Potent "clicked" MMP2 inhibitors: Synthesis, molecular modeling and biological exploration

Article abstract of DOI:10.1039/c0ob00852d

A new series of MMP2 inhibitors is described, following a fragment-based drug design approach. One fragment containing an azide group and a well known hydroxamate Zinc Binding Group in a α-sulfone, α-tetrahydropyrane scaffold, has been synthesized. Water-

Full text of DOI:10.1039/c0ob00852d

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 4587
www.rsc.org/obc
PAPER
Potent “Clicked” MMP2 Inhibitors: Synthesis, Molecular Modeling and
Biological Exploration†
Jose Mar´ıa Zapico,a Pilar Serra,a Josune Garc´ıa-Sanmart´ın,b Kamila Filipiak,a,c Rodrigo J. Carbajo,d
Anne K. Schott,d Antonio Pineda-Lucena,d Alfredo Mart´ınez,b Sonsoles Mart´ın-Santamar´ıa,a
Beatriz de Pascual-Teresa*a and Ana Ramos*a
Received 8th October 2010, Accepted 21st March 2011
DOI: 10.1039/c0ob00852d
A new series of MMP2 inhibitors is described, following a fragment-based drug design approach. One
fragment containing an azide group and a well known hydroxamate Zinc Binding Group in a a-sulfone,
a-tetrahydropyrane scaffold, has been synthesized. Water-LOGSY, STD and competition-STD
experiments indicate that this fragment binds to the active site of the enzyme. A click chemistry reaction
was used to connect the azide to lipophilic alkynes selected to interact selectively with the S1¢ subunit of
MMP2, as shown by docking and molecular dynamic experiments of the designed compounds. The
most potent compounds 18 and 19 displayed an IC50 of 1.4 and 0.3 nM against MMP2 respectively, and
showed negligible activity towards MMP1 and MMP7, two metalloproteinases which have a shallow
S1¢ subsite. Compound 18 also showed a promising selectivity profile against some antitarget
metalloproteinases, such as MMP8, and considerably less activity against MMP14 (IC50 = 65 nM), and
MMP9 (IC50 = 98 nM), other MMPs characterized by having a deep S1¢ pocket and, therefore, more
similar to MMP2.
Matrilysin: MMP7 and MMP26, Membrane-type MMPs:
Introduction
MMP14 to MMP17, MMP24 and MMP25 and other MMPs.
Matrix metalloproteinases (MMPs), also called matrixins, are a
Their role in the progression of cancer may involve several
familyofstructurallyrelatedzinc-containingenzymes thatmediate
mechanisms. Originally, MMPs were thought to mediate invasion
the breakdown of connective tissue and are, therefore, targets
and metastasis primarily by matrix remodelling, thereby allowing
for therapeutic inhibitors in many inflammatory, malignant, and
the tumour cells access to blood and lymphatic vessels. Evidence
for this mechanism is based largely on the increased invasiveness
degenerative diseases.1–3
The vertebrates MMP family includes at least 26 enzymes,4
of cell lines over-expressing MMPs. More recently, it has been
among which 23 have been known in humans. They are classified
shown that MMPs can play a role in primary tumour growth.
into six groups: Collagenases: MMP1, MMP8, MMP13 and
This involves the release of stroma-bound growth factors or MMP
MMP18 (not present in human gene), Gellatinases: MMP2
mediated tumour angiogenesis. Alternatively, remodelling of the
and MMP9, Stromelysins: MMP3, MMP10 and MMP11,
extracellular matrix in the vicinity of the primary tumour may
provide the special requirements necessary for tumour growth.5
Effective MMP inhibitors (MMPIs) are characterized by: 1)
aDepartamento de Qu´ımica, Facultad de Farmacia, Universidad San Pablo
a functional group chelating the catalytic zinc ion (zinc-binding
group, ZBG); 2) one or more side chains undergoing strong van
der Waals interactions with the the enzyme subsites, mainly with
S1¢; and 3) functional groups providing additional interactions
with enzyme backbone.
Zn2+-chelating hydroxamates have been the most frequently
used ZBG in MMPI design. However, binding of MMPIs and,
most importantly, selectivity have been explained by subtle
interactions with key residues belonging to different subsites.6
This issue has been deeply and accurately addressed by the
use of computational techniques, which have pointed to the
relevance of S1¢ subsite in the regulation of MMP selectivity.7
Thus, extensive hydrophobic interactions between the lipophilic
CEU, Urbanizacio´n Montepr´ıncipe, 28668, Boadilla del Monte, Madrid,
Spain. E-mail: aramgon@ceu.es, bpaster@ceu.es; Fax: (+34)913510496;
Tel: (+34)913724796
bCenter for Biomedical Research of La Rioja (CIBIR) C/Piqueras 98,
26006, Logron˜o, Spain
cDepartment of Molecular Biology, Faculty of Mathematics and Natural
Sciences, The John Paul II Catholic University of Lublin, 20-718, Lublin,
Poland
dStructural Biology Laboratory, Medicinal Chemistry Department, Centro
de Investigacio´n Pr´ıncipe Felipe, Avda. Autopista del Saler, 16, 46012,
Valencia, Spain
† Electronic supplementary information (ESI) available: IR, 1H and 13C
NMR, MS (ESI) and microanalysis data of compounds 14–22. Evolution
of the root-mean-square deviations (RMDS) and some distances moni-
tored for the MD simulations. See DOI: 10.1039/c0ob00852d
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4587
©

View Article Online
moieties of the ligand and the S1¢ pocket are largely responsible
for the binding potencies.8 Moreover, a novel series of highly
selective MMP13 inhibitors (pyrimidine dicarboxamides)9 with
low nanomolar affinities, and a novel series of MMP3 inhibitors
(amidinobenzisothiazoles)10 have been recently reported, which
exhibit facile and extensive hydrophobic binding in the S1¢ pocket
and no interaction with the catalytic Zn2+ ion.
After prolonged research, only MMP1, 2, and 7 have been
experimentally validated as cancer targets. Inhibition of MMP1
has been hypothesized to be the cause of the clinically observed
musculoskeletal syndrome when broad spectrum inhibitors are
used.11,12 On the other hand, reduction of some MMPs such as
MMP3, and 8 could enhance tumorgenesis and metastasis, so their
inhibition should be avoided.13,14 MMP9 is a tricky enzyme since
its inhibition might be useful in treating patients with early-stage
cancers, but MMP9 is an anti-target in patients with advanced
disease. So, MMP9 inhibition should also be prevented.4
In this work we have focused on the design of selective MMP2
inhibitors. MMP2 plays an important role in cancer, among
many other processes, by stimulating tumour growth, angiogenesis
and metastasis, through its involvement in the degradation of
extracellular matrix. It is overexpressed in human tumour samples
and has been identified in association with highly invasive cells.
For all these reasons, MMP2 has been considered for many
years an important target for the design of anticancer agents. No
MMPI has been licensed for the treatment of cancer to date, due
mainly to low selectivity, although there are many compounds in
clinical development, such as the MMP2 inhibitor incyclinide, a
tetracycline derivative which has been tested for cancer metastasis
and solid tumours, among other indications, such as allergic
conditions, inflammatory, and infectious (fungal) diseases.15
Some series of a-sulfone, a-piperidine and a-tetrahydopyran
hydroxamates of general structure 1, including compounds SC-
78080/SD-2590 and SC-7774 (Scheme 1) have been described
as potent MMP2, MMP9, and MMP13 inhibitors, while sparing
MMP1.12,16–18 For this reason we have used this scaffold in the
design of new inhibitors, with the aim of increasing the selectivity
for MMP2 over MMP9, while maintaining the MMP1-sparing
characteristic.
Our approach is based on the use of click chemistry to
connect the scaffold containing the hydroxamate ZBG with
appropriate subunits designed on the basis of reported selective
hydroxamate MMP2 inhibitors (Scheme 1). The Cu(I)-catalyzed
azide-alkyne 1,3-dipolar cycloaddition (CuAAC) has been used
by several research groups for the discovery of enzyme inhibitors
of acetylcholinesterase,19–21 HIV-1 protease,22 carbonic anhydrase
II,23 protein tyrosine phosphatases (PTPs),24 and caspases.25 It
has also been used in an attempt to detect selective MMP
inhibitors26 by the rapid assembly of a library of different succinyl
hydroxamates with a series of azides, followed by in situ screen-
ing for inhibition against MMP7, thermolysin and collagenase.
However, they only identified two compounds with activity in
the micromolar range for MMP7, inactive for thermolysin and
collagenase, but without information about other MMPs.
In this work we present the design and click synthesis of a new
potent MMP2 and MMP3 inhibitor (18) that is 70-fold more active
against MMP-2 than against MMP-9, and completely inactive
against MMP1, MMP7 and MMP8. A combination of molecular
modelling studies and NMR techniques has been used to reveal the
potential interactions that govern the recognition and the binding
to MMP2 of the best inhibitors in the synthesized series, and
to rationalize the structure-affinity relationships at the molecular
level.
Results and Discussion
Chemistry
A series of P1¢-diversified MMP2 inhibitors containing the
hydroxamate ZBG (Table 1) has been synthesized, using a CuAAC
click approach, as shown in Scheme 1. The corresponding azide
contains, apart from the ZBG, an a-tetrahydro-2H-pyranyl sul-
fone group (a-THP-sulfone) capable of establishing very relevant
hydrogen bond interactions with the enzyme backbone, driving
the P1¢ group into the S1¢ pocket. A series of P1¢ groups has been
chosen from reported selective MMP2 inhibitors, and have been
introduced in the final molecule via the corresponding alkyne.
Some of the alkynes used are commercially available (Fig. 1), and
some had to be synthesized (Scheme 2).
Scheme 1 (From top to bottom.) General structure of MMP2 inhibitors
1 containing a tetrahydropyrane or piperidine heterocycle, structures
of SC-78080/SD-2590 and SC-77774, and general structure of the
synthesized analogues based on a click synthetic approach.
Fig. 1 Commercially available alkynes used in this work.
