Identifying chelators for metalloprotein inhibitors using a fragment-based approach

Article abstract of DOI:10.1021/jm101266s

Fragment-based lead design (FBLD) has been used to identify new metal-binding groups for metalloenzyme inhibitors. When screened at 1 mM, a chelator fragment library (CFL-1.1) of 96 compounds produced hit rates ranging from 29% to 43% for five matrix meta

Full text of DOI:10.1021/jm101266s

J. Med. Chem. 2011, 54, 591–602 591
DOI: 10.1021/jm101266s
Identifying Chelators for Metalloprotein Inhibitors Using a Fragment-Based Approach
Jennifer A. Jacobsen,† Jessica L. Fullagar,† Melissa T. Miller, and Seth M. Cohen*
Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Drive, La Jolla, California 92093-0358,
United States. †Authors contributed equally to this work.
Received September 15, 2010
Fragment-based lead design (FBLD) has been used to identify new metal-binding groups for
metalloenzyme inhibitors. When screened at 1 mM, a chelator fragment library (CFL-1.1) of 96
compounds produced hit rates ranging from 29% to 43% for five matrix metalloproteases (MMPs),
24% for anthrax lethal factor (LF), 49% for 5-lipoxygenase (5-LO), and 60% for tyrosinase (TY). The
ligand efficiencies (LE) of the fragment hits are excellent, in the range of 0.4-0.8 kcal/mol. The MMP
enzymes all generally elicit the same chelators as hits from CFL-1.1; however, the chelator fragments
that inhibit structurally unrelated metalloenzymes (LF, 5-LO, TY) vary considerably. To develop more
advanced hits, one hit from CFL-1.1, 8-hydroxyquinoline, was elaborated at four different positions
around the ring system to generate new fragments. 8-Hydroxyquinoline fragments substituted at either
the 5- or 7-positions gave potent hits against MMP-2, with IC50 values in the low micromolar range. The
8-hydroxyquinoline represents a promising new chelator scaffold for the development of MMP
inhibitors that was discovered by use of a metalloprotein-focused chelator fragment library.
Introduction
MBG for the active site metal ion.26,28 It is possible that such
interactions might be optimized by finding the best MBG for a
given metalloenzyme, but to do so, a variety of MBGs would
be needed to address the structural and chemical space pre-
sented by the diverse range of metalloenzyme targets. While a
reasonable number of MBGs have been introduced for use in
MMPi,23,32 the study of different MBGs in inhibitors of other
metalloenzymes are relatively scarce.33-37 The limited atten-
tion given to MBGs is surprising in light of the effort put forth
to systematically vary all other components of metalloenzyme
inhibitors.38
In order to identify new chelating scaffolds for metalloen-
zyme inhibitors, fragment-basedlead design (FBLD) has been
proposed as a viable approach. FBLD is a method in which
low molecular weight compounds (fragments) are screened
against drug targets by use of sensitive techniques such as
nuclear magnetic resonance spectroscopy or affinity mass
spectrometry.39,40 Fragments are then linked or “grown” to
generate potent leads, which can be further optimized to
obtain improved solubility and pharmacokinetics via tradi-
tional medicinal chemistry. FBLD has several distinct advan-
tages over more traditional methods of lead development.
Active sites are more efficiently probed by small fragments
that are not limited by steric constraints or hydrogen-bonding
mismatches.39 Additionally, compounds of lower complexity
can more effectively sample the available chemical diversity.40
Leads developed by fragment methods typically have
good ligand efficiencies (LE), a standard measure used to
compare compound potency with molecular size by the for-
mula LE=-RT(ln[IC50])/HACinwhich IC50 isthe fragment
concentration at which enzyme activity is reduced 50% and
HAC is the heavy atom (or non-hydrogen atom) count.41
The identification of novel MBGs for metalloenzyme in-
hibitors is a task particularly well suited for FBLD.42 In fact,
Metalloenzymes represent at least a third of all proteins and
utilize a wide variety of metal ion cofactors for catalytic,
electron transfer, structural, or other key roles.1,2 As a result,
inhibitors of metalloenzymes are desired for mechanistic stud-
ies, as well as for applications including pesticides,3-5 pre-
servatives,6 cosmetics,7 and therapeutics. Pathogenic metal-
loenzyme activity is associated with many illnesses such as
cancer,8-11 inflammatory diseases,12,13 infectious diseases,14-19
cardiovascular diseases,20,21 and neurodegenerative dis-
eases.21,22 The presence of the metal ion in these enzymes has
been frequently exploited for the development of synthetic
inhibitors.18 Specifically, the metal ion cofactor can serve as
an anchoring site that may be easily targeted by metal-binding
groups (MBGs).a A range of MBGs has been used in the
inhibitors of different metalloenzymes, but among the most
common groups are hydroxamic acids, sulfonamides, and
carboxylic acids.23-25
More recently, reports of matrix metalloprotease inhibitors
(MMPi) with novel MBGs (often referred to as zinc-binding
groups for these inhibitors, ZBGs) suggest an expanded role
for the MBG in inhibitors of metalloenzymes. In particular,
the MBGs of MMPi have been found to influence inhibitor
potency26-28 and selectivity.29,30 These MBG effects have
been attributed to hydrogen-bonding interactions with the
protein,31 van der Waals contacts,30 and the affinity of the
*To whom correspondence should be addressed. E-mail:
scohen@ucsd.edu. Telephone: (858) 822-5596. Fax: (858) 822-5598.
a Abbreviations: CFL, chelator fragment library; FBLD, fragment-
based lead design; HAC, heavy atom count; iNOS, inducible nitric oxide
synthase; LF, lethal factor; LE, ligand efficiency; 5-LO, 5-lipoxygenase;
MMP, matrix metalloproteinase; MMPi, matrix metalloproteinase in-
hibitor; MBGs, metal-binding groups; TY, tyrosinase; ZBG, zinc-bind-
ing group.
r
2010 American Chemical Society
Published on Web 12/28/2010
pubs.acs.org/jmc

592 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
Figure 1. Structures of all fragments in CFL-1.1. The library is organized with each row corresponding to a MBG class: picolinic acids
(1a-12a), quinolines (1b-12b), pyrimidines (1c-12c), hydroxypyrones (1d-12d), hydroxypyridinones (1e-12e), and salicylic acids (1f-12f).
Compounds 1g-12g and 1h-12h are classified as miscellaneous.
many common metal chelators may be considered fragments
that bind to metals with affinities in the micro- to millimolar
range, which are suitable for screening in many bioassays. In
addition, the binding mode of many chelators may be mod-
eled by complexes that mimic metalloenzyme active sites or
may be inferred from published crystal structures of the
chelator bound to the metal of interest.26,43-45 As a result,
screening methods that produce structural binding informa-
tion, which typically require large quantities of protein and
time-consuming data analysis (e.g., X-ray- or NMR-based
screening), may not be as essential when dealing with MBG-
based library screening. Finally, chelators may function as
good molecular anchors, where strongly bound fragments are
unlikely to dramatically change their binding mode upon
elaboration, thereby facilitating lead development.39 Several
studies have applied FBLD to the development of metalloen-
zyme inhibitors,42,46-52 although in most cases the MBG was
not varied as a fragment.
In this report, FBLD is used to identify new MBGs for
metalloenzyme inhibitors. A small chemical library of 96
metal chelators (Figure1) was assembledand screened against
five MMP family members to identify new MBGs for MMP
inhibitors.42 This library was also screened against an unre-
lated zincenzyme (anthrax lethal factor, LF), a non-heme iron
enzyme (5-lipoxygenase, 5-LO), a dinuclear copper enzyme
(tyrosinase, TY), and a heme iron enzyme (inducible nitric
oxide synthase, iNOS) in order to probe the applicability
of this approach to generate new MBGs for a wide range of
metalloenzymes. Finally, we demonstrate the advancement of
a hit from the library by producing a focused library based on
an 8-hydroxyquinoline scaffold. The findings presented show
that chelators with multiple points of attachment offer more
synthetic opportunities that are important for the successful
development of metalloprotein inhibitors.
Results
Design and Evaluation of a Chelator Fragment Library
(CFL). A fragment library consisting of metal chelators
(CFL-1) was assembled as previously described.42 All but
five fragments from CFL-1 have a molecular weight <200
amu. The fragments all have known metal-binding modes
using two to four nitrogen, oxygen, and sulfur heteroatoms.
CFL-1 contains a total of 96 fragments with six metal-
binding classes including picolinic acids, quinolines, pyrimi-
dines, hydroxypyrones, hydroxypyridinones, salicylic acids,
and 24 miscellaneous compounds. All of the fragments are
soluble up to 50 mM in DMSO and are commercially
available or readily synthesized.30,53-61 Previously reported
screening of CFL-1 against MMPs identified four com-
pounds that could not be screened in either colorimetric or
fluorometric MMP assays due to problems with solubility
or interference with the assay conditions.42 These four
compounds were replaced (6b, 10b, 7f, and 2h) and a new
generation of this library (CFL-1.1, Figure 1) was used for
the experiments reported here.
The fragments of CFL-1.1 were screened against MMP-1,
MMP-2, MMP-3, MMP-8, MMP-9, LF, 5-LO, TY, and

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2 593
Figure 2. Thermoplot representing the results from the screens of CFL-1.1 against various metalloenzymes. Cells are color-coded by percent
inhibition: black (0-25%), red (25-50%), orange (51-75%), and yellow (76-100%). Gray cells indicate compounds that interfered with the
assay.
iNOS using reported assay conditions (Tables S1-S9, Sup-
porting Information).26,42,62,63 The results of the screens are
depicted in a thermoplot (Figure 2) representing the percent
inhibition of the fragments against each metalloenzyme at a
fragment concentration of 1 mM. Hits were defined as
fragments that inhibit enzyme activity >50% at 1 mM.
Based on the screening results of CFL-1.1, the IC50 values
of select hits against MMP-2, LF, 5-LO, and TY were
determined and corresponding LEs were calculated (Table 1).
Hits against MMPs. CFL-1.1 produced a significant
number of hits against MMP-1, MMP-2,42 MMP-3,
MMP-8, and MMP-9 (Figure 2). The percent of fragments
identified as hits (hit rate) ranged from 29% to 43% for the
MMPs screened. 3-Hydroxypicolinic acid (5a) decreased
the activity of MMP-1, MMP-2, MMP-3, and MMP-8
by approximately 50%. Derivatives of 8-hydroxyquinoline
(2b, 7b, 8b, and 9b) strongly inhibited all five MMPs.
8-Hydroxyquinoline (7b) was found to have an IC50 value
of 130 μM against MMP-2 (Table 1). Two other quinolines
stood out as MMP hits: 8-hydroxyquinoline-N-oxide (5b)
and the sulfonamide quinoline (10b).61 Two pyrimidines (4c
and 5c) were MMP hits, with compound 4c having an IC50
value of 121 μM against MMP-2. The sulfur-containing
hydroxypyrothiones (2d and 5d) completely inhibit all
MMPs, while most of the hydroxypyrones (1d, 7d-11d)
showed ∼50% inhibition of the MMPs as well. The IC50
value of hydroxypyrothione 2d against MMP-2 was found to
be 76 μM. The sulfur-containing hydroxypyridinethiones
(2e, 7e, 9e, 11e, and 12e) also completely inhibit all the
MMPs. Among the hydroxypyridinones only 1e and 10e,
both of which are based on a 1,2-hydroxypyridinone core,
showed >50% inhibition of the MMPs. In general, salicylic
acids did not effectively inhibit any MMPs, while several mis-
cellaneous compounds were found to be hits. Of particular

