Design, modelling, synthesis and biological evaluation of peptidomimetic phosphinates as inhibitors of matrix metalloproteinases MMP-2 and MMP-8

Article abstract of DOI:10.1016/j.bmc.2005.04.079

Three novel peptidomimetic phosphinate inhibitors have been synthesized and evaluated as inhibitors of matrix metalloproteinases MMP-2 and MMP-8. Their IC₅₀ values are in the micromolar range, and one of them showed to be the most effective inh

Full text of DOI:10.1016/j.bmc.2005.04.079

Bioorganic & Medicinal Chemistry 13 (2005) 4740–4749
Design, modelling, synthesis and biological evaluation of
peptidomimetic phosphinates as inhibitors of matrix
metalloproteinases MMP-2 and MMP-8
Gianluca Bianchini,a Massimiliano Aschi,a Giancarlo Cavicchio,a Marcello Crucianelli,a,*
Serena Preziuso,b Carlo Gallina,b Adele Nastari,c Enrico Gavuzzod and
Fernando Mazzaa,*
a
`
Dipartimento di Chimica, Ingegneria Chimica e Materiali, Universita dell’Aquila, via Vetoio, I-67010-Coppito (AQ), Italy
b
`
Dipartimento di Scienze del Farmaco, Universita G. d’Annunzio, Via dei Vestini 31, I-66013 Chieti, Italy
cPolifarma Research Center, Via Tor Sapienza 138, I-00185 Roma, Italy
dIstituto di Chimica Biomolecolare Sez. Roma, P.le Aldo Moro 5, I-00184 Roma, Italy
Received 4 January 2005; accepted 28 April 2005
Available online 1 June 2005
Abstract—Three novel peptidomimetic phosphinate inhibitors have been synthesized and evaluated as inhibitors of matrix metal-
loproteinases MMP-2 and MMP-8. Their IC50 values are in the micromolar range, and one of them showed to be the most effective
inhibitor of MMP-2. The differences in binding affinities for MMP-2 and MMP-8 of the three phosphinates have been rationalized
by means of modelling studies and molecular dynamics simulations.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
the enzyme active site and a zinc-binding function (ZBF)
capable of coordinating the catalytic zinc ion. The
hydroxamate group is considered the most powerful
ZBF for MMPIs,14 since its simple replacement by other
zinc-binding groups causes a 100- to 2000-fold decrease
in potency.15 Hydroxamate inhibitors, however, are gen-
erally affected by lack of specificity,16 owing to the over-
whelming chelating power of the hydroxamate, which
often prevails over the contribution of the peptidomi-
metic moiety. They show, in addition, poor pharmacoki-
netic properties17 and can entail toxicity in long-term
therapy, due to the release of hydroxylamine, a known
carcinogenic agent.18 Novel MMPIs containing different
ZBF such as carboxylate, thiolate, phosphonate and
phosphinate are therefore currently investigated.
Matrix metalloproteinases (MMPs) form a growing fam-
ily of zinc endopeptidases that play important roles in
physiological processes such as ovulation, embryogenic
growth, angiogenesis, and differentiation.1–4 Imbalances
in the expression of MMPs and their endogenous specific
inhibitors (TIMPs)5 are implicated in a number of patho-
logical states such as tumour growth and metastasis,6,7
multiple sclerosis,8 rheumatoid arthritis,9 Alzheimerꢀs
10a
disease
and heart failure.10b Although TIMPs have
been successfully tested in animal experiments,11 they
could not be developed as therapeutic agents, owing
to inadequate pharmacological stability.12 Therefore,
design and synthesis of potent, small molecule MMP
inhibitors (MMPIs) endowed with adequate pharmaco-
logical and selectivity profile are actively investigated.13
2. Inhibitor design
General requirements for MMPIs are a peptidomimetic
backbone mimicking the substrate moiety recognized in
We have previously studied some tripeptide phospho-
nate MMPIs structurally related to endogenous
inhibitors isolated from snake venom.19 Their mode of
binding in the active site of zinc endopeptidases has
been determined by X-ray structure resolution of their
complexes. While the phosphonate Pro-Leu-L-Trp(P)-
Keywords: Matrix metalloproteinases inhibitors; Phosphinic acids;
Peptidomimetics; Molecular dynamics and modelling.
*
Correspondingauthors. Tel.:+39(0)862433780;fax:+39(0)862433753
(M.C.); e-mail addresses: cruciane@univaq.it; mazza@univaq.it
0968-0896/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.04.079

G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4741
(OH)2 binds in the S region of the MMP-8 active site,
3. Chemistry
the analogue containing a pyroglutamate in place of
the Pro residue is accommodated in the S0 region of
the active site of adamalysin,19c a snake venom endopep-
tidase structurally related to human MMPs.
Despite the interesting biological properties of pepti-
dyl phosphinates,23 difficulties often encountered in
the phosphorous alkylation steps, during their synthe-
sis, have greatly limited the study of this kind of
inhibitors.
Superposition of these two crystal structures shows that
the [PO2ꢀ] anion, ligating the catalytic zinc ion, main-
tains exactly the same position relative to the metal, in
both the complexes accommodating the peptidomimetic
chains in opposite directions. This finding prompted us
to start the design of novel MMP-8 phosphinic inhibi-
tors from the coordinates of the [PO2ꢀ] chelating group.
Our efforts pointed to select peptidomimetic chains that,
extending from the phosphinate group, could establish
useful binding interactions in both the S and S0 region
of the enzyme active site. While occupation of the S0 re-
gion usually guarantees adequate potency of the inhibi-
tor, binding interactions in the S region could be chiefly
exploited to improve selectivity.20 Within the class of
phosphinic acid-based pseudopeptide inhibitors, two
main typologies of frameworks have been so far synthe-
tized (A and B, Fig. 1), following receptor-based drug
design strategies.20c,21
Synthesis of the peptidyl phosphinates 1a–1c required
preparation of the biphenylalkyl iodides 4a–4c (Scheme
1), powerful alkylating reagents necessary for an efficient
phosphorous alkylation. The sequential double alkyl-
ation of the phosphorus atom, to give the protected
intermediates 8a–8c containing the unsymmetrically
disubstituted phosphinate core, followed by coupling
with the Pro-Leu dipeptide chain and protecting groups
removal, is summarized in Scheme 2.
Alkylation of the phosphorus atom of ammonium
ipophosphite has been performed by a modification of
the Regan–Boyd24 procedure. In the first alkylation
step, pyrophoric bistrimethylsilylphosphonite (BTSP)25
(5) was generated in situ by treatment of ammonium
ipophosphite with hexamethyldisilazane (HMDS), and
reacted with the appropriate biphenylalkyl iodides 4a–
4c, to give the expected monoalkylphosphinic acids
6a–6c in 58–70% yields. These, in turn, were converted
into the bistrimethylsilyl phosphinates 7a–7c by
treatment with HMDS and reacted overnight in
CH2Cl2, under reflux, with commercially available
phthalimidomethyl bromide, to perform the second
alkylation step at the phosphorus atom. After conven-
tional workup, the expected dialkylphosphinic acids,
without further purification, were converted by treat-
ment with diazomethane into the corresponding methyl
esters 8a–8c that can be easily purified and completely
characterized.