4588 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Calculated IC50 inhibitory concentrations towards MMP2 and MMP9 for compounds 12–22 and BiPS
Zymogram MMP2
IC50, nM
Zymogram
MMP9 IC50, nM
Selectivity
MMP9/MMP2
Compound
P1¢
13
10 ¥ 10-3
2.4 ¥ 10-3
7.5 ¥ 10-3
6.8
32.7 ¥ 10-3
11.1 ¥ 10-3
13.6 ¥ 10-3
3.3
4.5
1.8
—
14
15
16
17
18
19
a
—
1.0 ¥ 10-2
1.9 ¥ 10-3
9.5 ¥ 10-3
3.1 ¥ 10-2
18.8 ¥ 10-3
51.5 ¥ 10-3
3.0
9.7
5.4
20
21
CH3(CH2)3–
53 ¥ 10-3
127.5 ¥ 10-3
>50.00
2.4
—
1.1
22
1.2
>50.00
—
0.68
4.4
5.9
22.3 ¥ 10-3
53.4 ¥ 10-3
2.4
a MMP9 data behaved erratically and it was not possible to obtain a specific IC50 value. Nevertheless, MMP9 inhibition was lower than the one elicited
for MMP2, as shown in Fig. 6A
The detailed synthesis of the ZBG containing azide is shown
in Scheme 3. Alkylation of 4-nitrobenzenethiol with methyl 2-
bromoacetate gave sulfide 5, which was oxidized with oxone
to give sulfone 6. The acidic methylene was dialkylated with
bis(2-bromoethyl)ether to give sulfone 7. Saponification of the
methyl ester with a H2O/THF solution of sodium hydroxide and
subsequent catalytic hydrogenation of the nitro group in 8 gave
amino acid 9. This was transformed into azide 10 by reaction
with tert-butyl nitrite and azidotrimethylsilane. Coupling of the
carboxylic acid with O-(tetrahydro-2H-pyranyl)hydroxylamine
(THP-hydroxylamine) using EDC as amide coupling agent16 gave
the THP-protected hydroxamate 11, which was subsequently
deprotected with a solution of HCl in dioxane to give the desired
azide 12.
For the CuAAC reaction, we used the Cu(II)/ascorbate system,
and a variety of conditions were investigated. The use of catalytic
amounts of CuSO4 (0,25–2 mol%) in t-BuOH/H2O, and ascorbic
acid (5–10 mol%) brought about a very slow transformation of
the starting material, without apparent progress of the reaction,
even after 3 days. Consequently, the amount of reactants was
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4589
©

View Article Online
only one isomer in all CuAAC reactions, which in the case of
triazole 14 has been characterized as the 1,4-disubstituted isomer
by 1H-NMR (nOe) experiments (see the Experimental Section).
Molecular Modeling
For the designed compounds shown in Table 1, we first evaluated
the suitability to act as MMP2 inhibitors by means of docking
techniques. There are not many experimental 3D structures of
MMP2 available, PDB 1hov being the only complex among
MMP2 catalytic domain and an inhibitor, hydroxamate i52
(Fig. 2). Therefore, this structure was considered to perform
docking studies. PDB 1hov is an NMR structure composed of 11
models, and the superimposition of all of them showed no relevant
changes around the ligand binding region. Previous studies of the
binding mode of a set of putative MMP2 inhibitors, including i52,
showed no difference in the docking results performed on the 11
models, so we considered only model 1 to carry out the docking
studies.28
Scheme 2 Synthesis of alkynes 2–4.
Scheme 3 Synthesis of azide 12. (a) methyl 2-bromoacetate, K2CO3,
DMF, 98%; (b) oxone, MeOH, H2O, 90%; (c) O(CH2CH2Br)2, K2CO3,
DMAP, NBu4I, DMF, 76%; (d) NaOH, H2O/THF, 90%; (e) H2, Pd/C,
95%; (f) t-BuONO, TMSN3, 94%;(g) NH2OTHP, EDC, HOBT, DMAP,
94%; (h) HCl, dioxane, 91%.
Fig. 2 Chemical structure of compound i52 showing the interactions with
the main subsites of MMP2 active site.
progressively increased, and the best results were obtained when
0.5 equivalents of CuSO4 and two equivalents of ascorbic acid were
used, leading to compounds 13–22 with moderate yields. Other
procedures have been developed where the catalyst is introduced
in the form of Cu(0),27 or CuI.25 Attempts to carry out the
click reaction between azide 12 and phenylacetylene using these
methodologies have been unsuccessful, and the starting material
was recovered unreacted, even after five days of reaction.
Compounds were docked into MMP2 using different ap-
proaches (see the Experimental Section), combining rigid and
flexible docking procedures and, in some cases, the use of two
docking programs, AutoDock and Glide. The predicted binding
modes for azide 12, and triazoles 13, 14, 15 (which bears the P1¢
group present in compound i52), 18 and 19, were very similar
to that of i52, that is, with the two hydroxamate oxygen atoms
coordinating the catalytic Zn2+ ion, in a divalent fashion, adopting
a distorted trigonal bipyramidal geometry, together with the
P1¢ side chain inside the S1¢ binding pocket, and in complete
agreement with the experimental data relative to the MMP2-i52
The Cu(I)-catalyzed Huisgen 1,3-dipolar cycloaddition of azides
and terminal alkynes is a regioselective process, forming exclu-
sively the 1,4-substituted product. Accordingly, we have obtained
4590 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
complex. Final end of the P1¢ chain occupies a hydrophobic pocket
defined by Phe148, Phe115, Leu150 and Thr145, establishing
efficient hydrophobic interactions. These interactions are of great
relevance for selectivity as Thr145 is an exclusive residue for
gelatinases, while Phe148 makes MMP2 distinct from MMP9.
The sulfone group perfectly mimics the sulfonamide group from
i52, establishing hydrogen bonds with the NH groups of Leu83
(S3¢) and Ala84 (S1 pocket) (Fig. 3). Ala84 is highly conserved in
MMPs and this interaction provides a key anchorage point to the
active site.
Fig. 4 Chemical structure of compound 23.29
As it will be discussed below, compounds 18 and 19 showed the
best profiles of MMP2 inhibition and MMP2 selectivity. Thus, we
were prompted to carry out molecular dynamics (MD) simulations
of the complexes obtained from the docking studies to validate the
proposed binding modes.
Accordingly, MD simulations of both MMP2-18 and MMP2-19
complexes (3 ns) were performed. Interactions of metalloproteins
with ligands are often difficult to describe effectively due to the
limited availability of appropriate force fields. For the treatment
of zinc, we have successfully followed the “cationic dummy atom”
approach used to impose orientational requirements for zinc
ligands, while still allowing flexibility.30 Globally, all the MMP2-
ligand interactions were in agreement with the observations from
the docking studies (Fig. 3 and Fig. 5).
Fig. 3 The binding modes of compound 18 in the active site of MMP2
obtained from docking (grey), and MD simulation (green, minimized
average structure for the 2 ns–2.5 ns interval of the MD simulation).
Exceptionally, for compound 17, the predicted binding pose
did not show interaction between the hydroxamate group and the
catalytic Zn2+ ion. The shorter length of the P1¢ group allows a
deep entrance to the end of the S1¢ pocket, so the hydroxamate
group establishes a hydrogen bond with NH from Leu83 instead.
For compound 22, which is one carbon longer than 13, some
docked orientations were predicted, but none of the docking
procedures placed the side chain inside the S1¢ pocket. Also,
compound 16, which is one carbon longer than 15, showed
predicted binding poses with the P1¢ group inside the S1¢, but
none of them with Zn2+ coordination. These results suggest that
the introduction of a methylene group between the triazole and
the phenyl ring in the P1¢ group brings about a loss of activity,
most probably related to an increase in flexibility that does not
allow proper binding to MMP2.
Compound 21 was designed based on sulfonylurea derivative 23
(Fig. 4), one of the most selective and potent MMP2 hydroxamate
inhibitors found in the literature.29 For this compound AutoDock
was not able to find any solution within the binding site and Glide
predicted one pose in which the chain is not able to fill the S1¢
pocket. The energetically stable conformations of the sulfonylurea
group orient the final phenyl ring towards a neighbour pocket
defined by the alkyl sp3 carbons of Arg149 and Leu137, but
without establishing the efficient hydrophobic interactions that
characterize the recognition inside the S1¢ pocket.
Fig. 5 Minimized average structure for the 2 ns–2.5 ns interval of the MD
simulation for the MMP2-19 complex.
The P1¢ chain occupies the S1¢ pocket lined with hydrophobic
residues Phe148, Phe115, Leu150 and Thr145. Also, it is worth
mentioning that the sulfone group establishes stable hydrogen
bonds with NH groups from Leu83 and Ala84. In the case of
19, the P2¢ phenyl ring is stacked with His120, while in the case
of 18, the triazol ring is the one which is interacting with His120.
These interactions contribute to anchoring the inhibitor within the
MMP2 active site and, simultaneously, result in a good bidentated
orientation of the hydroxamate group towards the zinc atom.
This divalent coordination is maintained along the simulations
supporting the proposed binding mode for this novel class of
MMP2 ligands (RMSD and monitored distances are available
at the ESI†).
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4591
©

View Article Online
Table 2 Inhibitory activity of compounds 18 and 19 towards selected
MMPs
Biological Evaluation
Zymography. Zimography is a widely used technique to study
extracellular matrix-degrading enzymes, such as MMPs, from
tissue extracts, cell cultures, serum and urine.31 It is a simple
and very sensitive technique, with a detection limit for MMP2
in gelatin zymography of 10 pg, and able to identify MMPs by
the degradation of their substrate. Therefore, zimography allows
the detection of highly potent inhibitors, with IC50 in the low
picomolar range.
Inhibition of MMP2 and MMP9 activities by increasing
concentrations of MMPIs was measured by gelatin zymography
using human serum as the source of gelatinases. The synthe-
sized compounds were added to the digestion buffer during
an overnight incubation, and zymogram bands were analyzed
by densitometry. Their inhibitory activity was compared to
the one elicited by 2R-[(4-biphenylsulfonyl)amino]-N-hydroxy-3-
phenylpropinamide (BiPS), a commercially available MMP2/9
inhibitor (Table 1). Representative zymograms showing the dose-
dependent inhibition of MMP2 and MMP9 activities elicited by
compounds 16 and 18 are represented in Fig. 6.