594 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
Table 1. Calculated IC50 Values, Heavy Atom Count (HAC), And
Ligand Efficiency (LE) Values for Select Hits from CFL-1.1a
thiol at the 2-position. All of the tested hydroxypyrones
(1d-3d, 5d, 7d-9d, 11d, 12d) and hydroxypyridinones
(2e-9e, 11e, 12e) inhibit 5-LO >70%. The IC50 value of
5d against 5-LO was found to be 11 μM. Unlike the zinc-
dependent MMPs and LF that were more effectively inhib-
ited by the sulfur (thione) derivatives of the hydroxypyrones
(1d-12d) and hydroxypyridinones (1e-12e), 5-LO was in-
hibited by both O,O and O,S chelators of these compound
classes (compare 1d and 2d; 8e and 9e). Consistent with our
findings, fragment 1e has appeared in the literature as
part of 5-LO inhibitors and dual inhibitors of 5-LO and
COX.66-68 Furthermore, two salicylic acid derivatives
(6f, 9f) were identified as 5-LO hits with compound 9f having
an IC50 value of 75 μM against 5-LO. Among the miscella-
neous fragments, catechol derivatives (8g, 1h-4h) and acet-
ylacetone derivatives (8h-12h) also inhibit 5-LO.
enzyme
MMP-2b
compound
IC50 value (μM)
HAC
LE (kcal/mol)
7b
4c
130 ( 28
121 ( 37
76 ( 1
11
8
0.50
0.70
0.73
0.52
0.61
0.44
0.63
0.53
0.75
0.51
0.52
0.82
0.55
0.97
2d
1g
11g
3h
2d
1g
5d
9f
8
1500064
146 ( 71
389 ( 134
20465
5
9
11
8
LFc
1140065
5
5-LOc
TYc
11 ( 2
75 ( 5
159 ( 28
3.8 ( 0.5
100 ( 9
0.462
9
11
10
9
6a
5d
1f
10
9
11g
a Compound 1g (acetohydroxamic acid) is included as a reference.
b Ligand efficiency calculated at 37 °C. c Ligand efficiency calculated at
25 °C.
CFL-1.1 has a very high hit rate of 60% against the
copper-dependent TY. Several of the chelators identified
from the screening of CFL-1.1 have been previously reported
as inhibitors of TY including tropolone (11g, IC50
=
interest, two tropolone derivatives (11g, 12g) stand out as
new MBGs for MMPs, with an IC50 value of 146 μM (11g)
against MMP-2.
0.4 μM),62 kojic acid (7d, IC50 = 23 μM),62 1,2-hydroxypyr-
idinone (1e, IC50 = 1.2 μM)69 and L-mimosine,70 an amino
acid derivative of 3,4-hydroxypyridinones (3e-5e). The IC50
values of maltol (4d) and several other hydroxypyridinones
(3e, 4e, 6e, and 8e) against TY have also been reported.71 All
of the aforementioned fragments and structurally related
compounds were confirmed as TY hits from CFL-1.1
(Figure 2). Novel hits against TY from CFL-1.1 included
thiopyrones (2d and 5d) and thiopyridinones (2e, 7e, 9e, 11e,
and 12e). Compound 5d was found to have an IC50 value of
3.8 μM against TY (Table 1). Similar to the MMPs, the
pyrones and pyridinones with O,S donor atoms are more
potent against TY than the corresponding compounds with
O,O donor atoms (compare 1d and 2d, 4d and 5d, 8e and 9e,
4e and 12e; Figure 2). However, like 5-LO, many O,O donor
atom chelators (3d, 8d, 9d, 11d, 1e, and 10e) did show some
inhibition against TY. Other new hits included the picolinic
acids (1a-12a), with all but one fragment (2a) inhibiting
TY >75%. Quinoline fragments containing carboxylic acids
(2b-4b) were found to be hits against TY. Also, salicylic
acids (1f, 3f-5f, and 7f-12f) were very effective at inhibiting
TY activity, showing more than 90% inhibition, with sal-
icylic acid (1f) having an IC50 value of 100 μM against TY
(Table 1). A wide variety of miscellaneous compounds were
also found to be hits against TY (4h, 7h-12h).
The hit rate of CFL-1.1 against the heme iron protein
iNOS at 1 mM was 4% (3 out of 73 fragments). Several
compounds were excluded from screening based on insolu-
bility in the assay buffer or incompatibility with the assay
reagents. There are only six compounds (5a, 1b, 6c, 8d, 12d,
and 12h) that inhibit iNOS greater than 40%, and these
fragments cover nearly all the metal-binding classes repre-
sented in CFL-1.1. Overall, CFL-1.1 did not reveal any good
fragment hits against iNOS.
8-Hydroxyquinoline Sublibrary Synthesis and Screening. In
order to demonstrate ( a) an ability to further develop novel
hits from CFL-1.1 and (b) the versatility offered by new
chelator scaffolds, a sublibrary of fragments was developed
based on a single hit from CFL-1.1. The 8-hydroxyquinoline
scaffold (7b) was selected for development into a small
sublibrary because it was found to be a potent hit against
the MMPs in our screening of CFL-1.1. To the best of our
knowledge, hydroxyquinolines have seen very limited use as
a MBG in MMP inhibitors.23,24,31,72,73 A focused library of
In general, the fragments from CFL-1.1 did not show
specificity among the MMPs, that is, fragments found as hits
were potent against all of the MMPs tested. One notable
exception was the activity of several picolinic acids against
MMP-3. In general, the picolinic acids were not active
against the MMPs, but MMP-3 was inhibited by six picolinic
acid derivatives (1a, 4a, 5a, 8a, 10a, and 12a). Of these, only
5a and 12a were found to show any significant activity
against other MMPs. Compound 1a, which does not come
up as a hit against MMP-1, MMP-2, MMP-8, or MMP-9,
was previously reported to have an IC50 value of 181 μM
against MMP-3.28 To date, the origin of the selectivity of the
picolinic acids for MMP-3 is unclear. Nonetheless, this
finding validates the robust nature of this library screen-
ing approach and suggests that appropriate selection of
the chelator fragment in combination with a suitable back-
bone group, may be able to augment the specificity of an
MMPi.
Hits against Other Metalloenzymes. Screening of CFL-1.1
against LF at 1 mM yielded 22 hits, giving a hit rate of 24%.
Four compounds from CFL-1.1 were not screened because
they were found to precipitate or interfere with the assay
(10b, 11b, 10c, 5e). In contrast to the MMPs, several of the
picolinic acids (1a-3a, 5a, 9a-11a) inhibited LF at close to
50%. Three picolinic acids (6a-8a) inhibit LF >70% and
are all substituted at the 6-position, unlike the picolinic acid
hits against MMP-3. Quinoline compounds 2b, 5b, 7b, and
9b were hits against LF. The hydroxypyrothiones (2d and 5d)
completely inhibit LF at 1 mM; however, the hydroxypyr-
idinethiones (7e, 9e, 11e, and 12e) only inhibit LF ∼50%.
Fragment 2d was previously reported to have an IC50 value
of 204 μM against LF,65 again validating the screening
results performed here. In addition, several miscellaneous
compounds were found to be hits against LF (4g, 5g, 7g, 11g,
12g, 1h, and 3h).
The hit rate of CFL-1.1 against the non-heme iron protein
5-LO was 49%. Several of the 8-hydroxyquinoline deriva-
tives (2b, 7b-9b) including 8-hydroxyquinoline-N-oxide (5b)
hit 5-LO. Unlike the zinc-dependent enzymes, a number of
pyrimidine derivatives (2c, 3c, 5c, 7c, 8c, and 10c) hit 5-LO at
1 mM. Of these compounds, 3c, 5c, 7c, and 8c each have a

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2 595
Figure 3. Synthesis of the components of the 8-hydroxyquinoline sublibrary. Substituents were appended to four different positions (2-, 4-, 5-,
and 7-positions) around the hydroxyquinoline ring.
Figure 4. Results of hydroxyquinoline sublibrary screen against MMP-2 at 25 μM. Only compounds with substituents at the 5- and 7-positions
gave significant inhibition activity.
16 fragments based on 8-hydroxyquinoline was prepared by
derivatizing this scaffold around the ring at positions 2, 4, 5,
and 7. These four positions were selected in order to place
substituents thoroughly around the ring system, as well as for
the synthetic accessibility of these particular sites.74 At each
of the four positions, an amine group was installed that was
coupled with one of four different sulfonyl chlorides to
generate sulfonamide analogues (Figure 3). Substituents
were loosely selected based on efficacy of previously reported
MMP inhibitors.47,61
target) at a concentration of 25 μM; the results are shown in
Figure 4. The assay results clearly show that the derivatives
substituted at the 2- and 4-positions (72a-d and 74a-d) are
ineffective against MMP-2 (only 72c has an IC50 value of
48 ( 1 μM). In contrast, the compounds substituted at the
5- and 7-positions (75a-d and 77a-d) all showed >65%
inhibition of MMP-2, regardless of the specific R-group. The
IC50 values of these compounds against MMP-2 were found
to be in the low micromolar range (Table 2). The most potent
hit from this sublibrary, 77b, has an IC50 value of
3 μM, representing a >40-fold improvement over fragment
7b. This improvement clearly demonstrates that hits from
The 8-hydroxyquinoline-focused library was initially
screened against MMP-2 (as a representative metalloprotein

