Exploitation of Conformational Dynamics in Imparting Selective Inhibition for Related Matrix Metalloproteinases

Article abstract of DOI:10.1021/acsmedchemlett.7b00130

Matrix metalloproteinases (MMPs) have numerous physiological functions and share a highly similar catalytic domain. Differential dynamical information on the closely related human MMP-8, -13, and -14 was integrated onto the benzoxazinone molecular template. An in silico library of 28,099 benzoxazinones was generated and evaluated in the context of the molecular-dynamics information. This led to experimental evaluation of 19 synthesized compounds and identification of selective inhibitors, which have potential utility in delineating the physiological functions of MMPs. Moreover, the approach serves as an example of how dynamics of closely related active sites may be exploited to achieve selective inhibition by small molecules and should find applications in other enzyme families with similar active sites.

Full text of DOI:10.1021/acsmedchemlett.7b00130

Letter
pubs.acs.org/acsmedchemlett
Exploitation of Conformational Dynamics in Imparting Selective
Inhibition for Related Matrix Metalloproteinases
Kiran V. Mahasenan,† Maria Bastian,† Ming Gao,† Emma Frost,† Derong Ding,†
Katerina Zorina-Lichtenwalter,† John Jacobs,† Mark A. Suckow,‡ Valerie A. Schroeder,‡
William R. Wolter,‡ Mayland Chang,*,† and Shahriar Mobashery*,†
†Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556, United States
‡Freimann Life Science Center and Department of Biological Sciences, University of Notre Dame, Notre Dame, Indiana 46556,
United States
S
* Supporting Information
ABSTRACT: Matrix metalloproteinases (MMPs) have nu-
merous physiological functions and share a highly similar
catalytic domain. Differential dynamical information on the
closely related human MMP-8, -13, and -14 was integrated
onto the benzoxazinone molecular template. An in silico library
of 28,099 benzoxazinones was generated and evaluated in the
context of the molecular-dynamics information. This led to
experimental evaluation of 19 synthesized compounds and
identification of selective inhibitors, which have potential utility
in delineating the physiological functions of MMPs. Moreover,
the approach serves as an example of how dynamics of closely related active sites may be exploited to achieve selective inhibition
by small molecules and should find applications in other enzyme families with similar active sites.
KEYWORDS: Matrix metalloproteinases, Molecular dynamics, Virtual library design, Molecular docking, Enzyme kinetics,
Animal studies
enzymes), which culminated in failure in clinical trials.9,10
atrix metalloproteinases (MMPs) comprise a family of
M23 closely related human zinc-dependent endopepti-
The poor outcome of these trials is reflected in the reality that
in the diseased tissue one sees expressions of both the beneficial
and the detrimental MMPs. A compelling example of this
contrast is our recent finding that MMP-8 has a healing
(beneficial) and MMP-9 a pathological (detrimental) effect in
wound healing in diabetes.11,12 This is paradoxical. Expression
of MMP-8 is the body’s response to the need for wound repair,
yet that of MMP-9 is a consequence of the aforementioned
dysregulation. It is of interest that broad-spectrum MMP
inhibition in a clinical trial of nondiabetic wounds actually
showed delayed wound healing.13 However, highly selective
MMP-9 inhibition in a rodent model of diabetic wounds sped
up the healing process.11,12 As the details of the involvement of
MMPs in these beneficial and detrimental activities will be
elucidated for various MMP-mediated diseases in the future, the
need for highly selective inhibitors for the MMPs as both
mechanistic tools and potential therapeutics is self-evident. The
present study addresses this need for highly selective MMP
inhibitors.
dases with many physiological functions spanning tissue
development and growth, remodeling and repair, angiogenesis,
and wound healing, among others.1,2 The activities of these
enzymes are highly regulated, and when this regulation is
disrupted, MMPs manifest pathological outcomes such as
cancer development and metastasis, rheumatoid arthritis, and
cardiovascular and neurological diseases, to name a few.3,4
MMPs are evolutionarily related to each other, and their active
sites share significant structural similarities.5,6 Indeed, these
similarities explain the overlapping activities that the various
members of the family exhibit, a testament to the critical
functions that MMPs perform, mandating redundancy in
catalysis.