Compound 18
Compound 19
IC50, nMa
IC50, nMa
MMP1
MMP2
MMP3
MMP7
MMP8
MMP9
MMP10
MMP12
MMP13
MMP14
>1000
1.4
17.2
>100
>100
98
51
3.2
0.9
65
>1000
0.3
9.6
70
6.1
11.3
7.8
1.2
1.4
8.2
a Concentration required for 50% inhibition of enzyme activity. Standard
deviations are less than 10% of mean values in three experiments
12. These results are in agreement with the docking studies that
predicted that these compounds are not able to establish efficient
hydrophobic interactions inside the S1¢ pocket (21 and 22) or only
in orientations where the hydroxamate group is not coordinating
the Zn2+ ion (16).
Specific MMP2 IC50 values of all compounds tested were lower
than the corresponding values for MMP9, putting forward a
promising MMP2/MMP9 selectivity profile. Taking into account
these interesting results, we then proceeded to select the most active
and selective compounds (18 and 19) in order to submit them for
a complete MMP inhibitory profile.
MMPs inhibition profile
Compounds 18 and 19 were selected to carry out studies of
selectivity in a panel of ten different MMPs to obtain a complete
inhibitory profile. The two compounds were tested in a MMP
Inhibitor Profiling Kit (Enzo Life Science, Inc), which allows the
quantitative measurement of MMP inhibition and was used as
described in the Experimental Section.
Table 2 reports MMP inhibitory activities of these compounds
as IC50 values. Both 18 and 19 show potency as MMP2 and
MMP13 inhibitors in the low nanomolar range. Compound 18
showed an IC50 = 1.4 nM towards MMP2, and IC50 = 0.9 nM
towards MMP13, and selectivity over MMP1 and MMP7. The
negligible activity of these inhibitors against MMP-1 and MMP-
7 is not unexpected. They have been designed to interact with
MMP-2, an enzyme characterized by a large adaptable S1¢
pocket, while MMP-1 and MMP-7 display short and narrow S1¢
pockets.33,34 Moreover, 18 is also selective versus MMP8 and shows
a reduced activity against MMP14 (IC50 = 65 nM), and MMP9
(IC50 = 98 nM), other MMPs characterized by having a deep S1¢
pocket and, therefore, more similar to MMP2. The introduction of
the rigid biphenyl moiety attached to the triazole subunit, together
with the proper length and hydrophobic character of P1¢ may be
favouring the interaction with the deep hydrophobic S1¢ pocket,
leading to high inhibitory potency and selectivity for MMP2 and
MMP13.
Fig. 6 Representative zymograms showing the dose-dependent inhibition
of MMP2 and MMP9 activities elicited by compounds 16 (A) and 18 (B).
Bands were quantified by densitometry and graphically represented to
obtain the IC50 value for each compound and MMP type. For compound
18 (D), the IC50 for both MMPs (green diamonds for MMP2, blue dots for
MMP9) were calculated, whereas for 16 (C) the values for MMP9 (blue
dots) were too erratic and did not allow for a good calculation of the IC50
value and only the IC50 for MMP2 (green triangles) could be calculated.
Using this methodology all compounds tested showed sig-
nificant inhibitory activity on MMP2 and MMP9. Our results
demonstrate that compounds 13–15, 17–19 inhibit MMP2 with an
IC50 in the picomolar range, values which are remarkably higher
than those found for BiPS in the same experiment (Table 1).32
For azide 12, where the side chain is absent, the inhibition
value achieved is three orders of magnitude lower, showing the
key role that this group is having in the molecular recognition
process. Compounds 16, 21 and 22 exhibited lower inhibition
values than the rest of the triazoles, with IC50 in the nanomolar
range, and of the same order of magnitude as the value for azide
Compound 19, where the rigid biphenyl has been replaced by
the more flexible p-methoxyphenyl unit, also exhibits an increased
potency with IC50 values for MMP2 of 0.3 nM. However, while
selectivity versus MMP1 is similar to that shown for 18, the
selectivity versus the rest of the MMPs is diminished.
4592 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
NMR studies
These experimental results are in complete agreement with our
docking predictions, demonstrating the appropriate selection of
the azide fragment to carry out fragment-based drug design.
Attachment of carefully selected alkynes has led to compounds
13–22 which were also submitted to the same NMR experiments.
Compounds 20 and 21 showed positive results in the interaction
with MMP2 and in the competition experiments. Compounds
14, 17, 19 and 22 showed positive results in the NMR binding
experiments, although their competition assays did not provide
conclusive results due to the low solubility of the compounds in
water. The rest of the molecules could not be assayed by NMR,
not even after the addition of up to 10% DMSO to improve the
solubility of the sample.
STD experiments can provide structural information by identi-
fying the epitope of the molecule interacting with the protein, as
its signals will see a relatively larger increase in intensity compared
to those signals belonging to a region of the molecule not directly
involved in the binding interaction. However, from the analysis of
the interaction experiments of 12, 20 and 21, we could not find a
region showing a binding preference over the rest.
NMR is a very powerful biophysical technique to study protein-
ligand interactions and, in addition to the analysis of binding
events, it can provide relevant structural information. Also, NMR
is very sensitive when analysing weak intermolecular interactions.
In the current study, two complementary ligand-based NMR
techniques covering a wide range of binding affinities, water-
LOGSY35 and STD (Saturation Transfer Difference)36 have been
used. These NMR experiments are designed to evaluate the effect
on the ligand upon binding to the protein, which is not required in
large amounts nor isotopically labelled. To improve the stability
of the recombinantly expressed catalytic domain of MMP237 and
prevent auto-proteolytic activity, it was necessary to mutate the
active site glutamic acid to glutamine, as shown by Rowsell et al.38
This conservative mutation does not affect inhibitor binding in the
active site.
All compounds shown in Table 1 were tested for binding to
MMP2 by NMR, and competition experiments were recorded
using BiPS, as a positive control. Fig. 7 shows the water-LOGSY,
STD and competition-STD spectra acquired for the azide frag-
ment 12 containing the hydroxamate ZBG. Both water-LOGSY
and STD yield unambiguous positive results. The decrease in the
intensity of the signals for 12 in the STD experiment acquired in
the presence of BiPS inhibitor, indicates that both molecules bind
to the same site on MMP2.
Conclusions
A series of a-sulfone, a-tetrahydropyrane hydroxamates was
designed based on data from the literature about selective
MMP2/MMP9 inhibitors, and exploiting the potential of click
chemistry in a fragment based drug design approach. One of the
designed fragments contains a hydroxamate ZBG, and its ability
to bind the active site of the enzyme was demonstrated by water-
LOGSY and STD NMR experiments. A CuAAC reaction was
used to connect the ZBG fragment bearing an azide to several
carefully selected hydrophobic alkynes. These alkynes were chosen
to selectively interact with the hydrophobic S1¢ pocket of MMP2.
The obtained compounds were evaluated as MMP2 and MMP9
inhibitors in a gelatin zymography using human serum as the
source of gelatinases and in a colorimetric assay against a panel of
ten MMPs. Among the novel inhibitors, compound 18 presented
a promising profile, with low nanomolar activity against MMP2
and MMP13, negligible activity towards MMP1 and MMP7, and
a 70-fold higher activity compared to MMP9. It also showed
lower activity against MMP10 and MMP14. Docking, molecular
dynamics simulations and NMR experiments have been used to
analyze the mode of binding and to rationalize the structure-
affinity relationships at the molecular level.
The described fragment-based design has proved to be a useful
method to detect highly MMP2 active inhibitors, selective over
MMP1, with a promising selectivity profile against some antitarget
metalloproteinases, such as MMP8, and considerably less activity
against MMP9. This approach, together with the information
obtained from the computational study, is currently being used
in the design of new inhibitors making use of different ZBGs. The
lack of water solubility of these compounds could preclude their
use in clinics. Current work is directed towards the introduction
of basic groups which could improve their pharmacokinetic
properties.
Fig. 7 NMR binding experiments of MMP2 with fragment 12. (a)
Aromatic proton region of the monodimensional 1H NMR of MMP2
(5 mM) and 12 (100 mM). At this protein concentration the signals from
MMP2 are not observed. (b) Water-LOGSY and (c) STD experiments
show positive signals from the ligand, thus proving the interaction between
the protein and 12. (d) Competition STD experiment after the addition
of 200 mM BiPS. The reduction of the signals compared to spectrum (c)
indicates that both 12 and BiPS bind to the same MMP2 site. The extra
signals on the spectrum correspond to the BiPS molecule.
Additionally, this approach is especially attractive as it permits
the design and efficient synthesis of inhibitors with improved
selectivity for one or another MMP by a careful selection of the
appropriate alkynes.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4593
©

View Article Online
added. The mixture was extracted with Et2O, the organic extracts
were washed with water and brine, dried (MgSO4) and evaporated
to dryness. The residue was purified by column chromatography
on silica gel, using hexane as eluent to afford 3 as a colourless oil
(430 mg, 97%) which decomposes on standing. nmax/cm-1 2115; dH
(CDCl3) 0.92 (3H, t, J 6.4, CH3), 1.29–1.36 (4H, m, CH2CH2CH3),
1.56–1.67 (2H, m, ArCH2CH2), 2.20 (1H, t, J 2.9, C CH), 2.61
(2H, t, J 7.3, ArCH2CH2), 3.60 (2H, d, J 2.9, C C–CH2), 7.16
(2H, d, J 7.8, ArH), 7.29 (2H, d, J 7.8, ArH); dC (CDCl3) 14.0,
22.5, 24.3, 31.2, 31.4, 35.4, 70.2, 82.1, 127.6, 128.5, 133.1, 141.2.
