Investigating the selectivity of metalloenzyme inhibitors

Article abstract of DOI:10.1021/jm401053m

The inhibitory activity of a broad group of known metalloenzyme inhibitors against a panel of metalloenzymes was evaluated. Clinically approved inhibitors were selected as well as several other reported metalloprotein inhibitors in order to represent a broad range of metal binding groups (MBGs), including hydroxamic acid, carboxylate, hydroxypyridinonate, thiol, and N-hydroxyurea functional groups. A panel of metalloenzymes, including carbonic anhydrase (hCAII), several matrix metalloproteinases (MMPs), angiotensin converting enzyme (ACE), histone deacetylase (HDAC-2), and tyrosinase (TY), was selected based on their clinical importance for a range of pathologies. In addition, each inhibitor was evaluated for its ability to remove Fe³⁺ from holo-transferrin to gauge the ability of the inhibitors to access Fe ³⁺ from a primary transport protein. The results show that the metalloenzyme inhibitors are quite selective for their intended targets, suggesting that despite their ability to bind metal ions, metalloprotein inhibitors are not prone to widespread off-target enzyme inhibition activity.

Full text of DOI:10.1021/jm401053m

Subscriber access provided by RMIT University Library
Article
Investigating the Selectivity of Metalloenzyme Inhibitors
Joshua A. Day, and Seth M. Cohen
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 27 Sep 2013
Downloaded from http://pubs.acs.org on September 28, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Investigating the Selectivity of Metalloenzyme
Inhibitors
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Joshua A. Day and Seth M. Cohen*
†Department of Chemistry and Biochemistry, University of California, San Diego, La Jolla,
California 92093, United States
* To whom correspondence should be addressed. Eꢀmail: scohen@ucsd.edu. Telephone: (858)
822ꢀ5596.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 42
1
2
3
4
Abstract
5
6
7
8
The inhibitory activity of a broad group of known metalloenzyme inhibitors against a panel of
metalloenzymes was evaluated. Clinically approved inhibitors were selected as well as several
other reported metalloprotein inhibitors, in order to represent a broad range of metal binding
groups (MBGs), including hydroxamic acid, carboxylate, hydroxypyridinonate, thiol, and Nꢀ
hydroxyurea functional groups. A panel of metalloenzymes, including carbonic anhydrase
(hCAII), several matrix metalloproteinases (MMPs), angiotensin converting enzyme (ACE),
histone deacetylase (HDACꢀ2), and tyrosinase (TY) was selected based on their clinical
importance for a range of pathologies. In addition, each inhibitor was evaluated for its ability to
remove Fe3+ from holoꢀtransferrin to gauge the ability of the inhibitors to access Fe3+ from a
primary transport protein. The results show that the metalloenzyme inhibitors are quite selective
for their intended targets, suggesting that despite their ability to bind metal ions, metalloprotein
inhibitors are not prone to widespread offꢀtarget enzyme inhibition activity.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 42
Journal of Medicinal Chemistry
1
2
3
4
Introduction
5
6
7
8
Approximately oneꢀthird of proteins are metalloproteins, which serve to execute a wide array of
functions in vivo, including regulating blood pH, facilitating matrix degradation, modulating
DNA transcription, and many others.1 Given the importance of these functions, metalloenzyme
misregulation plays a significant role in human disease. Pathologies for which metalloenzymes
are implicated include cancer,2, 3 heart disease,4 and HIV/AIDS.5 Given the impact of these
diseases on human health, metalloenzyme inhibition offers an appealing approach to disease
treatment. Indeed, sales of metalloprotein inhibitors account for billions of dollars in
pharmaceutical sales annually.6
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Typically, metalloenzyme inhibitors are drugꢀlike small molecules that incorporate a
metal binding group (MBG) in order to coordinate the active site metal ion. The MBG is
appended to the drugꢀlike portion of the inhibitor (often referred to as a ‘backbone’ group) via a
linker. The backbone can take on many forms, reflective of the diversity in enzyme active sites.
The metalloenzyme inhibitors in Figure 1 illustrate the variety of backbones, customized for their
respective targets, which consist of assorted substituted aliphatic chains, aromatic rings, and
heterocycles. In contrast to the diversity found in the inhibitor backbones, most metalloenzyme
inhibitors incorporate a much narrower selection of MBGs. A search of the Protein Data Bank
(PDB) for metalloprotein inhibitors show that the hydroxamic acid is the most common MBG,
followed by carboxylic acids, thiols, and phosphonates. The use of only a few MBGs limits the
chemical space being explored for the development of new therapeutics, particularly in light of
the pharmacokinetic liabilities of inhibitors that contain hydroxamic acid3, 7ꢀ10 and thiol MBGs.11,
12
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 42
1
2
3
4
5
6
7
8
9
There appears to be a common opinion that inhibitors containing MBGs are more
promiscuous then other small molecule compounds leading to offꢀtarget effects. These effects
include indiscriminate metalloenzyme inhibition or metal ion disregulation by metal ion removal
from nonꢀtarget metalloproteins, thus a perception of greater risk in obtaining clinically
successful therapeutics.13ꢀ16 Although molecules that contain MBGs have the capacity to bind
metal ions and metalloenzymes, to the best of our knowledge, a rigorous examination of the
perceived liability of such compounds has not been reported in the scientific literature. Metal ion
removal is an unlikely prospect based on existing data generated from the bioinorganic research
community.17, 18 For example, transferrin, the major Fe3+ transport protein, which can exchange
Fe3+ >100 times during its biologically useful lifetime,19 does not readily liberate its Fe3+ cargo,
as it possesses an extremely high association constant (Ka) for Fe3+ (1022 Mꢀ1).20 Even at
millimolar concentrations, ironꢀscavenging siderophores, such as enterobactin, require nearly an
hour (in vitro) to remove iron from holoꢀtransferrin.21
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to test whether metalloenzyme inhibitors are more prone to inhibit offꢀtarget
enzymes and/or remove metal ions from accessible sources, this report has tested the capacity of
a select group of chelating inhibitors to inhibit a panel of metalloenzymes. In addition, the same
inhibitors were examined for their activity against trypsin (a metalꢀindependent protein) and for
their ability to remove Fe3+ from the transport protein transferrin, which represents an important,
exchangable pool of transition metal ions in the bloodstream. By examining cross inhibition
within a group of metalloenzymes, this study seeks to determine whether inhibitors of
metalloenzymes are less specific than other small molecule enzyme inhibitors. The results
presented here suggest that metalloenzyme inhibitors are not more promiscuous than other small
4
ACS Paragon Plus Environment