Notwithstanding the documented central roles that individ-
ual MMPs play in the onset and progression of various diseases,
inhibitors for MMPs have not been introduced to the market,
with one exception. Doxycycline, an antibacterial agent that
exhibits poor broad-spectrum MMP-inhibitory activity, is used
clinically for treatment of periodontitis.7,8 The experience of the
clinical trials of MMP inhibitors in treatment of cancer defined
the essential challenge. The key impediment in the progress of
these inhibitors to the clinic was the use of broad-spectrum
inhibitors, which complicated treatment due to indiscriminant
inhibition of many MMPs (and related zinc-dependent
MMPs remain as worthy targets for drug development, as
long as selectivity in targeting could be imparted onto the
Received: March 23, 2017
Accepted: May 1, 2017
Published: May 1, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
Figure 1. (A) Chemical structure of lead compound 1. (B) X-ray structure of inhibitor 1 bound to MMP-8, showing the displaced Tyr227. Residues
218−228 of the Ω-loop are shown as a tube. (C) Sequence alignment of the catalytic domain amino acids of the 23 members of the MMP family was
performed (see Figure S1), of which that for the segments of MMP-8, MMP-13, and MMP-14 are given here. The rectangular box highlights the Ω-
loop residues Pro217−Pro230 for MMP-8, Pro242−Pro255 for MMP-13, and Pro259−Pro272 for MMP-14, respectively. The sequence logo
(created with WebLogo v2.816) shows the residue conservation. The arrow points to the aromatic residues within the Ω-loop critical for
differentiation of the dynamics of the loop. (D) Superimposition of the X-ray crystal structures for the native Ω-loop conformations of MMP-8
(green; PDB ID: 2OY4), of MMP-13 (purple; PDB ID: 456C), and of MMP-14 (gray; PDB ID: 3MA2) shows nearly identical loop conformations
(root-mean-squared deviation (RMSD) for the Cα of loop residues of 1.6 Å for MMP-8 vs MMP-13, 0.8 Å for MMP-8 vs MMP-14, and 1.4 Å for
MMP-13 vs MMP-14). The arrow points to the middle of the Ω-loops. (E) The ligand-induced MMP-8 Tyr227-out conformation (in green; PDB
ID 3DNG) is superimposed to the Tyr-227-in the conformation found in the unlinganded MMP-8 (PDB ID: 2OY4). The arrow points to the Tyr-
227 site. (F) Time evolution of the backbone root-mean-square deviation (RMSD) of loop residues Pro242−Pro255 of MMP-13 (purple) and the
corresponding residues MMP-8 (green) and MMP-14 (gray) over 500 ns molecular-dynamics simulations. (See Figure S5 of the Supporting
Information for the corresponding plots for all the residues of the catalytic domain.) (G) Side-chain dihedral angle that defines rotamers based on
the Cα−Cβ bond torsional axis (calculated with C−Cα−Cβ-Cγ atoms) of Tyr227 over the simulation time, along with that of corresponding
residues of MMP-13 (Phe252) and of MMP-14 (Phe269). (H) Superimposition of the MMP-13 native conformation (PDB ID: 456C; purple) with
that of the MMP-8-inhibitor-induced conformation (PDB ID: 3DNG; green) showing the requirement for Phe252 in MMP-13 to be displaced to
accommodate the inhibitor. The structure of compound 1 is superimposed to the SiteMap binding-site-calculation result of the MMP-8 crystal
structure (PDB ID: 3DNG). Blue-dot points indicate the location where ligand occupancy is favored, while the gray mesh indicates hydrophobic
preference.
molecules. However, the similarities of these enzymes within
their active sites have presented a formidable challenge in
design of such inhibitors. Indeed, there are extraordinarily few
examples of highly selective inhibitors for MMPs.14,15 One is
the thiirane class of inhibitors from our laboratory, which shows
high selectivity for MMP-2 and/or MMP-9.17,18 Members of
this class of inhibitors undergo a ring-opening reaction within
the active sites of MMP-2 and MMP-9, which leads to a tight-
binding inhibitor species generated within the active site.19,20
These inhibitors have been used widely in many animal models
for MMP-2- and MMP-9-dependent diseases.11,21−25 Similarly,
MMP-13 selective inhibitors have potential use in the study of
cardiovascular diseases and osteoarthritis.26,27 We describe here
the design of specific benzoxazinones, which target MMP-13
for selective inhibition.