Experimental Section
Chemistry
General procedures. Melting points (uncorrected) were de-
termined on a Stuart Scientific SMP3 apparatus. Infrared (IR)
spectra were recorded with a Perkin-Elmer 1330 infrared spec-
trophotometer. 1H and 13C NMR data were recorded on a Bruker
300-AC instrument. Chemical shifts (d) are expressed in parts per
million relative to internal tetramethylsilane; coupling constants
(J) are in hertz. Mass spectra were run on a Bruker Esquire 3000
spectrometer. Elemental analyses (C, H, N) were performed on a
LECO CHNS-932 apparatus at the Microanalyses Service of the
University Complutense of Madrid; unless otherwise stated, all
reported values are within 0.4% of the theoretical compositions.
Thin-layer chromatography (TLC) was run on Merck silica gel 60
F-254 plates. Unless stated otherwise, starting materials used were
high-grade commercial products.
1-Phenoxy-4-(prop-2-yn-1-yl)benzene (4). The procedure de-
scribed above was used for the synthesis of 4. From 1-
bromo-4-phenoxybenzene (4,00 g, 16,06 mmol), magnesium turn-
ings (0.42 g, 17.29 mmol), iodine (10 mg) and (3-bromo-1-
propynyl)trimethylsilane (2.36 g, 12.35 mmol), and after chro-
matography using hexane/DCM (95 : 5) as eluent trimethyl-[3-(4-
phenoxyphenyl)prop-1-ynyl]silane was produced as a colourless
oil (1.15 g, 33%), which decomposes on standing. nmax/cm-1 2160;
N-(Prop-2-yn-1-ylcarbamoyl)benzenesulfonamide
(2).
d
H (CDCl3) 0.28 (9H, s, Si(CH3)3), 3.71 (2H, s, C C–CH2), 7.03–
Benzenesulfonyl isocyanate (5.09 g, 26.38 mmol) was added
to a solution of propargylamine (1.48 g, 26.38 mmol) in
anhydrous CH3CN (40 cm3) under argon. After stirring at room
temperature overnight, the reaction mixture was concentrated
under vacuum, and the obtained solid was recrystallized from
AcOEt to give 2 (5.47 g, 87%) as a white solid, mp 161.8–163.6 ◦C.
nmax (KBr)/cm-1 3354, 3269, 2125, 1705, 1667; dH (DMSO-d6)
3.10 (1H, t, J 2.4, C CH), 3.74 (2H, dd, J 2.4 and 5.5, CH2),
6.92 (1H, t, J 5.49, NH-CH2), 7.59–7.71 (3H, m, ArH), 7.89–7.92
(2H, m, ArH), 10.9 (1H, br s, SO2NH); dC (DMSO-d6) 28.9, 73.2,
81.1, 127.3, 129.1, 133.3, 140.2, 151.3. MS (ESI): m/z 260.96
[M+Na]+.
7.09 (4H, m, ArH), 7.13–7.19 (1H, m, ArH), 7.36–7.39 (4H, m,
ArH); dC (CDCl3) 0.1, 25.4, 86.9, 104.3, 118.6, 119.0, 123.1, 129.1,
129.7, 131.1, 155.8, 157.3.
From
trimethyl-[3-(4-phenoxyphenyl)prop-1-ynyl]silane
(1,15 g, 4,10 mmol), AgNO3 (1.05 g, 6.15 mmol) and KCN
(2,67 g, 41,01 mmol) and after column chromatography using
hexane/DCM (9 : 1) as eluent, 4 was produced as a colourless
oil (734 mg, 86%) which decomposes on standing. dH (CDCl3)
2.22 (1H, t, J 2.9, CH), 3.61 (2H, d, J 2.9, CH2), 6.97–7.04 (4H,
m, ArH), 7.08–7.14 (1H, m, ArH), 7.31–7.36 (4H, m, ArH); dC
(CDCl3) 24.0, 70.5, 81.2, 118.6, 119.1, 123.1, 129.1, 129.7, 130.9,
155.8, 157.3.
1-Pentyl-4-(prop-2-yn-1-yl)benzene (3). A portion (1 cm3) of
a solution of 1-bromo-4-pentylbenzene (3.00 g, 12.81 mmol) in
anhydrous THF (10 cm3) was added to a mixture of magnesium
turnings (0.34 g, 14.04 mmol) and iodine (10 mg) in anhydrous
THF (2 cm3) under argon. When the Grignard reaction began,
and after addition of 10 cm3 of dry THF, the remaining solution
of aryl bromide was added dropwise. The reaction mixture was
refluxed for 2 h. After cooling to room temperature, the rest of the
magnesium was removed by filtration via cannula. To the resulting
solution was added (3-bromoprop-1-ynyl)trimethylsilane (2.00 g,
10.25 mmol) and the mixture was refluxed for 12 h, cooled and
concentrated under vacuum. A solution of 1 N HCl (50 cm3)
was added and the crude was extracted with Et2O. The organic
extracts were dried (MgSO4), filtered and evaporated to dryness.
The residue was chromatographed on silica gel using hexane as
eluent to obtain trimethyl-[3-(4-pentylphenyl)prop-1-ynyl]silane
as a colourless oil (708 mg, 21%). nmax/cm-1 2165; dH (CDCl3)
0.20 (9H, s, Si(CH3)3), 0.90 (3H, t, J 6.8, CH3), 1.29–1.38 (4H, m,
CH2CH2CH3), 1.56–1.66 (2H, m, ArCH2CH2), 2.60 (2H, t, J 7.8,
ArCH2CH2), 3.64 (2H, s, C C–CH2), 7.15 (2H, d, J 8.3, ArH),
7.26 (2H, d, J 8.3, ArH); dC (CDCl3) 0.1, 14.0, 22.6, 25.7, 31.3,
31.5, 35.5, 86.5, 104.6, 127.7, 128.5, 133.4, 141.1.
Methyl [(4-nitrophenyl)sulfanyl]acetate (5). K2CO3 (5.39 g,
39.02 mmol) was added to a solution of◦ 4-nitrobenzenethiol
(5.05 g, 26.01 mmol) in DMF (50 cm3) at 0 C. The reaction was
◦
stirred for 15 min at 0 C and methyl bromoacetate (2.65 cm3,
28.62 mmol) was added. The reaction mixture was stirred
overnight at room temperature and then diluted with ethyl acetate
(200 cm3) and washed successively with saturated aqueous NH4Cl
and brine. The extract was dried (MgSO4), filtered and evaporated
to dryness, and the residue which was chromatographed on silica
gel (hexane/AcOEt 85 : 15) to give 5 (5.80 g, 98%) as a white solid,
◦
mp 69.8–71.2 C (EtOH). (Found: C, 47.43; H, 3.93; N, 6.78; S,
14.30. C9H9NO4S requires C, 47.57; H, 3.99; N, 6.16; S, 14.11%);
n
max (KBr)/cm-1 1724; dH (CDCl3) 3.78 (3H, s, OCH3), 3.80 (2H,s,
CH2), 7.43 (2H, d, J = 8.8, ArH), 8.16 (2H, d, J = 8.8, ArH); dC
(CDCl3) 34.2, 53.0, 124.0, 126.6, 145.3, 145.6, 169.1; MS (ESI):
m/z 249.93 [M+Na]+.
R
Methyl [(4-nitrophenyl)sulfonyl]acetate (6). Oxoneꢀ (potas-
sium hydrogen persulfate) (38.69 g 62.93 mmol) was added to
a solution of 5 (5.72 g, 25.17 mmol) in MeOH (50 cm3) and water
(5 cm3) at 0 ◦C. After stirring at room temperature for 2 h, the
crude reaction was filtered and the solid washed with MeOH.
The solution was concentrated in vacuo, solved in AcOEt, washed
with saturated aqueous NaHCO3 and brine. The extract was dried
(MgSO4), filtered and evaporated to dryness, to give 6 (5.84 g,
A solution of silver nitrate (609 mg, 3,59 mmol) in water (3 cm3)
and EtOH (7 cm3) was added dropwise to a solution of the TMS-
protected alkyne (618 mg, 2.39 mmol) in EtOH (10 cm3). The
solution was stirred at room temperature for 30 min and then, a
solution of and KCN (1,58 g, 23,91 mmol) in water (3 cm3) was
◦
90%) as a yellowish solid, mp 130.3–131.8 C (AcOEt). (Found:
C, 41.74; H, 3.50; N, 5.81; S, 12.47. C9H9NO6S requires C, 41.70;
4594 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
H, 3.54; N, 5.40; S, 12.37%); nmax (KBr)/cm-1 1740; dH (CDCl3)
3.75 (3H, s, OCH3), 4.21 (2H, s, CH2), 8.18 (2H, d, J 8.8, ArH),
8.46 (2H, d, J 8.8, ArH); dC (CDCl3) 53.4, 60.4, 124.4, 130.3, 143.9,
151.1, 162.4; MS (ESI): m/z 281.95 [M+Na]+.
(DCM/MeOH 97.5 : 2.5) to give 10 (2.04 g, 94%) as a yellowish
solid, mp 191.4–192.4 ◦C (dec). (Found: C, 46.47; H, 4.24; N,
13.44; S, 10.28. C12H13N3O5S requires C, 46.30; H, 4.21; N, 13.50;
S, 10.30%); nmax (KBr)/cm-1 3443, 2124, 2099, 1735; dH (DMSO-
d6) 1.90–2.07 (4H, m, 2CH2), 3.16 (2H, t, J 12.2 Hz, 2OCHH),
3.93 (2H, dd, J 12.2 and 3.7, 2OCHH), 7.37 (2H, d, J 8.6, ArH),
7.77 (2H, d, J 8.6, ArH); dC (DMSO-d6) 28.0, 64.2, 70.9, 119.7,
130.5, 132.1, 146.1, 167.7; MS (ESI): m/z 333.98 [M+Na]+.