Page 5 of 42
Journal of Medicinal Chemistry
1
2
molecule therapeutics. Furthermore, these inhibitors are incapable of removing Fe3+ from
transferrin, even at concentrations far above any clinically relevant dose.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Metalloenzyme inhibitors examined in this study (above line) and other compounds
used as positive controls (below line). Metalꢀbinding groups (MBGs) are highlighted in red.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 42
1
2
3
4
5
6
7
8
Inhibitor Screening – Target Selection. A group of ten metalloprotein inhibitors were
selected for evaluation (Figure 1). The molecules were chosen to represent a diverse range of
MBGs (7 distinct MBGs), target proteins (7 distinct targets), and therapeutic indications (at least
8 distinct pathologies). The inhibitors selected included both FDAꢀapproved drugs as well as
potent inhibitors from the scientific literature that were either purchased or synthesized according
to published methods (Table 1).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Metalloenzyme inhibitors examined in this study.
FDA
Inhibitor
Source
Target Protein
IC50 (nM)
Indication
Glaucoma
Approved
Yes
Acetazolamide Commercial22
hCAII
ACE
25
21
Captopril
SAHA
Commercial23
Hypertension Yes
Cutaneous Tꢀ
Synthesized24, 25 HDAC
10
cell
Yes
lymphoma
Asthma
Zileuton
RCDꢀ1
BOTI
Commercial26
Synthesized27
Synthesized28
5ꢀLOX
HIV IN
BoNT/A
600
60
Yes
No
No
HIV/AIDS
Botulism
Tumor
410
NSA
Synthesized29
Synthesized30
Synthesized31
MMPꢀ2, ꢀ9
MMPꢀ12
MMPs
240, 310
18
No
No
No
malignancy
Tumor
1,2ꢀHOPOꢀ2
malignancy
Tumor
CGS
8ꢀ43
malignancy
Deferoxamine Commercial32
Transferrin/Ferritin N/A
Iron overload Yes
6
ACS Paragon Plus Environment

Page 7 of 42
Journal of Medicinal Chemistry
1
2
Discovered in the 1930s,33 carbonic anhydrase (CA) represents the oldest known class of
Zn2+ metalloenzyme. The active site lies at the bottom of an open, coneꢀshaped pocket in the
protein structure. The enzyme possesses a His3 binding motif coordinated to the active site Zn2+
ion with a nucleophilic hydroxide occupying the fourth coordination site (Figure 2). Human CA
(hCA) regulates blood pH, reversibly catalyzing the dehydration of bicarbonate to yield carbon
dioxide (Figure 2). Systemic inhibition of hCA can have deleterious effects on the body in the
form of acidosis as blood pH homeostasis is disrupted. Acetazolamide (diamox) is an FDAꢀ
approved drug whose clinical efficacy for treating glaucoma was first determined in 1954 by
Becker.22 Acetazolamide inhibits many hCA isoforms (IC50 ~25 nM) by binding the catalytic
Zn2+ ion through a sulfonamide moiety (Figure 1). The drug disrupts the homeostasis between
carbon dioxide and bicarbonate and as the concentration of bicarbonate decreases, water is drawn
from the eye by osmosis, decreasing intraocular pressure, which proves to be an effective
treatment for many cases of glaucoma.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Matrix metalloproteinases (MMPs) are a group of Zn2+ꢀdependent endopeptidases known
for their potential to disrupt angiogenesis in malignant tumors and represent a prototypical
metalloenzyme target in medicinal chemistry.34ꢀ36 MMP enzymes are capable of breaking down
many forms of connective tissue, including gelatin (MMPꢀ2 and ꢀ9) and elastin (MMPꢀ12). The
MMP active site contains a catalytic Zn2+ ion coordinated by a His3 motif and a water molecule
(Figure 2).37 Three MMP inhibitors (Figure 1) were chosen for this study based on their
different MBGs as well as known isoform selectivity. Developed by Tamura et al,29 NSA (for Nꢀ
sulfonylamino acid) is an MMPꢀ2 and ꢀ9 selective inhibitor (IC50 240 and 310 nM, respectively)
that uses a carboxylic acid moiety as the MBG. A second MMP inhibitor, 1,2ꢀHOPOꢀ2, uses a
hydroxypyridinone MBG, which is quite distinct from other MMP inhibitors. 1,2ꢀHOPOꢀ2 was
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 42
1
2
3
4
5
6
7
8
9
prepared by Agrawal et al as an MMPꢀ12 inhibitor, but shows a semiꢀselective profile with
activity against other isoforms.38 The third MMP inhibitor included in the study is the potent,
broad spectrum compound CGS 27023A.31 Developed by Novarits Pharmaceuticals, CGS
27023A incorporates a hydroxamic acid MBG, the most common MBG for targeting MMPs and
most other metalloenzymes. Despite the discovery of several potent broadꢀspectrum MMP
inhibitors, no inhibitors have successfully completed clinical trials for any major disease
indications (e.g. cancer, arthritis, etc.), due to problematic side effects, lack of efficacy in human
trials, and other barriers.3, 7 Currently the only FDAꢀapproved MMP inhibitor, doxycycline, is a
treatment for periodontitis.39
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Angiotensin converting enzyme (ACE) is a Zn2+ dipeptidyl carboxypeptidase in which
the catalytic Zn2+ ion is bound to the protein by a His2Glu motif (Figure 2). ACE functions
endogenously to generate the biologically active polypeptide angiotensin II from angiotensin I.40
The cleaved product induces a vasoconstriction response that manifests as increased blood
pressure. An important metalloenzyme inhibitor for treating hypertension, captopril (capoten,
Squibb, patented in 1976),23 incorporates a thiol group capable of coordinating the catalytic Zn2+
ion in the active site of ACE (Figure 1). Subsequent ACE inhibitors moved away from the thiol
MBG, in order to avoid the deleterious clinical effects associated with some thiols. Secondꢀ
generation ACE inhibitors generally utilized carboxylates as MBGs, requiring further backbone
enhancements to achieve comparable levels of inhibition to captopril.41 Due to the similarity in
function and active site between MMPs and ACE, ACE inhibitors have been previously studied
and observed to crossꢀinhibit MMPs.42 Captopril has also been observed to inhibit mushroom
tyrosinase, another metalloprotein examined in this study (see below).43
8
ACS Paragon Plus Environment

