-cyclohexylethylamines.
O
Scheme 1.
The only exception was the case of 4-methylcyclohexyl-
amine, which was available commercially as an (E) and
(Z) mixture and was reacted with 1 as such. Separation
of the reaction mixture yielded the pure (E) and (Z)
products, 3i and 3j, as described in Experimental sec-
tion. The structure of the higher melting product was
determined by X-ray crystallography and found to be
the trans-isomer, namely, N-(E)-(4-methylcyclo-
hexyl)carbamoylphosphonic acid, 3i.10
Examination of the results in Table 1 revealed that the
two optically active 2-butylcarbamoylphosphonic acids,
3b and 3c, differ greatly in their inhibitory potencies on
the three enzymes, MMP-1, -2, and -8, and that they
also differ from those of the racemate 3a, in such a
way that in each case, one enantiomer is more potent
from the racemate while the other is less (inactive). Nei-
ther 3b nor 3c have any significant activity on MMP-3
and -9. The (S)-isomer, 3b, was a selective and potent
inhibitor of MMP-2, while the (R)-isomer, 3c, was a
selective and potent inhibitor of MMP-8. The latter
(also known as neutrophile collagenase) is associated
All carbamoylphosphonic acids, 3, were characterized
1
by elemental analysis and by their H and 31P NMR
spectra, the results of which fully confirmed their struc-
Table 1. Inhibition constants of carbamoylphosphonates on different MMPs
Symbol
Structure of R
in R–NHCOPO3H2
MMP-1
IC50 (lM)a
MMP-2
IC50 (lM)a
MMP-3
IC50 (lM)a
MMP-8
IC50 (lM)a
MMP-9
IC50 (lM)a
3ab
3b
(RS)-2-Butyl–
(S)-2-Butyl–
40
25
30
80
5
20
>100
>100
>100
2
0.8
>100
5
100
100
>100
>100
0.1
3c
(R)-2-Butyl–
2-Propyl–
3-Pentyl–
80
50
0.4
50
3db
3eb
3f
80
2
30
1
10
>100
10
(S)-1-Cyclohexylethyl–
(R)-1-Cyclohexylethyl–
Cyclohexylmethyl
(E)-4-Me-Cyclohexyl–
(Z)-4-Me-Cyclohexyl–
Cyclohexyl–
>100
>100
80
3g
3hb
>100
30
0.2
0.2
4
>100
1
>100
1
3i
>100
100
0.1
1
>100
50
1
20
>100
20
3j
3kb
20
3
15
3
>100
>100
>100
>100
>100
>100
>100
>100
3l
exo-2-Norbornyl–
endo-2-Norbornyl–
Cyclopentyl–
>100
>100
>100
3m
3nb
>100
0.5
0.1
0.080
a Errors for these measurements are 5% of the mean values.
b Reported in Ref. 7.

E. Breuer et al. / Tetrahedron: Asymmetry 15 (2004) 2415–2420
2417
with inflammatory conditions. Furthermore, the latter
two compounds were distinctly superior to the two
closely analogous and achiral 2-propyl and 3-pent-
ylcarbamoylphosphonic acids, 3d and 3e, respectively.
The structures of these compounds and their inhibition
constants on MMP-2 and MMP-8 are shown in
Figure 1.
itor of MMP-3 (an enzyme associated with mammary
development and tumor progression), the (R)-enantio-
mer 3g inhibited MMP-2 with high selectivity. Interest-
ingly, the inhibition constant (IC50) of 3g, on MMP-2 is
identical to that of the achiral analogue 3h, which is
lacking the methyl group. However homochiral 3g was
much better in its selectivity for MMP-2 relative to the
other MMP subtypes examined. The structures of 3f,
3g, and 3h, and their inhibition constants on MMP-2
and MMP-3 are shown in Figure 2.