596 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
Table 2. Calculated IC50 Values, Heavy Atom Count (HAC), and
Ligand Efficiency (LE) for 75a-d and 77a-d Fragments
While the screening results of CFL-1.1 are largely the same
across the different MMPs, MMP-3 did show some varia-
bility. Fifteen fragments were found to be significantly more
potent for MMP-3 than the other MMPs. This is most clearly
seen with the picolinic acids, where seven of these compounds
(1a, 2a, 4a, 8a, 9a, 10a, 12a) show improved inhibition of
MMP-3 relative to the other MMPs. In particular, 1a, 4a, 10a,
and 12a inhibit MMP-3 >80% at 1 mM but are not potent
against other MMPs. Unlike most other MMPs, which
operate best at pH 7.4, MMP-3 activity is optimal at pH 6,
which may be one factor that contributes to the difference in
potency of the picolinic acids against this particular MMP.75
In order to evaluate the use of CFL-1.1 against other
metalloenzymes, the library was screened against LF, 5-LO,
TY, and iNOS. All of these enzymespossess metal ions attheir
active sites and were chosen for their functional diversity and
biomedical importance. The fragments that hit each enzyme
vary significantly, which suggests that choosing specific MBG
scaffolds can be used as a starting point for developing
metalloenzyme-selective inhibitors. The LF hits of CFL-1.1
most closely align with the MMP hits, a fact that reflects that
they are both zinc(II)-dependent proteases. Even with their
similarities, the thermoplot (Figure 2) shows LF hits that are
not hits against the MMPs. For example, several picolinic
acids that do not inhibit MMPs (6a-8a, 11a) reduce LF
activity moderately at 1 mM. The different coordination
environment of the zinc(II) ion in these enzymes, as well as
divergence in the immediate active site environment between
LF and MMPs, likely contribute to the ability of picolinic
acids to selectively inhibit LF.76-79
The hits against the non-heme iron enzyme 5-LO are very
different from the hits against LF or the MMPs. Numerous
pyrimidines (2c, 3c, 5c, 7c, 8c, 10c) are 5-LO hits, and
essentially all hydroxypyrones and hydroxypyridinones effec-
tively inhibit 5-LO. Unlike the zinc(II)-dependent enzymes,
which show a strong preference for O,S donor ligands among
the hydroxypyrones and hydroxypyridinones, 5-LO shows no
pronounced preference, which likely reflects the greater affi-
nity of the iron(III) ion (which is generated in the 5-LO
catalytic cycle) for hard Lewis basic O,O donor ligands. For
the dinuclear copper enzyme TY, the most distinguishing hits
are the picolinic acids (1a-12a) and the salicylic acids
(1f-12f), the latter of which were found to be generally
ineffective against all the other metalloenzymes studied. It is
our contention that the differences between the hits found for
5-LO, and TY when compared with the MMPs and LF are
significant and support the use of CFLs in the discovery and
development of metalloprotein-selective inhibitors. These
varying preferences for MBGs likely arise from both the
ligand affinities for the different metal ions and differences
in the shape and size of the metal active sites. The high degree
of variability among metalloenzyme hits supports the notion
that metal-binding fragments may be optimized for a given
metalloenzyme, which represents an area of study that has not
been given substantial attention in the development of metal-
loenzyme inhibitors.
compound
IC50 value (μM)
HAC
LE (kcal/mol) (37 °C)
75a
75b
75c
75d
77a
77b
77c
77d
8.8 ( 0.6
7.2 ( 1.0
10.8 ( 0.4
8.4 ( 0.3
5.6 ( 1.9
3.0 ( 0.5
3.9 ( 0.4
4.3 ( 0.2
21
22
27
20
21
22
27
20
0.34
0.33
0.26
0.36
0.36
0.36
0.28
0.38
CFL-1.1 can be readily improved by derivatization in a
focused library. In addition, it is important to note that
effective inhibition in this focused library is highly dependent
on the position of the substituent with only two positions
giving potent derivatives. Furthermore, the position of the
substituent appears to be more important than the nature of
the substituent at this stage of fragment growth; that is, none
of the backbone substituents results in effective MMP-2
inhibition at all positions on the hydroxyquinoline ring. This
shows both the versatility and promise presented by identify-
ing new chelator leads from a fragment library; one hit
fragment can generate more than one subsequent focused
fragment (in this case both 75a-d and 77a-d fragments).
Discussion
Fragments identified as hits againstthe MMPs fromscreen-
ing CFL-1.1 included hydroxypyrothiones (2d and 5d), hy-
droxypyridinethiones (2e, 7e, and 9e), and nitrogen-based
ligands (1a, 3g, and 4g). Many of these compounds had been
identified in early studies,26,28,64 demonstrating that the CFL
approach was valid and could readily identify active frag-
ments. Consistent with prior reports, the O,S donor ligands
(2d, 5d, 2e, 7e, and 9e) are more potent against MMPs than
their structural analogs with O,O donor atoms (1d, 4d, 1e, and
8e).26,64 New MMP hits from CFL-1.1 include picolinic acids
(5a, 12a), quinolines (2b, 3b, 5b-10b, 12b), pyrimidines (1c,
3c-5c), hydroxypyrones (7d-12d), hydroxypyridinones
(10e-12e), salicylic acid (5f), hydroxyquinone (5g), a bipyr-
idine derivative (6g), tropolones (11g, 12g), naphthalene diol
(1h), and a benzoic acid derivative (3h). By comparing the
inhibitionofMMPsacrossa chelatorclass, informationabout
binding mode can be inferred. The most potent quinolines (2b,
7b-9b) are all substituted with a hydroxyl group at the
8-position. The similarities in structures and activities of these
compounds support a conserved binding mode with the
quinoline nitrogen and the hydroxyl oxygen atoms binding
to the zinc(II) ion.
Perhaps the most important result from the screens of CFL-
1.1 against the five MMPs is the commonality of hits across
the enzyme family. Compounds that inhibit one MMP tend to
inhibit all of the MMPs tested. Conversely, the set of frag-
ments that fail to inhibit at 1 mM are the same for all five
MMPs. This can be readily recognized in the thermoplot
(Figure 2) by comparing the inhibition of hits (9b, 4c, 2d,
12e, and 11g) and “non-hits” (6a, 6c, 8e, 1f, and 8h) across all
five MMPs. Because the zinc(II) active site is highly conserved
across MMP family members, few differences in chelator
potency are expected among different MMP enzymes. How-
ever, some recent reports have indicated that in combination
with an appropriate backbone group, the selection of a given
chelator can augment the selectivity of an MMPi.30
One thing the aforementioned proteins (MMPs, LF, 5-LO,
TY) have in common is that they are all inhibited by a high
percentage of fragments from CFL-1.1. In contrast, few
fragments from CFL-1.1 inhibit iNOS, which is consistent
with the fact that fragments chosen for CFL-1.1 are designed
to bind metals through two or more donor atoms. CFL-1.1
was designed specifically for metalloenzymes with more
than one adjacent, vacant coordination site on the metal ion

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2 597
(i.e., most fragments provide two or more donor atoms for
binding). Because iNOS is a heme enzyme, only one axial
binding site is accessible on the iron ion, rendering most of the
fragments of CFL-1.1 ineffective due to an inability to chelate
the heme center ion. The ability to target some metalloen-
zymes, but not others, by rational selection of the library
fragmentsagainillustratesthe abilitytotailortheselibraries to
specific classes of metalloenzyme targets.
be elaborated at multiple positions, as demonstrated by
the preparation of a small, focused library based on an
8-hydroxyquinoline fragment. This focused library showed
that substitution at either the 5- or 7-positions of the 8-hydro-
xyquinoline gave potent leads against MMP-2, while deriva-
tives at the 2- and 4-positions did not. Overall this study shows
the value of chelator libraries in FBLD against metalloprotein
targets and should open up a wide variety of new scaffolds
from which new inhibitors can be devised.
Examination of the LE values reveals that the hits from the
screen of CFL-1.1 are very high-quality fragments for FBLD.
LE values account for both ligand size and potency and are a
good way to compare the quality of initial hits. Generally a LE
of >0.3 kcal/mol is considered promising for a fragment hit.41
Table 1 shows the ligand efficiencies of a number of chelator
hits against MMP-2, LF, 5-LO, and TY. Compound 1g
(acetohydroxamic acid) represents the fragment equivalent
of the most common MBG for many metalloenzyme inhibi-
tors including MMPs and LF.24,26,31,65 The LE of 1g against
MMP-2 is excellent, at 0.52 kcal/mol, but other hits from
CFL-1.1 far exceed this value. For example, compounds 4c
and 2d demonstrate extraordinary LE values of 0.70 and 0.73
kcal/mol against MMP-2, respectively. Similarly, fragment 2d
has a higher LE for LF when compared with 1g (Table 1).
Several of the chelators identified as 5-LO and TY hits also
show very high LE values, demonstrating the usefulness of
CFL-1.1 for identifying scaffolds for elaboration into potent
metalloprotein inhibitors.
The ability to elaborate these MBG fragments into more
advanced hits is essential for inhibitor development. Unlike
hydroxamic acids, many of the chelators from CFL-1.1 may
be modified at more than one position due to their cyclic
structures. Asa proof-of-concept, derivativesof7b, which was
a potent hit against MMP-2 (Table 2), were synthesized with
substituents at one of four different positions around the
8-hydroxyquinoline ring. As seen in Figure 4, the positioning
of the backbone has a pronounced effect on the potency of the
inhibitor. The fragments with substituents on the 5- and
7-positions of 8-hydroxyquinoline inhibit MMP-2 with mi-
cromolar IC50 values, while derivatives with backbones at the
2-position or 4-position show little efficacy when screened at
25 μM. Interestingly, the structural differences between these
backbones do not affect inhibitor potency as significantly as
the position of the groups on the MBG. It is likely that
differences in the relative orientation of the fragment in the
active site, imposed by the metal-ligand interactions, con-
tribute to the observed differences in activity. The synthetic
flexibility offered by chelators with multiple sites of derivitza-
tion (such as 7b) may be used to probe productive binding
pockets surrounding the metal ion. In addition, having multi-
ple sites of substitution demonstrates that one hit from CFL-
1.1 can actually produce multiple sublibraries and advanced
hits of different connectivity (e.g., 5- and 7-substituted
fragments). Sublibraries can also be developed with fragments
bearing more than one substituent on these cycle scaffolds.
Experimental Section
General. Unless otherwise noted, starting materials were
purchased from commercial suppliers (Sigma-Aldrich, Chem-
Bridge, Acros Organics) and were used without further purifica-
tion. All commercial materials were listed as 95% purity or
greater and were used without further purification. The purity of
all synthesized compounds was determined to be g95% by
either elemental analysis or HPLC. Flash silica gel chromatog-
raphy was performed using Merck silica gel 40-63 μm mesh. 1H
NMR spectra were recorded on one of several Varian FT-NMR
spectrometers, property of the Department of Chemistry and
Biochemistry, University of California, San Diego. Mass spec-
trometry was performed at the Small Molecule Mass Spec-
trometry Facility in the Department of Chemistry and Biochem-
istry, University of California, San Diego. Microwave reactions
were performed in 10 mL vials using a CEM Discover S-class
microwave reactor. For all assays, IC50 values were obtained
for select fragments that showed g50% inhibition. IC50 values
of hits were obtained by preparing serial dilutions of DMSO
stock solutions for each compound. IC50 values were calculated
from the plotted dose-response curve using GraphPad Prism 5
software.
Chelator Fragment Library 1.1 (CFL-1.1): Synthesis and
Characterization. CFL-1.1 was assembled as previously de-
scribed for CFL-1 with the following exceptions:42 four com-
pounds in CFL-1 were replaced with those shown in Figure 1
(6b, 10b, 7f, and 2h). Compounds 6b, 7f, and 2h from CFL-1.1
were obtained from commercial suppliers, while compound 10b
was synthesized as described below. Compounds were stored as
solids or in DMSO stock solutions at a concentration of 50 mM.
Fragment 10b. 8-Aminoquinoline (200 mg, 1.39 mmol) and
methanesulfonyl chloride (162 μL, 2.08 mmol) were dissolved in
3 mL of pyridine. The solution was irradiated in a microwave
synthesizer for 3 min at 130 °C.61 The reaction mixture was
poured over 10 mL of H2O, and the precipitate was isolated by
1
vacuum filtration. Yield: 210 mg (68%). H NMR (500 MHz,
DMSO-d6): δ 9.34 (s, br, 1H), 8.90 (dd, J = 4.0, 1.7 Hz, 1H), 8.40
(dd, J = 8.6, 1.7 Hz, 1H), 7.70 (m, 2H), 7.62 (q, J = 4 Hz, 1H),
7.57 (t, J = 8 Hz, 1H), 3.12 (s, 3H). ESI-MS(þ): m/z 223.22 [M þ
H]þ. 13C NMR (125 MHz, DMSO-d6): δ 149.9, 139.3, 137.1,
134.7, 128.7, 127.3, 123.3, 122.9, 117.2. Anal. Calcd for C10H10-
N2O2S: C, 54.04; H, 4.53; N, 12.60. Found: C, 53.70; H, 4.37; N,
12.50.
N-(8-Hydroxyquinolin-2-yl)benzenesulfonamide (72a). 2-Amino-
8-hydroxyquinoline (100 mg, 0.624 mmol) and benzenesulfonyl
chloride (236 μL, 1.87 mmol) were dissolved in 3 mL of pyridine.
The reaction mixture was irradiated in a microwave synthesizer at
130 °C for 6 min. The reaction was quenched with 10 mL of water
and then extracted with 10 mL of CH2Cl2. The organic layer was
dried over MgSO4, filtered, and dried to yield a white powder. This
white powder is the intermediate 2-(phenylsulfonamido)quinolin-
8-yl benzenesulfonate, which contains a sulfonamide moiety at the
2-position as well as a sulfonate ester group at the 8-position of the
ring. Therefore, to remove the sulfonate ester, the intermediate was
dissolved in 2-3 mL of MeOH, followed by addition of 3 mL of 2
M NaOH dropwise, after which the mixture was heated to reflux for
4-6 h under N2. After heating, the solution was allowed to cool
to room temperature, and the MeOH was removed under vacuum.
Conclusions
This work describes the development and evaluation of a
fragment library based on metal chelators. A diverse collec-
tion of ligands is presented in order to offer alternatives to the
common metal-binding motifs found in most metalloenzyme
inhibitors. CFL-1.1 produced high hit rates against MMPs,
LF, 5-LO, and TY, but not against the heme-dependent
iNOS. Many of the chelators identified as new MBGs may