Our first approach towards phosphinate-based inhibi-
tors pointed to the synthesis and evaluation of the three
analogues 1a–1c (Fig. 2). They are based on the Pro-
Leu-NH-CH2-PO2ꢀ sequence, in accordance with previ-
ous structural studies on both phosphonate19c and
hydroxamate22 inhibitors, complexed in the S region of
the MMP-8 active site. The second chain bound to the
central phosphorus atom was limited to a highly lipo-
philic biphenyl group, useful for increasing the affinity
of the ligands by insertion into the deep S01 hydrophobic
pocket of the enzyme. The biphenyl group was connect-
ed to phosphorus by means of a methylene group in the
phosphinates 1a and 1b. An ethylene bridge has been
inserted in 1c, to evaluate the effect of chain elongation,
reorientation of the rigid biphenyl group, in view of
higher flexibility, and deeper insertion into the S10
pocket.
Removal of the phthalimido protecting group was
accomplished by treatment of an ethanol solution of
methyl esters 8a–8c with hydrazine, at room tempera-
ture, for 18 h. Finally, the a-aminophosphinates 9a–9c
were coupled with Z-Pro-Leu-OH by reaction with
dicyclohexylcarbodiimide (DCC) in the presence of
1-hydroxybenzotriazole (HOBT)26 to afford the
expected peptidomimetics 10a–10c. Conversion into
the final phosphinates 1a–1c was carried out by treat-
ment with bistrimethylsilylacetamide (BTSA) and trim-
ethylsilyl iodide (TMSI), which caused contemporary
removal of the carbobenzyloxy group from the termi-
nal proline residue and of the methyl group from the
phosphinate ester. Phosphinates 1a–1c were isolated,
purified and fully characterized by 1H, 13C and 31P
NMR, ESI-MS, MALDI-TOF-MS and elemental
analyses.
Figure 1. Selected frameworks of phosphinic acid-based pseudopeptide
MMPIs. (A) Reiter, L. A., et al.20c; Hagmann, W. K., et al.21c. (B)
Goulet, J. L., et al.21d; Yiotakis, A., et al.21a; Schiodt, C. B., et al.21b
.
4. Results and discussion
4.1. In vitro enzyme inhibition assays
Figure 2. Peptidyl biphenylalkylphosphinates designed on the basis of
the mode of binding of phosphonate inhibitors complexed in both S
and S0 subsites of MMP-8.
The inhibiting activities of the new compounds were
evaluated against MMP-2 and MMP-8, measuring the

4742
G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
Scheme 1. Synthesis of the biphenylalkyl iodides 4a–4c. Reagents and conditions: (i) LiAlH4, DT, THF, 12 h, 87% yield for 3b; 5 h, 99% yield for 3c;
(ii) HI, CHCl3, rt, 4 h, 82% yield for 4a; HI, rt, acetone, 12 h, 99% yield for 4b; I2, PPh3, imidazole, CH2Cl2, rt, 3 h, 93% yield for 4c.
Scheme 2. Synthesis of peptidyl phosphinates 1a–1c. Reagents and conditions: (i) (Me3Si)2NH, 110 °C, 1–2 h; (ii) CH2Cl2 reflux, 12 h, 58–70% yield;
(iii) (Me3Si)2NH, CH2Cl2, 0 °C, 1 h; (iv) phthalimidomethyl bromide, CH2Cl2, reflux, 12 h, and phosphinate methylation after quenching: CH2N2,
0 °C, toluene/MeOH, 49–57% yield; (v) H2N–NH2, EtOH, rt, 18 h, 70–78% yield; (vi) Z-Pro-Leu-OH, DCC, HOBT, THF, 0 °C to rt, 15 h, 67–80%
yield; (vii) BTSA, CH2Cl2, rt, 1 h; TMSI, CH2Cl2, ꢀ20 °C to rt, 3 h, 72–83% yield.
residual enzyme activity by means of continuous fluori-
metric assays in the presence of the fluorescent substrate
QF 24.27 The resulting IC50 values (Table 1) are in the
micromolar range, and represent a validation of our de-
sign of phosphinate inhibitors extending in both the S
and S0 region of the MMP active site.
and MMP-8 can offer an interpretation at molecular
level of their behaviour. The two ligands 1a and 1b,
differing only in the terminal hydroxyl group, show
similar binding with the two enzymes. However, the
following different interactions emerge at the S3 and S10
subsites of the putative complexes formed with the two
enzymes. The aromatic side chains of Tyr 155, His 166
and Phe 168 surrounding the S3 subsite of MMP-2 form
a hydrophobic cavity where favourable contacts with
the Pro ring of each inhibitor can take place. At the S3
subsite of MMP-8, the shorter and hydrophilic side
chain of Ser 151, replacing the Tyr 155 present in
MMP-2, prevents similar interactions probably forming
H-bonds with water molecules. Moreover, the S01 pocket
of MMP-2 has the form of a tunnel open to the solvent,
whereas in MMP-8 it is bottom panelled by the side
chains of Arg 222 and Tyr 227. As a consequence, the
biphenyl groups of both ligands can be easily inserted
in the S01 pocket of MMP-2, whereas the terminal hydro-
xyl of 1b can give rise to steric hindrance with the side
chains of the residues paving the S01 pocket of MMP-8.
The better interactions of both ligands at S3 and S01 sub-
sites of MMP-2 can explain their better activities shown
on this enzyme. The steric hindrance occurring between
the terminal hydroxyl of 1b with the bottom of the S10
pocket in MMP-8 can give reason for the lower activity
of 1b compared to 1a. The similar activities shown by
the two ligands on MMP-2 can be interpreted by lack
of ligand–protein H-bond through the additional
hydroxyl group, probably solvated in the open tunnel
of MMP-2.
Both phosphinates 1a and 1b, characterized by a meth-
ylene bridge between the phosphorus atom and the
biphenyl group, are more potent than 1c, which contains
an ethylene bridge, against MMP-2 and MMP-8. Fur-
thermore, while 1a and 1b present similar affinity to-
wards MMP-2, 1a is about three times more potent
than 1b against MMP-8. An attempt to rationalize the
above results by means of computational simulations
is described in the following paragraphs.
4.2. Molecular modelling study
The comparison of the four putative complexes formed
by the inhibitors 1a and 1b in the active site of MMP-2
Table 1. In vitro inhibition of MMP-2 and MMP-8 by phosphinates
1a–1c
Inhibitor
IC50 (lM)
MMP-2
MMP-8
1a
1b
1c
1.7
1.4
48
5.1
17.6
44.5

G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4743
4.3. Molecular dynamics simulation
drawn only in the presence of the complete picture
(i.e., the dynamical behaviour of 1c within the enzyme)
our results suggest that the inclusion of a further meth-
ylene group renders the confinement of 1c into the active
lock-and-key conformation, a high energy process.
The additional methylene group of 1c was unexpectedly
found to cause a decrease of inhibition (see Table 1), rel-
ative to other two ligands, of more than one order of
magnitude on both enzymes.
To better ascertain the physicochemical reasons for this
behaviour, we carried out molecular dynamics (MD)
simulations. Our strategy was to compare the free en-
ergy values of the conformations adopted by 1c in aque-
ous solution and in the putative binding with the MMP
active site. To overcome the multidimensional problem
due to the large number of internal degrees of freedom,
we decided to analyze the simulated trajectories by
essential dynamics.28 This procedure greatly facilitates
the description of conformational behaviour of the li-
gand, since it allows us to define a new set of coordinates
where large amplitude motions (essential motions)
responsible for the conformational changes can be ex-
tracted. Two essential coordinates were found to be suf-
ficient for describing conformational fluctuations of
ligand 1c. This result allowed us to define a plane (essen-
tial plane) where Helmholtz free energy minima can be
easily located as a function of only two conformational
(essential) coordinates.29 Five energy minima, denoted
as I–V, populate the free energy map of 1c reported
in Figure 3, where contour lines are drawn every
2 kJ/mol. The binding conformation, indicated by a
cross, is practically never sampled during the simulation,
being characterized by a high value of free energy. The
analysis of the conformations sampled in the free energy
minima reveals that all of them present at least one
intramolecular H-bond.