N-(RS)-2-butylcarbamoylphosphonic acid, 3a
IC50 = 30 µM on MMP-2, 5 µM on MMP-8
O
O
P
In addition to the two pairs of enantiomers, we have
also examined the effect of geometrical isomerism on
the inhibitory characteristics of carbamoylphospho-
nates. Examination of the (E)- and (Z)-4-methyl-
H
H
Et
Me
Me
Et
HO
HO
P
N
HO
HO
N
O
H
O
H
cyclohexylcarbamoylphosphonates
(3i
and
3j,
3b
3c
respectively), revealed that they differ in their inhibitory
profiles. Both 4-methylcyclohexyl derivatives (Fig. 3) are
less active than the parent cyclohexyl compound, 3k, on
all enzymes, except on MMP-8, which is inhibited only
by 3i.
N-(S)-2-butylcarbamoylphosphonic acid
IC50 = 0.8 µM on MMP-2
N-(R)-2-butylcarbamoylphosphonic acid
IC50 >100 µM on MMP-2
IC50 >100 µM on MMP-8
IC50 = 0.4 µM on MMP-8
O
O
Finally, examination of the two isomeric bicyclic deriv-
atives the exo- and endo-2-norbornylcarbamoylphos-
phonates, 3l and 3m, respectively, revealed that,
although the exo-derivative, 3l, inhibits MMP-1 (the
classical collagenase thought to be involved in chronic
wound formation and healing as well as in arthritis) with
an IC50 of 1lM, the endo-compound, 3m, is somewhat
more interesting, as it is 150 times more potent on
MMP-2 than the exo-phosphonate 3l, coupled with
impressive selectivity. As the norbornane system can
be viewed both as a substituted cyclopentane or cyclo-
hexane derivative, the comparison with these two can
tell us the effect of substitution in cyclopentane (or
cyclohexane) on the potency and selectivity. From this
Me
H
H
Et
HO
HO
P
N
HO
HO
P
N
Et
Me
O
H
O
H
3e
3d
N-2-propylcarbamoylphosphonic acid
IC50 = 5 µM on MMP-2
N-3-pentylcarbamoylphosphonic acid
IC50 = 30 µM on MMP-2
IC50 = 50 µM on MMP-8
IC50 = 10 µM on MMP-8
Figure 1.
A somewhat similar situation was seen in case of the
enantiomeric 1-cyclohexylethylcarbamoylphosphonates
3f and 3g. While the (S)-enantiomer 3f is a potent inhib-
O
O
P
O
P
H
H
N
Me
H
N
HO
HO
P
N
HO
HO
HO
HO
H
Me
O
H
O
H
O
H
3g
3f
3h
N-(S)-1-cyclohexylethylcarbamoyl-
phosphonic acid
N-(R)-1-cyclohexylethylcarbamoyl-
phosphonic acid
N-cyclohexylmethylcarbamoyl
phosphonic acid
IC50 = 1 µM on MMP-2
IC50 = 0.1 µM on MMP-3
IC50 = 0.2 µM on MMP-2
IC50 >100 µM on MMP-3
IC50 = 0.2 µM on MMP-2
IC50 = 1 µM on MMP-3
Figure 2.
O
O
CH3
HO
HO
HO
HO
C
C
P
P
NH
NH
CH3
O
O
3j
3i
(Z)-methylcyclohexylcarbamoylphosphonic
acid
(E)-methylcyclohexylcarbamoylphosphonic
acid
IC50 = 4 µM on MMP-2
IC50 = 1 µM on MMP-8
IC50 = 20 µM on MMP-2
IC50 = 20 µM on MMP-8
Figure 3.

2418
E. Breuer et al. / Tetrahedron: Asymmetry 15 (2004) 2415–2420
O
O
HO
HO
C
HO
HO
P
NH
C
P
NH
3m
O
3l
O
endo-2-Norbornylcarbamoylphosphonic
acid
exo-2-Norbornylcarbamoylphosphonic
acid
IC50 = 0.1 µM on MMP-2
IC50 > 100 µM on MMP-1
IC50 = 15 µM on MMP-2
IC50
= 1 µM on MMP-1
Figure 4.
comparison we see that the potencies of bicyclic 3m and
monocyclic 3n on MMP-2 are comparable, but the bicy-
clic 3m is more selective than either of the monocyclic
derivatives, 3n or 3k. Finally, it is important to empha-
size that as both norbornane derivatives are racemic
mixtures, it is quite likely that resolution of 3m to the
two enantiomers will yield one less potent and one more
potent and probably highly selective MMP-2 inhibitor.