598 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
To the aqueous solution was added 1 M HCl dropwise until a
precipitate formed. The precipitate was filtered, rinsed with water,
and purified by flash chromatography on silica with 0-3% MeOH
in CH2Cl2 as the eluant. A white solid was isolated. Yield: 40 mg
(21%). 1H NMR (400 MHz, DMSO-d6): δ 12.12 (s, br, 1H), 11.04
(s, br, 1H), 8.18 (d, J = 9.2 Hz, 1H), 7.89 (s, br, 2H), 7.59-7.53 (m,
3H), 7.29-7.22 (m, 2H), 7.12 (d, J = 6.4 Hz, 1H), 7.06 (d, J = 10
Hz, 1H). ESI-MS (þ): m/z 301.26 [M þ H]þ. 13C NMR (125 MHz,
DMSO-d6): δ 143.3, 141.9, 132.6, 132.6, 126.3, 125.3, 118.7, 115.4.
Anal. Calcd for C15H12N2O3S: C, 59.99; H, 4.03; N, 9.33. Found:
C, 60.23; H, 4.23; N, 9.39.
4-Fluoro-N-(8-hydroxyquinolin-2-yl)benzenesulfonamide (72b).
Compound 72b was synthesized according to the procedure
described for 72a starting from 2-amino-8-hydroxyquinoline
(100 mg, 0.624 mmol) and 4-fluorobenzene-1-sulfonyl chloride
(364 mg, 1.87 mmol). Purification of this compound did not
require column chromatography. Yield: 74 mg (37%). 1H NMR
(400 MHz, DMSO-d6): δ 12.17 (s, br, 1H), 11.05 (s, br, 1H), 8.19
(d, J = 9.2 Hz, 1H), 7.95 (s, br, 2H), 7.38 (t, J = 8.8 Hz, 2H),
7.30-7.23 (m, 2H), 7.13 (d, J = 7.2 Hz, 1H), 7.08 (d, J = 9.6 Hz,
1H). 13C NMR (125 MHz, DMSO-d6): δ 165.3, 163.3, 129.2,
125.4, 118.7, 116.6, 116.4. ESI-MS (þ): m/z 319.24 [M þ H]þ.
Anal. Calcd for C15H11FN2O3S: C, 56.60; H, 3.48; N, 8.80.
Found: C, 56.99; H, 3.82; N, 8.74.
described for 74a starting from 4-amino-8-hydroxyquinoli-
ne HBr (100 mg, 0.41 mmol) and 4-fluorobenzene-1-sulfonyl
3
chloride (126 mg, 0.65 mmol). Synthesis of this compound
required 20 min of microwave irradiation. A white solid was
isolated. Yield: 52 mg (40%). 1H NMR (400 MHz, CD3OD) δ
8.49 (s, 1H), 8.04, (t, J = 9.2 Hz, 1H), 7.87-7.84 (m, 2H), 7.59
(d, J = 6.9 Hz, 1H), 7.40 (t, J = 7.4 Hz, 1H), 7.16 (t, J = 6.9 Hz,
2H), 6.58 (d, J = 3.4 Hz, 1H). ESI-MS(þ): m/z 319.02 [M þ H]þ.
13C NMR (125 MHz, DMSO-d6) δ 166.48, 164.46, 160.9, 131.9,
131.8, 131.8, 127.9, 127.8, 122.2, 116.7, 116.5, 114.4, 114.2, 102.9.
Anal. Calcd for C15H11FN2O3S 0.73CH3OH 0.14H2O: C,
3
3
54.84; H, 4.89; N, 8.13. Found: C, 55.36; H, 4.02; N, 7.75.
N-(8-Hydroxyquinolin-4-yl)-[1,10-biphenyl]-4-sulfonamide (74c).
Compound 74c was synthesized according to the procedure de-
scribed for 74a starting from 4-amino-8-hydroxyquinoline HBr
3
(100 mg, 0.41 mmol) and [1,10-biphenyl]-4-sulfonyl chloride (157
mg, 0.62 mmol). Synthesis of this compound required 10 min of
microwave irradiation. A white solid was isolated. Yield: 30 mg
(20%). 1H NMR (400 MHz, CDCl3): δ 8.30 (s, 1H), 7.00 (d, J =
8.6 Hz, 2H), 7.69 (d, J= 6.9 Hz, 1H), 7.61 (d, J= 8.0 Hz, 2H), 7.55
(t, J = 8.5 Hz, 3H), 7.47 (t, J = 6.3 Hz, 2H), 7.40 (t, J = 7.4 Hz,
1H) 7.32 (t, J = 8.0 Hz, 1H), 6.49 (s, 1H). ESI-MS(þ): m/z 377.09
[M þ H]þ. 13C NMR (125 MHz, DMSO-d6): δ 146.2, 138.5, 134.7,
129.6, 129.4, 129.3, 128.6, 127.7, 127.6, 127.6, 127.5, 122.3, 103.3.
N-(8-Hydroxyquinolin-2-yl)-[1,10-biphenyl]-4-sulfonamide (72c).
Compound 72c was synthesized according to the procedure de-
scribed for 72a starting from 2-amino-8-hydroxyquinoline (100
mg, 0.624 mmol) and [1,10-biphenyl]-4-sulfonyl chloride (473 mg,
1.87 mmol). Yield: 26 mg (11%). 1H NMR (400 MHz, DMSO-d6):
δ 12.19 (s, br, 1H), 11.07 (s, br, 1H), 8.19 (d, J = 9.2 Hz, 1H), 7.95
(d, J = 5.6 Hz, 2H), 7.83 (d, J = 8 Hz, 2H), 7.69 (d, J = 7.6 Hz,
2H), 7.48 (t, J = 7.6 Hz, 2H), 7.41 (t, J = 6.8 Hz, 1H), 7.30-7.22
(m, 2H), 7.13(d, J= 7.6 Hz, 1H), 7.08 (d, J=10Hz, 1H). ESI-MS
(þ): m/z 377.28 [M þ H]þ. 13C NMR (125 MHz, DMSO-d6): δ
139.1, 129.5, 128.8, 127.7, 127.5, 125.3, 118.7, 118.7. Anal. Calcd
Anal. Calcd for C21H16N2O3S 0.92 CH3OH: C, 64.86; H, 4.89; N,
3
6.90. Found: C, 64.68; H, 4.59; N, 6.49.
N-(8-Hydroxyquinolin-4-yl)-thiophene-2-sulfonamide (74d).
Compound 74d was synthesized according to the procedure
described for 74a starting from 4-amino-8-hydroxyquinoli-
ne HBr (100 mg, 0.41 mmol) and thiophene-2-sulfonyl chloride
3
(112 mg, 0.61 mmol). A white solid was isolated. Yield: 54 mg
1
(44%). H NMR (400 MHz, CD3OD): δ 8.09 (d, J = 5.2 Hz,
1H), 8.01 (d, J = 8.6 Hz, 1H), 7.79 (d, J = 5.2 Hz, 1H), 7.61
(d, J = 3.4 Hz, 1H), 7.57 (d, J = 8.0 Hz, 1H), 7.38 (t, J = 8.0 Hz,
1H), 7.02 (t, J = 8.6 Hz, 1H), 6.55 (d, J = 5.8 Hz, 1H). 13C NMR
(125 MHz, DMSO-d6): δ 151.6, 150.6, 144.9, 142.0, 136.4, 136.0,
134.4, 127.8, 122.5, 122.1, 121.6, 120.1, 103.0. ESI-MS(þ): m/z
307.03 [M þ H]þ. Anal. Calcd for C13H10N2O3S2: C, 50.97; H,
3.29; N, 9.14. Found: C, 50.68; H, 3.64; N, 9.10.
for C21H16N2O3S 0.46 CH3OH: C, 65.89; H, 4.60; N, 7.16. Found:
C, 65.92; H, 4.67; N, 7.45.
3
N-(8-Hydroxyquinolin-2-yl)thiophene-2-sulfonamide (72d). Com-
pound 72d was synthesized according to the procedure described
for 72a starting from 2-amino-8-hydroxyquinoline (100 mg, 0.624
mmol) and thiophene-2-sulfonyl chloride (342 mg, 1.87 mmol).
Yield: 30 mg (16%). 1H NMR (400 MHz, CD3OD): δ 8.17 (d, J =
9.2 Hz, 1H), 7.72 (m, 2H), 7.29 (m, 2H), 7.15 (dd, J = 5.6, 2.8 Hz,
1H), 7.10 (m, 2H). ESI-MS (þ): m/z 307.30 [M þ H]þ. 13C NMR
(125 MHz, DMSO-d6) δ 148.8, 142.5, 132.1, 130.7, 127.8, 125.5,
118.7, 115.7. HPLC: 98.39% pure.
N-(8-Hydroxyquinolin-5-yl)benzenesulfonamide (75a). 5-Amino-
8-hydroxyquinoline 2HCl (100 mg, 0.429 mmol) and benzenesul-
3
fonyl chloride (81 μL, 0.644 mmol) were dissolved in 3 mL of
pyridine. The reaction was allowed to stir overnight under N2. The
reaction was quenched with 10 mL of water and extracted with
10 mL of CH2Cl2. The organic layer was dried over MgSO4 and
purified by flash silica chromatography in 0-3% methanol in
CH2Cl2. Yield: 22 mg (17%). 1H NMR (400 MHz, DMSO-d6): δ
9.98 (s, 2H), 8.80 (d, J = 4 Hz, 1H), 8.26 (d, J = 8.4 Hz, 1H), 7.59
(d, J = 8 Hz, 3H), 7.51-7.45 (m, 3H), 6.94 (s, 2H). 13C NMR (125
MHz, DMSO-d6): δ 152.7, 148.2, 139.4, 138.4, 132.7, 132.1, 129.1,
126.8, 126.5, 126.4, 122.4, 121.6, 110.5. ESI-MS (þ): m/z 301.19
[M þ H]þ. Anal. Calcd for C15H12N2O3S: C, 59.99; H, 4.03; N,
9.33. Found: C, 59.88; H, 4.40; N, 9.70.
4-Amino-8-hydroxyquinoline HBr. 4-Amino-8-hydroxyqui-
3
noline was prepared according to a literature procedure start-
ing from 4-amino-8-methoxyquinoline (100 mg, 0.57 mmol).80
Yield: 130 mg (84%). 1H NMR (400 MHz, D2O): δ 7.77 (d, J =
6.4 Hz, 1H), 7.10 (d, J = 6.4 Hz, 2H), 6.93 (dd, J = 6.8, 2.4 Hz,
1H), 6.40 (d, J = 7.2 Hz, 1H). ESI-MS(þ): m/z 161.33 [M þ H]þ.
13C NMR (125 MHz, DMSO-d6) δ 149.9, 139.3, 137.1, 134.7,
128.7, 127.3, 123.3, 122.9, 117.2.
4-Fluoro-N-(8-hydroxyquinolin-5-yl)benzenesulfonamide (75b).
Compound 75b was synthesized according to the procedure
described for 75a starting from 5-amino-8-hydroxyquinoli-
N-(8-Hydroxyquinolin-4-yl)benzenesulfonamide (74a). 4-Amino-
8-hydroxyquinoline HBr (50 mg, 0.21 mmol) and benzenesulfonyl
3
chloride (29 μL, 0.23 mmol) were dissolved in 3 mL of pyridine.
The reaction mixture was irradiated in a microwave synthesizer at
130 °C for 6 min. The reaction was quenched with 10 mL of water
and then extracted with 10 mL of CH2Cl2. The organic layer was
dried with MgSO4, filtered, and dried to yield a white solid. Yield:
16 mg (25%). 1H NMR (400 MHz, CD3OD): δ 8.05 (d, J = 6 Hz,
1H), 7.98 (d, J = 8.4 Hz, 1H), 7.82 (d, J = 8.4 Hz, 2H), 7.57 (q, J =
6.4 Hz, 2H), 7.42 (t, J = 8 Hz, 2H), 7.35 (t, J = 8 Hz, 1H), 6.53
(d, J = 5.6 Hz, 1H). ESI-MS(þ): m/z 301.05 [M þ H]þ. 13C NMR
(125 MHz, DMSO-d6) δ 167.6, 163.4, 160.6, 150.4, 148.8, 134.9,
134.5, 129.7, 128.7, 122.4, 119.8, 103.4. Anal. Calcd for C15H12-
N2O3S: C, 59.99; H, 4.03; N, 9.33. Found: C, 59.73; H, 4.08; N, 9.32.
4-Fluoro-N-(8-hydroxyquinolin-4-yl)benzenesulfonamide (74b).
Compound 74b was synthesized according to the procedure
ne 2HCl (100 mg, 0.429 mmol) and 4-fluorobenzenesulfonyl
chloride (125 mg, 0.644 mmol). Yield: 24 mg (21%). H NMR
3
1
(400 MHz, DMSO-d6): δ 10.03 (s, br, 2H), 8.81 (dd, J = 4, 1.6Hz,
1H), 8.26 (dd, J = 8.8, 1.6 Hz, 1H), 7.63 (dd, J = 8.8, 5.2 Hz, 2H),
7.49 (dd, J = 8.4, 4 Hz, 1H), 7.34 (t, J = 8.8 Hz, 2H), 6.95 (s, 2H).
ESI-MS (þ): m/z 319.24 [M þ H]þ. 13C NMR (125 MHz,
DMSO-d6): δ 163.2, 152.8, 148.3, 148.3, 138.4, 135.8, 132.1,
129.9, 129.8, 126.7, 126.4, 122.2, 121.7, 116.35, 116.2, 110.5. Anal.
Calcd for C15H11FN2O3S 0.29CH3OH: C, 56.05; H, 3.74; N,
8.55. Found: C, 55.88; H, 3.77; N, 8.91.
3
N-(8-Hydroxyquinolin-5-yl)-[1,10-biphenyl]-4-sulfonamide (75c).
Compound 75c was synthesized according to the procedure de-
scribed for 75a starting from 5-amino-8-hydroxyquinoline 2HCl
3
(100 mg, 0.429 mmol) and [1,10-biphenyl]-4-sulfonyl chloride (163 mg,