5. Conclusion
Three novel peptidyl phosphinates 1a–1c have been
designed and synthesized as inhibitors of MMPs on
the basis of the structures of phosphonate inhibitors
complexed with MMPs. These ligands, extending in
both the S and S0 subsites of MMPs, are based on the
common phosphinic tripeptide moiety Pro-Leu-Gly(P)-
(OH) linked to a biphenyl group through a methylene
or ethylene bridge. Inhibition of MMP-2 and MMP-8,
in the micromolar range, was confirmed by kinetic mea-
surements and validates our structure-based design of
this model of phosphinate inhibitors. Molecular model-
ling and molecular dynamics studies offer an interpreta-
tion at molecular level of their different behaviour and
give reason for the lower activity shown by the ligand
containing the longer ethylenic bridge 1c. These results
can be regarded as a useful starting point for future
improvements in the development of more potent and
selective phosphinate inhibitors of MMPs.
6. Experimental
6.1. Molecular dynamics
For the MD simulation of structure 1c, we have adopted
the gromos96 force field implemented in a modified ver-
sion of the Gromacs package.30 The point charges were
calculated using the CHelpG protocol on the Gamess
package.31
Moreover, the torsion angles involving the ethylenic
bridge undergo an extent of fluctuations around the
equilibrium positions, much larger than those around
the other rotational degrees of the ligand (Table 2). Con-
sequently, even though a conclusive assessment can be
The structure in its ꢁactiveꢀ configuration, after energy
minimization, was put at the centre of a cubic box of
3.5 nm length filled with 868 water single point charge
molecules,32 and with one Clꢀ counterion to ensure elec-
troneutrality of the system.
After a short solvent relaxation at the desired density,
the overall system was gently heated to 300 K and a sim-
ulation of 20 ns was finally carried out under isothermal/
isochoric conditions (canonical ensemble) with a stan-
dard protocol.33
The simulation and the analysis of the trajectory were
performed using the Gromacs software package and
our own programs.
For checking the actual equilibration of the simulated
system, the root mean square deviation (RMSD) with
respect to the initial structures of 1c was evaluated. On
the equilibrated portion of the trajectory we finally car-
ried out essential dynamics analysis for evaluating the
conformationally relevant coordinates. This analysis,
due to the limited dimensions of the investigated sys-
tems, was performed including all the atoms.
Figure 3. Three hundred kelvin free energy plot for 1c on the plane of
the first two essential motions.

4744
G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
Table 2. Free energy minima, their values and relevant backbone torsion angles together with their RMSFs (°) for 1c
Energy minima
DA0 (kJ/mol)
u1
w1
u2
79 20
w2
u3
w3
h1
h2
69 21
h3
I
0
0
127 23
124 21
127 22
127 23
134 25
150 20
133 19
148 19
145 21
143 24
126 15
102 14
116 18
115 15
52 16
79 14
97 19
77 16
76 14
77 12
114
116
113
114
118
7
7
8
7
7
165 12
119 16
166 46
176 59
122 67
174 90
93 66
162 92
163 91
91 18
II
III
IV
V
87 19
90 27
87 11
103 59
178 11
172 14
+1
+2
+3
101 24
111 26
6.2. Molecular modelling
for 31P are given in parts per million relative to phos-
phoric acid as the internal standard. Peak multiplici-
ties are abbreviated: singlet, s; doublet, d; triplet, t;
quartet, q; multiplet, m; broad, br. [a]D values were
determined with a Perkin-Elmer 241 polarimeter. Mass
spectra were performed on an Applied Biosystems
Voyager DE-PRO MALDI-TOF and on a Micromass
Quattro triple-quadruple ESI-MS instrument. MAL-
DI-TOF-MS analyses were performed on a matrix of
a-cyano-4-hydroxycinnamic acid in acetonitrile/water
50:50, while for ESI-MS analyses samples were dis-
solved in acetonitrile solution containing 0.1% of tri-
fluoroacetic acid (TFA). Elemental microanalyses (C,
H, N) were within 0.4% of the calculated values.
The ligands herein described have been designed follow-
ing structure-based drug-design criteria maximizing
their putative structural affinity within the enzyme ac-
tive site, using the program SYBYL on a Silicon Graph-
ics workstation O2. The coordinates of the enzymes
have been taken from PDB (entries 1QIB and 1I73
for MMP-2 and MMP-8, respectively). The same pro-
gram has been used for the minimization of the putative
complexes formed by 1a and 1b with MMP-2 and
MMP-8.
The following constraints have been imposed on the
starting position of each ligand in the enzyme active site:
the two phosphinic oxygens have been positioned at 2.1
6.4. 2-(1,10-Biphenyl-4-yl)ethanol (3c)
˚
and 2.5 A from the catalytic zinc ion; moreover, the
˚
2.5 A Zn-coordinated oxygen atom has been H-bound
To a cooled (0 °C) 250 mL two-necked flask containing
a solution of 2c (3.0 g, 14.1 mmol) in 70 mL of anhy-
drous THF, 1.6 g (42.4 mmol) of LiAlH4 was added
portionwise while stirring, under nitrogen. When effer-
vescence ceased, the mixture was heated at reflux for
5 h until disappearance of starting material. The mix-
ture was diluted with ethyl acetate, washed with
1.0 M aqueous solution of HCl, water, and brine,
and the organic layer dried over anhydrous sodium sul-
fate. Evaporation of the solvent afforded 2.76 g (99%
yield) of product 3c.
˚
(2.8 A) to Glu 202 (MMP-2) and Glu 198 (MMP-8).
The Leu NH and CO groups of inhibitor have been
˚
H-bound (2.8 A) to CO and NH groups of Ala 167
(MMP-2) and Ala 163 (MMP-8), respectively. The
biphenyl substituent has been inserted in the S01 pocket.
Free rotation has been left to all torsion angles of the
inhibitors.
6.3. Chemistry
All reagents were purchased from commercial suppli-
ers and used without further purification. Dry tetra-
hydrofuran (THF) was freshly distilled from sodium
and benzophenone; dry dichloromethane was distilled
from P2O5. In all other cases, commercially available
reagent-grade solvents were employed without purifi-
cation. Reactions performed in dry solvents were car-
ried out in nitrogen atmosphere. Melting points are
uncorrected and were obtained on a capillary appara-
1
Compound 3c: mp (CHCl3) = 95–96 °C; H NMR (ace-
tone-d6): d 7.55–7.15 (m, 9H), 3.65 (t, J = 7 Hz, 2H),
2.80 (br signal, 1H), 2.72 (t, J = 7 Hz, 2H).
6.5. 40-(Hydroxymethyl)-1,10-biphenyl-4-ol (3b)
Compound 3b (87% yield) was prepared similarly to 3c.
tus (Buchi 510). Analytical thin-layer chromatography
¨
Compound 3b: mp (CHCl3) = 181–183 °C; 1H NMR
(acetone-d6): d 8.46 (s, 1H), 7.60–7.35 (m, 6H), 6.92
(d, J = 8 Hz, 2H), 4.64 (s, 2H), 4.18 (br s, 1H).