The literature on inhibitors of metalloproteinases, inclusive
of MMPs, is extensive.14,28 However, the vast majority of the
inhibitor classes are zinc-ion chelators, which often suffer cross-
reactivity with other enzymes. We were drawn to the previously
reported benzoxazinone class of inhibitors for the reasons given
below.29 An X-ray structure of inhibitor 1 (Figure 1A) bound to
the active site of MMP-8 (Figure 1B) showed that the
compound was ensconced deep within the S1′ pocket and did
not directly interact with the catalytic zinc ion. The S1′ pocket
is present in all MMPs. It is defined by a surface, which is
capped by an Ω-loop (Figure 1C). The length of the loop is
variable among the MMPs (Figure S1), and from the known X-
ray crystal structures, it appears to be a mobile element.30,31
Yet, MMP-8, MMP-13, and MMP-14 have essentially S1′ loops
with the identical length of 12 amino acids flanked by proline
residues at each end (Figure 1C). Compound 1 inhibits MMP-
B
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
8 and MMP-13, among the eight MMPs that were
evaluated.29,32 This observation is easily explained by the
similarity of the S1′ pockets in these MMPs (Table S1 and
Figure S2), but that the inhibitor does not inhibit MMP-14
stood out as a noteworthy curiosity, to which we will return. In
the present report, we exploit the properties of the S1′ loops of
the closely related MMP-8, MMP-13, and MMP-14 in arriving
at molecules of the benzoxazinone class with selectivity for
either MMP-8 or MMP-13. This work also sheds light on how
the active site of MMP-14 is distinct, notwithstanding the
seemingly high structural similarities.
underwent significant dynamics in the cases of MMP-8 and
MMP-13 (Figure 1H), were promising regions for inhibitor
design. The matching sites on the chemical structure of 1 were
targeted for structure elaboration by molecular modeling. In
preparation for this effort, a synthetically accessible virtual
library of 28,099 compounds was constructed by the use of the
in silico combinatorial library enumeration method imple-
mented in the CombiGlide program (v. 3.2, Schrodinger, LLC,
̈
NY, USA). The analysis was based on the structural template of
benzoxazinone (1), where the structural variations were
introduced at the highlighted points (Figure 1A). Meanwhile,
we computationally clustered snapshots from the 500 ns of the
dynamics trajectory of MMP-13, which resulted in identi-
fication of a total of six conformations in which R1 and R2
pockets were created (See Supporting Information for method
details). We docked compound 1 into these structures by the
These observations argue for potential differential structural
flexibility among the three active sites. In the native states of
MMP-8, MMP-13, and MMP-14, with the S1′ sites unoccupied,
the Ω-loop adopts a comparable conformation in all three
(Figure 1D). Binding of benzoxazinone 1 to the active site of
MMP-8 causes a unique conformational change in the Ω-loop
according to the X-ray structure (the Arg222−Ser228 stretch;
Figure 1E). This observation is evidence for molecular
flexibility within the active site in the case of MMP-8, but the
question becomes if MMP-13 and MMP-14 also exhibit this
property, or alternatively, do they have differential flexibility
that could be exploited in the design of selective inhibitors. In
order to assess this, we carried out molecular-dynamics
simulations of MMP-8, MMP-13, and MMP-14 under identical
conditions. The starting points for the simulations were the
crystal structures of MMP-8, MMP-13, and MMP-14 with the
Ω-loop conformations in the apoenzyme positions (Figure 1D;
PDB structures 2OY4, 456C, and 3MA2, respectively). The
structures were solvated with explicit water and were subjected
to molecular-dynamics simulations for 500 ns using the
AMBER14 suite.33,34 As observed by the RMSD of the Ω-
loop residues over the simulation time (Figure 1F), MMP-14
showed the most stable Ω-loop. The molecular-dynamics
simulation trajectory of MMP-14 revealed unique interactions
of the Ω-loop residues Trp263 and Glu267 (Figure S2). The
residue Trp263 of MMP-14 formed a consistent hydrogen-
bonding interaction with Asn225 during the time course, while
Glu267 maintained a salt-bridge with Lys146 (Figure S3). The
corresponding amino acids in MMP-8 and MMP-13 are unable
to make similar interactions. The MMP-13 Ω-loop showed
considerable flexibility in comparison to that of MMP-8 and
MMP-14. Interestingly, Phe252 of MMP-13, which corre-
sponds to Tyr227 in MMP-8, made a dramatic motion during
the simulation (see Supporting Information movie), as
demonstrated by the plot of the dihedral angle corresponding
to the atoms C−Cα−Cβ−Cγ that define the side-chain
rotamers (Figure 1G and Figure S4). These simulations
indicated that all three enzymes are different within the S1′
cavity, notwithstanding the seeming similarities from the X-ray
structures. The structure for MMP-14 is the most stable
(static), but those of MMP-8 and MMP-13 are more flexible
(dynamic). The flexibility for MMP-13 is more exaggerated
compared to that for MMP-8. Such differences present
challenges in inhibitor discovery, as the static crystal structure
cannot guide the design. Yet, once identified, the nature of the
dynamics could be exploited in opportunities for selective
inhibitor design. As we describe below, this task has been
accomplished in the design of selective MMP-13 inhibitors.