Page 9 of 42
Journal of Medicinal Chemistry
1
2
The histone deacetylase (HDAC) family of Zn2+ metalloenzymes, in conjunction with
histone acetyltransferases (HAT), regulate gene expression. HDACs can be segregated into
Zn2+ꢀdependent metalloenzymes (Classes I, IIa, IIb, and IV) and the metalꢀfree sirtuins (Class
III). For this study, only the Zn2+ metalloenzymes were considered. Deacetylation results in the
positively charged εꢀamines of lysine interacting strongly with the negatively charged phosphate
backbone of DNA, serving to coil DNA tighter onto the histone scaffold comprising chromatin.
At the site of increased DNA constriction translation is turned off and genes are repressed. The
genes downꢀregulated by HDACs include those coding for cancer cell apoptosis, supplying the
impetus for inhibitor development.44, 45 The HDAC family incorporates an active site consisting
of a Zn2+ ion bound by a HisAsp2 motif (Figure 2). Inhibitors of HDACs, such as
suberoylanilide hydroxamic acid (SAHA, vorinostat, Merck), activate tumor suppression genes
within cancerous cells leading to growth arrest and cell death.46 SAHA, a firstꢀinꢀclass, broad
spectrum HDAC inhibitor, is a relatively simple molecule incorporating a hydroxamic acid
MBG, an aliphatic linker, and an aromatic capping group (Figure 1). SAHA was discovered in
1996 by Richon et al25 and approved by the FDA as a treatment for cutaneous Tꢀcell lymphoma
in 2006.47 Currently, there is only one other HDAC inhibitor that has received FDA approval,
romidepsin, also for treating cutaneous T cell lymphoma.48
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mushroom tyrosinase (TY) is a dinuclear Cu2+ oxidase controlling the production of
melanin. The two active site Cu2+ ions are bound by six histidine residues (three to each metal
ion, Figure 2) and reversibly bind either a bridging ꢁꢀoxide or peroxide species.49 TY converts
tyrosine to LꢀDOPA and subsequently LꢀDOPA to dopaquinone. Dopaquinone then proceeds
through a series of intermediates, eventually polymerizing to generate melanin.49 Inhibition of
TY decreases melanin production and subsequently results in skin whitening. Consequently, TY
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 42
1
2
inhibitors are of interest to the cosmetics industry.50 The natural product tropolone was used as
an inhibitor of TY for this study (IC50 ~400 nM); notably, a crystal structure of tropolone bound
to TY shows the molecule occupying the substrate channel of the enzyme without coordinating
the active site metal ions (PDB: 2Y9X).51 However, the tropolone MBG is known to inhibit
several other metalloenzymes49 and has been used has been used successfully as a lead for
inhibitor development.52
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Finally, transferrin, the primary iron transport protein, was included in this study to
represent metal transport and homeostasis proteins. Transferrin serves a dual purpose,
transporting iron throughout the body while scavenging any free or loosely bound iron in the
serum. There are two Fe3+ ion binding sites in transferrin, an Nꢀterminal domain and a Cꢀ
terminal domain, with both sites possessing a high affinity for Fe3+ ions.53 Transferrin is
extremely effective at binding Fe3+ ions, possessing an association constant of 1022 Mꢀ1.20 As a
result, there is essentially no free iron in the serum of healthy humans. Pathogens have
developed potent Fe3+ chelators known as siderophores in order to obtain the necessary Fe3+ ions
for proliferation.54 The aim of the transferrin assays was to determine if any of the
metalloenzyme inhibitors studied possess the capacity to remove Fe3+ ions from holoꢀtransferrin.
Deferoxamine (desferal) is a siderophore originally isolated from Streptomyces pilosus by Bickel
et al in 1960.32 Deferoxamine (Figure 1) is a clinically approved therapeutic for iron overload
and was chosen for screening against the panel of metalloenzymes in this study. Despite the
efficacy of deferoxamine for eliminating iron, it does not remove Fe3+ ions from holoꢀtransferrin
on a rapid timescale55 (although many factors such as ionic strength, pH, and others can
influence removal kinetics).56, 57 Deferiprone (L1, Figure 1), another clinically approved Fe3+
10
ACS Paragon Plus Environment