(TLC) was routinely used to monitor reactions: plates
precoated with E. Merck silica gel 60 F254 or RP-8
F254S of 0.25 mm thickness were used. Merck silica
gel 60 (230–400 ASTM mesh) or Aldrich RP-18 silica
gel (Cat. No. 377635) was employed for column flash-
6.6. 4-(Iodomethyl)-1,10-biphenyl (4a)
1
chromatography (FC). H, 13C and 31P NMR spectra
To a solution containing 3a (2 g, 10.9 mmol) in 10 mL of
chloroform, 14 mL of an aqueous solution of HI (57%)
was added dropwise, under stirring. The mixture was al-
lowed to react at room temperature (rt) for 4 h and later
worked-up as follows: the mixture was extracted with
were recorded on a Varian VXR 300 or a Bruker AC
1
200 spectrometer. Chemical shifts (d) for 13C and H
are given in parts per million relative to (CH3)4Si
(TMS) as the internal standard. Chemical shifts (d)

G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4745
chloroform, the separated organic layers were washed
with a (10%) aqueous solution of NaHCO3 and dried
over anhydrous sodium sulfate. Evaporation of the
solvent afforded 2.63 g (82% yield) of product 4a.
J = 6.4 Hz), 129.84, 128.24, 127.95 (d, J = 3.2 Hz),
127.84 (d, J = 0.6 Hz), 35.48 (d, J = 134.8 Hz). 31P
NMR (CD3OD): d 33.25. MS-ESI+: m/z = 233 [M+H]+.
6.10. 2-(1,10-Biphenyl-4-yl)ethylphosphinic acid (6c) and
(40-hydroxy-1,10-biphenyl-4-yl)methylphosphinic acid (6b)
Compound 4a: mp (n-hexane/EtOAc) = 104–106 °C;
1H NMR (CDCl3): d 7.60–7.38 (m, 9H), 4.51 (s, 2H).
13C NMR (CDCl3): d 140.75, 140.41, 138.27, 129.16,
128.79, 127.53, 127.00, 5.65.
Compounds 6c (58% yield) and 6b (60% yield) were pre-
pared starting, respectively, from 4c and 4b, analogously
as described for 6a.
6.7. 4-(2-Iodoethyl)-1,10-biphenyl (4c)
Compound 6c: Rf = 0.29 (EtOAc/AcOH 8:2); mp
1
To a solution containing PPh3 (2.42 g, 9.23 mmol), imid-
azole (681 mg, 12 mol), and iodine (2.34 g, 9.23 mmol)
dissolved in 50 mL of dry dichloromethane, 1.52 g
(7.69 mmol) of alcohol 3c was added portionwise with
stirring, under nitrogen. The reaction was stirred for
3 h at rt, and then crude oil was purified by FC (n-hex-
ane/EtOAc 80:20) affording 2.20 g (93% yield) of prod-
uct 4c.
(CH2Cl2) = 203–205 °C; H NMR (CDCl3): d 9.75 (br
s, 1H), 7.60–7.22 (m, 9H), 7.15 (d, J = 548 Hz, 1H),
3.05–2.88 (m, 2H), 2.25–2.05 (m, 2H). 13C NMR
(CDCl3): d 140.74, 139.53, 139.18 (d, J = 15.8 Hz),
128.80, 128.57, 127.43, 127.24, 127.01, 30.94 (d,
J = 92.5 Hz), 26.48. 31P NMR (CD3OD): d 37.90. MS-
ESI+: m/z = 247 [M+H]+.
Compound 6b: Rf = 0.26 (EtOAc/AcOH 8:2); mp
(CHCl3) = 211–213 °C; 1H NMR (CD3OD): d 7.55–
7.28 (m, 6H), 6.84 (d, J = 8.6 Hz, 2H), 6.98 (d,
J = 550 Hz, 1H), 3.20 (d, J = 18.5 Hz, 2H). 13C NMR
(CD3OD): d 158.20, 141.01 (d, J = 3.9 Hz), 133.20,
131.25, (d, J = 6.1 Hz), 130.61 (d, J = 7.5 Hz), 128.90,
127.68 (d, J = 3.1 Hz), 116.63, 38.77 (d, J = 86.3 Hz).
31P NMR (CD3OD): d 32.52. MALDI-TOF-MS:
m/z = 249 [M+H+].
Compound 4c: mp (n-hexane/EtOAc) = 58–59 °C; 1H
NMR (CDCl3): d 7.60–7.20 (m, 9H), 3.45–3.15 (m, 4H).
13C NMR (CDCl3): d 140.64, 139.64, 139.54, 128.67,
127.26, 127.17, 126.91, 39.88, 5.50.
6.8. 40-(Iodomethyl)-1,10-biphenyl-4-ol (4b)
Compound 4b (99% yield) was prepared similarly to 4a.
Compound 3c: mp (Et2O) = 153–155 °C; 1H NMR
(CDCl3): d 7.52–7.40 (m, 7H), 6.90 (d, J = 8.8 Hz, 2H),
4.51 (s, 2H). 13C NMR (acetone-d6): d 158.14, 141.22,
138.85, 132.41, 130.17, 128.78, 127.40, 116.56, 6.84.
6.11. Methyl-1,10-biphenyl-4-yl-methyl[(1,3-dioxo-1,3-
dihydro-2H-isoindol-2-yl)methyl]phosphinate (8a)
A two-necked flask containing a suspension of 6a
(500 mg, 2.17 mmol) in 6 mL of dry dichloromethane
was added with 0.5 mL (2.4 mmol) HMDS, under nitro-
gen. The resulting solution was stirred for 1 h at rt,
added with phthalimidomethyl bromide (521 mg,
2.17 mmol) and refluxed overnight. After TLC control
(n-hexane/ethyl acetate 7:3) showed the disappearance
of reagent 6a, solvent was evaporated in vacuo and
the residue, dissolved in 5 mL of a 1:1 methanol/toluene
mixture, was cooled to 0 °C and added with an excess of
an ethereal solution of diazomethane under stirring.
After 1 h, TLC control (toluene/acetone 1:1) showed
the end of reaction: solvent was evaporated in a rotava-
por containing some drops of acetic acid in the receiver
flask and crude oil was purified by FC (toluene/acetone
70:30 as eluent mixture) affording 500 mg (57% yield) of
product 8a as a white solid.
6.9. 1,10-Biphenyl-4-yl-methylphosphinic acid (6a)
A
two-necked flask containing ammonium hypo-
phosphite24b (564 mg, 6.8 mmol) was added with
2.13 mL (10.2 mmol) HMDS and the mixture was
heated at 105 °C (temperature value controlled with
Vertex) for 2 h, under nitrogen in order to synthesize
BTSP 5.25 The system was cooled at rt, added via syr-
inge with 2 g (6.8 mmol) of 4a previously dissolved in
5 mL of anhydrous dichloromethane and refluxed un-
der nitrogen for 12 h. The reaction mixture was fil-
tered, and the solvent was removed to yield an oil,
which was dissolved in dichloromethane and added
with a stoichiometric amount (6.8 mmol) of methanol
(to achieve desilylation and solution). The mixture
was newly filtered and the solution was washed with
2.0 M aqueous solution of HCl, dried over anhydrous
sodium sulfate, followed by removal of the solvent
affording 1.03 g (70% yield) of product 6a as a white
solid; the latter was carefully dried over P2O5 over-
night, under vacuum. The product was used without
further purification.
Compound 8a: Rf = 0.38 (toluene/acetone 6:4); mp
1
(EtOAc) = 146–148 °C; H NMR (CDCl3): d 7.91–7.82
(m, 2H), 7.79–7.71 (m, 2H), 7.59–7.30 (m, 9H), 4.1 (d,
J = 7.9 Hz, 2H), 3.78 (d, J = 10.7 Hz, 3H), 3.36 (ABX
dd, J = 15.65 and 23.86 Hz, 1H), 3.28 (ABX dd,
J = 15.65 and 23.86 Hz, 1H). 13C NMR (CDCl3): d
167.39, 140.62, 139.96, 134.35, 131.85, 130.47 (d,
J = 5.8 Hz), 129.32 (d, J = 9.3 Hz), 128.74, 127.38 (d,
J = 2.7 Hz), 127.29, 127.02, 123.64, 52.23 (d,
J = 6.9 Hz), 35.93 (d, J = 89.7 Hz), 34.73 (d,
J = 95.5 Hz). 31P NMR (CDCl3): d 45.03. MS-ESI+:
m/z = 406 [M+H]+.