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2 599
0.644 mmol). Yield: 60 mg (37%). 1H NMR (400 MHz, DMSO-
d6):δ10.03 (s, br, 1H), 9.99 (s, br, 1H), 8.79 (dd, J=4, 1.6Hz, 1H),
8.31 (dd, J = 8.4, 1.2 Hz, 1H), 7.79 (d, J = 8.4 Hz, 2H), 7.68
(dd, J = 13.2, 7.2 Hz, 4H), 7.50-7.43 (m, 4H), 6.98 (q, J = 8 Hz,
2H). 13C NMR (125 MHz, DMSO-d6):δ152.7, 148.2, 144.0, 138.4,
138.3, 132.2, 129.1, 128.6, 127.5, 127.2, 127.0, 126.5, 126.4, 122.4,
122.4, 121.6, 110.5. ESI-MS (þ): m/z 377.16 [M þ H]þ. Anal.
Calcd for C21H16N2O3S: C, 67.00; H, 4.28; N, 7.45. Found: C,
66.77; H, 4.66; N, 7.45.
(400 MHz, DMSO-d6): δ 9.93 (s, br, 2H), 8.81 (m, 1H), 8.29 (dd,
J = 8.4, 2 Hz, 1H), 7.83 (q, J = 8 Hz, 4H), 7.71 (d, J = 6.4 Hz,
2H), 7.57 (dd, J = 8.8, 3.6 Hz, 1H), 7.51-7.47 (m, 3H), 7.44 (dd,
J = 8, 3.2 Hz, 1H), 7.38 (dd, J = 8.8, 3.2 Hz, 1H). 13C NMR (125
MHz, DMSO-d6): δ 148.5, 145.7, 143.9, 139.6, 138.4, 138.3, 136.0,
129.1, 128.5, 127.3, 127.1, 127.0, 126.3, 124.8, 121.3, 120.7, 117.1.
ESI-MS (þ):m/z 377.10 [M þ H]þ. Anal. Calcd for C21H16N2O3S:
C, 67.00; H, 4.28; N, 7.45. Found: C, 66.78; H, 4.68; N, 7.28.
N-(8-Hydroxyquinolin-7-yl)thiophene-2-sulfonamide (77d).
Compound 77d was synthesized according to the procedure
described for 74a starting from 7-amino-8-hydroxyquino-
line (100 mg, 0.625 mmol) and thiophene-2-sulfonyl chloride
(113 mg, 0.618 mmol). The product was purified by flash silica
chromatography in 0-3% MeOH in CH2Cl2 followed by
N-(8-Hydroxyquinolin-5-yl)thiophene-2-sulfonamide (75d).
Compound 75d was synthesized according to the procedure
described for 75a starting from 5-amino-8-hydroxyquinoli-
ne 2HCl (100 mg, 0.429 mmol) and thiophene-2-sulfonyl chlor-
3
ide (118 mg, 0.644 mmol). Yield: 21 mg (16%). 1H NMR
(400 MHz, CD3OD): δ 8.77 (dd, J = 4, 1.6 Hz, 1H), 8.33 (dd,
J = 8.4, 1.6 Hz, 1H), 7.71 (dd, J = 4.8, 1.6 Hz, 1H), 7.43 (dd,
J = 8.4, 4 Hz, 1H), 7.33 (dd, J = 4, 1.6 Hz, 1H), 7.10 (d, J = 8.4
Hz, 1H), 7.03 (dd, J = 5.2, 3.6 Hz, 1H), 6.98 (d, J = 8 Hz, 1H).
13C NMR (125 MHz, DMSO-d6): δ 153.3, 148.7, 140.3, 138.8,
133.7, 132.8, 132.4, 128.1, 127.2, 126.9, 122.6, 122.2, 110.9. ESI-
MS (þ): m/z 307.21 [M þ H]þ. Anal. Calcd for C13H10N2O3S2:
C, 50.97; H, 3.29; N, 9.14. Found: C, 50.82; H, 3.60; N, 9.48.
5-Chloro-8-hydroxy-7-nitroquinoline. 5-Chloro-8-hydroxy-
7-nitroquinoline was prepared according to a literature pro-
cedure.74 Yield: 1.6 g (66%). 1H NMR (400 MHz, DMSO-d6): δ
9.10 (d, J = 4.4 Hz, 1H), 8.61 (d, J = 9.2 Hz, 1H), 8.21 (s, 1H),
7.95 (dd, J = 8.4, 4.4 Hz, 1H). ESI-MS(-): m/z 223.35 [M -
H]-. 13C NMR (125 MHz, DMSO-d6): δ 150.9, 150.5, 140.4,
134.2, 132.8, 129.0, 126.4, 122.4, 118.4.
1
recrystallization in EtOH. Yield: 27 mg (14%). H NMR (400
MHz, DMSO-d6): δ 9.97 (s, br, 1H), 8.80 (d, J = 4.4 Hz, 1H),
8.28 (d, J = 8 Hz, 1H), 7.83 (d, J = 3.6 Hz, 1H), 7.49 (d, J = 9.2
Hz, 4H), 7.35 (d, J = 8.8 Hz, 1H), 7.06 (t, J = 4.4 Hz, 1H). ESI-
MS (þ): m/z 306.98 [M þ H]þ. 13C NMR (125 MHz, DMSO-d6):
δ 148.5, 146.5, 141.1, 138.4, 136.1, 132.9, 131.8, 127.4, 126.6,
125.4, 121.5, 120.2, 117.1. Anal. Calcd for C15H12N2O3S: C,
49.85; H, 4.06; N, 8.39. Found: C, 49.38; H, 4.24; N, 8.05.
Fluorometric Screening against MMPs. CFL-1.1 was
screened against MMP-1, -2, -3, -8, and -9 at a concentration
of 1 mM for each fragment. The assay was carried out in white
NUNC 96-well plates as previously described.64 Each well
contained a total volume of 90 μL including buffer (50 mM
HEPES, 10 mM CaCl2, 0.05% Brij-35, pH 7.5), human
recombinant MMP (ENZO Life Sciences; 15.3 U of MMP-
1, 1.16 U of MMP-2, 2 U of MMP-3, 1.84 U of MMP-8, or
0.9 U of MMP-9), and the fragment solution (1 mM final
concentration). After a 30 min incubation period at 37 °C,
the reaction was initiated by the addition of 10 μL of the
fluorogenic MMP substrate (4 μM final concentration, Mca-
7-Amino-8-hydroxyquinoline. 7-Amino-8-hydroxyquinoline
was prepared according to a literature procedure.74 Yield: 406
mg (95%). 1H NMR (300 MHz, DMSO-d6): δ 10.16 (s, br, 1H),
8.77 (d, J = 7.5 Hz, 1H), 8.71 (d, J = 5.7 Hz, 1H), 8.34 (d, J = 3
Hz, 1H), 7.66 (d, J = 8.7 Hz, 1H), 7.48 (t, J = 5.4 Hz, 1H), 7.39
(d, J = 8.7 Hz, 1H). ESI-MS(þ): m/z 161.32 [M þ H]þ. 13C
NMR (125 MHz, DMSO-d6) δ 145.3, 142.7, 131.0, 123.2, 122.5,
122.2, 121.2, 117.1, 115.8.
N-(8-Hydroxyquinolin-7-yl)benzenesulfonamide (77a). Com-
pound 77a was synthesized according to the procedure de-
scribed for 74a starting from 7-amino-8-hydroxyquinoline
(100 mg, 0.625 mmol) and benzenesulfonyl chloride (78 μL,
0.618 mmol). The product was purified by flash silica chroma-
tography in 0-3% MeOH in CH2Cl2 followed by recrystalliza-
tion in EtOH. Yield: 11 mg (6%). 1H NMR (400 MHz, DMSO-
d6): δ 9.81 (s, br, 2H), 8.76 (s, 1H), 8.24 (d, J = 8.4 Hz, 1H), 7.74
(d, J = 6.8 Hz, 2H), 7.56-7.51 (m, 1H), 7.47 (d, J = 7.6 Hz, 4H),
7.31 (dd, J = 8.8, 2.8 Hz, 1H). ESI-MS (þ): m/z 301.19 [M þ
H]þ. 13C NMR (125 MHz, DMSO-d6): δ 148.9, 146.2, 141.2,
138.8, 136.5, 133.0, 129.4, 127.0, 126.8, 125.2, 121.8, 121.1,
117.5. Anal. Calcd for C15H12N2O3S: C, 59.99; H, 4.03; N,
9.33. Found: C, 59.97; H, 4.34; N, 9.20.
4-Fluoro-N-(8-hydroxyquinolin-7-yl)benzenesulfonamide (77b).
Compound 77b was synthesized according to the procedure
described for 74a starting from 7-amino-8-hydroxyquinoline
(100 mg, 0.625 mmol) and 4-fluorobenzene-1-sulfonyl chloride
(120 mg, 0.618 mmol). The product was purified by recrystalliza-
tion in EtOH. Yield: 52 mg (26%). 1H NMR (300 MHz, DMSO-
d6): δ 9.84 (s, br, 1H), 8.78 (dd, J = 3.6, 0.9 Hz, 1H), 8.27 (dd, J =
8.4, 1.8 Hz, 1H), 7.77 (dd, J = 9, 5.7 Hz, 2H), 7.51-7.46 (m, 2H),
7.36-7.28 (m, 3H). ESI-MS(þ): m/z 319.03 [M þ H]þ. 13C NMR
(125 MHz, DMSO-d6): δ 148.9, 146.7, 138.8, 137.5, 136.5, 130.1,
130.0, 127.0, 125.9, 121.9, 120.8, 117.6, 116.6, 116.4. Anal. Calcd
Pro-Leu-Gly-Leu-Dpa-Ala-Arg-NH2 AcOH, ENZO Life
3
Sciences). Fluorescence measurements were recorded using
a Bio-Tek Flx 800 fluorescence plate reader every minute for
20 min with excitation and emission wavelengths at 320 and
400 nm, respectively. The rate of fluorescence increase was
compared for samples versus negative controls (no inhibitor,
arbitrarily set as 100% activity). Eighteen compounds from
CFL-1.1 were omitted from the screening due to excessive
background fluorescence that interfered with the assay read-
ings. A separate buffer (50 mM MES, 10 mM CaCl2, 0.05%
Brij-35, pH 6.0) was used for all experiments with MMP-3.
The 8-hydroxyquinoline sublibrary was screened against
MMP-2 using the procedure described above with a final
fragment concentration in each well of 25 μM.
Colorimetric Screening against MMPs. Due to excessive
background fluorescence, 18 compounds from CFL-1.1 were
screened in a colorimetric assay against MMP-1, -2, -3, -8, and -9
at a concentration of 1 mM for each fragment. The assay was
carried out in clear Costar 96-well, half-area, flat-bottom assay
plates. Each well contained a total volume of 90 μL including
buffer (50 mM HEPES, 10 mM CaCl2, 0.05% Brij-35, 1 mM
DTNB, pH 7.5), human recombinant MMP (ENZO Life Scien-
ces), and the fragment solution (1 mM final concentration).
After a 30 min incubation period at 37 °C, the reaction was
initiated by the addition of 10 μL of chromogenic MMP sub-
strate (500 μM final concentration, Ac-Pro-Leu-Gly-[2-mercap-
to-4-methyl-pentanoyl]-Leu-Gly-OC2H5, ENZO Life Sciences).
Absorbance was monitored at 405 nm using a Bio-Tek ELx 808
colorimetric plate reader, and measurements were recorded
every minute for 20 min. The rate of absorbance increase was
compared for samples versus negative controls (no inhibitor,
arbitrarily set as 100% activity). A separate buffer (50 mM
MES, 10 mM CaCl2, 0.05% Brij-35, 1 mM DNTB, pH 6.0) was
used for all experiments with MMP-3.
for C15H11FN2O3S 0.29H2O: C, 55.74; H, 3.60; N, 8.67. Found:
3
C, 55.54; H, 4.02; N, 9.11.
N-(8-Hydroxyquinolin-7-yl)-[1,10-biphenyl]-4-sulfonamide (77c).
Compound 77c was synthesized according to the procedure de-
scribed for 74a starting from 7-amino-8-hydroxyquinoline (100
mg, 0.625 mmol) and [1,10-biphenyl]-4-sulfonyl chloride (156 mg,
0.618 mmol). The product was purified by flash silica chromatog-
raphy in 0-3% MeOH in CH2Cl2. Yield: 30 mg (13%). 1H NMR
Screening against LF. CFL-1.1 was screened against LF at a
concentration of 1 mM for each fragment. The assay was carried