Page 11 of 42
Journal of Medicinal Chemistry
1
2
chelating agent was also included as a positive control, as it is known to rapidly remove Fe3+ ions
from holoꢀtransferrin.21
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 42
1
2
3
4
5
6
7
8
9
Figure 2. Reactions catalyzed by and active sites of the metalloenzymes studied (top to
bottom):58 MMPꢀ2 (1QIB),59 hCAII (3KS3),60 HDACꢀ2 (3MAX),61 ACE (1O8A),62 and TY
(2Y9W).51 Residues involved in ligating the metal ion are explicitly shown.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Other Inhibitors Tested. Human immunodeficiency virus integrase (HIVꢀ1 IN)
incorporates viral DNA into the host DNA, generating reactive CpA 3'ꢀhydroxyl ends on the
viral cDNA and subsequently fusing the viral DNA into the host genome in a process known as
strand transfer. The two Mg2+ ions in the dinuclear active site are responsible for activating the
DNA primer 3'-hydroxyl group.63 Raltegravir is a firstꢀinꢀclass inhibitor of HIVꢀ1 IN developed
by Merck that received FDA approval in 2007 (Figure 1).64 Inhibitors based on the MBGs
similar to that of raltegravir were developed by Agrawal et al to elucidate the effect of the MBG
on HIV integrase inhibition efficacy.27 One compound used in the latter study, RCDꢀ1,
possesses the same 5ꢀhydroxyꢀ3ꢀmethylpyrimidinꢀ4(3H)ꢀone (HMPO) MBG found in
raltegravir, and shows effective strand transfer inhibition. RCDꢀ1 (IC50 ~60 nM) was chosen as
representative of raltegravir, because it is more synthetically accessible and shares a common
MBG and backbone with the FDA approved drug.
An extremely potent family of toxins, botulinum neurotoxins (BoNTs) are the Zn2+ꢀ
dependent metalloenzyme toxins produced by Clostridium botulinum. Due to the paucity of
available treatments for botulism, synthetic efforts have been directed towards developing a
potent inhibitor of botulinum neurotoxin. The BoNT inhibitor selected for this study was
designed by the Janda group28 (BOTI, Figure 1) around the ubiquitous hydroxamic acid MBG
with a backbone to impart potency against BoNT serotype A (BoNT/A) with an IC50 of 410 nM.
12
ACS Paragon Plus Environment

Page 13 of 42
Journal of Medicinal Chemistry
1
2
5ꢀLipoxygenase (5ꢀLOX) is a nonꢀheme Fe3+ꢀdependent dioxygenase responsible for
smooth muscle contractions observed in asthma and allergic reactions.65 5ꢀLOX functions
endogenously to convert cisꢀpolyunsaturated fatty acids into leukotrienes, first adding molecular
oxygen to the fifth carbon on the fatty acid, generating a hydroperoxide, and subsequently
dehydrating the hydroperoxide to yield an epoxideꢀcontaining leukotriene.66 The leukotrienes
trigger an inflammatory response leading to bronchoconstriction. Correspondingly, inhibitors of
5ꢀLOX activity are sought for their therapeutic applications towards treating asthma. One such
inhibitor, zileuton (zyflo), is an FDAꢀapproved drug for the prophylactic treatment of asthma
(IC50 410 nM).26 Zileuton was developed by Abbot Laboratories in 199126 and was approved for
distribution in 1996.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results
Inhibition of MMPs. The activity of MMPꢀ2 and ꢀ12 were monitored via a kinetic assay
that measures the increase in fluorescence upon cleavage of a peptidic substrate (OmniMMP).67
The assay was performed in buffer (50 mM HEPES, 10 mM CaCl2, 0.05% Brijꢀ35, pH 7.5) in
which the substrate was added to a mixture of protein and inhibitor which had been preincubated
at 37 °C for 30 min. The effect of all ten inhibitors against these proteinases is compared in
Figure 3. At 10 ꢂM, CGS exhibited greater than 95% inhibition of both MMPs. Although NSA
is reported to be an MMPꢀ2 and ꢀ9 isoform inhibitor (IC50 240 and 310 nM, respectively) at high
concentrations, such as 10 ꢁM used here, isoform selectivity was abolished, resulting in total
inhibition of MMPꢀ2 and ꢀ12. The third MMP inhibitor, 1,2ꢀHOPOꢀ2, retained some selectivity
towards MMPꢀ12, achieving 100% inhibition against MMPꢀ12, but only 80% inhibition against
MMPꢀ2. At a concentration of 10 ꢁM, the other hydroxamic acidꢀbased inhibitors in the study,
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 42
1
2
3
4
5
6
7
8
9
SAHA and BOTI, were observed to decrease MMP activity by <25%. RCDꢀ1 also decreased
MMPꢀ2 activity by ~50%, but this is not entirely surprising based on overall structural
similarities (MBG, linker, and backbone) to 1,2ꢀHOPOꢀ2. The remaining compounds
demonstrated little activity against MMPꢀ2 and ꢀ12 (Figure 3). It is particularly interesting to
note that despite previous reports in which captopril was observed to inhibit MMPꢀ2 in cell
studies,42 less than 10% inhibition was observed in our experiments.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Percent enzyme activity of MMPꢀ2 (black) and MMPꢀ12 (gray) in the presence of 10
µM of each metalloenzyme inhibitor.
Inhibition of hCAII. The activity of hCAII was evaluated using a common procedure by
monitoring the esterase activity of the enzyme with pꢀnitrophenyl acetate as a substrate.68 The
assay was performed in buffer (50 mM TrisꢀSO4, pH 8.0) in which the substrate was added to a
mixture of protein and inhibitor which had been preincubated at 30 °C for 10 min. Upon
14
ACS Paragon Plus Environment