Compound 6a: Rf = 0.39 (EtOAc/AcOH 8:2); mp
1
(CH2Cl2) = >220 °C; H NMR (CD3OD): d 7.62–7.25
(m, 9H), 6.8 (d, J = 666.6 Hz, 1H), 3.16 (d, J = 21.7 Hz,
2H). 13C NMR (CD3OD): d 142.14 (d, J = 1.5 Hz),
140.8 (d, J = 3.8 Hz), 133.47 (d, J = 9.4 Hz), 131.4 (d,

4746
G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
6.12. Methyl-2-(1,10-biphenyl-4-yl)ethyl[(1,3-dioxo-1,3-
dihydro-2H-isoindol-2-yl)methyl]phosphinate (8c) and
methyl (1,3-dioxo-1,3-dihydro-2H-isoindol-2-yl)methyl-
[(40-hydroxy-1,10-biphenyl-4-yl)methyl]phosphinate (8b)
139.87, 139.75, 128.97, 128.71, 127.58, 127.46, 127.17,
51.75 (d, J = 6.9 Hz), 39.05 (d, J = 97.3 Hz), 27.45 (d,
J = 87.1 Hz), 27.45 (d, J = 2.3 Hz). 31P NMR (CDCl3):
d 57.38.
Compounds 8c (55% yield) and 8b (49% yield) were pre-
pared starting, respectively, from 6c and 6b, analogously
as described for 8a.
Compound 9b: Rf = 0.34 (EtOAc/MeOH 6:4); sticky oil;
1H NMR (CD3OD): d 7.54–7.41 (m, 6H), 6.83 (d,
J = 8.6, 2H), 3.85 (d, J = 8.2 Hz, 2H), 3.74 (d,
J = 10.2 Hz, 3H), 3.44 (d, J = 17.6 Hz, 2H). 13C NMR
Compound 8c: Rf = 0.27 (toluene/acetone 7:3); sticky
oil; H NMR (CDCl3): d 7.88–7.81 (m, 2H), 7.76–7.69
(CD3OD):
d
158.36, 141.48, 133.0, 131.42 (d,
1
J = 5.7 Hz), 129.85, 128.92, 127.81, 116.68, 52.91 (d,
J = 7.1 Hz), 37.28 (d, J = 99.3 Hz), 34.28 (d,
J = 86.9 Hz). 31P NMR (CD3OD): d 55.67.
(m, 2H), 7.57–7.26 (m, 9H), 4.10 (d, J = 8.2 Hz, 2H),
3.83 (d, J = 10.7 Hz, 3H), 3.14–2.93 (m, 2H), 2.22–2.04
(m, 2H). 13C NMR (CDCl3): d 167.25, 140.78, 139.99,
139.66, 139.30, 134.34, 131.75, 128.67 (d, J = 4.25 Hz),
127.29, 127.14, 126.94, 123.62, 51.83 (d, J = 6.7 Hz),
34.77 (d, J = 92.9 Hz), 29.90 (d, J = 91.8 Hz), 26.85
(d, J = 3.5 Hz). 31P NMR (CDCl3): d 49.84. MS-ESI+:
m/z = 420 [M+H]+.
6.15. 1-[(Benzyloxy)carbonyl]prolyl-N1-{[(1,10-biphenyl-4-
yl-methyl)(methoxy)phosphoryl] methyl}leucinamide (10a)
To a cooled (0 °C) funnel-equipped flask containing a
dry THF (4 mL) solution of 9a (118 mg, 0.43 mmol)
and Z-Pro-Leu-OH (155 mg, 0.43 mmol), a solution
of DCC (97 mg, 0.47 mmol) and HOBT (64 mg,
0.47 mmol) dissolved in 2 mL of the same solvent was
added dropwise under stirring. After standing at rt over-
night, N,N0-dicyclohexylurea was filtered off and the
solution was evaporated under reduced pressure. The
residue, dissolved in EtOAc, was washed with saturated
solution of NaHCO3 and brine, and organic layer dried
over anhydrous Na2SO4. Crude oil was purified by FC
(EtOAc/MeOH 95:5) affording 212 mg (80% yield) of
product 10a as 1:1 mixture of diastereomers.
Compound 8b: Rf = 0.4 (toluene/acetone 1:1); mp = 215–
217 °C; 1H NMR (CDCl3): d 7.91–7.87 (m, 2H), 7.78–7.74
(m, 2H), 7.44–7.33 (m, 6H), 6.86 (d, J = 8.7 Hz, 2H), 4.10
(d, J = 7.7 Hz, 2H), 3.77 (d, J = 10.8 Hz, 3H), 3.38–3.26
(m, 2H). 13C NMR (CDCl3): d 167.59, 156.59, 140.05,
134.58, 132.07, 131.81, 130.44 (d, J = 5.9 Hz), 128.10,
127.85, 126.88 (d, J = 2.7 Hz), 123.75, 115.71, 52.46
(d, J = 7.1 Hz), 35.73 (d, J = 89.3 Hz), 34.62 (d,
J = 95.5 Hz). 31P NMR (CDCl3): d 51.88. MALDI-
TOF-MS: m/z = 422 [M+H+].
6.13. Methyl-aminomethyl(1,10-biphenyl-4-yl-
methyl)phosphinate (9a)
Compound 10a: Rf = 0.60 (EtOAc/MeOH 9:1); 1H
NMR (CDCl3): d 7.60–7.20 (m, 14H), 7.05 (br signal,
1H), 6.80 (br signal, 1H), 5.15 (s, 2H), 4.58–4.40 (m,
1H), 4.40–4.28 (m, 1H), 3.85–3.35 (br m, 7H), 3.26 (d,
J = 17.4 Hz, 2H), 2.25–1.10 (br m, 7H), 1.05–0.75 (br
signal, 6H). 13C NMR (CDCl3): d 172.31, 172.02,
156.41, 140.64, 139.95, 136.23, 130.39, 129.63, 128.76,
128.60, 128.26, 127.87, 127.47, 127.30, 127.00, 67.62,
60.97, 52.03, 47.24, 40.20, 36.3 (d, J = 94 Hz), 34.42
(d, J = 78 Hz), 29.00, 25.02, 24.70, 22.99, 21.65. 31P
NMR (CDCl3): d 48.95, 48.62. MS-ESI+: m/z = 620
[M+H+], 642 [M+Na+].
To a flask containing 8a (225 mg, 0.56 mmol) dissolved
in 3 mL of absolute ethanol, 556 lL (2.80 mmol) of
hydrazine monohydrate was added under stirring and
the reaction mixture was allowed to stir, at room tem-
perature, for 18 h. After filtration, the solvent was evap-
orated and crude oil was purified by FC (AcOEt/MeOH,
60:40) affording 120 mg of product 9a (78% yield).
Compound 9a: Rf = 0.21 (EtOAc/MeOH 7:3); sticky oil;
1H NMR (CDCl3): d 7.51–7.25 (m, 9H), 3.66 (d,
J = 10.3 Hz, 3H), 3.22 (d, J = 16.7 Hz, 2H), 3.10–2.75
(m, 2H), 1.64 (br s, 2H). 13C NMR (CDCl3): d 140.57,
139.98, 130.37 (d, J = 7.4 Hz), 130.07 (d, J = 5.5 Hz),
128.82, 127.58 (d, J = 2.8 Hz), 127.38, 127.01, 51.75 (d,
J = 7 Hz), 38.69 (d, J = 101.2 Hz), 33.51 (d,
J = 82.5 Hz). 31P NMR (CDCl3): d 52.51.