The structures of 3l and 3m and their inhibition con-
stants on MMP-1 and MMP-2 are shown in Figure 4.
4.2. N-(S)-(+)-2-Butylcarbamoylphosphonic acid, 3b
(representative procedure)
A solution of diethyl N-[(S)-2-butyl]carbamoylphospho-
nate, 2b, obtained in the previous step (1.013g,
4.27mmol) and bromotrimethylsilane (TMSBr, 2.76
mL, 21.3mmol) in MeCN (10mL) was kept at room
temperature overnight. The volatiles were evaporated
and the residue taken up in MeOH (10mL), which
was evaporated to leave behind a white solid, which
was recrystallized from EtOH to yield 0.48g (62%)
crystals mp 149–151°C, ½aŠ ¼ þ13:5 (c 0.0185,
D
3. Conclusion
MeOH). NMR (D2O) 31P ꢀ2.66ppm, 1H, 0.66 (t,
J = 7.2Hz, 3H), 0.95 (d, J = 6.6Hz, 3H), 1.30 (m,
2H), 3.55 (m, 1H). Anal. Calcd for C5H12NO4P:
C, 33.15; H, 6.63; N, 7.73. Found: C, 33.21; H, 6.71;
N, 7.38.
Herein we have examined the effect of chirality and geo-
metrical isomerism in some alkyl and cyclo-
alkylcarbamoylphosphonic acids on the potency and
selectivity of MMP inhibition. The inhibitory potencies
of pairs of enantiomeric and diastereoisomeric alkyl
and cycloalkylcarbamoylphosphonic acids were evalu-
ated on recombinant MMP-1, MMP-2 MMP-3,
MMP-8, and MMP-9 enzymes. The results show that
pairs of enantiomers and diastereoisomers had in all
cases different inhibitory potencies and different
selectivity profiles on the enzymes studied. Such results
are consistent with the assumption that the carbamoyl-
phosphonates indeed interact with a chiral environment
such as an enzyme. We are currently continuing our
efforts to elucidate the mode of interaction of car-
bamoylphosphonates with MMPs.
4.3. Diethyl N-[(R)-2-butyl]carbamoylphosphonate 2c
1
Oil (0.74g, 91%). NMR (CDCl3) 31P ꢀ0.90ppm, H,
0.88 (t, J = 7.5Hz, 3H), 1.15 (d, J = 6.9Hz, 3H), 1.34
(t, J = 7.2Hz, 6H), 1.49 (quin, 2H, J = 7.2Hz), 3.98
(m, 1H), 4.18 (m, 4H), 6.88 (br s, 1H).
4.4. N-[(R)-(ꢀ)-2-Butyl]carbamoylphosphonic acid 3c
White solid 0.5g (99%) mp 142–145°C, ½aŠ ¼ ꢀ12:6 (c
D
1
0.0193, MeOH). NMR (D2O) 31P ꢀ2.66ppm, H, 0.68
(t, J = 7.2Hz, 3H), 0.97 (d, J = 6.6Hz, 3H), 1.30 (m,
2H), 3.55 (m, 1H). Anal. Calcd for: C5H12NO4P: C,
33.15; H, 6.63; N, 7.73. Found: C, 30.61; H, 6.42; N,
6.77.
4. Experimental section
4.1. Diethyl N-[(S)-2-butyl]carbamoylphosphonate, 2b
(representative procedure)
4.5. Diethyl N-[(S)-(+)-1-cyclohexylethyl]carbamoylphos-
phonate 2f
1
A solution of triethyl phosphonothiolformate 1 (1.74g,
7.7mmol) and (S)-(+)-2-butylamine (0.77mL, 7.61
mmol) in MeCN (15mL) was kept at room temperature
for 3days. The volatiles were removed and the residue
dried in high vacuum to yield 1.67g colorless oil
(92%), which was purified by column chromatography
(AcOEt/PE, 8:2) to give 1.48g, 82% of oil. NMR
(CDCl3) 31P ꢀ0.88ppm, 1H, 0.88 (t, J = 7.5Hz, 3H),
1.13 (d, J = 6.9Hz, 3H), 1.32 (t, J = 7.2Hz, 6H), 1.47
(quin, 2H, J = 7.2Hz), 3.97 (m, 1H), 4.18 (m, 4H),
6.90 (br s 1H). Anal. Calcd for C9H20NO4P: C, 45.49;
H, 8.42; N, 5.89. Found: C, 45.23; H, 8.45; N, 5.75.