Page 15 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
cleavage of the acetate group, pꢀnitrophenolate anion absorbs at 405 nm allowing for observation
of hCAII activity as a function of increasing absorbance over time. The effects of the inhibitors
against hCAII are compared in Figure 4. Of the inhibitors screened, only the onꢀtarget inhibitor,
acetazolamide, significantly inhibited hCAII at a concentration of 10 ꢂM, resulting in essentially
complete (100%) inhibition. All the remaining inhibitors were inactive against hCAII. Despite
the fact that seven of the ten inhibitors in this study are designed for Zn2+ metalloenzymes, and
hCAII has a relatively open active site (Figure 2), only minor crossꢀinhibition was observed.
These results suggest that even potent MBGs for Zn2+ ions such as hydroxamic acids and thiols
will not indiscriminately inhibit Zn2+ metalloenzymes in the absence of favorable inhibitor
backboneꢀprotein interactions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Percent enzyme activity of hCAII in the presence of 10 µM of each metalloenzyme
inhibitor.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 42
1
2
3
4
5
6
7
8
Inhibition of HDAC-2. The activity of HDACꢀ2 was measured following the
deacetylation of a small, dyeꢀcontaining peptide. The assay was performed using a
commercially available kit according to manufacturer instructions. The assay was performed in
buffer (supplied by the manufacturer) in which the substrate was added to a mixture of protein
and inhibitor that had been preincubated at room temperature for 5 min. The natural product
trichostatin A (TSA) is utilized as part of the endpoint assay to quench enzyme activity and was
also used as an added positive control in this study. At 10 ꢂM, SAHA and TSA completely
prevent substrate deacetylation (Figure 5). BOTI also demonstrated some capacity for inhibiting
HDAC, presumably due to the structural similarities between the two inhibitors. BOTI is a
relatively simple hydroxamic acid, possessing a MBG, linker, and capping group typically seen
in HDAC inhibitor design.45 Interestingly, the third hydroxamic acid, CGS, did not demonstrate
any inhibitory activity against HDACꢀ2. The active site in HDACꢀ2 is accessed through a
narrow pore in the protein, typically filled by the linker of an HDAC inhibitor that presumably
cannot accommodate the steric bulk of CGS. No inhibition was observed among the remaining
inhibitors.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16
ACS Paragon Plus Environment

Page 17 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Percent enzyme activity of HDACꢀ2 in the presence of 10 µM of each metalloenzyme
inhibitor.
Inhibition of ACE. ACE activity was determined as previously reported using the
OmniMMP fluorogenic substrate.69 In addition to MMPs, OmniMMP has been determined to be
a suitable substrate for a variety of Zn2+ꢀdependent metalloproteases.69 The assay was performed
in buffer (50 mM MES, pH 6.5) in which the substrate was added to a mixture of protein and
inhibitor which had been preincubated at 37 °C for 10 min. The FDAꢀapproved ACE inhibitor
captopril served as the positive control in this assay. Two other inhibitors from this study were
observed to inhibit at 10 ꢂM, 1,2ꢀHOPOꢀ2 and SAHA (Figure 6). Inhibition by 1,2ꢀHOPOꢀ2 is
not particularly surprising based on the degree of substrate overlap between MMPs and ACE (i.e.
the fluorogenic OmniMMP substrate is used in both assays).69 Substrate overlap also explains
why CGS, a potent, broad spectrum MMP inhibitor, also showed weak inhibition of ACE;
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 42
1
2
3
4
5
6
however, it is interesting to note that the MMP inhibitor NSA did not inhibit ACE, showing that
high selectivity between MMPs and ACE inhibitors can be achieved.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Percent enzyme activity of ACE in the presence of 10 µM of each metalloenzyme
inhibitor.
Inhibition of TY. TY catalyzes two sequential reactions, oxidation of tyrosine to
LꢀDOPA
(L
ꢀ3,4ꢀdihydroxyphenylalanine) and oxidation of
LꢀDOPA to
L
ꢀdopaquinone.67 The product,
Lꢀ
dopaquinone, polymerizes to generate melanin, which can be observed spectrophotometrically
by an increase in absorbance at 475 nm.70 In this study,
LꢀDOPA was used as the substrate for
assaying TY activity. The assay was performed in buffer (10 mM NaH2PO4, pH 6.8) in which
the substrate was added to a mixture of protein and inhibitor that had been preincubated at room
temperature for 10 min. The reaction was allowed to proceed for 10 min before recording the
18
ACS Paragon Plus Environment