use of the Glide program36 (Schrodinger, LLC, NY, USA), and
̈
the complexes were energy-minimized. We could reproduce the
crystallographic fit seen for MMP-8 in two of these conforma-
tional states within 1 Å of the root-mean-squared deviation.
These two conformations were then used for screening of the
28,099-strong virtual library using Glide. The compounds were
ranked for the goodness of fit, but were eliminated, if they did
not conform to Lipinski’s rule of five37 and Jorgensen’s rule of
three.38−40 With synthetic access (and availability of the starting
materials) being a final filter, we selected for synthesis 19 of the
high-ranking compounds (See Supporting Information for
method details).
Five of the target compounds are given in Table 1; an
additional 14 compounds are shown in the Supporting
Information (Figure S6). We highlight the five-step synthesis
Table 1. Kinetic Parameters for Inhibition of MMPs
The use of the SiteMap program35 (Schrodinger, LLC, NY,
̈
USA) on the crystal structure of the MMP-8 complex with
compound 1 (PDB ID: 3DNG) indicated that the protein loci
in the vicinity of the labels R1 and R2 (Figure 1A), which
a
b
*
Inhibition at 10 μM. Inhibition at 50 μM. Racemic mixture (R, S).
C
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
of compound 8 below (Scheme 1). The synthesis is a variation
of a reported method for compound 1.32 1-(4-
and thiophene, respectively). These compounds showed
weaker inhibition for MMP-8, likely due to their less flexible
and thus unaccommodating S1′ site. These findings collectively
suggest the requirement of a bulkier rigid hydrophobic group
that can replace the Tyr/Phe for the S1′ site and accommodate
the pocket for inhibition. We note that not all the synthesized
compounds showed inhibitory properties for the panel of the
enzymes that was screened, which is the limitation of the
method, but the properties of compound 8 fit the traits that we
set out to identify: selectivity for MMP-13 over MMP-8 and
lack of inhibition of MMP-14. Although our repeated efforts in
crystallography of compound 8 with MMP-13 did not succeed,
the potent in vitro inhibition data support direct binding of the
compound to the target.
a
Scheme 1. Synthesis of Inhibitor 8
The pharmacokinetic properties of compound 8 were
assessed in mice (n = 3 mice per time point per route of
administration, total 54 mice) after intravenous (iv) and oral
(po) dose administration (Figure 2, and Supporting Informa-
a
Reagents and conditions: (a) NCCH2COOEt, toluene, 140 °C. (b)
S, HNEt2, MeOH, 50 °C. (c) NCCO2Et, HCl, dioxane, 50 °C. (d)
LiOH, THF, rt. (e) HATU, NMM, DMF, rt.
(Trifluoromethyl)phenyl)ethan-1-one (2) and ethyl cyanoace-
tate were heated in toluene to afford the α,β-unsaturated ethyl
ester 3 as a mixture of E- and Z-isomers. This isomeric mixture
was allowed to react with elemental sulfur to give 4. The
reaction of ethyl cyanoformate in the presence of acid afforded
compound 5. Subsequent hydrolysis of the ethyl ester of 5
resulted in compound 6. Coupling of 6 and amine 7, which
itself was synthesized in three steps according to the
literature,32 afforded the desired product 8.