Methyl 4-[(4-nitrophenyl)sulfonyl]tetrahydro-2H -pyran-4-car-
boxylate (7). K2CO3 (4.00 g, 28.93 mmol) was added to a solution
of 6 (3.00 g, 11.57 mmol) in DMF (30 cm3) at 0 ◦C. The reaction
mixture was stirred for 15 min at room temperature and bis-(2-
bromoethyl)ether (2.95 g,12.73 mmol), DMAP (85 mg, 0.69 mmol)
and tetra-n-butylammonium iodide (256 mg, 0.69 mmol) were
added. Then the mixture was stirred at room temperature 24 h and
poured into 1 N HCl (100 cm3). The solid obtained was filtered,
washed with hexane to give 7 as a pale yellow solid (4.66 g, 76%)
after recrystallization, mp 192.8–194.6 ◦C (AcOEt). (Found: C,
47.47; H, 4.56; N, 4.51; S, 9.66. C13H15NO7S requires C 47.41; H,
4.59; N, 4.25; S, 9.74%); nmax (KBr)/cm-1 3104, 2967, 2867, 1739;
dH (CDCl3) 2.20–2.34 (4H, m, 2CH2), 3.29 (2H, td, J 12.2, 3.4,
2OCHH), 3.81 (3H, s, CH3), 4.01–4.07 (2H, m, 2OCHH), 8.00
(2H, d, J 8.8, ArH), 8.42 (2H, d, J 8.8 Hz, ArH); dC (CDCl3) 28.0,
53.6, 64.6, 72.2, 123.9, 131.7, 140.6, 151.2, 167.2; MS (ESI): m/z
352.06 [M+Na]+.
4-[(4-Azidophenyl)sulfonyl]-N-(tetrahydro-2H-pyran-2-yloxy)-
tetrahydro-2H-pyran-4-carboxamide (11). HOBT (859 mg,
6.36 mmol), NMM (1.75 cm3, 15.90 mmol), O-tetrahydro-2H-
pyran-2-yl-hydroxylamine (1.24 g, 10.60 mmol) and EDCI (1.42 g,
7.42 mmol) were added to a solution of 10 (1.65 g, 5.30 mmol) in
DMF (20 cm3). The reaction mixture was stirred overnight at room
temperature and then diluted with AcOEt (60 cm3) and washed
successively with a saturated aqueous solution of NH4Cl and brine.
The organic layer was dried (MgSO4), evaporated to dryness and
the solid obtained was recrystallized from EtOH and water to give
11 as an orange solid (1.97 g, 91%), mp 194.2–195.8 ◦C. (Found:
C, 49.83; H, 5.39; N, 13.58; S, 7.82. C17H22N4O6S requires C, 49.75;
H, 5.40; N, 13.65; S, 7.81%); nmax (KBr)/cm-1 3323, 2139, 1693; dH
(CDCl3) 1.61–1.91 (6H, m, 3CH2), 2.08 (2H, d, CH2) 2.28 (2H, td
J 13.2 and 4.9, CH2), 3.46–3.54 (2H, “t”, CH2O) 3.70–3.75 (1H,
m, 1/2CH2O), 3.97–4.05 (3H, m, 3/2 CH2O), 5.01 (1H, br s, CH),
7.37 (2H, d, J 8.8, ArH), 7.85 (2H, d, J 8.8, ArH), 9.42 (1H, s,
NH); dC (CDCl3) 18.2, 24.9, 27.7, 28.3, 28.5, 62.3, 64.2, 64.3, 70.0,
101.9, 119.6, 129.6, 132.0, 147.0, 163.9; MS (ESI): m/z 433.12
[M+Na]+.
4-[(4-Nitrophenyl)sulfonyl]tetrahydro-2H -pyran-4-carboxylic
acid (8). 1 N NaOH (120 cm3) was added to a solution of 7
(3.94 g, 11.96 mmol) in THF (40 cm3) and the mixture was stirred
at room temperature for 4 h. The crude reaction was concentrated,
suspended in H2O and washed with AcOEt. The aqueous layer
was acidified to pH 2 with HCl 3 N and extracted with AcOEt.
The organic layer was dried (MgSO4), filtered and evaporated to
dryness to give 8 as a white solid (3.40 g, 90%), mp 239.3–240.3 ◦C
(dec). (Found: C, 45.75; H, 4.16; N, 4.72; S, 10.16. C12H13NO7S
requires C, 45.71; H, 4.16; N, 4.44; S, 10.17%); nmax (KBr)/cm-1
3445, 1732; dH (DMSO-d6) 1.96–2.09 (4H, m, 2CH2), 3.14–3.23
(2H, m, CH2O), 3.92–3.97 (2H, m, CH2O), 8.07 (2H, d, J 9.1,
ArH), 8.46 (2H, d, J 9.1, ArH); dC (DMSO-d6) 27.9, 64.1, 71.5,
124.3, 132.1, 140.2, 151.1, 167.5; MS (ESI): m/z 338.03 [M+Na]+.
4-[(4-Azidophenyl)sulfonyl]-N-hydroxytetrahydro-2H-pyran-4-
carboxamide (12). To a solution of 11 (1.92 g, 4.69 mmol) in
dioxane (10 cm3) was added 4 N HCl in dioxane (5.86 cm3,
23.43 mmol) followed by MeOH (10 cm3). After stirring at room
temperature for 2 h, the reaction mixture was concentrated under
vacuum and the obtained solid was recrystallized from EtOH and
water to give 12 (1.36 g, 89%) as an orange solid, mp 197.2–
198.1 ◦C (dec). (Found: C, 44.37; H, 4.47; N, 16.80; S, 9.70.
C12H14N4O5S requires C, 44.17; H, 4.32; N, 17.17; S, 9.83%); nmax
(KBr)/cm-1 3349, 3194, 2142, 1681; dH (DMSO-d6) 1.90 (2H, td,
J 12.8 and 4.3, 2CHH), 2.20 (2H, d, J 12.8 H, 2CHH), 3.14 (2H,
t, J 11.6, 2OCHH), 3.88 (2H, dd, J 11.6 and 3.7, 2OCHH), 7.35
(2H, d, J 8.6, ArH), 7.77 (2H, d, J 8.6, ArH), 9.23 (1H, s, NH),
11.03 (1H, s, OH); dC (DMSO-d6) 27.4, 63.9, 69.3, 119.6, 130.3,
132.1, 145.9, 160.6; MS (ESI): m/z 349.06 [M+Na]+.
4-[(4-aminophenyl)sulfonyl]tetrahydro-2H -pyran-4-carboxylic
acid (9). A solution of 8 (3.94 g, 12.50 mmol) in EtOH (150 cm3)
and H2O (50 cm3) and 10% Pd/C (250 mg) were introduced into a
Parr shaker apparatus, and maintained under a hydrogen pressure
of 60 p.s.i for 5 h at room temperature. Palladium was filtered off
and the solvent was removed to obtain 9 as a white solid (3.40 g,
◦
95%), mp 227.6–228.8 C. (Found: C, 50.36; H, 5.25; N, 5.05; S,
11.11. C12H15NO5S requires C, 50.52; H, 5.30; N, 4.91; S, 11.24%);
n
max (KBr)/cm-1 3444, 3363, 3298, 1735; dH (DMSO-d6) 1.82 (2H,
General procedure for the preparation of triazoles. To a suspen-
sion, of azide 12 (1 equiv) and the corresponding alkyne (1.2 equiv.)
in t-BuOH/H2O (1 : 1,5) was added sodium ascorbate (2 equiv. of
freshly prepared 1 M solution in water) and copper(II) sulfate
pentahydrate (0.5 equiv. of a 0.25 M solution in water) under
argon. The mixture was stirred vigorously overnight, and then
diluted with water (20 cm3) and ice. The precipitate formed was
collected by filtration and washed with cold water and hexane.
The solid was solved in a DCM : MeOH : NH3 (aqueous) 6 : 3 : 1
mixture, filtered through silica gel and the filtrate was concentrated
under vacuum.
td, J 12.8 and 4.9, 2CHH), 2.04 (2H, J 12.8, 2CHH), 3.15 (2H, t,
J 11.6, 2OCHH), 3.87 (2H, dd, J 11.6 and 3.7 Hz, 2OCHH), 6.25
(2H, NH2), 6.61 (2H, d, J 8.6, ArH), 8.46 (2H, d, J 8.6, ArH); dC
(DMSO-d6) 28.4, 64.3, 70.4, 112.4, 118.7, 131.9, 154.3, 168.1; MS
(ESI): m/z 308.06 [M+Na]+.
4-[(4-Azidophenyl)sulfonyl]tetrahydro-2H -pyran-4-carboxylic
acid (10). Compound 9 (1.99 g, 6.98 mmol) was dissolved in
anhydrous CH3CN (50 cm3) under argon and cooled to 0 ◦C
in an ice bath. To this solution was added t-BuONO (1.24 cm3,
10.47 mmol), followed by TMSN3 (1.10 cm3, 8.38 mmol) dropwise.
The reaction mixture was stirred at 0 ◦C for 30 min, and at room
temperature for 3 h. Then, it was evaporated to dryness, and
the residue was purified by column chromatography on silica gel
N-Hydroxy-4-({[4-(4-phenoxyphenyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (13). From
12 (57.7 mg, 0.18 mmol), 1-ethynyl-4-phenoxybencene (41.2 mg,
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4595
©

View Article Online
0.21 mmol), sodium ascorbate (354 mL, 0.35 mmol) and copper(II)
sulfate pentahydrate (354 mL, 0.09 mmol) 13 (60 mg, 65%) was
produced as a white solid, mp 216.4–217.9 C (dec). (Found: C,
N-hydroxy-4-[(4-{4-[((phenylsulfonyl)carbamoylamino)methyl]-
1H -1,2,3-triazole-1-yl}phenyl)sulfonyl]-tetrahydro-2H -pyran-4-
carboxamide (21). From 12 (105.4 mg, 0.32 mmol), 2 (92.3 mg,
0.39 mmol), sodium ascorbate (646 mL, 0.65 mmol) and copper(II)
sulfate pentahydrate (646 mL, 0.16 mmol), 21 (98 mg, 79%) was
produced as a white solid m.p. 185.0 ◦C (dec).