Page 19 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
change in absorbance. Interestingly, none of the inhibitors tested were found to reduce TY
activity at 10 ꢂM (Figure 7). Only the positive control, tropolone (IC50 ~400 nM), was effective
at shutting down TY activity. This is somewhat surprising, as studies on a simple fragment
library showed that TY is readily inhibited by a wide range of MBGs.67 Even RCDꢀ1, which is a
dinuclear metalloenzyme inhibitor, did not inhibit TY. These findings reinforce the notion that
enzymatic selectivity is imparted to a metalloenzyme inhibitor by both the MBG and backbone
constituents, as was recently demonstrated by Fullagar et al52 in which a tropolone derivative
designed as an inhibitor of the Zn2+ꢀdependent metalloenzyme LasB simultaneously lost efficacy
for TY inhibition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Percent enzyme activity of TY in the presence of 10 µM of each metalloenzyme
inhibitor.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 42
1
2
3
4
5
6
7
8
9
Inhibition of Trypsin. Trypsin is a serine endopeptidase (metalꢀindependent enzyme) that
participates in the degradation of proteins in the digestive system. We sought to include the
possible crossꢀreactivity of the metalloprotein inhibitors on one, common metalꢀindependent
enzyme. Typsin was an ideal choice because of its availability, facile assay, and similar function
(proteinase) to many of the other enzymes (e.g. MMPs) examined in this study. The assay was
performed using a commercially available kit according to manufacturer instructions. The assay
was performed in buffer (supplied by the manufacturer) in which the substrate was added to a
mixture of protein and inhibitor that had been preincubated at room temperature for 10 min. An
irreversible inhibitor of trypsin, tosyllysine chloromethyl ketone hydrochloride (TLCK), was
included as a positive control. None of the metalloprotein inhibitors showed any inhibition of
trypsin at a concentration of 10 ꢂM (Figure 8). Notably, trypsin is an integral part of the HDACꢀ
2 activity assay used in this study, confirming that inhibitor activity against HDACꢀ2 is not due
to inhibition of trypsin (Figure 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Percent enzyme activity of trypsin in the presence of 10 µM of each metalloenzyme
inhibitor.
Transferrin Fe3+ Removal Assay. Biological access to transition metal ions in vivo is
highly regulated and restricted. To explore whether the potent inhibitors in Figure 1 could access
biological sources of transition metal ions, their ability to extract Fe3+ from the serum
metalloprotein transferrin was examined. The capacity of an inhibitor to remove Fe3+ from
transferrin was determined by using standard assay conditions,57, 71, 72 where the change in
absorbance at 466 nm of a 100 ꢁM solution (50 mM Tris, pH 7.4, 150 mM NaCl, 20 mM
NaHCO3) of Fe3+ꢀsaturated transferrin (holoꢀtransferrin) is monitored over the course of 1 hour
in the presence of an inhibitor. The only inhibitor observed to remove Fe3+ from transferrin was
1,2ꢀHOPOꢀ2 (Table 2), which is consistent with observations that HOPOꢀbased ligands are
known to effect iron removal from transferrin. Indeed, the positive control for the metal removal
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 42
1
2
3
4
5
6
7
8
9
assays, deferiprone, is an Nꢀmethylated 3,4ꢀHOPO derivative. Deferoxamine is known to
remove Fe3+ ions from ferritin, but not transferrin, so it is not surprising that deferoxamine
remained innocuous in this assay. No data was obtained for RCDꢀ1, as it interfered with the
assay conditions.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Pseudoꢀfirstꢀorder rate constants for iron removal. 100 ꢂM Human holoꢀtransferrin
and 1 mM of each compound is used in the assay.
Inhibitor
Acetazolamide
Captopril
SAHA
k
obs (x 102 minꢀ1)
0.00
0.00
0.00
0.00
ꢀ
Zileuton
RCDꢀ1
BOTI
0.00
0.00
0.92
0.00
0.00
4.06
NSA
1,2ꢀHOPOꢀ2
CGS
Deferoxamine
Deferiprone
Discussion
As more metalloproteins are identified and validated as clinical targets, the potential for
crossꢀinhibition by inhibitors will become an increasing concern. Given the prevalence of metal
ion cofactors in biology, the possibility for indiscriminate metal binding could have a range of
deleterious effects. Inhibitors bound to unintended metalloproteins decreases the bioavailability
of inhibitor for the target enzyme, as well as potentially causing undesired offꢀtarget effects. The
22
ACS Paragon Plus Environment