6.16. 1-[(Benzyloxy)carbonyl]prolyl-N1-{[[2-(1,10-biphenyl-
4-yl)ethyl](methoxy)phosphoryl] methyl}leucinamide (10c)
and 1-[(benzyloxy)carbonyl]prolyl-N1-{[[(40-hydroxy-1,10-
biphenyl-4-yl)methyl](methoxy)phosphoryl]methyl}leucin-
amide (10b)
Tripeptides 10c (67% yield) and 10b (71% yield) were
prepared starting, respectively, from 9c and 9b, analo-
gously as described for 10a.
6.14. Methyl-aminomethyl[2-(1,10-biphenyl-4-yl)ethyl]-
phosphinate (9c) and methyl-aminomethyl[(40-hydroxy-
1,10-biphenyl-4-yl)methyl]phosphinate (9b)
Compound 10c: Rf = 0.69 (EtOAc/MeOH 8:2); 1H NMR
(CDCl3): d 7.56–7.26 (m, 14H), 7.05 (br signal, 1H), 6.70
(br signal, 1H), 5.15 (s, 2H), 4.42–4.36 (m, 1H), 4.32–4.26
(m, 1H), 3.76–3.48 (m, 7H), 3.00–2.92 (m, 2H), 2.16–2.05
(m, 3H), 1.96–1.90 (m, 3H), 1.73–1.66 (m, 3H), 0.91–0.86
(br signal, 6H). 13C NMR (CDCl3): d 172.09, 157.12,
141.09, 140.23 (d, J = 15.5 Hz), 139.54, 136.41, 128.96,
128.83, 128.79, 128.53, 128.12, 128.04, 127.53, 127.37,
127.21, 67.86, 61.22, 52.28, 47.47, 40.28, 36.6 (d,
Compounds 9c (70% yield) and 9b (74% yield) were pre-
pared starting, respectively, from 8c and 8b, analogously
as described for 9a.
Compound 9c: Rf = 0.35 (EtOAc/MeOH 1:1); sticky oil;
1H NMR (CDCl3): d 7.60–7.26 (m, 9H), 3.76 (d,
J = 10.3 Hz, 3H), 3.05–2.92 (m, 4H), 2.25–2.10 (m,
2H), 1.58 (br s, 2H). 13C NMR (CDCl3): d 140.82,

G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4747
J = 98 Hz), 29.56 (d, J = 53 Hz), 27.22 (d, J = 3.7 Hz),
25.28, 24.91, 23.19, 21.82. 31P NMR (CDCl3): d 53.31,
53.02. ESI+-MS: m/z = 634.8 [M+H+], 656.7 [M+Na+].
129.37, 128.07, 127.71, 60.72, 52.33, 41.11, 32.33 (d,
J = 85.8 Hz), 30.85, 29.28 (d, J = 35.5 Hz), 27.59, 25.90,
24.79, 23.33, 21.61. 31P NMR (CD3OD): d 64.01. ESI+-
MS: m/z = 486.7 [M+H+], 508.7 [M+Na+]. Calcd for
C26H36N3O4PÆH2O: C, 62.01; H, 7.61; N, 8.34. Found:
C, 62.38; H, 7.42; N, 8.60.
Compound 10b: Rf = 0.33 (EtOAc/MeOH 9:1); 1H
NMR (CD3OD): d 7.53–7.25 (m, 11H), 6.81 (d,
J = 8.9 Hz, 2H), 5.14–5.06 (m, 2H), 4.43–4.25 (m, 2H),
3.69–3.43 (m, 9H), 2.32–1.48 (m, 7H), 0.96–0.80 (m,
6H). 13C NMR (CD3OD): d 174.89, 174.54, 157.11,
156.12, 141.61, 141.07, 137.56, 131.49 (d, J = 5.4 Hz),
130.55 (d, J = 8.9 Hz), 129.66, 129.34, 128.94, 128.79,
128.65, 128.19, 128.04 (d, J = 2.5 Hz), 127.70, 68.31,
61.92, 60.71, 53.05, 41.45, 40.92, 36.70 (d,
J = 98.5 Hz), 31.81 (d, J = 54 Hz), 25.30, 24.37, 23.45,
21.72. 31P NMR (CDCl3): d 51.12, 50.10. MALDI-
TOF-MS: m/z = 636 [M+H+], 658 [M+Na+].
20
D
1
Compound 1b: ½aꢁ ꢀ59.2 (c 1.35, CH3OH); H NMR
(CD3OD): d 7.51–7.32 (m, 6H), 6.83 (d, J = 8.6 Hz, 2H),
4.48–4.31 (m, 2H), 3.59 (d, J = 9.2 Hz, 2H), 3.34–3.31
(m, 2H), 3.18 (d, J = 16 Hz, 2H), 2.53–2.36 (m, 1H),
2.16–1.97 (m, 3H), 1.68–1.61 (m, 3H), 0.99–0.94 (m,
6H). 13C NMR (CD3OD): d 174.51, 169.95, 158.23,
148.32, 134.02, 131.60, 128.87, 127.49, 126.54, 116.64,
60.89, 53.93, 41.60, 38.82 (d, J = 98.6 Hz), 35.54 (d,
J = 82.3 Hz), 31.10, 25.91, 24.96, 23.41, 21.88. 31P NMR
(CD3OD):
d
62.86. MALDI-TOF-MS: m/z = 488
6.17. Prolyl-N1-{[(1,10-biphenyl-4-yl-methyl)(hydroxy)-
phosphoryl]methyl}leucinamide (1a)
[M+H+]. Calcd for C25H34N3O5PÆ2H2O: C, 57.35; H,
7.32; N, 8.03. Found: C, 57.02; H, 7.52; N, 8.23.
To a solution containing tripeptide 10a (190 mg,
0.307 mmol) dissolved in 4 mL of dry dichloromethane
was added an excess of BTSA (0.825 mL, 3.37 mmol)
while stirring, under nitrogen. After 1 h at rt, the reac-
tion mixture was cooled (ꢀ20 °C) and an excess of
TMSI (0.349 mL, 2.45 mmol) was added dropwise.
The solution was allowed to warm at rt within 1 h and
after 2 h solvent was evaporated under reduced pressure.
Crude oil was taken up with 3.0 mL of a 7:3 CH3CN/
H2O mixture. After removal of the solvents, the residue
was purified by RP-18 FC (CH3CN/H2O, from 60/40 to
80/20) affording 109 mg (75% yield) of free pseudo-tri-
peptide 1a as white solid.38
6.19. MMPs activation and inhibition assays
The proenzymes27 (pro-MMP-2 and -8) were activated
immediately before their use with p-aminophenylmercu-
ric acetate (APMA, 1 mM) for 1 h at 37 °C.39 For assay
measurements, 10ꢀ2 mM solutions of the inhibitors (in
MeOH for 1a, 1b and in a 90:10 mixture of MeOH/
DMSO for 1c) were further diluted as required in the
assay buffer (50 mM Tris–HCl, 5 mM CaCl2, 0.02%
NaN3, 0.05% Brij 35, pH 7.4, for MMP-2, and 50 mM
Tris–HCl, 10 mM CaCl2, 150 mM NaCl, pH 7.5, for
MMP-8). The activated enzyme added with inhibitor
solution was incubated in the assay buffer for 3 h, the
temperature being maintained at 25 °C for MMP-2
20
D
Compound 1a: ½aꢁ ꢀ69 (c 0.29, DMSO); 1H NMR
and at 37 °C for MMP-8. After addition of
a
(DMSO-d6, 600 MHz): d 9.0 (br signal, 1H), 8.38 (br
s, 1H), 7.61 (d, J = 8 Hz, 2H), 7.53 (d, J = 8 Hz, 2H),
7.48–7.30 (m, 5H), 4.42 (br signal, 1H), 4.12 (br signal,
1H), 3.38–3.05 (m, 4H), 2.93 (d, 2H), 2.23–2.18 (m,
1H), 1.90–1.83 (m, 1H), 1.82–1.75 (m, 2H), 1.65–1.43
(m, 3H), 0.91–0.83 (m, 6H). 13C NMR (DMSO-d6): d
171.48, 167.97, 140.03, 137.78, 130.44, 130.33, 128.95,
127.27, 126.5, 126.33, 58.91, 51.65, 45.69, 29.35, 24.24,
23.40, 23.09, 21.54. 31P NMR (DMSO-d6): d 67.86.