◦
59.58; H, 4.72; N, 10.35; S, 5.91. C26H24N4O6S requires C, 59.99; H,
4.65; N, 10.76; S, 6.16%); nmax (KBr)/cm-1 3230, 1669; dH (DMSO-
d6) 1.99 (2H, td, J 12.8 and 4.9, 2CHH), 2.26 (2H, d, J 12.8,
2CHH), 3.17 (2H, t, J 11.6, 2OCHH), 3.91 (2H, dd, J 11.6 and
3.7, 2OCHH), 7.08–7.22 (5H, m, ArH), 7.44 (2H, t, ArH), 7.96–
8.04 (4H, m, ArH), 8.24 (2H, d, J 8.6, ArH), 9.29 (1H, br s, NH),
9.48 (1H, s, Triazole-H) 11.10 (1H, br s, OH). dC (DMSO-d6) 27.4,
64.0, 69.7, 119.2, 119.3, 119.7, 120.0, 124.0, 125.1, 127.4, 130.3,
132.4, 134.0, 140.6, 147.5, 156.4, 157.2, 160.5; MS (ESI): m/z
521.18 [M+H]+.
N-Hydroxy-4-({[4-(4-phenoxybenzyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (22). From
12 (80 mg, 0.25 mmol), 4 (61.3 mg, 0.29 mmol), sodium ascorbate
(490 mL, 0.49 mmol) and copper(II) sulfate pentahydrate (490 mL,
0.12 mmol), 22 (36 mg, 27%) was produced as a white solid mp
138.0–139.3 ◦C (dec).
Molecular Modelling. Ligand processing
N-Hydroxy-4-({4-[4-(4-methylphenyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (14). From
12 (66.3 mg, 0.20 mmol), 1-ethynyl-4-methylbenzene (28.3 mg,
0.24 mmol), sodium ascorbate (406 mL, 0.41 mmol) and copper(II)
sulfate pentahydrate (406 mL, 0.10 mmol) 14 (57 mg, 63%) was
produced as a yellow solid, m.p. 208.1–209.4 ◦C (dec).
All ligands (together with compound i52) were first built using
CORINA,39 in its neutral protonation state and extended confor-
mation. Assignment of the atom types and charge calculations
were performed with Sybyl 7.3 (Gasteiger-Marsili charges).40
A
minimization for each compound was first carried out using the
program Macromodel.41 Water was set as solvent, and charges
selected from the mol2 file created with Sybyl. The program was
set to explore through 500 steps. The force field selected was
AMBER. A conformational analysis was then performed to all
compounds to be docked by use of the program Macromodel41
and the Monte Carlo methodology42 that allows the random
variation of the selected torsional angles. The parameters given
to Macromodel were set to default with some changes: the force
field selected was AMBER, as explained before, and a distance-
dependent dielectric constant of 1.0 GB/SA solvation model43
was selected. The program was set to explore through 500 steps
modifying three torsional angles each step. The limit of acceptance
was set to 50 kJ mol-1 above the instant minimum found. The
minimization method selected was conjugated gradient.44 The
convergence RMS to evaluate if two conformers are similar was
N-Hydroxy-4-({4-[4-(4-pentylphenyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (15). From
12 (45.3 mg, 0.14 mmol), 1-ethynyl-4-pentylbenzene (28.7 mg,
0.17 mmol), sodium ascorbate (278 mL, 0.28 mmol) and copper(II)
sulfate pentahydrate (278 mL, 0.07 mmol) 15 (45 mg, 65%) was
produced as a brown solid mp 198.0–199.7 ◦C (dec).
N-Hydroxy-4-({4-[4-(4-pentylbenzyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (16). From
12 (80 mg, 0.25 mmol), 3 (54.8 mg, 0.29 mmol) sodium ascorbate
(490 mL, 0.49 mmol) and copper(II) sulfate pentahydrate (490 mL,
0.12 mmol), 16 (60 mg, 48%) was produced as a white solid, mp
167.8–168.9 ◦C.
N-Hydroxy-4-{[4-(4-phenyl-1H-1,2,3-triazole-1-yl)phenyl]sul-
fonyl}tetrahydro-2H-pyran-4-carboxamide
(17). From
12
˚
set to a limit of 2 A. The criteria of minimization convergence
(71.6 mg, 0.22 mmol), phenylacetylene (26.9 mg, 0.26 mmol),
sodium ascorbate (438 mL, 0.44 mmol) and copper(II) sulfate
pentahydrate (438 mL, 0.11 mmol), 17 (26.7 mg, 28%) was
produced as a white solid mp 228.9–230.5 ◦C (dec).
-1
-1
˚
was set to 0.05 (kJ mol )A , that is, if a minimization step of a
structure is not able to improve by more than 0.05 (kJ mol )A-1 the
-1
˚
predecessor structure, the energetic minimum has been achieved.
Superimposition. All the conformers resulting from the con-
formational search were then superimposed to the ligand i52
is present in PDB 1hov (model 1). This ligand was previously
submitted to the same conformational search analysis and fol-
lowed the same superimposition procedure in order to validate the
methodology. The criteria selected for the superimposition was the
maximum number of atoms fitted using the CSR program,45 and
the minimum value of the Root Mean Square Deviation (RMSD).
Only the conformers with the higher number of fitted atoms were
selected for the docking studies.
N-Hydroxy-4-{[4-(4-biphenyl-4-yl-1H-1,2,3-triazole-1-yl)phen-
yl]sulfonyl}tetrahydro-2H-pyran-4-carboxamide (18). From 12
(60 mg, 0.18 mmol), 4-ethynylbiphenyl (39.3 mg, 0.22 mmol)
sodium ascorbate (368 mL, 0.37 mmol) and copper(II) sulfate
pentahydrate (368 mL, 0.09 mmol) 18 (56 mg, 60%) was produced
as a yellow solid, mp 227.7–229.0 ◦C (dec).
N-hydroxy-4-({4-[4-(4-methoxyphenyl)-1H-1,2,3-triazole-1-yl]-
phenyl}sulfonyl)tetrahydro-2H-pyran-4-carboxamide (19). From
12 (100 mg, 0.31 mmol), 4-ethynylanisol (48.6 mg, 0.37 mmol)
sodium ascorbate (612 mL, 0.61 mmol) and copper(II) sulfate
pentahydrate (612 mL, 0.15 mmol), 19 (86 mg, 61%) was produced
as a yellow solid, mp 217.1–218.3 ◦C (dec).
Docking Studies with AutoDock346
MMP2 catalytic domain from PDB 1hov (model 1) was em-
ployed as macromolecule. For the zinc ion, the parameter set
4-{[4-(4-Butyl-1H-1,2,3-triazole-1-yl)phenyl]sulfonyl}-N -hy-
˚
droxytetrahydro-2H-pyran-4-carboxamide
(20). From
12
reported by Stote et al.47 was used, which means r = 1.1 A, e =
(82.9 mg, 0.25 mmol), 1-hexine (54.8 mg, 0.30 mmol), sodium
ascorbate (508 mL, 0.51 mmol) and copper(II) sulfate pentahydrate
(508 mL, 0.13 mmol) 20 (67 mg, 65%) was produced as a white
solid mp 164.5–166.0 ◦C (dec).
0.25 kcal mol-1 and a formal charge of +2e. Affinity grid files
were generated using the auxiliary program AutoGrid.48 The grid
size was set to 128 ¥ 60 ¥ 60 points with a grid spacing of
˚
0.375 A containing the S1¢, S2¢ and S1 subsites, together with
4596 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
the catalytic zinc. The Lamarckian Genetic Algorithm (LGA)
was used as the search engine. The original Lennard-Jonnes and
hydrogen-bonding potentials provided ◦by the program were used.
˚
Step sizes of 2 A for translation and 60 for rotation were chosen.
A maximum number of energy evaluations was set to 250 000.
For each of the 100 independent runs, a maximum number of
27 000 LGA operations were generated on a single population
of 100 individuals. Operator weights for crossover, mutation, and
elitism were set to 0.80, 0.20 and 1, respectively. The parameters
for the docking using LGA were identical for all docking jobs.
After docking, the 100 solutions were clustered in groups with
Fig. 8 Different geometries for the coordination of the Zn2+ ion found in
the citliography for zinc-dependent proteases. MMPIs are found to coordi-
nate by square-based pyramidal or trigonal bipyramidal geometry.53,54 Zinc
ion is represented in red colour and the coordinating atoms are represented
in grey (coordinating atoms in MMP are three histidines and one or two
atoms of the ligand).
˚
Root Mean Square (RMS) deviations less than 1.0 A. The clusters
were ranked by the lowest energy representative of each cluster.
The compounds were studied with no restraints between the
ZBG and the zinc atom. Other variables were set to default values
using AutoDock Tools (ADT).46
All docking results not matching all these specific characteristics
(distances, angles and docked energy) were rejected.
We have followed several docking protocols, which include:
random docking, in which all possible torsionals of each com-
pound were activated, and the starting point of each conformation
was randomly selected by the program; superimposed docking,
in which the nitrogen atom from the hydroxamate group was
superimposed with the same nitrogen atom of compound i52,
and selected as starting point of the docking study (in this case all
possible torsionals were activated); rigid random docking, in which
the starting point of each conformation was randomly selected by
the program, but all torsionals were inactivated, so that the docked
conformation was the one obtained from the superimposition with
ligand i52.
Molecular Dynamics Simulations
Charge calculation. Geometries of 18 and 19 were optimized
at the B3LYP/3-21G* level of theory and basis set, followed by
a B3LYP/6-31G* single-point energy calculation to determine
the electrostatic potential around the molecule, which was subse-
quently used in the two-stage RESP fitting procedure in order to
obtain point charges.55 All ab initio calculations were carried out
using Gaussian 03 suite of programs.56
Molecular Dynamics Simulations. The following parameters
were assigned to zinc atom: van der Waals parameters of r =
Docking Studies with Glide49–51
-6
3.1 A, and e = 1E kcal mol-1, charge of zero, and a mass of
˚
50.38 u.m.a. To each of the five dummy atoms covalently bonded
to zinc, the following parameters were assigned: charge of +0.4,
van der Waals parameters r and e of zero, and mass of 0.1 u.m.a.
Protein-ligand system was first minimized in vacuum for 1000 steps
of steepest descent followed by 4000 steps of conjugate gradient
using AMBER 9 program.57 All Ca atoms were restrained to their
initial coordinates. Distance and angle restrictions were applied
on the catalytic zinc, the dummy atoms and the ZBG atoms.