600 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
out in white NUNC 96-well plates as previously described.42
Each well contained a volume of 90 μL including buffer (20 mM
HEPES, pH 7.4), recombinant LF (10 nM final concentration,
List Biological Laboratories), and the fragment solution (1 mM
final concentration). After a 20 min incubation period at 25 °C,
the reaction was initiated by the addition of 10 μL of fluorogenic
LF substrate (2 μM final concentration, MAPKKide DAB-
CYL/FITC, List Biological Laboratories). Fluorescence mea-
surements were recorded using a Bio-Tek Flx 800 fluorescence
plate reader every minute for 20 min with excitation and
emission wavelengths at 485 and 528 nm, respectively. The rate
of fluorescence increase was compared for samples versus
negative controls (no inhibitor, arbitrarily set as 100% activity).
During this screen, four compounds (10b, 11b, 10c, 5e) were
excluded from the assay due to precipitation or interference with
the assay.
Screening against 5-LO. CFL-1.1 was screened against 5-LO
at a concentration of 1 mM for each fragment. The assay was
performed according to a literature procedure at room
temperature.63 Each well contained a volume of 80 μL including
buffer (50 mM Tris, 2 mM EDTA, 2 mM CaCl2, pH 7.5), human
recombinant 5-LO (0.2 U, Cayman Chemicals), reporter dye
(20,70-dichlorofluorescin diacetate; H2DCFDA, 10 μM, In-
vitrogen), fragment solution (1 mM), arachidonic acid (AA,
3 μM, Fischer Scientific), and adenosine triphosphate (ATP,
10 μM, Sigma-Aldrich). H2DCFDA and 5-LO were incubated
for 5 min prior to the addition of the fragment solution. This was
followed by a second incubation for 10 min. The reaction was
initiated by the addition of a substrate solution containing AA
and ATP. The reaction was monitored using a Bio-Tek Flx 800
fluorescence plate reader. Fluorescence measurements were
recorded every minute for 20 min with excitation and emission
wavelengths at 485 and 528 nm, respectively. The rate of
fluorescence increase was compared for samples versus negative
controls (no inhibitor, arbitrarily set as 100% activity). The
5-LO screen of CFL-1.1 resulted in nine compounds (3b, 4d, 6d,
10d, 10e, 8f, 1g, 10g, and 12g) that significantly increased the
fluorescence of the assay relative to the control with no inhi-
bitor. The origin of this interference was not identified, and these
compounds were excluded from further examination with re-
spect to 5-LO.
solution (prepared as directed by supplier) was added (10 μL)
followed by 10 μL of a freshly made stock of 1 mM NADPH to
initiate the reaction. After 40 min incubation, the reaction was
stopped by heat inactivation of the iNOS (incubation for 30 s in
boiling water). In order to destroy excess NADPH, 10 μL of a
lactate dehydrogenase (LDH) solution and 10 μL of an LDH
cofactor solution were added followed by 20 min incubation.
Following this incubation, a background reading of absorbance
was taken at 540 nm using a Bio-Tek ELx 808 colorimetric plate
reader. Griess Reagent 1 (1% sulfanilamide in 5% phosphoric
acid) was added (50 μL) to each well followed by 50 μL of Griess
Reagent 2 (0.1% N-(1-naphthyl)ethylenediamine 2HCl in
3
water). The reagents were allowed to develop for 10 min prior
to collecting a second absorbance measurement at 540 nm. After
subtracting the background absorbance, the remaining absor-
bance of the negative controls (no inhibitor) was arbitrarily set
as 100% activity. The ratio of absorbance between inhibitor and
control wells was defined as percent iNOS activity.
Acknowledgment. We thank Dr. Y. Su (U.C.S.D.) for
performing mass spectrometry experiments and Dr. Noeris
€
K. Salam (Schrodinger, Inc.) for helpful discussions. This
work was supported by the University of California, San
Diego, the National Science Foundation (instrumentation
grants CHE-9709183, CHE-0116662, and CHE-0741968),
and the National Institutes of Health (Grants R01 HL080049;
R21 HL094571).
Supporting Information Available: Tables of percent inhibi-
tion of CFL-1.1 fragments against the tested proteins. This
material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Lu, Y. Metalloprotein and metallo-DNA/RNAzyme design: Cur-
rent approaches, success measures, and future challenges. Inorg.
Chem. 2006, 45, 9930–9940.
(2) McRae, R.; Bagchi, P.; Sumalekshmy, S.; Fahrni, C. J. In situ
imaging of metal in cells and tissues. Chem. Rev. 2009, 109, 4780–
4827.
(3) Seffernick, J. L.; McTavish, H.; Osborne, J. P.; de Souza, M. L.;
Sadowsky, M. J.; Wackett, L. P. Atrazine chlorohydrolase from
Pseudomonas sp. strain ADP is a metalloenzyme. Biochemistry
2002, 41, 14430–14437.
(4) Hou, C.-X.; Dirk, L. M. A.; Goodman, J. P.; Williams, M. A.
Metabolism of the peptide deformylase inhibitor actinonin in
tobacco. Weed Sci. 2006, 54, 246–254.
Screening against TY. CFL-1.1 was screened against TY at a
concentration of 1 mM for each fragment. The assay was
performed according to a literature procedure at room
temperature.62 Each well contained a volume of 100 μL includ-
ing buffer (50 nM phosphate, pH 6.8), mushroom TY (30 U,
Sigma-Aldrich), fragment solution (1 mM), and L-dopamine
(0.5 mM, Sigma-Aldrich). Mushroom TY and the fragment
solution were incubated for 10 min. A background absorbance
reading at 475 nm was recorded using a Bio-Tek ELx 808
colorimetric plate reader. L-Dopamine was added to initiate
the reaction, which was allowed to proceed for 10 min before a
second absorbance reading at 475 nm was taken. After sub-
tracting the background absorbance, the remaining absorbance
of the negative controls (no inhibitor) was arbitrarily set as
100% activity. The ratio of absorbance between inhibitor and
control wells was defined as percent TY activity. Several com-
pounds from CFL-1.1 (6b-10b, 6g-9g, and 3h) interfered with
the TY assay, resulting in an increase in absorbance relative to
the negative control. The origin of this interference was not
identified, and these compounds were excluded from further
examination with respect to TY.
(5) Biou, V.; Dumas, R.; Cohen-Addad, C.; Douce, R.; Job, D.;
Pebay-Peyroula, E. The crystal structure of plant acetohydroxy
acid isomeroreductase complexed with NADPH, two magnesium
˚
ions and a herbicidal transition state analog determined at 1.65 A
resolution. EMBO J. 1997, 16, 3405–3415.
(6) Parvez, S.; Kang, M.; Chung, H.-S.; Bae, H. Naturally occuring
tyrosinase inhibitors: Mechanism and applications in skin health,
cosmetics and agriculture industries. Phytother. Res. 2007, 21, 805–
816.
(7) Smit, N.; Vicanova, J.; Pavel, S. The hunt for natural skin whiten-
ing agents. Int. J. Mol. Sci. 2009, 10, 5326–5349.
(8) Ma, X.; Ezzeldin, H. H.; Diasio, R. B. Histone deacetylase
inhibitors: Current status and overview of recent clinical trials.
Drugs 2009, 69, 1911–1934.
(9) Coussens, L. M.; Fingleton, B.; Matrisian, L. M. Matrix metallo-
proteinase inhibitors and cancer: trials and tribulations. Science
2002, 295, 2387–2392.
(10) Sousa, S. F.; Fernandes, P. A.; Ramos, M. J. Farnesyltransferase
inhibitors: A detailed chemical view on an elusive biological
problem. Curr. Med. Chem. 2008, 15, 1478–1492.
(11) Shao, J.; Zhou, B.; Chu, B.; Yen, Y. Ribonucleotide reductase
inhibitors and future drug design. Curr. Cancer Drug Targets 2006,
6, 409–431.
(12) Cristalli, G.; Costanzi, S.; Lambertucci, C.; Lupidi, G.; Vittori, S.;
Volpini, R.; Camaioni, E. Adenosine deaminase: Functional im-
plications and different classes of inhibitors. Med. Res. Rev. 2001,
21, 105–128.
Screening against iNOS. CFL-1.1 was screened against iNOS
at a concentration of 1 mM for each fragment. iNOS assays were
performed at 37 °C using a commercially available colorimetric
assay kit purchased from Calbiochem. Murine recombinant
iNOS protein (0.1 U, Calbiochem), substrate L-arginine
(80 μmol, Sigma-Aldrich), and inhibitor fragments (80 μmol)
were incubated in assay buffer (Calbiochem, Catalog No.
482702) at a total volume of 60 μL for 5 min. A nitrate reductase