The synthetic compounds were evaluated for their in vitro
enzyme inhibition on a panel of eight representative MMPs. A
rapid up-front screening was conducted by incubating the
compounds at 10 μM with the corresponding enzymes and the
requisite substrates. Those compounds that showed >50%
inhibition were further evaluated, and their dissociation
constants (Ki) were determined (Table 1 and Table S2).
These compounds showed marginal to no inhibition of MMP-
1, MMP-2, MMP-3, MMP-7, MMP-9, and MMP-14.
Compound 8 emerged as the most selective for MMP-13
among the synthesized compounds. This compound is a
competitive inhibitor with 240-fold selectivity for MMP-13 over
MMP-8 (Ki of 36 11 nM vs 8500 1400 nM). Similarly,
compounds 9 and 10 showed 35- and 40-fold selectivity,
respectively, for MMP-13 over MMP-8 (compound 9, 190
14 nM vs 6500 900 nM; compound 10, 0.47 0.06 nM vs
19 5 nM). The activities of these two compounds indicated
the importance of hydrophobic bulky substitution in this region
for MMP-13 selectivity. Selectivity in inhibition by compound
11 was switched in favor of MMP-8, as opposed to MMP-13,
but only modestly so. On the other hand, compound 12
inhibited MMP-8 and MMP-13 equally well and equally
potently (subnanomolar inhibition).
The thienopyrimidinone cyclic system, connected to the
benzoxazinone scaffold via the amide linker was substituted
with other diverse groups in the rest of the compounds
reported here (13−26, Supporting Information Figure S6 and
Table S2). Pyrazole substitution (compound 13, 14, 15)
resulted in loss in binding affinity from nanomolar to
micromolar level, but demonstrated weak selectivity for
MMP-13. Compound 13, with bulkier naphthalene ring
occupying the S1′ site, showed 4-fold preference for MMP-13
over compounds 14 and 15, with smaller substitutions (phenyl
Figure 2. Plasma concentration−time curves after a single 1 mg/kg
intravenous (iv, red) dose and 10 mg/kg oral (po, blue) dose
administration of compound 8 in mice (n = 3 per time point per route
of administration, total 54 mice).
tion). After a single 1 mg/kg iv dose of 8, the plasma
concentration was 4.4 1.6 μM at 2 min and remained above
the Ki for MMP-13 at 24 h (0.062 0.078 μM). Compound 8
had AUC0‑∞ of 889 μM·min, low clearance (CL) of 2.2 mL/
min/kg (3% of hepatic blood flow), high volume of distribution
(Vd) of 1.0 L/kg, and an elimination half-life of 5.3 h. After a 10
mg/kg po dose, the maximum plasma concentration was 0.40
0.13 μM at 6 h and declined to 0.042 0.032 μM at 48 h;
however, it was above Ki for MMP-13. The elimination half-life
was 17.5 h. AUC0‑∞ was 415 μM·min. Comparison of the AUC
after po to that after iv gave an absolute oral bioavailability of
4.7%, after dose adjustment.
In conclusion, we report here an example of how
crystallographically similar binding sites may be exploited for
selective inhibitor design based on their dynamic nature. Our
computational simulations offered insights on the flexibility of
the structurally similar binding sites of MMP-8, MMP-13, and
MMP-14. Encouraged by the opportunity offered by this
simulation, a virtual focused library was designed and evaluated
in silico. Synthesis and in vitro evaluation of a subset of the
library successfully identified compounds with desired activity.
The kinetic data clearly document that selectivity based on the
design paradigm is achievable. The compound disclosed herein
has favorable in vivo pharmacokinetic properties in mice and
can potentially serve as a useful tool for delineating the
functions of the MMP family of enzymes.
D
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
metalloproteinases: contribution to pathogenesis, diagnosis and
treatment of periodontal inflammation. Ann. Med. 2006, 38, 306−321.
(9) Cathcart, J. M.; Cao, J. MMP Inhibitors: Past, present and future.
Front. Biosci., Landmark Ed. 2015, 20, 1164−1178.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsmedchem-
lett.7b00130.