This procedure allowed readjustment of covalent bonds and van
der Waals contacts without changing the overall conformation
of the protein. The resulting minimized complex was solvated by
Glide program was used to develop docking studies for all
compounds. First, structure of MMP2 catalytic domain obtained
from PDB 1hov (model 1) was optimized and minimized using
the protein preparation wizard application integrated in the Glide
package. Then, interaction maps of the protein binding site
were created using the Receptor grid generation application, also
included in the Glide package, and selecting ligand i52 as the
centre of the grid box. All compounds were prepared using the
Ligprep application. OPLS-2005 force field was selected for both
protein and ligands preparation. Docking studies were performed
randomly. For each docking study the program was set to explore
through 10 000 steps and a maximum number of 10 poses per
ligand. The selected docking method was XP (Extra-Precision).52
˚
a box of TIP3P waters which extended at least 8 A away from
any given protein atom (~ 9500 TIP3P water molecules). Periodic
boundary conditions were applied and electrostatic interactions
were represented using the smooth Particle Mesh Ewald (PME)
Data analysis. All docking results were manually analyzed.
The cut off value for coordinating distances, between the catalytic
Zn2+ ion and the oxygen atoms (or atom) from the hydroxamic
method58 with a grid spacing of ~1 A. First, all hydrogen atoms
˚
were minimized for 1000 steps of steepest descent. Next, the
position of water molecules was relaxed for 2000 steps of steepest
descent followed by 3000 steps of conjugate gradient. Finally,
the whole system, including the solute, was energy minimized
for 2000 steps of steepest descent plus 3000 steps of conjugate
gradients. At this point, the complex was kept frozen and started
the thermalization of the solvent by running three 30-ps MD
simulations to increase the temperature up to 300 K. Positional
constraints were only applied on the ZBG (His120, His124,
His130, and the two coordinating oxygen atoms of 18 and 19)
during 500 ps. During the next 500 ps restraints were gradually
lowered, followed by 3-ns unrestrained MD simulation (RMSD,
Supporting Information†).
˚
group of the inhibitor, was set to 2.53 A, as this is the maximum
reported distance between a coordinating atom and the MMP zinc
ion.53 For each potential docked solution, angles were measured to
confirm they are the optimal ones for the coordination geometries
and the conformers were classified into three different categories
depending on their coordination geometry with the catalytic zinc
ion: a trigonal bipyramidal coordination, a square-based pyramid
coordination and no coordinating pattern, using the mean atom
distance and angles displacement published by Alberts et al.
(Fig. 8).53
Energies were scrutinized, selecting only those docking solu-
tions with a favourable binding energy (Edocked < 0 Kcal mol-1).
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4597
©

View Article Online
Zymography
concentration of ligand for the Saturation Transfer Difference
(STD) experiments kept a ratio protein : ligand 1 : 100. 200 mM
of BiPS was added for the competition experiments. For each
sample, a 1D 1H reference, a water-LOGSY and a STD experiment
were recorded. 8 K points were used for a sweep width of
9,600 Hz and a total of 1 K and 512 scans were accumulated
for the water-LOGSY and STD experiments, respectively. In the
water-LOGSY experiments, the large bulk water magnetization
is partially transferred via the protein–ligand complex to the
free ligand in a selective manner. A non interacting compound
results in negative resonances, whereas protein–ligand interactions
are characterized by positive signals or by a reduction in the
negative signals obtained in the absence of the protein. In the STD
experiment, a positive interaction is recognized by the presence of
positive signals, a negative interaction yielding no signals. BiPS
was purchased from VWR (Barcelona, Spain).
Human blood serum from a healthy volunteer was used as the
source of gelatinases.12 Equal amounts of serum (1.5 ml) were
loaded on each well of a 10% gelatin zymogram gel (Invitrogen,
Carlsbad, CA, USA). After electrophoresis, the gel was incubated
for 30 min in renaturing buffer and then in developing buffer
for another 30 min, both at room temperature. At this point,
gels were cut into individual strips which were incubated in the
presence of the inhibitors in developing buffer overnight at 37◦
C. Following several washes, the gel fragments were stained with
Simply Blue (Invitrogen), destained, and scanned with Odyssey
(Li-cor, Lincoln, Nebraska, USA). Gelatinolytic activity for each
enzyme (MMP2, 72 kDa; MMP9, 92 kDa) was quantified by
image analysis (Image-J, NIH, Bethesda, MD, USA) and the IC50
values were calculated with the MS Excel-add-in XLfit (IDBS,
UL, Vs. 5.0).
Abbreviations
MMPs Inhibition Assays
Abbreviations: DTNB, 5,5¢-dithiobis-(2-nitrobenzoic acid);
HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid;
Brij-35, polyoxyethyleneglycol dodecyl ether; PDB, Protein
Data Bank; DCM, dichloromethane; THP, tetrahydropyrane;
THF, tetrahydrofurane; EDC, N-(3-dimethylaminopropyl)-
N’-ethylcarbodiimidehydrochloride; TFA, trifluoroacetic acid;
DMAP, 4-(dimethylamino)pyridine.
MMPs activity measurements were performed by using MMP
Inhibitor Profiling Kit purchased from Enzo Life Science In-
ternational, Inc., following the manufacturer’s protocol with
slight modifications. Proteolytic activity was measured using a
thiopeptide substrate (Ac-PLG-[2-mercapto-4-methylpentanoyl]-
LG—OC2H5) where the MMP cleavage site peptide bond has been
replaced by a thioester bond.59,60 Hydrolysis of this bond by MMP
produces a sulfhydryl group that reacts with DTNB to form 2-
nitro-5-thiobenzoic acid, which was detected by its absorbance
at 414 nm (microplate photometer Thermo Scientific Multiscan
FC). Enzyme reactions were carried out at 37 ◦C in a 100 ml
final volume of solutions, where the catalytic domains of the
corresponding MMP were incubated in triplicate with at least
six concentrations of inhibitors. The assay buffer contained the
following components: 50 mM HEPES, 10 mM CaCl2, 0.05% Brij-
35 and 1mM DTNB at pH 7.5. After addition of substrate, the
increase of absorbance was recorded at 1 min time intervals for 20
min. Data were plotted as OD versus time for each sample, in order
to obtain the reaction velocity (V) in OD min-1. The percentage
of residual activity for each compound was calculated using the
following formula: % of remaining activity = (V in the presence of
inhibitor/Vcontrol) ¥ 100. An inhibitor, NNGH, was included as
a prototypic control inhibitor.61 The concentration of compound
that provided 50% inhibition of enzymatic activity (IC50) was
determined by semilogarithmic dose-response plots (Graph Pad
Prism 5.0).62
Acknowledgements
This work was supported by the Spanish Ministry of Science and
Innovation (SAF2005-02608, SAF2008-00945, SAF2008-01845,
and SAF2009-13240). R. J. C. thanks the Spanish Ministry of
Science for a Ramo´n y Cajal contract. Grants to P. S. from
the Spanish Ministry of Education (FPU program), and to J.
M. Z. from Fundacio´n Universitaria San Pablo CEU are also
acknowledged. We thank EADS-CASA for a fellowship to K.
F. This work was awarded a “Premio del Colegio Oficial de
Farmace´uticos de Madrid”, given by the Real Academia Nacional
de Farmacia (Spain, 2010).
References
1 T. E. Cawston, Pharmacol. Ther., 1996, 70, 163–182.
2 K. M. Bottomley, W. H. Johnson and D. S. Walter, J. Enzyme Inhib.
Med. Chem., 1998, 13, 79–101.
3 I. Massova, L. P. Kotra, R. Fridman and S. Mobashery, FASEB J.,
1998, 12, 1075–1095.
4 C. M. Overall and O. Kleifeld, Nat. Rev. Cancer, 2006, 6, 227–239.
5 J. B. Summers, In Annual Reports in Medicinal Chemistry, D. Robertson
and J. Plattner, ed. Academic Press, San Diego, 1998, Vol. 33, pp 131–
149.
6 I. Bertini, V. Calderone, M. Fragai, A. Giachetti, M. Loconte, C.
Luchinat, M. Maletta, C. Nativi and K. J. Yeo, J. Am. Chem. Soc.,
2007, 129, 2466–2475.
7 O. Nicolotti, T. F. Miscioscia, F. Leonetti, G. Muncipinto and A.
Carotti, J. Chem. Inf. Model., 2007, 47, 2439–2448.
8 J. W. Skiles, N. C. Gonnella and A. Y. Jeng, Curr. Med. Chem., 2001, 8,
425–474.
9 C. K. Engel, B. Pirard, S. Schimanski, R. Kirsch, J. Habermann, O.
Klingler, V. Schlotte, K. U. Weithmann and K. U. Wendt, Chem. Biol.,
2005, 12, 181–189.
NMR experiments
All spectra were recorded at 300 K either with a Bruker Ultrashield
Plus Avance II 600 MHz or an Avance III 500 MHz spectrometer.
The 600 MHz spectrometer is equipped with a 5 mm cryogenically
cooled TCI probe, the 500 MHz with a broadband room-
temperature TBI probe. A typical NMR sample contained a
concentration of 1–5 mM of MMP2 (expression and purification
details of this MMP2 construct will be published elsewhere) and
20–100 mM of ligand, in an approximate ratio protein:ligand of
1 : 20, optimal for the water-LOGSY experiments. The sample
buffer was 10 mM deuterated-Tris/HCl pH 7.4 with 50 mM
NaCl, 0.02% NaN3, 100 mM CaCl2 and 100 mM ZnCl2. The
10 A. Amadasi, P. Cozzini, M. Incerti, E. Duce, E. Fisicaro and P. Vicini,
Bioorg. Med. Chem., 2007, 15, 1420–1429.
4598 | Org. Biomol. Chem., 2011, 9, 4587–4599
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
11 T. Tuccinardi, A. Martinelli, E. Nuti, P. Carelli, F. Balzano, G. Uccello-
Barretta, G. Murphy and A. Rossello, Bioorg. Med. Chem., 2006, 14,
4260–4276.