Colorless oil (76%). NMR (CDCl3) 31P ꢀ0.78ppm, H,
0.80–1.80 (mÕs, 11H), 1.10 (t, J = 6.9Hz, 3H), 1.34 (t,
J = 7.2Hz, 6H), 3.90 (m, 1H), 4.20 (m, 4H), 6.86 (br,
1H).
4.6. N-(S)-(ꢀ)-(1-Cyclohexylethyl)carbamoylphosphonic
acid 3f
White solid (99%) mp 155–156°C, ½aŠ ¼ ꢀ20:2 (c 0.07,
D
MeOH). NMR (D2O) 31P ꢀ2.62ppm, 1H, 0.60–1.56
(mÕs, 11H), 0.90 (t, J = 6.9Hz, 3H), 3.52 (quin,
J = 6.9Hz, 1H). Anal. Calcd for C9H18NO4P: C,

E. Breuer et al. / Tetrahedron: Asymmetry 15 (2004) 2415–2420
2419
45.96; H, 7.66; N, 5.96. Found: C, 46.09; H, 7.66; N,
5.79.
C8H16NO4P: C, 43.44; H, 7.24; N, 6.33. Found: C,
43.04; H, 7.38; N, 5.97.
4.7. Diethyl N-[(R)-1-cyclohexylethyl]carbamoylphos-
phonate 2g
4.12. N-(Z)-4-(Methylcyclohexyl)carbamoylphosphonic
acid 3j
1
1
Oil, NMR (CDCl3) 31P ꢀ0.78ppm, H, 0.80–1.80 (mÕs,
Solid, mp 149–151°C. NMR (D2O) 31P ꢀ2.63ppm, H,
11H), 1.08 (d, J = 6.9Hz, 3H), 1.32 (t, J = 7.2Hz, 6H),
3.90 (m, 1H), 4.17 (m, 4H), 6.90 (br d).
0.73 (d, J = 6Hz, 3H), 0.98–1.15 (m, 2H), 1.3–1.58 (m,
7H), 3.74 (broad, 1H). Anal. Calcd for C8H16NO4P:
C, 43.44; H, 7.24; N, 6.33. Found: C, 43.72; H, 7.52;
N, 6.03.
4.8. N-[(R)-1-Cyclohexylethyl]carbamoylphosphonic acid
3g
4.13. Diethyl [exo-2-norbornyl]carbamoylphosphonate 2l
Crystals from EtOH, mp 154–155°C, ½aŠ ¼ þ21:0 (c
D
1
1
0.07, MeOH). NMR (D2O) 31P ꢀ2.60ppm, H, 0.60–
1.56 (mÕs, 11H), 0.94 (d, J = 6.9Hz, 3H), 3.56 (quin,
J = 6.9Hz, 1H). Anal. Calcd for C9H18NO4P: C,
45.96; H, 7.66; N, 5.96. Found: C, 45.69; H, 7.55; N,
5.66.
Oil, NMR (CDCl3) 31P ꢀ1.09ppm, H, 1.00–1.50 (mÕs),
1.75 (m, 1H), 2.18 (d, J = 3.3Hz, 1H), 2.24 (s, 1H), 3.74
(m, 1H), 4.10–4.25 (m, 4H), 6.97 (br d (1H).