ESI+-MS: m/z = 472 [M+H+], 494 [M+Na+]. Calcd for
C25H34N3O4PÆ2H2O: C, 59.16; H, 7.55; N, 8.28. Found:
C, 58.76; H, 7.38; N, 8.14.
0.054 mM DMSO solution of fluorogenic substrate
QF24, hydrolysis was monitored by recording the
increase in fluorescence (kex 328 nm, kem 393 nm), during
30 min, using a Perkin-Elmer spectrofluorimeter LS
50B. The IC50 values were calculated from control reac-
tions without the inhibitor. The mean IC50 was obtained
from at least three independent measurements, and the
standard deviation of the mean was less than 10%.
Acknowledgments
The authors are grateful to Dr. Andrea Aramini and Dr.
Manuel Pascasi for their support in the experimental
work. Financial support by Italian MIUR and by Polif-
arma Research Center is gratefully acknowledged.
6.18. Prolyl-N1-{[[2-(1,10-biphenyl-4-yl)ethyl](hydroxy)-
phosphoryl]methyl}leucinamide (1c) and prolyl-N1-({hydr-
oxy[(40-hydroxy-1,10- biphenyl-4-yl)methyl]phosphoryl}
methyl)leucinamide (1b)
Final tripeptides 1c (83% yield) and 1b (72% yield) were
prepared starting, respectively, from 10c and 10b, anal-
ogously as described for 1a.
References and notes
1. (a) Coussens, L. M.; Fingleton, B.; Matrisian, L. M.
Science 2002, 295, 2387–2392; (b) Foda, H. D.; Zucker, S.
Drug Discovery Today 2001, 6, 478–482; (c) Leung, D.;
Abbenante, G.; Fairlie, D. P. J. Med. Chem. 2000, 43,
305–341.
2. Woessner, J. F., Jr. FASEB J. 1991, 5, 2145–2154.
3. Murphy, G.; Docherty, A. J. P. J. Res. Cell Mol. Biol.
1992, 7, 120–125.
20
D
1
Compound 1c: ½aꢁ ꢀ54.3 (c 0.35, CH3OH); H NMR
(CD3OD): d 7.59–7.30 (m, 9H), 4.49–4.42 (m, 1H),
4.35–4.28 (m, 1H), 3.38–3.31 (m, 4H), 3.01–2.87 (m,
2H), 2.51–2.37 (m, 1H), 2.17–2.00 (m, 5H), 1.71–1.68
(m, 3H), 0.99–0.93 (m, 6H). 13C NMR (CD3OD): d
175.20, 169.66, 141.95, 140.96, 140.54, 129.67, 129.54,

4748
G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4. Birkedal-Hansen, H.; Moore, W. G. I.; Bodden, M. K.;
M.; Larsen, Y. B.; Meldal, M.; Foged, N. T. Curr. Med.
Chem. 2001, 8, 967; (c) Caldwell, C. G.; Sahoo, S. P.; Polo,
S. A.; Eversole, R. R.; Lanza, T. J.; Mills, S. G.;
Niedzwiecki, L. M.; Izquierdo-Martin, M.; Chang, B.
C.; Harrison, R. K.; Kuo, D. W.; Lin, T.-Y.; Stein, R. L.;
Durette, P. L.; Hagmann, W. K. Bioorg. Med. Chem. Lett.
1996, 6, 323–328; (d) Goulet, J. L.; Kinneary, J. F.;
Durette, P. L.; Stein, R. L.; Harrison, R. K.; Izquierdo-
Martin, M.; Kuo, D. W.; Lin, T.-Y.; Hagmann, W. K.
Bioorg. Med. Chem. Lett. 1994, 4, 1221–1224.
Windsor, L. J.; Birkedal-Hansen, B.; De Carlo, A.; Engler,
J. A. Crit. Rev. Oral Biol. Med. 1993, 4, 197–250.
5. (a) Woessner, J. F.; Nagase, H. Matrix Metalloproteinases
and TIMPs; Oxford University Press: New York, 2000; (b)
Denhardt, D. T.; Feng, B.; Edwards, D. R.; Cocuzzi, E.
T.; Malyankar, U. M. Pharmacol. Ther. 1993, 59, 329–341.
6. Pyke, C.; Ralfkiaer, E.; Huhtala, P.; Hurskainen, T.;
Dano, K.; Tryggvason, K. Cancer Res. 1992, 52, 1336–
1341.
7. Stetler-Stevenson, W. G.; Aznavoorian, S.; Liotta, L. A.
Annu. Rev. Cell Biol. 1993, 9, 541–573.
8. Gijbels, K.; Masure, S.; Carton, H.; Opdenakker, G.
J. Neuroimmunol. 1992, 41, 29–34.
9. Murphy, G.; Hembry, R. M. J. Rheumatol. 1992, 19, 61–
64.
10. (a) Peress, N.; Perillo, E.; Zucher, S. J. Neuropathol. Exp.
Neurol. 1995, 54, 16–22; (b) Lopez, B.; Gonzalez, A.; Diez,
J. Curr. Opin. Nephrol. Hypertens. 2004, 13, 197–204.
11. (a) Martin, D. C.; Sanchez-Sweatman, O. H.; Inderdeo, D.
S.; Tsao, M. S.; Khokha, R. Lab. Invest. 1999, 79, 225–
22. (a) Netzel-Arnett, S.; Fields, G.; Birkedal-Hansen, H.;
Van Wart, H. E. J. Biol. Chem. 1991, 266, 1; (b) Netzel-
Arnett, S.; Sang, Q.-X.; Moore, W. G. I.; Navre, M.;
Birkedal-Hansen, H.; Van Wart, H. E. Biochemistry 1993,
32, 6427; (c) Bode, W.; Reinemer, P.; Huber, R.; Kleine,
T.; Schnierer, S.; Tschesche, H. EMBO J. 1994, 13, 1263;
(d) Grams, F.; Reinemer, P.; Powers, J. C.; Kleine, T.;
Pieper, M.; Tschesche, H.; Huber, R.; Bode, W. Eur. J.
Biochem. 1995, 228, 830.
23. For an excellent discussion concerning the chemistry and
biological activity of aminophosphonic and aminophos-
phinic acids see: Aminophosphonic and Aminophosphinic
Acids; Kukhar, V. P., Ed.; Wiley: Chichester, 2000.
24. (a) Boyd, E. A.; Regan, A. C.; James, K. Tetrahedron Lett.
1994, 35, 4223–4226; (b) Boyd, E. A.; Regan, A. C.; James,
K. Tetrahedron Lett. 1992, 33, 813–816.
25. CAUTION: due to its pyrophoric character, BTSP was
prepared in situ and directly used in the reaction with alkyl
iodides.