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2 601
(13) Moss, M. L.; Sklair-Tavron, L.; Nudelman, R. Drug insight:
Tumor necrosis factor-converting enzyme as a pharmaceutical
target for rheumatoid arthritis. Nat. Clin. Pract. Rheumatol.
2008, 4, 300–309.
(14) Burnett, J. C.; Henchal, E. A.; Schmaljohn, A. L.; Bavari, S. The
evolving field of biodefence: Therapeutic developments and diag-
nostics. Nat. Rev. Drug Discovery 2005, 4, 281–297.
(15) Hernick, M.; Fierke, C. A. Zinc hydrolases: The mechanisms of
zinc-dependent deacetylases. Arch. Biochem. Biophys. 2005, 433,
71–84.
Jones, P.; Kinzel, O.; Laufer, R.; Monteagudo, E.; Muraglia, E.;
Nizi, E.; Orvieto, F.; Pace, P.; Pescatore, G.; Scarpelli, R.;
Stillmock, K.; Witmer, M. V.; Rowley, M. Discovery of Raltegra-
vir, a potent, selective orally bioavailable HIV-integrase inhibitor
for the treatment of HIV-AIDS infection. J. Med. Chem. 2008, 51,
5843–5855.
(38) Johnson, S. L.; Chen, L.-H.; Barile, E.; Emdadi, A.; Sabet, M.;
Yuan, H.; Wei, J.; Guiney, D.; Pellecchia, M. Structure-activity
relationship studies of a novel series of anthrax lethal factor
inhibitors. Bioorg. Med. Chem. 2009, 17, 3352–3368.
(39) Congreve, M.; Chessari, G.; Tisi, D.; Woodhead, A. J. Recent
developments in fragment-based drug discovery. J. Med. Chem.
2008, 51, 3661–3680.
(16) Villain-Guillot, P.; Bastide, L.; Gualtieri, M.; Leonetti, J.-P. Pro-
gress in targeting bacterial transcription. Drug Discovery Today
2007, 12, 200–208.
(40) Leach, A. R.; Hann, M. M.; Burrows, J. N.; Griffen, E. J. Fragment
screening: An introduction. Mol. Biosyst. 2006, 2, 429–446.
(41) Hopkins, A. L.; Groom, C. R.; Alex, A. Ligand efficiency: A
useful metric for lead selection. Drug Discovery Today 2004, 9,
430–431.
(42) Agrawal, A.; Johnson, S. L.; Jacobsen, J. A.; Miller, M. T.; Chen,
L.-H.; Pellecchia, M.; Cohen, S. M. Chelator fragment libraries for
targeting metalloproteinases. ChemMedChem 2010, 5, 195–199.
(43) Puerta, D. T.; Cohen, S. M. Elucidating drug-metalloprotein
interactions with tris(pyrazolyl)borate model complexes. Inorg.
Chem. 2002, 41, 5075–5082.
(44) Puerta, D. T.; Cohen, S. M. Examination of novel zinc-binding
groups for use in matrix metalloproteinase inhibitors. Inorg. Chem.
2003, 42, 3423–3430.
(45) Puerta, D. T.; Schames, J. R.; Henchman, R. H.; McCammon,
J. A.; Cohen, S. M. From model complexes to metalloprotein
inhibition: A synergistic approach to structure-based drug discov-
ery. Angew. Chem., Int. Ed. 2003, 42, 3772–3774.
(17) Dubey, S.; Satyanarayana, Y. D.; Lavania, H. Development of
integrase inhibitors for treatment of AIDS: An overview. Eur. J.
Med. Chem. 2007, 42, 1159–1168.
(18) White, R. J.; Margolis, P. S.; Trias, J.; Yuan, Z. Targeting
metalloenzymes: A strategy that works. Curr. Opin. Pharmacol.
2003, 3, 502–507.
(19) Vaughan, M. D.; Sampson, P. B.; Honek, J. F. Methionine in and
out of proteins: Targets for drug design. Curr. Med. Chem. 2002, 9,
385–409.
(20) Dive, V.; Chang, C.-F.; Yiotakis, A.; Sturrock, E. D. Inhibition of
zinc metallopeptidases in cardiovascular disease: From unity to
trinity, or duality? Curr. Pharm. Des. 2009, 15, 3606–3621.
(21) Dunkel, P.; Gelain, A.; Barlocco, D.; Haider, N.; Gyires, K.;
Sperlagh, B.; Magyar, K.; Maccioni, E.; Fadda, A.; Matyus, P.
Semicarbazide-sensitive amine oxidase/vascular adhesion protein
1: Recent developments concerning substrates and inhibitors of a
promising therapeutic target. Curr. Med. Chem. 2008, 15, 1827–
1839.
(22) Mai, A.; Rotili, D.; Valente, S.; Kazantsev, A. G. Histone deace-
tylase inhibitors and neurodegenerative disorders: Holding the
promise. Curr. Pharm. Des. 2009, 15, 3940–3957.
(23) Jacobsen, J. A.; Jourden, J. L. M.; Miller, M. T.; Cohen, S. M. To
bind zinc or not to bind zinc: An examination of innovative
approaches to improved metalloproteinase inhibition. Biochim.
Biophys. Acta 2010, 1803, 72–94.
(24) Puerta, D. T.; Cohen, S. M. A bioinorganic perspective on matrix
metalloproteinase inhibition. Curr. Top. Med. Chem. 2004, 4, 1551–
1573.
(25) Muri, E. M. F.; Nieto, M. J.; Sindelar, R. D.; Williamson, J. S.
Hydoxamic acids as pharmacological agents. Curr. Med. Chem.
2002, 9, 1631–1653.
(26) Puerta, D. T.; Lewis, J. A.; Cohen, S. M. New beginnings for matrix
metalloproteinase inhibitors: Identification of high-affinity zinc-
binding groups. J. Am. Chem. Soc. 2004, 126, 8388–8389.
(27) Puerta, D. T.; Mongan, J.; Tran, B. L.; McCammon, J. A.; Cohen,
S. M. Potent, selective pyrone-based inhibitors of stromelysin-1.
J. Am. Chem. Soc. 2005, 127, 14148–14149.
(28) Jacobsen, F. E.; Lewis, J. A.; Cohen, S. M. A new role for old
ligands: Discerning chelators for zinc metalloproteinases. J. Am.
Chem. Soc. 2006, 128, 3156–3157.
(46) Hohwy, M.; Spadola, L.; Lundquist, B.; Hawtin, P.; Dahmen, J.;
Groth-Clausen, I.; Nilsson, E.; Persdotter, S.; von Wachenfeldt,
K.; Folmer, R. H. A.; Edman, K. Novel prostaglandin d synthase
inhibitors generated by fragment-based drug design. J. Med. Chem.
2008, 51, 2178–2186.
(47) Hajduk, P. J.; Sheppard, G.; Nettesheim, D. G.; Olejniczak, E. T.;
Shuker, S. B.; Meadows, R. P.; Steinman, D. H.; Carrera, G. M.,
Jr.; Marcotte, P. A.; J, S.; Walter, K.; Smith, H.; Gubbins, E.;
Simmer, R.; Holzman, T. F.; Morgan, D. W.; Davidsen, S. K.;
Summers, J. B.; Fesik, S. W. Discovery of potent nonpeptide
inhibitors of stromelysin using SAR by NMR. J. Am. Chem. Soc.
1997, 119, 5818–5827.
(48) Charton, J.; Deprez, B. P.; Deprez-Poulain, R. F. Synthesis of a
200-member library of squaric acid N-hydroxylamide amides.
Bioorg. Med. Chem. Lett. 2008, 18, 4968–4971.
(49) Wang, J.; Uttamchandani, M.; Li, J.; Mingyu, H.; Yao, S. Q. Rapid
assembly of matrix metalloprotease inhibitors using click chemis-
try. Org. Lett. 2006, 8, 3821–3824.
(50) Hu, M.; Li, J.; Yao, S. Q. In situ “click” assembly of small molecule
matrix metalloprotease inhibitors containing zinc-chelating
groups. Org. Lett. 2008, 10, 5529–5531.
(51) Hofmann, B.; Franke, L.; Proschak, E.; Tanrikulu, Y.; Schneider,
P.; Steinhilber, D.; Schneider, G. Scaffold-hopping cascade yields
potent inhibitors of 5-lipoxygenase. ChemMedChem 2008, 3, 1535–
1538.
(29) Jacobsen, F. E.; Buczynski, M. W.; Dennis, E. A.; Cohen, S. M. A
macrophage cell model for selective metalloproteinase inhibitor
design. ChemBioChem 2008, 9, 2087–2095.
(30) Agrawal, A.; Romero-Perez, D.; Jacobsen, J. A.; Villarreal, F. J.;
Cohen, S. M. Zinc-binding groups modulate selective inhibition of
MMPs. ChemMedChem 2008, 3, 812–820.
(31) Whittaker, M.; Floyd, C. D.; Brown, P.; Gearing, A. J. H. Design
and therapeutic application of matrix metalloproteinase inhibitors.
Chem. Rev. 1999, 99, 2735–2776.
(32) Brown, S.; Meroueh, S. O.; Fridman, R.; Mobashery, S. Quest for
selectivity in inhibition of matrix metalloproteinases. Curr. Top.
Med. Chem. 2004, 4, 1227–1238.
(52) Nordstrom, H.; Gossas, T.; Hamalainen, M.; Kallblad, P.; Nystrom,
S.; Wallberg, H.; Danielson, U. H. Identification of MMP-12
inhibitors by using biosensor-based screening of a fragment library.
J. Med. Chem. 2008, 51, 3449–3459.
(53) Liu, Z. D.; Khodr, H. H.; Liu, D. Y.; Lu, S. L.; Hider, R. C.
Synthesis, physicochemical characterization, and biological eva-
luation of 2-(10-hydroxyalkyl)-3-hydroxypyridin-4-ones: Novel
iron chelators with enhanced pFe3þ values. J. Med. Chem. 1999,
42, 4814–4823.
(33) Suzuki, T.; Miyata, N. Rational design of non-hydroxamate
histone deacetylase inhibitors. Mini-Rev. Med. Chem. 2006, 6,
515–526.
(34) Winum, J.-Y.; Scozzafava, A.; Montero, J.-L.; Supuran, C. T.
Design of zinc binding functions for carbonic anhydrase inhibitors.
Curr. Pharm. Des. 2008, 14, 615–621.
(54) Lewis, J. A.; Puerta, D. T.; Cohen, S. M. Metal complexes of the
trans-influencing ligand thiomaltol. Inorg. Chem. 2003, 42, 7455–
7459.
(55) Lewis, J. A.; Cohen, S. M. Addressing lead toxicity: Complexation
of lead(II) with thiopyrone and hydroxypyridinethione O,S mixed
chelators. Inorg. Chem. 2004, 43, 6534–6536.
(35) Johnson, S. L.; Jung, D.; Forino, M.; Chen, Y.; Satterthwait, A.;
Rozanov, D. V.; Strongin, A. Y.; Pellecchia, M. Anthrax lethal
factor protease inhibitors: Synthesis, SAR, and structure-based 3D
QSAR studies. J. Med. Chem. 2006, 49, 27–30.
(36) Hajduk, P. J.; Shuker, S. B.; Nettesheim, D. G.; Craig, R.; Augeri,
D. J.; Betebenner, D.; Albert, D. H.; Guo, Y.; Meadows, R. P.; Xu,
L.; Michaelides, M.; Davidsen, S. K.; Fesik, S. W. NMR-based
modification of matrix metalloproteinase inhibitors with improved
bioavailability. J. Med. Chem. 2002, 45, 5628–5639.
(56) Yan, Y.-L.; Miller, M. T.; Cao, Y.; Cohen, S. M. Synthesis of
hydroxypyrone- and hydroxythiopyrone-based matrix metallopro-
teinase inhibitors: Developing a structure-activity relationship.
Bioorg. Med. Chem. Lett. 2009, 19, 1970–1976.
(57) Yan, Y.-L.; Cohen, S. M. Efficient synthesis of 5-amido-3-hydro-
xy-4-pyrones as inhibitors of matrix metalloproteinases. Org. Lett.
2007, 9, 2517–2520.
(58) Liu, Z. D.; Piyamongkol, S.; Liu, D. Y.; Khodr, H. H.; Lu, S. L.;
Hider, R. C. Synthesis of 2-amido-3-hydroxypyridin-4(1H)-ones:
Novel iron chelators with enhanced pFe3þ values. Bioorg. Med.
Chem. 2001, 9, 563–573.
(37) Summa, V.; Petrocchi, A.; Bonelli, F.; Crescenzi, B.; Donghi, M.;
Ferrara, M.; Fiore, F.; Gardelli, C.; Paz, O. G.; Hazuda, D. J.;