4.14. N-[exo-2-Norbornyl]carbamoylphosphonic acid 3l
Crystals from EtOH (1g, 86%) mp 170–172°C. NMR
4.9. Diethyl N-(E)-(4-methylcyclohexyl)carbamoylphos-
phonate 2i
(D2O) 31P ꢀ2.79ppm, H, 0.85–1.05 (mÕs, 3H), 1.05–
1
1.47 (m, 4H), 1.50–1.60 (m, 1H), 1.98 (d, J = 3Hz,
1H), 2.07 (s, 1H), 3.41 (br d, 1H). Anal. Calcd for
C8H14NO4P: C, 43.83; H, 6.39; N, 6.39. Found: C,
43.87; H, 6.67; N, 6.22.
A
solution of triethyl phosphonothiolformate
1
(2.12g, 9.38mmol) and 4-methylcyclohexylamine (a
mixture of cis and trans, 1.27mL, 9.57mmol) in MeCN
(15mL) was kept at room temperature overnight.
Examination of the reaction mixture by 31P NMR
spectroscopy showed two peaks at ꢀ0.82 (trans) and
ꢀ0.89 (cis) ppm. The volatiles were removed and the
residue dried in high vacuum to yield a semi-solid resi-
due, which was dissolved in hexane to yield white crys-
talline needles, 0.625g (24%), mp 77–79°C. NMR
4.15. Diethyl N-[endo-2-norbornyl]carbamoylphosphonate
2m
Solid, mp 48–50°C. NMR (CDCl3) 31P ꢀ1.13ppm, H,
1
0.82 (dt, J = 13.5, 4.5Hz, 1H), 1.10–1.65 (m, 12H), 2.06
(tt, J = 13.5, 4.5Hz, 1H), 2.22 (t, 1H), 2.45 (s, 1H), 4.10–
4.25 (m, 5H), 7.05 (br d (1H, NH). Anal. Calcd for
C12H22NO4P: C, 52.28; H, 7.98; N, 5.08. Found: C,
52.01; H, 7.85; N, 5.32.
1
(CDCl3) 31P ꢀ0.82ppm, H, 0.89 (d, J = 6.5Hz, 3H),
1.04 (dq, J = 11.9, 3Hz, 2H), 1.12 (dq, J = 11.9,
3.5Hz, 2H), 1.37 (dt, J = 6.5, 0.55Hz, 6H), 1.73 (m,
3H), 1.95 (m, 2H), 3.80 (m, 1H), 4.29–4.15 (m, 4H),
6.85 (br d, 1H). Anal. Calcd for: C12H24NO4P: C,
51.98; H, 8.66; N, 5.05. Found: C, 51.76; H, 8.53; N,
5.35.
4.16. N-[endo-2-Norbornyl]carbamoylphosphonic acid 3m
A solid from EtOH (0.727g, 73%) mp 164–165°C.
NMR (D2O) 31P ꢀ2.79ppm, 1H, 0.76 (dt, J = 13.2,
3.3Hz, 1H), 1.0–1.4 (mÕs, 6H), 1.81 (m, 1H), 2.02 (s,
1H), 2.21 (s, 1H), 3.80 (br d, 1H). Anal. Calcd for
C8H14NO4P: C, 43.83; H, 6.39; N, 6.39. Found: C,
43.79; H, 6.18; N, 6.19.
4.10. Diethyl N-(Z)-(4-methylcyclohexyl)carbamoylphos-
phonate 2j
The hexane solution obtained after separation of the
trans-isomer, described in the previous experiment,
was evaporated to a semi-solid mass, which was shown
by 1H NMR to contain a minor proportion of the
trans-ester and a different product. The mixture was sep-
arated on silica column (ether/petroleum ether) to give
the cis-ester, the less polar component, followed by frac-
tions containing a cis–trans-mixture and finally, the pure
trans-product. The cis-ester was an oil, NMR (CDCl3)
31P ꢀ0.89ppm.
4.17. Analysis of MMP activity: use of recombinant
MMPs and relevant substrates
Commercial recombinants MMP-1, MMP-2, MMP-3,
MMP-8, and MMP-9 (R&D Systems, Minneapolis,
MN) were incubated at four different concentrations
(1–50ng) with their respective substrates for 3h. Fluoro-
genic Peptide Substrate I (ES001, R&D Systems) was
used for MMP1 and MMP-8. Fluorogenic Peptide Sub-
strate II ES002, R&D Systems) was used for MMP-3.