26. (a) Ko¨nig, W.; Geiger, R. Chem. Ber. 1970, 103, 788–798;
(b) Solas, D.; Hale, R. L.; Patel, D. V. J. Org. Chem. 1996,
61, 1537–1539.
27. Recombinant Human pro-MMP-2 and -8 were purchased
from R&D Systems, catalogue numbers, respectively, 902-
MP and 908-MP; the fluorogenic peptide QF24, used as a
very sensitive substrate for in situ determination of MMP
activity, was purchased from Bachem, Cat. No. M-1895.
28. Amadei, A.; Linssen, A. B. M.; Berendsen, H. J. C.
Proteins 1993, 17, 412–425.
29. The free energy calculations were carried out by preven-
tively dividing the essential plane in 2500 intervals. The
trajectory was then projected on this plane and the
probability (qn) of finding the trajectory within each nth
234; (b) Kruger, A.; Fata, J. E.; Khokha, R. Blood 1997,
¨
90, 1993–2000.
´
12. Overall, C. M.; Lopez-Otin, C. Nat. Rev. Cancer 2002, 2,
657–672.
13. (a) Peterson, J. T. Heart Fail. Rev. 2004, 9, 63–79; (b)
Skiles, J. W.; Gonnella, N. C.; Jeng, A. Y. Curr. Med.
Chem. 2004, 11, 2911–2977; (c) Michaelides, M. R.;
Curtin, M. L. Curr. Pharm. Design 1999, 5, 787–819; (d)
Whittaker, M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H.
Chem. Rev. 1999, 99, 2735–2776.
14. Muri, E. M. F.; Nieto, M. J.; Sindelar, R. D.; Williamson,
J. S. Curr. Med. Chem. 2002, 9, 1631–1653.
15. Babine, E. R.; Bender, S. L. Chem. Rev. 1997, 97, 1359–
1472.
16. Freskos, J. N.; McDonald, J. J.; Mischke, B. V.; Mullins,
P. B.; Shieh, H. S.; Stegeman, R. A.; Stevens, A. M.
Bioorg. Med. Chem. Lett. 1999, 9, 1757–1760.
17. Tamura, Y.; Watanabe, F.; Nakatani, T.; Yasui, K.; Fuji,
M.; Komurasaki, T.; Tsuzuki, H.; Maekawa, R.;
Yoshioka, T.; Kawada, K.; Sugita, K.; Ohtani, M. J.
Med. Chem. 1998, 41, 640–649.
18. Hodgson, J. Biotechnology 1995, 13, 554–557.
19. (a) DꢀAlessio, S.; Gallina, C.; Gavuzzo, E.; Giordano, C.;
Gorini, B.; Mazza, F.; Paglialunga Paradisi, M.; Panini,
G.; Pochetti, G.; Sella, A. Bioorg. Med. Chem. 1999, 7,
389–394; (b) Cirilli, M.; Gallina, C.; Gavuzzo, E.; Giord-
ano, C.; Gomis-Ruth, F. X.; Gorini, B.; Kress, L. F.;
Mazza, F.; Paglialunga Paradisi, M.; Pochetti, G.; Politi,
V. FEBS Lett. 1997, 418, 319–322; (c) Gavuzzo, E.;
Pochetti, G.; Mazza, F.; Gallina, C.; Gorini, B.; DꢀAlessio,
S.; Pieper, M.; Tschesche, H.; Tucker, P. A. J. Med. Chem.
2000, 43, 3377–3385; (d) Aschi, M.; Roccatano, D.; Di
Nola, A.; Gallina, C.; Gavuzzo, E.; Pochetti, G.; Pieper,
M.; Tschesche, H.; Mazza, F. J. Comput.-Aided Mol. Des.
2002, 16, 213–225.
20. (a) Finzel, B. C.; Baldwin, E. T.; Bryant, G. L., Jr.; Hess,
G. F.; Wilks, J. W.; Trepod, C. M.; Mott, J. E.; Marshall,
V. P.; Petzold, G. L.; Poorman, R. A.; OꢀSullivan, T. J.;
Schostarez, H. J.; Mitchell, M. A. Protein Sci. 1998, 7,
2118–2126; (b) Yeh, L. A.; Chen, J.; Baculi, F.; Gingrich,
D. E.; Shen, T. Y. Bioorg. Med. Chem. Lett. 1995, 5, 1637–
1642; (c) Reiter, L. A.; Mitchell, P. G.; Martinelli, G. J.;
Lopresti-Morrow, L. L.; Yocum, S. A.; Eskra, J. D.
Bioorg. Med. Chem. Lett. 2003, 13, 2331–2336.
interval was evaluated. The Helmholtz free energy ðDAi0ꢀj
Þ
for the transition from an interval ꢁiꢀ to an interval ꢁjꢀ can
be simply calculated using the standard relation
DA0iꢀj ¼ ꢀkT lnðqj=qiÞ where k is the Boltzmann constant.
30. Van der Spoel, D.; van Drunen R.; Berendsen, H.J.C.,
‘‘GRoningen MAchine for Chemical Simulations,’’ Dept.
of Biophysical Chemistry, BIOSON Research Institute,
Nijenborgh 4 NL-9717 AG Groningen, 1994, e-mail to
gromacs@chem.rug.nl.
31. Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S.
T.; Gordon, M. S.; Jensen, J. J.; Koseki, S.; Matsunaga,
N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupuis, M.;
Montgomery, J. J. Comput. Chem. 1993, 14, 1347.
32. Feenstra, K. A.; Hess, B.; Berendsen, H. J. C. J. Comput.
Chem. 1999, 20, 786.
33. The temperature was kept constant by the isogaussian
algorithm.34 Periodic boundary conditions (pbc) were
systematically applied and a cut-off of 1.10 nm was
considered when Lennard-Jones (LJ) forces were com-
puted. The long range electrostatics was calculated using
the particle mesh ewald method (PME),35 and the algo-
rithm LINCS36 was used to constrain all bond lengths. We
also used roto-translational constraints to stop the solute
in the simulated box.37
21. The following references are cited as selected examples:
(a) Matziari, M.; Beau, F.; Cuniasse, P.; Dive, V.;
Yiotakis, A. J. Med. Chem. 2004, 47, 325–336; (b) Schiødt,
C. B.; Buchardt, J.; Terp, G. E.; Christensen, U.; Brink,
34. Evans, D. J.; Morriss, G. P. In Statistical Mechanics of
Non-Equilibrium Liquids; Academic Press: London, 1990.

G. Bianchini et al. / Bioorg. Med. Chem. 13 (2005) 4740–4749
4749
35. Darden, T. A.; York, D. M.; Pedersen, L. G. J. Chem.
Phys. 1993, 98, 10089.
36. Bekker, B.; Berendsen, H. J. C.; Fraije, J. G. E. M.
J. Comput. Chem. 1997, 18, 1463.
37. Amadei, A.; Chillemi, G.; Ceruso, M. A.; Grottesi, A.;
Di Nola, A. J. Chem. Phys. 2000, 112, 9.
38. Occasionally free pseudo tripeptides 1a and 1c showed
a dark-red colour probably due to the formation of a
charge-transfer complex with elemental iodine. This
complex can be destroyed by the following work-up:
a water suspension of product was slowly added with
the smallest amount of a 5% aqueous solution of
Na2S2O3 while stirring, until a white solid appeared;
then, after centrifugation and water washing, carefully
drying over P2O5 afforded pure product as a white
solid.
39. (a) Murphy, G.; Crabbe, T. Methods Enzymol. 1995, 248,
470; (b) Tschesche, H. Methods Enzymol. 1995, 248, 431.