40 J. Gasteiger and M. Marsili, Tetrahedron Lett., 1978, 19, 3181–3184.
41 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M. Lipton,
C. Caufield, G. Chang, T. Hendrickson and W. C. Still, J. Comput.
Chem., 1990, 11, 440–467.
12 D. P. Becker, T. E. Barta, L. J. Bedell, T. L. Boehm, B. R. Bond, J.
Carroll, C. P. Carron, G. A. DeCrescenzo, A. M. Easton, J. N. Freskos,
C. L. Funckes-Shippy, M. Heron, S. L. Hockerman, C. P. Howard, J.
R. Kiefer, M. H. Li, K. J. Mathis, J. J. McDonald, P. P. Mehta, G.
H. Munie, Y. Sunyer, C. Swearingen, C. I. Villamil, D. Welsch, J. M.
Williams, Y. Yu and J. Yao, J. Med. Chem., 2010, 53, 6653–6680.
13 M. D. Martin and L. M. Matrisian, Cancer Metastasis Rev., 2007, 26,
717–724.
14 K. Almholt, A. Juncker-Jensen, O. D. Laerum, K. Dano, M. Johnsen,
L. R. Lund and J. Romer, Mol. Cancer Ther., 2008, 7, 2758–2767.
15 J. Hu, P. E. Van Den Steen, Q. X. Sang and G. Opdenakker, Nat. Rev.
Drug Discovery, 2007, 6, 480–498.
16 D. P. Becker, C. I. Villamil, T. E. Barta, L. J. Bedell, T. L. Boehm,
G. A. DeCrescenzo, J. N. Freskos, D. P. Getman, S. Hockerman, R.
Heintz, S. C. Howard, M. H. Li, J. J. McDonald, C. P. Carron, C. L.
Funckes-Shippy, P. P. Mehta, G. E. Munie and C. A. Swearingen, J.
Med. Chem., 2005, 48, 6713–6730.
17 S. A. Kolodziej, S. L. Hockerman, T. L. Boehm, J. N. Carroll, G. A.
DeCrescenzo, J. J. McDonald, D. A. Mischke, G. E. Munie, T. R.
Fletcher, J. G. Rico, N. W. Stehle, C. Swearingen and D. P. Becker,
Bioorg. Med. Chem. Lett., 2010, 20, 3557–3560.
42 G. Chang, W. C. Guida and W. C. Still, J. Am. Chem. Soc., 1989, 111,
4379–4386.
43 W. C. Still, A. Tempczyk, R. C. Hawley and T. Hendrickson, J. Am.
Chem. Soc., 1990, 112, 6127–6129.
44 E. Polak and G. Ribiere, Revue Franc¸aise d’Informatique et de la
Recherche Ope´rationnelle, 1969, 16–R1.
45 M. Petitjean, Comput. Chem., 1998, 22, 463–465.
46 Autodock 3.0, The Scripps Institute, 1998.
47 R. H. Stote and M. Karplus, Proteins: Struct., Funct., Bioinf., 1995,
23, 12–31.
48 GRID v. 22a, Molecular Discovery Ltd., 215 Marsh Road, Pinner,
Middlesex UK, 2004.
49 Glide v 2.5021. Schro¨dinger, L.L.C., New York, USA, 2003.
50 R. A. Friesner, J. L. Banks, R. B. Murphy, T. A. Halgren, J. J. Klicic,
D. T. Mainz, M. P. Repasky, E. H. Knoll, M. Shelley, J. K. Perry, D.
E. Shaw, P. Francis and P. S. Shenkin, J. Med. Chem., 2004, 47, 1739–
1749.
51 T. A. Halgren, R. B. Murphy, R. A. Friesner, H. S. Beard, L. L. Frye,
W. T. Pollard and J. L. Banks, J. Med. Chem., 2004, 47, 1750–1759.
52 R. A. Friesner, R. B. Murphy, M. P. Repasky, L. L. Frye, J. R.
Greenwood, T. A. Halgren, P. C. Sanschagrin and D. T. Mainz, J.
Med. Chem., 2006, 49, 6177–6196.
18 S. A. Kolodziej, S. L. Hockerman, G. A. DeCrescenzo, J. J. McDonald,
D. A. Mischke, G. E. Munie, T. R. Fletcher, N. Stehle, C. Swearingen
and D. P. Becker, Bioorg. Med. Chem. Lett., 2010, 20, 3561–3564.
19 R. Manetsch, A. Krasinski, Z. Radic, J. Raushel, P. Taylor, K. B.
Sharpless and H. C. Kolb, J. Am. Chem. Soc., 2004, 126, 12809–12818.
20 W. G. Lewis, L. G. Green, F. Grynszpan, Z. Radic, P. R. Carlier, P.
Taylor, M. G. Finn and K. B. Sharpless, Angew. Chem. Int. Ed. Engl.,
2002, 41, 1053–1057.
53 I. L. Alberts, K. Nadassy and S. J. Wodak, Protein Sci., 1998, 7, 1700–
1716.
54 X. Hu, S. Balaz and W. H. Shelver, J. Mol. Graphics Modell., 2004, 22,
293–307.
55 C. I. Bayly, P. Cieplak, W. Cornell and P. A. Kollman, J. Phys. Chem.,
1993, 97, 10269–10280 .
21 A. Krasinski, Z. Radic, R. Manetsch, J. Raushel, P. Taylor, K. B.
Sharpless and H. C. Kolb, J. Am. Chem. Soc., 2005, 127, 6686–6692.
22 M. Whiting, J. Muldoon, Y. C. Lin, S. M. Silverman, W. Lindstrom, A.
J. Olson, H. C. Kolb, M. G. Finn, K. B. Sharpless, J. H. Elder and V. V.
Fokin, Angew. Chem., Int. Ed., 2006, 45, 1435–1439.
23 V. P. Mocharla, B. Colasson, L. V. Lee, S. Roper, K. B. Sharpless, C. H.
Wong and H. C. Kolb, Angew. Chem., Int. Ed., 2005, 44, 116–120.
24 R. Srinivasan, M. Uttamchandani and S. Q. Yao, Org. Lett., 2006, 8,
713–716.
56 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
57 D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling,
J. Wang, R. E. Duke, R. E. Luo, K. M. Merz, D. A. Pearlman, M.
Crowley, R. C. Walker, W. Zhang, B. Wang, S. Hayik, A. Roitberg,
G. Seabra, K. F. Wong, F. Paesani, X. Wu, S. Brozell, V. Tsui, H.
Gohlke, C. Yang, J. Tan, J. Mongan, V. Hornak, G. Cui, P. Beroza, D.
H. Mathews, C. Schafmeister, W. S. Ross, P. A. Kollman AMBER 9,
University of California, San Francisco, 2006.
25 S. L. Ng, P. Y. Yang, K. Y. T. Chen, R. Srinivasan and S. Q. Yao, Org.
Biomol. Chem., 2008.
26 J. Wang, M. Uttamchandani, J. Q. Li, M. Y. Hu and S. Q. Yao, Org.
Lett., 2006, 8, 3821–3824.
27 F. Himo, T. Lovell, R. Hilgraf, V. V. Rostovtsev, L. Noodleman, K. B.
Sharpless and V. V. Fokin, J. Am. Chem. Soc., 2005, 127, 210–216.
28 M. A. Garc´ıa, University San Pablo-CEU, Madrid, PhD Thesis 2007.
29 A. Scozzafava and C. T. Supuran, J. Med. Chem., 2000, 43, 1858–1865.
30 Y. P. Pang, J. Mol. Model., 1999, 5, 196–202.
31 K. Kupai, G. Szucs, S. Cseh, I. Hajdu, C. Csonka, T. Csont and P.
Ferdinandy, J. Pharmacol. Toxicol. Methods, 2010, 61, 205–209.
32 Y. Tamura, F. Watanabe, T. Nakatani, K. Yasui, M. Fuji, T. Ko-
murasaki, H. Tsuzuki, R. Maekawa, T. Yoshioka, K. Kawada, K.
Sugita and M. Ohtani, J. Med. Chem., 1998, 41, 640–649.
33 E. Dragoni, V. Calderone, M. Fragai, R. Jaiswal, C. Luchinat and C.
Nativi, Bioconjugate Chem., 2009, 20, 719–727.
34 C. M. Overall and C. Lopez-Otin, Nat. Rev. Cancer, 2002, 2, 657–672.
35 C. Dalvit, P. Pevarello, M. Tato`, M. Veronesi, A. Vulpetti and M.
Sundstro¨m, J. Biomol. NMR, 2000, 18, 65–68.
36 B. Meyer and T. Peters, Angew. Chem., Int. Ed., 2003, 42, 864–890.
37 Y. Q. Feng, J. Likos, L. M. Zhu, H. Woodward, J. McDonald, A.
Stevens, S. Howard and D. Welsch, J. Biomol. NMR, 2000, 17, 85–86.
38 S. Rowsell, P. Hawtin, C. A. Minshull, H. Jepson, S. M. V. Brockbank,
D. G. Barratt, A. M. Slater, W. L. McPheat, D. Waterson, A. M. Henney
and R. A. Pauptit, J. Mol. Biol., 2002, 319, 173–181.
58 T. A. Darden, D. York and L. G. Pedersen, J. Chem. Phys., 1993, 98,
10089–10092.
59 H. Weingarten and J. Feder, Anal. Biochem., 1985, 147, 437–440.
60 H. Weingarten, R. Martin and J. Feder, Biochemistry-Moscow, 1985,
24, 6730–6734.
61 L. J. MacPherson, E. K. Bayburt, M. P. Capparelli, B. J. Carroll, R.
Goldstein, M. R. Justice, L. Zhu, S. Hu, R. A. Melton, L. Fryer, R.
L. Goldberg, J. R. Doughty, S. Spirito, V. Blancuzzi, D. Wilson, E. M.
O’Byrne, V. Ganu and D. T. Parker, J. Med. Chem., 1997, 40, 2525–
2532.
39 J. Gasteiger, C. Rudolph and J. Sadowski, Tetrahedron Comput.
Methodol., 1990, 3, 537–547.
62 Graph Pad Prism version 5.0 for Windows, Graph Pad Software Inc.,
San Diego, California, USA, 2007.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4587–4599 | 4599
©