EnzChek gelatinase Assay Kit was used as the substrate
for MMP-2 and MMP-9. The examined compounds
were added at four to six different concentrations (0.1–
100lM) to the recombinant enzymes and the inhibitory
potencies expressed in a colorimetric change, were meas-
ured by an ELISA reader. The inhibitory potency (IC50)
was calculated from the kinetic data obtained.
4.11. N-(E)-(4-Methylcyclohexyl)carbamoylphosphonic
acid 3i
Compound 3i was made solid from EtOH by allowing
it to evaporate slowly, mp 169–171°C. NMR (D2O)
1
31P ꢀ2.61ppm, H, 0.60 (d, J = 5.7Hz, 3H), 0.77 (m,
2H), 1.05 (m, 3H), 1.46 (d, J = 12.9Hz), 1.57 (d,
J = 11.4Hz, 2H), 13.36 (m, 1H). Anal. Calcd for

2420
E. Breuer et al. / Tetrahedron: Asymmetry 15 (2004) 2415–2420
3. Kleiner, D. E.; Stetler-Stevenson, W. G. Cancer Chemo-
Acknowledgements
ther. Pharmacol. 1999, 43(Suppl.), S42–51.
4. Billinghurst, R. C.; Dahlberg, L.; Ionescu, M.; Reiner, A.;
Tanzer, M.; Zukor, D.; Chen, J.; Van Wart, H.; Poole, A.
R. Arthritis Rheum. 2000, 43, 673.
5. Ikeda, U.; Shimada, K. Clin. Cardiol. 2003, 26, 55–
59.
6. Pilcher, B. K.; Wang, M.; Qin, X.-J.; Parks, W. C.; Senior,
R. M.; Welgus, H. G. Ann. N.Y. Acad. Sci. 1999, 878,
12–24.
This work was supported in part by the Ministry of Sci-
ence of Israel and by the Israel Science Foundation. E.B.
and R.R. are affiliated with the Bloom center for Phar-
macy, and with the Alex Grass Center for Drug Design
and Delivery in the School of Pharmacy at The Hebrew
University of Jerusalem.
7. Breuer, E.; Salomon, C. J.; Katz, Y.; Chen, W.; Lu, S.;
Ro¨schenthaler, G.-V.; Hadar, R.; Reich, R. J. Med. Chem.
2004, 47, 2826–2832.
References
1. For recent reviews, see: Whittaker, M.; Floyd, C. D.;
Brown, P.; Gearing, A. J. H. Chem. Rev. 1999, 99,
2735–2776; Wojtowicz-Praga, S. Drugs R&D 1999, 1,
117–129; De, B.; Natchus, M. G.; Cheng, M.; Pikul, S.;
Almstead, N. G.; Taiwo, Y. O.; Snider, C. E.; Chen, L.;
Barnett, B.; Gu, F.; Dowty, M. Ann. N.Y. Acad. Sci. 1999,
878, 40–60; Johnson, L. L.; Dyer, R.; Hupe, D. J. Curr.
Opin. Chem. Biol. 1998, 2, 466.
¨
8. Farkas, E.; Katz, Y.; Bhusare, S.; Reich, R.; Roschent-
haler, G.-V.; Konigsmann, M.; Breuer, E. J. Biol. Inorg.
¨
Chem. 2004, 9, 307–315.
9. Salomon, C. J.; Breuer, E. Synlett 2000, 815–816.
10. Cohen, S.; Katz, Y.; Reich, R.; Breuer, E. In preparation.
11. Baragi, V. M.; Shaw, B. J.; Renkiewicz, R. R.; Kuipers, P.
J.; Welgus, H. G.; Mathrubutham, M.; Cohen, J. R.; Rao,
S. K. Matrix Biol. 2000, 19, 267–273.
2. Hidalgo, M.; Eckhardt, S. G. J. Nat. Cancer Inst. 2001, 93,
178–193.
12. Overall, C.; Lopez-Otin, C. Nature Rev. Cancer 2002, 9,
657–672.