■
S
(10) Dufour, A.; Overall, C. M. Missing the target: matrix
metalloproteinase antitargets in inflammation and cancer. Trends
Pharmacol. Sci. 2013, 34, 233−242.
The Ω-loop sequence alignment of 23 members of the
human MMP family, binding-site comparison of MMP-8,
MMP-13, and MMP-14, structures and activities of
compounds, methods for MD simulations, in silico library
design, virtual screening, syntheses and characterization
of compounds, enzyme inhibition studies, and pharma-
cokinetic studies (PDF)
(11) Gao, M.; Nguyen, T. T.; Suckow, M. A.; Wolter, W. R.; Gooyit,
M.; Mobashery, S.; Chang, M. Acceleration of diabetic wound healing
using a novel protease-anti-protease combination therapy. Proc. Natl.
Acad. Sci. U. S. A. 2015, 112, 15226−15231.
(12) Gooyit, M.; Peng, Z.; Wolter, W. R.; Pi, H.; Ding, D.; Hesek, D.;
Lee, M.; Boggess, B.; Champion, M. M.; Suckow, M. A.; Mobashery,
S.; Chang, M. A chemical biological strategy to facilitate diabetic
wound healing. ACS Chem. Biol. 2014, 9, 105−110.
Movie of MD simulation (AVI)
(13) Agren, M. S.; Mirastschijski, U.; Karlsmark, T.; Saarialho-Kere,
U. K. Topical synthetic inhibitor of matrix metalloproteinases delays
epidermal regeneration of human wounds. Exp. Dermatol. 2001, 10,
337−348.
(14) Fisher, J. F.; Mobashery, S. Recent advances in MMP inhibitor
design. Cancer Metastasis Rev. 2006, 25, 115−136.
AUTHOR INFORMATION
Corresponding Authors
*M.C.: E-mail, mchang@nd.edu; Phone, +1-574-631-2965.
*S.M.: E-mail, mobashery@nd.edu; Phone, +1-574-631-2933.
■
(15) Jacobsen, J. A.; Major Jourden, J. L.; 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, Mol. Cell Res. 2010, 1803, 72−94.
ORCID
Shahriar Mobashery: 0000-0002-7695-7883
Funding
This work was supported in part by the Craig H. Neilsen
Foundation (grant 282987 to M.C.).
(16) Crooks, G. E.; Hon, G.; Chandonia, J. M.; Brenner, S. E.
WebLogo: a sequence logo generator. Genome Res. 2004, 14, 1188−
1190.
Notes
(17) Gooyit, M.; Song, W.; Mahasenan, K. V.; Lichtenwalter, K.;
Suckow, M. A.; Schroeder, V. A.; Wolter, W. R.; Mobashery, S.;
Chang, M. O-phenyl carbamate and phenyl urea thiiranes as selective
matrix metalloproteinase-2 inhibitors that cross the blood-brain
barrier. J. Med. Chem. 2013, 56, 8139−8150.
(18) Gooyit, M.; Peng, Z.; Mobashery, S.; Chang, M. Thiirane Class
of Gelatinase Inhibitors as a Privileged Template that Crosses the Blood-
Brain Barrier; Royal Society of Chemistry Publishing, 2016; pp 262−
286.
(19) Forbes, C.; Shi, Q.; Fisher, J. F.; Lee, M.; Hesek, D.; Llarrull, L.
I.; Toth, M.; Gossing, M.; Fridman, R.; Mobashery, S. Active site ring-
opening of a thiirane moiety and picomolar inhibition of gelatinases.
Chem. Biol. Drug Des. 2009, 74, 527−534.
(20) Tao, P.; Fisher, J. F.; Shi, Q.; Vreven, T.; Mobashery, S.;
Schlegel, H. B. Matrix metalloproteinase 2 inhibition: combined
quantum mechanics and molecular mechanics studies of the inhibition
mechanism of (4-phenoxyphenylsulfonyl)methylthiirane and its
oxirane analogue. Biochemistry 2009, 48, 9839−9847.
(21) Cui, J.; Chen, S.; Zhang, C.; Meng, F.; Wu, W.; Hu, R.; Hadass,
O.; Lehmidi, T.; Blair, G. J.; Lee, M.; Chang, M.; Mobashery, S.; Sun,
G. Y.; Gu, Z. Inhibition of MMP-9 by a selective gelatinase inhibitor
protects neurovasculature from embolic focal cerebral ischemia. Mol.