Page 23 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
liberation of loosely coordinated metal ions could also lead to Fenton chemistry, resulting in
damaged biomolecules and possible lossꢀofꢀfunction. To the best of our knowledge, no study
has been performed to investigate the degree of interaction between a metal chelating,
metalloprotein inhibitor and a broad range of metalloproteins. A prior study investigated small
(<175 amu) metal binding fragments (i.e. MBGs) against a similar panel of metalloenzymes and
found that these fragments showed a range of inhibitory behavior, from broadꢀspectrum to highly
selective.67
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The hydroxamic acid MBG is the prototypical chelator for metalloenzyme inhibitor
design, and consequently three hydroxamateꢀbased compounds were included in this study
(Figure 1). The hydroxamic acid CGS is an extremely potent, broadꢀspectrum MMP inhibitor,
and yet shows little offꢀtarget metalloenzyme inhibition (Figure 9). In contrast, the other
hydroxamates, SAHA and BOTI, both showed some limited offꢀtarget inhibition. The selectivity
displayed by these hydroxamate inhibitors presumably stems from the composition of the
backbone group. The backbone of CGS is complex and has been finely tuned to fit into the
active site of MMPs, imparting both potency and selectivity to the target. In contrast, SAHA and
BOTI are much simpler inhibitors that possess backbones designed to fill a narrow channel in the
enzymes, with little specificity built in beyond the linker length between the MBG and the
capping group. Although SAHA and BOTI do exhibit offꢀtarget inhibition, it is still important to
note that the degree of inhibition is weak. Perhaps the same factors that result in their lower
specificity (i.e. their simplicity), also restricts these inhibitors to being weak offꢀtarget inhibitors,
hence incapable of matching the activity of onꢀtarget inhibitors, and not exhibiting problematic
crossꢀreactivity at therapeuticallyꢀrelevant concentrations.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 42
1
2
3
4
5
6
7
8
9
Three different MMP inhibitors, NSA, CGS, and 1,2ꢀHOPOꢀ2, include three different
MBGs: a carboxylic acid, hydroxamic acid, and a hydroxypyridinone, respectively (Figure 1).
These inhibitors were selective for MMPs with the exception of 1,2ꢀHOPOꢀ2, which was
observed to inhibit ACE, an enzyme with great similarity to the MMPs. This offꢀtarget
inhibition might be attributed to the more tightly coordinating MBG found in 1,2ꢀHOPOꢀ2
versus NSA or CGS.67 Indeed, CGS and NSA are highly active MMP inhibitors, but neither
showed activity against ACE (Figure 9). Hence, the importance of the MBG to inhibition is best
considered within the context of the full structure of the inhibitor when optimizing potency and
selectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although still quite low, the predominant offꢀtarget inhibition is observed in the cases of
RCDꢀ1, BOTI, and 1,2ꢀHOPOꢀ2, all of which are not FDA approved compounds and thereby
share a degree of simplicity relative to approved compounds. RCDꢀ1 is a substantially truncated
version of raltegravir, BOTI is a small compound that bears a strong resemblance to simple
HDAC inhibitors, and 1,2ꢀHOPOꢀ2 relies on a simple biphenyl moiety as the backbone. Also,
all three compounds possess different MBGs, precluding the possibility that an MBG alone will
impart promiscuity. Based on the high concentrations of compounds used in these experiments,
it is clear that none of these inhibitors exhibit significant offꢀtarget inhibition of the enzymes
tested.
In an effort to probe the most extreme form of metalloprotein and metal ion homeostasis
disruption, Fe3+ stripping studies from holoꢀtransferrin were performed. No inhibitor, with the
exception of 1,2ꢀHOPOꢀ2, could remove iron from transferrin (Table 2). It is also important to
stress that millimolar concentrations of 1,2ꢀHOPOꢀ2 were required to observe iron removal on a
reasonable timescale. Given that all of the metalloenzyme inhibitors studied are nanomolar
24
ACS Paragon Plus Environment

Page 25 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitors of their respective targets, and, at most, reach peak serum concentrations in the low
micromolar range,73ꢀ77 the possibility of metal stripping from transferrin is highly unlikely.
Advancements in bioinorganic chemistry indicate that metal ion homeostasis is tightly regulated
and that intracellular concentrations of free transition metal ions are very low, further suggesting
that metalloprotein inhibitors are unlikely to perturb metal trafficking.78, 79
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 9. Percent enzyme activity for each metalloenzyme in the presence of 10 µM captopril
(black) or CGS (gray) highlighting the selectivity of each inhibitor.
Conclusion
A variety of known, potent metalloenzyme inhibitors were screened against a
representative panel of metalloproteins. The results indicate that at very high concentrations of
inhibitor (10 µM), offꢀtarget inhibition is limited, even when inhibitors contain similar or
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
identical MBGs. Not surprisingly, the limited offꢀtarget inhibition observed is greatest among
metalloenzymes incorporating the same catalytic metal, active site structure, and functionality.
In addition, offꢀtarget inhibition was more common among earlyꢀstage lead compounds (e.g. 1,2ꢀ
HOPOꢀ2, BOTI) for which complete drug development efforts have not been applied. Almost
all FDAꢀapproved, fullꢀlength inhibitors showed no offꢀtarget inhibition, regardless of the MBG,
suggesting that inhibitor selectivity is a combination of the MBG and the backbone. Therefore,
the findings presented here strongly suggest that metal chelating inhibitors of metalloproteins
present no particular risk for indiscriminate offꢀtarget inhibition of metalloenzymes as a result of
nonꢀspecific metal chelation. Therefore, it appears that the development of metalloprotein
inhibitors present no more risk for nonꢀspecific activity than small molecule inhibitors of metalꢀ
independent enzymes. These results should help to spark renewed interest in such targets, where
a wealth of therapeutic possibilities await.14, 80
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
Materials. All reagents and acetazolamide, captopril, zileuton, and deferoxamine were obtained
from SigmaꢀAldrich. RCDꢀ1, 1,2ꢀHOPOꢀ2, BOTI, CGS, SAHA, and TSA were prepared by
literature methods.24, 27ꢀ31 The purity of all synthesized compounds was determined to be ≥ 95%
by analytical HPLC. UV/Visible spectra were recorded with a PerkinꢀElmer Lambda 25
spectrophotometer. Absorbance assays were performed using a BioTek ELx800 microplate
reader. Fluorescence assays were performed using either a BioTek FLx800 microplate reader or
a BioTek Synergy HT microplate reader.
26
ACS Paragon Plus Environment