602 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 2
Jacobsen et al.
(59) Wiley, R. H.; Jarboe, C. H. 2-Pyrones. XIX. 3-Hydroxy-2-pyrone
and 4-arylhydrazono-2,3-pyranones. J. Am. Chem. Soc. 1956, 78,
2398–2401.
(60) Smith, A. P.; Corbin, P. S.; Fraser, C. L. An improved synthesis of
hydroxymethyl bipyridines. Tet. Lett. 2000, 41, 2787–2789.
(61) Rouffet, M.; de Oliveira, C. A. F.; Udi, Y.; Agrawal, A.; Sagi, I.;
McCammon, J. A.; Cohen, S. M. From sensors to silencers:
Quinoline- and benzimidazole-sulfonamides as inhibitors for zinc
proteases. J. Am. Chem. Soc. 2010, 132, 8232–8233.
on dicopper(II) centers: Insights into tyrosinase inhibition mechan-
ism by transition-state analogs. Inorg. Chem. 2009, 48, 10874–
10876.
(70) Jimenez, M.; Garcia-Carmona, F. 4-Substituted resorcinols (sulfite
alternatives) as slow-binding inhibitors of tyrosinase catecholase
activity. J. Agric. Food Chem. 1997, 45, 2061–2065.
(71) Hider, R. C.; Lerch, K. The inhibition of tyrosinase by pyridinones.
Biochem. J. 1989, 257, 289–290.
(72) Bursavich, M. G.; Lombardi, S.; Gilbert, A. M.; Laakso, L. M.;
Khafizova, G.; How, D. B.; Sabatini, J. J.; Sum, P.-E.; Clemens,
J. J. U.S. Patent WO/2008/024922, 2008.
(62) Liu, J.; Yi, W.; Wan, Y.; Ma, L.; Song, H. 1-(1-Arylethylide-
ne)thiosemicarbazide derivatives: A new class of tyrosinase inhibi-
tors. Bioorg. Med. Chem. 2008, 16, 1096–1102.
(73) Li, J. J.U.S. Patent WO/2004/014377, 2004.
(63) Pufahl, R. A.; Kasten, T. P.; Hills, R.; Gierse, J. K.; Reiz, B. A.;
Weinberg, R. A.; Masferrer, J. L. Development of a fluorescence-
based assay of human 5-lipoxygenase. Anal. Biochem. 2007, 364,
204–212.
(74) Musser, J. H.; Jones, H.; Sciortino, S.; Bailery, K.; Coutts, S. M.;
Khandwala, A.; Sonnino-Goldman, P.; Leibowitz, M.; Wolf, P.;
Neiss, E. S. Synthesis and antiallergic activities of 1,3-oxazolo[4,5-
h]quinolines. J. Med. Chem. 1985, 28, 1255–1259.
(75) Johnson, L. L.; Pavlovsky, A. G.; Johnson, A. R.; Janowicz, J. A.;
Man, C.-F.; Ortwine, D. F.; Purchase, C. F.; White, A. D.; Hupe,
D. J. A rationalization of the acidic pH dependence for stromely-
sin-1 (matrix metalloprotease-3) catalysis and inhibition. J. Biol.
Chem. 2000, 275, 11026–11033.
(76) Aureli, L.; Gioia, M.; Cerbara, I.; Monaco, S.; Fasciglione, G. F.;
Marini, S.; Ascenzi, P.; Topai, A.; Coletta, M. Structural bases for
substrate and inhibitor recognition by matrix metalloproteinases.
Curr. Med. Chem. 2008, 15, 2192–2222.
(64) Puerta, D. T.; Griffin, M. O.; Lewis, J. A.; Romero-Perez, D.; Garcia,
R.; Villarreal, F. J.; Cohen, S. M. Heterocyclic zinc-binding groups for
use in next-generation matrix metalloproteinase inhibitors: potency,
toxicity, and reactivity. J. Biol. Inorg. Chem. 2006, 11, 131–138.
(65) Lewis, J. A.; Mongan, J.; McCammon, J. A.; Cohen, S. M.
Evaluation and binding-mode prediction of thiopyrone-based in-
hibitors of anthrax lethal factor. ChemMedChem 2006, 1, 694–697.
(66) Chowdhury, M. A.; Abdellatif, K. R. A.; Dong, Y.; Das, D.;
Suresh, M. R.; Knaus, E. E. Synthesis of celecoxib analogues
possessing a N-difluoromethyl-1,2-dihydropyrid-2-one 5-lipoxy-
genase pharmacophore: Biological evaluation as dual inhibitors
of cyclooxygenases and 5-lipoxygenase with anti-inflammatory
activity. J. Med. Chem. 2009, 52, 1525–1529.
(77) Collier, R. J.; Young, J. A. T. Anthrax toxin. Annu. Rev. Cell Dev.
Biol. 2003, 19, 45–70.
(78) Massova, I.; Kotra, L. P.; Mobashery, S. Structural insight into the
binding motifs for the calcium ion and the non-catalytic zinc in
matrix metalloproteases. Bioorg. Med. Chem. Lett. 1998, 8, 853–
858.
(79) Turk, B. E.; Wong, T. Y.; Schwarzenbacher, R.; Jarrell, E. T.;
Leppla, S. H.; Collier, R. J.; Liddington, R. C.; Cantley, L. C. The
structural basis for substrate and inhibitor selectivity of the anthrax
lethal factor. Nat. Struct. Mol. Biol. 2004, 11, 60–66.
(80) Milbank, J. B. J.; Stevenson, R. J.; Ware, D. C.; Chang, J. Y. C.;
Tercel, M.; Ahn, G.-O.; Wilson, W. R.; Denny, W. A. Synthesis
and evaluation of stable bidentate transition metal complexes
of 1-(chloromethyl)-5-hydroxy-3-(5,6,7-trimethoxyindol-2-ylcar-
bonyl)-2,3-dihydro-1H-pyrrolo[3,2-f]quinoline(seco-6-azaCBI-TMI)
as hypoxia selective cytotoxins. J. Med. Chem. 2009, 52, 6822–6834.
(67) Chowdhury, M. A.; Chen, H.; Abdellatif, K. R. A.; Dong, Y.;
Petruk, K. C.; Knaus, E. E. Synthesis and biological evaluation of
1-(benzenesulfonamido)-2-[5-(N-hydroxypyridin-2(1H)-
one)]acetylene regioisomers: A novel class of 5-lipoxygenase in-
hibitors. Bioorg. Med. Chem. Lett. 2008, 18, 4195–4198.
(68) Chowdhury, M. A.; Chen, H.; Abdellatif, K. R. A.; Dong, Y.;
Petruk, K. C.; Knaus, E. E. Synthesis of new 1-(2-, 3-, or
4-methanesulfonylphenyl)-2-[5-(N-hydroxypyridin-2(1H)-
one)]acetylene regioisomers: A search for novel cyclooxygenase
and lipoxygenase inhibitors. J. Heterocycl. Chem. 2009, 46, 58–61.
(69) Peyroux, E.; Ghattas, W.; Hardre, R.; Giorgi, M.; Faure, B.; Simaan,
A. J.; Belle, C.; Reglier, M. Binding of 2-hydroxypyridine-N-oxide