Neurodegener. 2012, 7, 21.
(22) Gu, Z.; Cui, J.; Brown, S.; Fridman, R.; Mobashery, S.; Strongin,
A. Y.; Lipton, S. A. A highly specific inhibitor of matrix metal-
loproteinase-9 rescues laminin from proteolysis and neurons from
apoptosis in transient focal cerebral ischemia. J. Neurosci. 2005, 25,
6401−6408.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge computing resources and assistance provided
by Center for Research Computing of the University of Notre
Dame.
ABBREVIATIONS
■
HATU, 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo-
[4,5-b]pyridinium 3-oxide hexafluorophosphate; iv, intrave-
nous; MD, molecular dynamics; MMP, matrix metalloprotei-
nase; NMM, 4-methylmorpholine; po, per os; RMSD, root-
mean-squared deviation; UPLC, ultraperformance liquid
chromatography; UV, ultraviolet
REFERENCES
■
(1) Klein, T.; Bischoff, R. Physiology and pathophysiology of matrix
metalloproteases. Amino Acids 2011, 41, 271−290.
(2) Pulkoski-Gross, A. E. Historical perspective of matrix metal-
loproteases. Front. Biosci., Scholar Ed. 2015, 7, 125−149.
(3) Tauro, M.; McGuire, J.; Lynch, C. C. New approaches to
selectively target cancer-associated matrix metalloproteinase activity.
Cancer Metastasis Rev. 2014, 33, 1043−1057.
(4) Vandenbroucke, R. E.; Libert, C. Is there new hope for
therapeutic matrix metalloproteinase inhibition? Nat. Rev. Drug
Discovery 2014, 13, 904−927.
(23) Hadass, O.; Tomlinson, B. N.; Gooyit, M.; Chen, S.; Purdy, J. J.;
Walker, J. M.; Zhang, C.; Giritharan, A. B.; Purnell, W.; Robinson, C.
R., 2nd; Shin, D.; Schroeder, V. A.; Suckow, M. A.; Simonyi, A.; Sun,
G. Y.; Mobashery, S.; Cui, J.; Chang, M.; Gu, Z. Selective inhibition of
matrix metalloproteinase-9 attenuates secondary damage resulting
from severe traumatic brain injury. PLoS One 2013, 8, e76904.
(24) Lee, M.; Chen, Z.; Tomlinson, B. N.; Gooyit, M.; Hesek, D.;
Juarez, M. R.; Nizam, R.; Boggess, B.; Lastochkin, E.; Schroeder, V. A.;
Wolter, W. R.; Suckow, M. A.; Cui, J.; Mobashery, S.; Gu, Z.; Chang,
M. Water-Soluble MMP-9 Inhibitor Reduces Lesion Volume after
Severe Traumatic Brain Injury. ACS Chem. Neurosci. 2015, 6, 1658−
1664.
(5) Fanjul-Fernandez, M.; Folgueras, A. R.; Cabrera, S.; Lopez-Otin,
C. Matrix metalloproteinases: evolution, gene regulation and func-
tional analysis in mouse models. Biochim. Biophys. Acta, Mol. Cell Res.
2010, 1803, 3−19.
(6) Massova, I.; Kotra, L. P.; Fridman, R.; Mobashery, S. Matrix
metalloproteinases: structures, evolution, and diversification. FASEB J.
1998, 12, 1075−1095.
(7) Peterson, J. T. Matrix metalloproteinase inhibitor development
and the remodeling of drug discovery. Heart Failure Rev. 2004, 9, 63−
79.
(8) Sorsa, T.; Tjaderhane, L.; Konttinen, Y. T.; Lauhio, A.; Salo, T.;
Lee, H. M.; Golub, L. M.; Brown, D. L.; Mantyla, P. Matrix
E
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(25) Marusak, C.; Bayles, I.; Ma, J.; Gooyit, M.; Gao, M.; Chang, M.;
Bedogni, B. The thiirane-based selective MT1-MMP/MMP2 inhibitor
ND-322 reduces melanoma tumor growth and delays metastatic
dissemination. Pharmacol. Res. 2016, 113, 515−520.