Page 27 of 42
Journal of Medicinal Chemistry
1
2
Human Carbonic Anhydrase II Activity Assays.68 hCAII was expressed and purified as
previously described.81 Assays were carried out in clear Costar 96ꢀwell plates. Each well
contained a volume of 100 ꢂL including buffer (50 mM TrisꢀSO4, pH = 8.0), hCAII (200 nM),
inhibitor (100 ꢂM), and pꢀnitrophenyl acetate (500 ꢂM). The enzyme and inhibitor were
incubated in solution at 30 °C for 10 min prior to the addition of the substrate to initiate the
reaction. The change in absorbance was monitored for 20 min at 405 nm. The negative control
wells, containing no inhibitor, were arbitrarily set as 100% activity. Readings from background
wells, which did not contain protein, were substracted from the active assay wells in order to
account for background hydrolysis activity caused by the buffer. The assays were performed in
triplicate, and each assay contained each inhibitor in triplicate. The data were normalized to
values measured for uninhibited enzyme. Assays were reported as mean Ə standard deviation.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MMP Assay.30 MMPꢀ2 and ꢀ12, and OmniMMPTM fluorogenic substrate were purchased from
Enzo Life Sciences (Farmingdale, NY). The assays were carried out in white NUNC 96ꢀwell
plates. Each well contained a volume of 100 ꢂL including buffer (50 mM HEPES, 10 mM
CaCl2, 0.05% Brijꢀ35, pH 7.5), human recombinant MMP (1.16 U of MMPꢀ2, 0.007 U of MMPꢀ
12, ENZO Life Sciences), inhibitor (100 uM), and fluorogenic OmniMMP substrate (4 ꢂM Mcaꢀ
ProꢀLeuꢀGlyꢀLeuꢀDpaꢀAlaꢀArgꢀNH2•AcOH, ENZO Life Sciences). The enzyme and inhibitor
were incubated in solution at 37 °C for 30 min, followed by the addition of the substrate to
initiate the reaction. The change in fluorescence was monitored for 30 minutes on a BioTek Flx
800 fluorescence plate reader with excitation and emission wavelengths at 320 and 400 nm,
respectively. The negative control wells, containing no inhibitor, were arbitrarily set as 100%
activity. MMP activity was defined as the ratio of fluorescence increase in the inhibitor wells
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 42
1
2
3
4
5
6
7
8
9
relative to the negative control wells, expressed as a percentage. The assays were performed in
triplicate, and each assay contained each inhibitor in triplicate. The data were normalized to
values measured for uninhibited enzyme. Assays were reported as mean Ə standard deviation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HDAC Assay. The HDAC assay kit (green) was purchased from BPS Bioscience (San Diego,
CA). The assays were carried out in black, low binding NUNC 96ꢀwell plates. Each well
contained a volume of 50 ꢂL including buffer (BPS Bioscience, Catalog No. 50031), HDACꢀ2
(30 ng, BPS Bioscience, Catalog No. 50002), inhibitor (10 ꢂM), BSA (1 mg/mL, Sigmaꢀ
Aldrich), and HDAC substrate 2 (20 ꢂM, BPS Bioscience, Catalog No. 50038). Prior to adding
substrate, the plate was preincubated for 5 min. Upon addition of substrate the plate was
incubated at 37 °C for 30 min. HDAC assay developer (50 ꢂL, BPS Bioscience, Catalog No.
50030) was added to each well and the plate incubated for 15 min at room temperature. The
fluorescence was recorded at excitation and emission wavelengths of 485 and 528 nm,
respectively. Blank wells containing no inhibitor or protein were subtracted from all wells. The
assays were performed in triplicate, and each assay contained each inhibitor in triplicate. The
data were normalized to values measured for uninhibited enzyme. Assays were reported as mean
Ə standard deviation.
ACE Assay.69 ACE was purchased from R&D Systems (Minneapolis, MN). The assays were
carried out in white NUNC 96ꢀwell plates. Each well contained 100 ꢂL including buffer (50 mM
MES, pH 6.5), human recombinant ACE/CD143 somatic form (20 ng, R&D Systems), inhibitor
(10 uM), and fluorogenic OmniMMP substrate (4 ꢂM McaꢀProꢀLeuꢀGlyꢀLeuꢀDpaꢀAlaꢀArgꢀ
NH2•AcOH, ENZO Life Sciences). The enzyme and inhibitor were incubated in solution at 37
28
ACS Paragon Plus Environment

Page 29 of 42
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
°C for 10 min prior to substrate addition. The change in fluorescence was monitored for 30 min
with excitation and emission wavelengths at 320 and 400 nm, respectively. The assays were
performed in triplicate, and each assay contained each inhibitor in triplicate. The data were
normalized to values measured for uninhibited enzyme. Assays were reported as mean Ə
standard deviation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Tyrosinase Assay.70 The assays were carried out in clear Costar 96ꢀwell plates. Each well
contained a volume of 100 ꢂL including buffer (10 mM NaH2PO4, pH=6.8), mushroom
tyrosinase (30 U, SigmaꢀAldrich), inhibitor (10 ꢂM), and LꢀDOPA (0.5 mM). The enzyme and
inhibitor were incubated in solution at room temperature for 10 min. Longer incubation times
(up to 30 min) did not significantly change the assay results. A background absorbance was
recorded at 475 nm using a BioTek ELx 808 colorimetric plate reader.
LꢀDOPA was added to
each well concurrently to initiate the reaction. After 10 min a second absorbance reading was
recorded at 475 nm. The background absorbance was subtracted from the second absorbance.
The negative control wells, containing no inhibitor, were arbitrarily set as 100% activity. The
assays were performed in triplicate, and each assay contained each inhibitor in triplicate. The
data were normalized to values measured for uninhibited enzyme. Assays were reported as mean
Ə standard deviation.
Trypsin Assay. The trypsin assay kit was purchased from BioVision (San Francisco, CA). The
assays were performed using a commercially available colorimetric assay kit purchased from
BioVision. The assays were carried out in clear Costar 96ꢀwell plates. Each well contained 100
ꢂL including buffer (BioVision, Catalog No. K771ꢀ100ꢀ1), trypsin (100 mU, TPCK treated,
29
ACS Paragon Plus Environment