(26) Austin, K. M.; Covic, L.; Kuliopulos, A. Matrix metalloproteases
and PAR1 activation. Blood 2013, 121, 431−439.
(27) Li, N. G.; Shi, Z. H.; Tang, Y. P.; Wang, Z. J.; Song, S. L.; Qian,
L. H.; Qian, D. W.; Duan, J. A. New hope for the treatment of
osteoarthritis through selective inhibition of MMP-13. Curr. Med.
Chem. 2011, 18, 977−1001.
(28) Georgiadis, D.; Yiotakis, A. Specific targeting of metzincin family
members with small-molecule inhibitors: progress toward a multi-
farious challenge. Bioorg. Med. Chem. 2008, 16, 8781−8794.
(29) Pochetti, G.; Montanari, R.; Gege, C.; Chevrier, C.; Taveras, A.
G.; Mazza, F. Extra binding region induced by non-zinc chelating
inhibitors into the S1′ subsite of matrix metalloproteinase 8 (MMP-8).
J. Med. Chem. 2009, 52, 1040−1049.
(30) Fabre, B.; Ramos, A.; de Pascual-Teresa, B. Targeting matrix
metalloproteinases: exploring the dynamics of the s1′ pocket in the
design of selective, small molecule inhibitors. J. Med. Chem. 2014, 57,
10205−10219.
(31) Devel, L.; Czarny, B.; Beau, F.; Georgiadis, D.; Stura, E.; Dive,
V. Third generation of matrix metalloprotease inhibitors: Gain in
selectivity by targeting the depth of the S1′ cavity. Biochimie 2010, 92,
1501−1508.
(32) Ding, D. R.; Lichtenwalter, K.; Pi, H. L.; Mobashery, S.; Chang,
M. Characterization of a selective inhibitor for matrix metal-
loproteinase-8 (MMP-8). MedChemComm 2014, 5, 1381−1383.
(33) Gotz, A. W.; Williamson, M. J.; Xu, D.; Poole, D.; Le Grand, S.;
Walker, R. C. Routine Microsecond Molecular Dynamics Simulations
with AMBER on GPUs. 1. Generalized Born. J. Chem. Theory Comput.
2012, 8, 1542−1555.
(34) Case, D. A.; Babin, V.; Berryman, J. T.; Betz, R. M.; Cai, Q.;
Cerutti, D. S.; Cheatham, T. E.; Darden, T. A.; Duke, R. E.; Gohlke,
H.; Goetz, A. W.; Gusarov, S.; Homeyer, N.; Janowski, P.; Kaus, J.;
́
Kolossvary, I.; Kovalenko, A.; Lee, T. S.; LeGrand, S.; Luchko, T.; Luo,
R.; Madej, B.; Merz, K. M.; Paesani, F.; Roe, D. R.; Roitberg, A.; Sagui,
C.; Salomon-Ferrer, R.; Seabra, G.; Simmerling, C. L.; Smith, W.;
Swails, J.; Walker; Wang, J.; Wolf, R. M.; Wu, X.; Kollman, P. A.
AMBER 2014; University of California, San Francisco, 2014.
(35) Halgren, T. A. Identifying and characterizing binding sites and
assessing druggability. J. Chem. Inf. Model. 2009, 49, 377−389.
(36) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new
approach for rapid, accurate docking and scoring. 1. Method and
assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739−1749.
(37) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J.
Experimental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv. Drug
Delivery Rev. 2001, 46, 3−26.
(38) Jorgensen, W. L.; Duffy, E. M. Prediction of drug solubility from
structure. Adv. Drug Delivery Rev. 2002, 54, 355−366.
(39) Roe, D. R.; Cheatham, T. E., 3rd PTRAJ and CPPTRAJ:
Software for Processing and Analysis of Molecular Dynamics
Trajectory Data. J. Chem. Theory Comput. 2013, 9, 3084−3095.
(40) Page-McCaw, A.; Ewald, A. J.; Werb, Z. Matrix metal-
loproteinases and the regulation of tissue remodelling. Nat. Rev. Mol.
Cell Biol. 2007, 8, 221−233.
F
DOI: 10.1021/acsmedchemlett.7b00130
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX