Design, synthesis, and biological evaluation of novel matrix metalloproteinase inhibitors as potent antihemorrhagic agents: From hit identification to an optimized lead

Article abstract of DOI:10.1021/jm501940y

Growing evidence suggests that matrix metalloproteinases (MMP) are involved in thrombus dissolution; then, considering that new therapeutic strategies are required for controlling hemorrhage, we hypothesized that MMP inhibition may reduce bleeding by delaying fibrinolysis. Thus, we designed and synthesized a novel series of MMP inhibitors to identify potential candidates for acute treatment of bleeding. Structure-based and knowledge-based strategies were utilized to design this novel chemical series, α-spiropiperidine hydroxamates, of potent and soluble (>75 μg/mL) pan-MMP inhibitors. The initial hit, 12, was progressed to an optimal lead 19d. Racemic 19d showed a remarkable in vitro phenotypic response and outstanding in vivo efficacy; in fact, the mouse bleeding time at 1 mg/kg was 0.85 min compared to 29.28 min using saline. In addition, 19d displayed an optimal ADME and safety profile (e.g., no thrombus formation). Its corresponding enantiomers were separated, leading to the preclinical candidate 5 (described in Drug Annotations series, J. Med. Chem. 2015, 10.1021/jm501939z).

Full text of DOI:10.1021/jm501940y

Subscriber access provided by RENSSELAER POLYTECH INST
Article
Design, synthesis and biological evaluation of novel matrix
metalloproteinase inhibitors as potent antihemorrhagic
agents: from hit identification to an optimized lead.
Josune Orbe, Juan A Sanchez-Arias, Obdulia Rabal, Jose A Rodriguez, Agustina Salicio, Ana Ugarte,
Miriam Belzunce, Musheng Xu, Wei Wu, haizhong Tan, hongyu Ma, Jose A Paramo, and Julen Oyarzabal
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm501940y • Publication Date (Web): 16 Feb 2015
Downloaded from http://pubs.acs.org on February 18, 2015
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 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, synthesis and biological evaluation of novel
matrix metalloproteinase inhibitors as potent
antihemorrhagic agents: from hit identification to an
optimized lead.
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
Authors:
1
2
2
1
1
Josune Orbe, Juan A. SánchezꢀArias, Obdulia Rabal, José A. Rodríguez, Agustina Salicio,
2
1
3
3
3
3
Ana Ugarte, Miriam Belzunce, Musheng Xu, Wei Wu, Haizhong Tan, Hongyu Ma, José A.
1
,4,*
and Julen Oyarzabal2,∗
Páramo,
1
2
Atherosclerosis Research Laboratory, Small Molecule Discovery Platform, Molecular
Therapeutics Program, Center for Applied Medical Research (CIMA), University of Navarra,
3
Avda. Pio XII 55, Eꢀ31008 Pamplona, Spain; WuXi Apptec (Tianjin) Co. Ltd., No 111
th
4
HuangHai Road, 4 Avenue, TEDA, Tianjin 300456, PR China; Hematology Service, Clínica
Universidad de Navarra. University of Navarra. Avda. Pio XII 36, Eꢀ31008 Pamplona, Spain.
∗
To whom correspondence should be addressed. For J.A.P.: +34 948 194700 (ext. 3015), eꢀmail,
japaramo@unav.es. For J.O.: phone, +34 948 194700 (ext 2044); eꢀmail,
julenoyarzabal@unav.es.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
Growing evidence suggests that matrix metalloproteinases (MMP) are involved in thrombus
dissolution; then, considering that new therapeutic strategies are required for controlling
hemorrhage, we hypothesized that MMP inhibition may reduce bleeding by delaying
fibrinolysis. Thus, we designed and synthesized a novel series of MMP inhibitors to identify
potential candidates for acute treatment of bleeding. Structureꢀbased and knowledgeꢀbased
strategies were utilized to design this novel chemical series, αꢀspiropiperidine hydroxamates, of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
potent and soluble (>75 ꢀg/mL) panꢀMMP inhibitors. The initial hit, 12, was progressed to an
optimal lead 19d. Racemic 19d showed a remarkable in vitro phenotypic response and
outstanding in vivo efficacy; in fact, the mouse bleeding time at 1 mg/kg was 0.85 min compared
to 29.28 min using saline. In addition, 19d displayed an optimal ADME and safety profile (e.g.,
no thrombus formation). Its corresponding enantiomers were separated, leading to the preclinical
candidate 5 (described in drug annotations series, ref18).
ACS Paragon Plus Environment
2

Page 3 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Bleeding represents a dangerous complication that can result in the need for blood transfusion
and is associated with substantial morbidity and mortality. Hemorrhage is responsible for almost
50% of deaths occurring within the first 24 hours of traumatic injury and for up to 80% of
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
1
intraoperative trauma mortality. Physiologically, coagulation occurs to prevent excesive blood
loss after injury and fibrinolysis is required to remove the blood clots after wound repair.
Pathological alterations of this delicate balance can lead to thrombosis or hemorrhage. Plasmin,
the key protein in fibrinolysis, results from plasminogen activation and dysregulation of this
2
system can lead to hyperfibrinolysis, which plays an important role in major bleeding events.
Thus, antifibrinolytic agents are widely used in major surgeries and trauma episodes to reduce
3
blood loss and transfusion and in the treatment of heavy menstrual bleeding. Two synthetic
lysine analogs, εꢀaminocaproic acid (EACA, 1) and tranexamic acid (TXA, 2), are the only
antifibrinolytics commercially available to control bleeding (chemical structures described in
Chart 1); in fact, they are the current standard of care (SoC). Recently, AZD6564 (3, Chart 1), a
piperidylꢀisoxazolone with improved potency over compound 2, was reported as a preclinical
4
candidate for further development. Antifibrinolytics elicit their effects by competitively
reducing the binding of plasminogen to fibrin, thus inhibiting the activation of plasminogen to
5
plasmin, an enzyme that degrades fibrin clots, fibrinogen and other plasma proteins. Molecule 2
6
has been associated with adverse side effects and requires a high oral dose, and 1 is less
7
effective. Moreover, there are some concerns with these compounds due to the potential risk of
8
,9
thrombotic complications.
Although aprotinin, a polypeptidic, bovineꢀderived protease
inhibitor, was an effective agent used to reduce bleeding during complex surgery, safety issues
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
10
led to its temporary withdrawal from the market in 2008. However, very recently, the European
Medicines Agency (EMA) has recommended that the suspension be lifted for a restricted range
11
of indications. Thus, in this scenario, new therapeutic strategies are required for the prevention
of major bleeding events.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Chart 1. Known antifibrinolytics (1, 2 and 3), MMP inhibitors (4, 6, 7, 8, 9 and 10) and structure
of a novel antihemorrhagic agent CMꢀ352 (5)
EACA (1)
TXA (2)
AZD6564 (3)
R
SC-78080/SD-3590 (6)
CM-352 (5)
GM6001/Ilomastat (4)
RS-130830 (8)
SC-77774 (7)
(10)
(9)
ACS Paragon Plus Environment
4

Page 5 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The fibrinolytic and matrix metalloproteinase (MMP) systems cooperate in thrombus
1
2
dissolution. MMPs can act either through direct fibrin/ogen targeting or by enhancing tissue
1
3
plasminogen activator (tPA)ꢀinduced fibrinolysis. This provides a new pharmacological target
for fibrinolysis control (Figure 1), as demonstrated by the markedly reduced fibrinolytic activity
and bleeding time found in MMPꢀ10 null mice, that can be reversed by administration of active
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
1
3
recombinant human MMPꢀ10. Moreover, MMPꢀ3, another fibrinolytic MMP 82% homologous
1
2
14
to MMPꢀ10, has also been associated with intracranial hemorrhage. Thus, the inhibition of
MMPꢀ10, MMPꢀ3 and/or other MMPs may provide a new opportunity for controlling bleeding.
In spite of limited clinical success of hydroxamateꢀbased MMP inhibitors in human cancer trials,
mainly due to lack of efficacy and to reports of sideꢀeffects (i.e., musculoskeletal pain and
1
5,16
inflammation) which occurred after 2ꢀ3 months of treatment
and subsided within 1ꢀ3 weeks
16
17
of discontinuation, there is still hope for MMP inhibition as a therapeutic approach. In fact,
the doses of MMP inhibitors required for acute use in antihemorrhagic treatment are much lower
1
8,19
than those for reducing metastasis;
and, in addition, this indication only requires shortꢀterm
MMP inhibition. Thus, these conditions lead to a new therapeutic opportunity for MMP
inhibitors, as antihemorrhagic agents. In this regard, we have recently demonstrated this
hypothesis with GM6001 (Ilomastat, 4), a synthetic broadꢀspectrum MMP inhibitor, and a novel,
18
preclinical antihemorrhagic candidate, CMꢀ352 (5, Chart 1). This novel compound (5) showed
remarkable efficacy and safety; thus, 5 became a preclinical candidate for the acute treatment of
hemorrhage. Herein we present a detailed account of the discovery of this series from hit
identification to an optimized lead.
Figure 1. Simplified scheme for the proposed cooperation between fibrinolytic and MMP
systems in thrombus dissolution.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PAIs: plasminogen activator inhibitors.
Results and discussion
Validation of a novel
α-spiropiperidine hydroxamate chemical series as MMP inhibitors.
MMP inhibitors (MMPi) have been explored as anticancer, antiꢀinflammatory and antiviral
agents, and different structures have been designed to target the zinc catalytic center of these
enzymes, with the majority of them containing a hydroxamic acid group acting as a chelating
2
0
zinc binding group (ZBG). Particularly, αꢀsulfonylꢀ4ꢀtetrahydropyranyl (THP) and αꢀsulfonylꢀ
αꢀpiperidinyl hydroxamates have been explored as potent MMPꢀ2, ꢀ9 and ꢀ13 inhibitors,
including the development candidate 6 (SCꢀ78080/SDꢀ2590) and the αꢀTHP 7 (SCꢀ77774) by
21
Pfizer (Chart 1). These compounds spare MMPꢀ1. The αꢀsulfone αꢀTHP series possesses better
ADME and PK properties than the closely related βꢀsulfone αꢀTHP core of the Roche clinical
22
candidate RSꢀ130830 (8). Wyeth’s initial work on the αꢀsulfone αꢀpiperidinyl series described
compound 9 as a potent, selective and orally active MMP inhibitor in the clinically relevant
2
3
advanced rabbit osteoarthritis model. Recent work on this chemical series by Wyeth has
ACS Paragon Plus Environment
6

Page 7 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
evolved towards selective TNFꢀα converting enzyme (TACE/ADAM17) inhibitors over MMPs
2
4
by incorporating a butynyloxyphenyl P1’ moiety.
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
a
Figure 2 Structural information for MMPꢀ13 and MMPꢀ3 isoforms, in complex with ligands,
guided the design of novel chemical series 5’.
C)
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
’
a
(
A) Complex of SCꢀ78080 6 with MMPꢀ13 (PDB accession code: 3KRY). (B) Complex of 10 with MMPꢀ3 (PDB
accession code: 1D8F). The molecular surface is drawn to demonstrate the narrowness of the S1’ pocket. (C)
Markush formula of novel αꢀspiropiperidine hydroxamates 5’.
2
1b
According to the Xꢀray structure of 6 with MMPꢀ13, PDB accession code 3KRY, and the Xꢀ
ray structure of MMPꢀ3 complexed with the 1ꢀsulfonylꢀ4ꢀbenzyloxycarbonylꢀpiperazine 10,
2
5
PDB accession code 1D8F, the piperidine Nꢀsubstituent is directed towards the solvent region
(
Figures 2a and 2b). In fact, structural modifications in this region, the solvent accessible area,
generally have a minor effect on potency and can provide a handle for modulating solubility and
pharmacokinetic parameters. From this structural information, we focused our efforts on
replacing the piperidine ring with a conformationally constrained ring system. Conformationally
2
6
constrained ring systems are considered privileged structures. Among them, the azaspirocyclic
system and, in particular spiropiperidines, have emerged as valuable fragments in drug discovery
27
programs. Potentially, the reduced molecular flexibility of these fragments and their spatial
conformation may have a favorable impact on the molecular properties as well as their offꢀtarget
profile (i.e., hERG and P450s). Thus, we designed a new series of MMP inhibitors incorporating
ACS Paragon Plus Environment
8

Page 9 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a spiropiperidine unit (5’, Figure 2C) while maintaining the hydroxamic acid moiety as a zinc
binding group (ZBG). The presence of an ionizable basic nitrogen in our molecules has a
positive impact on the water solubility of these compounds, an issue of the THP chemical
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
28,29
series.
In addition, proposed molecules bearing this chemotype assure an intellectual property
3
0
position for this novel chemical series of αꢀspiropiperidine hydroxamates (5’).
To validate our hypothesis, compounds 11ꢀ18, bearing different S1’ fragments selected
according to structureꢀbased and knowledgeꢀbased approaches, together with their synthetic
accessibility, were synthesized and evaluated for their inhibition against MMPꢀ3 and MMPꢀ10
(Table 1). The simple 4ꢀmethoxyphenyl derivative 12 exhibited nanomolar potency, with IC50
values of <100 nM against MMPꢀ3 and ꢀ10. Replacement of the methyl group in the ether with a
4
ꢀmethoxyphenyl moiety (13) resulted in 10ꢀfold higher IC values against both isoforms (IC
50 50
= 9 and 12 nM against MMPꢀ3 and MMPꢀ10, respectively). The corresponding 4ꢀ
methoxyphenyl ethers at the metaꢀ (14) and orthoꢀ (15) positions of the first aryl ring reduced the
potency (14) or even led to an inactive compound (15), most likely due to steric factors given the
narrowness of the S1’ pocket. Thus, we exclusively concentrated on substitutions at the 4ꢀ
position of the first phenyl ring. Replacement of the 4ꢀmethoxy group of compound 13 with 4ꢀ
OCF (compound 16) improved the activity, lowering the IC50 values to 4 nM against both
3
isoforms; the homologation of the distal phenyl by replacement with a benzyl ring slightly
attenuated the potency (compound 17), with IC values of 20 nM (MMPꢀ3) and 32 nM (MMPꢀ
5
0
10). Replacement of the ether linker with an amino linker (compound 18) decreased the activity
against MMPs. Thus, a pairwise comparison between 17 and 18 reveals that the NHꢀlinked
compound is more than 1.5 and 0.80 log units less potent as an MMPꢀ3 and MMPꢀ10 inhibitor,
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
respectively, than the corresponding oxygenꢀlinked derivative 17. Consequently, the phenotypic
functional assay clearly shows lower efficacy in the corresponding antifibrinolytic activity.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The similar trend in structure activity relationship (SAR) against MMP3 and MMP10 can be
explained by the high sequence similarity between both isoforms, with 86% of identical residues
in the catalytic domain. A superposition of the MMP3 and MMP10 crystal structures (PDB
3
1
1
D8F and 1Q3A, respectively) revealed minor differences between both proteins (rms
superposition of main chain atoms of 0.80 Å); the main difference is the replacement of His724
MMP3) for Asn240 (MMP10), which have their side chains oriented outside the S1’ pocket.
(
Table 1. Initial set of αꢀspiropiperidine hydroxamates
a
Cpd
R1
MMP3 IC50 MMP10
Thromboelastography
EC50 (nM)
(
nM)
IC (nM)
50
11
18900
11400
>15000
F
12
13
14
96
9
84
≤400
≤400
≤400
O
12
O
OMe
178
189
O
OMe
ACS Paragon Plus Environment
10

Page 11 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
15
20000
25400
3700
O
OMe
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
1
6
7
4
4
≤400
≤400
O
OCF3
OCF3
1
20
32
O
18
114
184
900
OCF
3
N
H
a
Human whole blood
Following our screening funnel, compounds with inhibition activities in the low nM range (IC50
100 nM) were tested as antifibrinolytic agents in a phenotypic functional assay using
<
thromboelastometry of whole blood in the presence of tPA. The effective concentration of a
compound to delay clot lysis time (CLT) by 50% (EC ) was considered to represent
5
0
18
antifibrinolytic activity (Table 1). The EC determined for compound 2 was 3 ꢀM; therefore,
5
0
our initial aim was the identification of compounds with improved functional efficacy (EC50)
compared to reference molecule 2. In a first assessment, the lowest concentration tested by
thromboelastometry was 400 nM. Compounds 12, 13, 16 and 17 in Table 1 had EC values
5
0
below this threshold value, showing excellent antifibrinolytic activity compared to 2. Hence,
modifications of the ether linker moiety was our initial favorite strategy for exploring different
substituents on the distal ring, with the most potent compounds, 13, 16 and 17, being our initial
reference points.
Exploration of phenoxyphenyl and benzyloxyphenyl sulfone moieties in
α-spiropiperidine
hydroxamates. Diverse substituents on the distal aryl moiety, entering deep into the hydrophobic
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
32
selectivity S1’ pocket, were selected by following our recently reported strategy for navigating
3
3
the biologically relevant chemical space (BRCS). This approach relies on the concept of
sampling Rꢀgroups according to the different potential interactions they may establish with the
protein (ligandꢀreceptor interactions, LRI) depending on their nature: alkyl, aryl, heteroaryl,
primary/secondary amine (with hydrogen bond donor), tertiary amine, amide (with hydrogen
bond acceptor and donor), ether, and keto (with hydrogen bond acceptor) and up to 14 classes of
different substitution patterns. Representative analogs of the substituted aryloxyphenyl αꢀ
spiropiperidine hydroxamates are shown in Table 2. A wide variety of substituents at the 4ꢀ
position, such as alkyl (19a, 19b), alkoxy (13), and fluorinated alkoxy groups (16, 19c) as well
as monoꢀ (19d) and disubstituted amides (19e), are well tolerated; in fact, these molecules
exhibit IC values in the low nanomolar range. According to the proposed binding site (Figure
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
2
b), 19f should fit and lead to a potent molecule; however, surprisingly, this compound, bearing
an ionizable tertiary amine at the para position that might interact with Phe732 (cationꢀπ), was
inactive. Replacement of the phenyl ring with a 3ꢀpyrido ring (19g) selectively attenuated the
potency against MMPꢀ10 (IC = 62 nM) compared to compound 13, which was 5ꢀfold more
5
0
potent, without significantly altering the potency against MMPꢀ3. Hence, the potent compounds
9aꢀ19e and 19g performed excellently in the functional antifibrinolytic assay, with EC s
1
5
0
≤400 nM. Meta substitution on the distal phenyl ring led to compounds with equivalent
compound 19h compared to 13) or slightly reduced potency (compound 19j compared to 19d)
(
for MMP inhibition. However, surprisingly, both meta derivatives showed poor efficacy in the
functional assay, with EC s of 3.7 ꢀM (19h) and 7.5 ꢀM (19j).
50
The introduction of substitution at the ortho position of the distal phenyl ring resulted in a
significant drop in MMP inhibitory activity (compound 19i in the low micromolar range) or in its
ACS Paragon Plus Environment
12

Page 13 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
total abolishment (compound 19k). Similarly, the incorporation of a chlorine at the 3ꢀposition of
the first phenyl ring in the derivative of 16 (compound 19p) abolished its MMP inhibitory
activity. These substitutions induced nonꢀplanarity in the R substructure; thus, this loss of
1
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
primary activity can be explained by the spatial conformations of these fragments placed in the
S1` pocket and the narrowness of S1’ (Figure 2b). This negative impact on the primary activity
was confirmed in their corresponding phenotypic functional assay data. The introduction of a
methoxy group, at the meta position, in the derivative 19d led to compound 19l. This new
molecule had a decreased potency against the two MMP isoforms (ꢀ3 and ꢀ10) by more than 1
log unit. In addition, the functional response of 19l , with an EC value of 900 nM, confirmed
5
0
the detrimental impact of meta substituents on the phenotypic response (as illustrated above for
19h and 19j). However, as we previously described for compound 17, the replacement of the
terminal phenyl ring with a benzyl ring resulted in minor differences in inhibitory activity. These
results perfectly fit the structural description of the S1’ pocket (Figure 2b) – narrow and long;
thus, the methylene bridge is accommodated within S1’, for example, 19d compared with 19n
(21 nM vs. 39 nM, respectively, against MMPꢀ3; and, 21 nM vs. 68 nM, respectively, against
MMPꢀ10). Finally, replacement of the aryl moiety with nonꢀaromatic and polar large fragments
(
19o) appears to be detrimental to biochemical activity, leading to IC values greater than 400
50
nM, most likely due to the hydrophobic nature of the S1’ pocket.
Table 2. Exploration of phenoxyphenyl and benzyloxyphenyl sulfone moieties in αꢀ
spiropiperidine hydroxamates.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
O
O
HO
S
R2
O
N
H
R1
N
H
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Thromboelastographya
EC50 (nM)
Cpd
R1
R2 MMP3 IC50
nM)
MMP10
IC (nM)
(
5
0
1
9a
9b
9c
9d
H
H
H
6
21
4
7
≤400
≤400
≤400
1
42
3
OCF
2
1
O
1
H
H
21
19
21
13
≤400
≤400
N
H
N
O
1
9e
N
1
9f
H
H
7270
16
7810
62
7500
≤400
3700
OMe
1
9g
N
1
9h
H
H
7
14
OMe
1
9i
916
739
3700
7500
MeO
1
9j
H
H
65
89
N
H
O
1
9k
H
N
38600
20700
>15000
O
O
1
9l
N
H
H
H
314
46
713
54
900
OMe
19m
≤400
ACS Paragon Plus Environment
14

Page 15 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
19n
19o
H
39
68
≤400
HN
N
H
476
435
900
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
O
OCF3
Cl
9600
8700
>15000
1
9p
a
Human whole blood
Exploration of N-substituted aryloxyphenyl
α-spiropiperidine hydroxamates. The next stage
involved an exploration around the basic nitrogen of the spiropiperidine in our chemical series
5’. As anticipated and according to the proposed binding mode (Figure 2b), the impact on
primary activity due to growing the vector (R’) from the nitrogen facing the solventꢀaccessible
area was minimal. Thus, the corresponding secondary amine of the spiro ring in compound 13
was converted into a tertiary amine (e.g., R = Me, 20a), a fluorinated alkylꢀcontaining tertiary
1
amine (20b) and an amide (20c), and their corresponding IC values against MMPs, as well as
5
0
their antifibrinolytic activities, were essentially unchanged (Table 3). These compounds were
initially designed to explore a variety of pKa values and to evaluate their impact on ADME
properties, mainly focused on solubility and hERG binding. Therefore, considering that the
primary and functional activities are maintained, these molecules, covering a diversity in
physicochemical properties, may provide alternative profiles from an ADME perspective. In fact,
based on the estimated pKa’s (Table 3), compounds 13, 20a and 20b will be ionized to different
extents at physiological pH, but molecule 20c, bearing a tertiary amide considered as "nonꢀbasic"
and "nonꢀacidic" under physiologic conditions, will not be protonated. However, because the
synthesized molecules bearing a secondary amine did not show any hERGꢀrelated issues and
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
optimal solubility was achieved as described below, no further exploration around this position
has been performed.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Exploration of Nꢀsubstituted αꢀspiropiperidine hydroxamates
a
Cpd
13
R1
MMP3 IC50
MMP10
Thromboelastography
EC50 (nM)
b
Estimated pKa
(
nM)
IC (nM)
50
-H
9
12
≤400
7.35
7.33
20a
10
17
≤
400
CF
3
20b
6
16
≤
400
6.66
O
20c
5
9
400
a
b
Human whole blood; In silico estimation performed using the algorithm implemented in
Pipeline Pilot v 8.5.
Exploration of acyl linker aryl moieties. Therapeutic scenarios where the stoppage of bleeding
is required (surgery, trauma and first aid, etc.) involve an acute treatment; therefore, we also
designed molecules that have very short halfꢀlives, i.e., immediate elimination after achieving
their goal, allowing for subsequent wound healing. Thus, taking into account the structural
description of the S1’ pocket (Figure 2b), we explored esterꢀlinked analogs (21), and those
corresponding to compounds 13 and 19d (21a and 21b) were synthesized (Table 4). Compound
21a exhibited potency and functional activity similar to compound 13.
ACS Paragon Plus Environment
16

Page 17 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Some of synthesized compounds (13, 16, 19d, 19p and 20a) were screened against other MMP
isoforms (Supplementary Information). In general, no selectivity was observed against any
isoform at 10 ꢀM, with inhibitory activities greater than 90% for most of the screened isoforms
(MMPꢀ1, ꢀ2, ꢀ8, ꢀ9, ꢀ12, ꢀ13, ꢀ14, and ꢀ20), except for MMPꢀ1 and MMPꢀ7 vs compound 19p,
whose inhibition were 9% and 26%. Nonꢀplanarity induced by chlorine atom at meta position led
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
to this selective MMP inhibitor, 19p, inactive against MMPꢀ1, ꢀ3, ꢀ7 and ꢀ10; its R substructure
1
did not accommodate within S1’ pocket. Unfortunately, this molecule did not interact with the
MMPs of interest (MMPꢀ3 and MMꢀ10) and, consequently, its functional antifibrinolytic
response was really poor (EC > 15ꢀM).
5
0
Table 4. Exploration of esterꢀlinked analogs
21
a
MMP3
MMP10
Thromboelastography
Cpd
21a
R1
IC (nM)
IC (nM)
EC50 (nM)
50
50
OMe
O
15
18
≤400
21b
132
103
7500
N
H
a
Human whole blood
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In vitro ADME data (Table 5) guided us during the lead optimization process to prioritize which
compounds were to progress to the in vivo assay. Our compounds showed low permeability
(PAMPA Pe < 1 nm/s), which is a desirable feature; in fact, to be efficacious as antifibrinolytics,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the compounds must be maintained in the bloodstream, minimizing tissue distribution and
avoiding undesirable side effects. Additionally, all of these compounds showed low inhibition of
the five major cytochrome P450 isoforms (1A2, 2C9, 2C19, 2D6, 3A4); in fact, the percentage
inhibition values, at 10 ꢀM, were less than 50% for all tested compounds. The metabolic stability
of this set of compounds was evaluated in human, mouse and rat liver microsomes after 20 min
of incubation at a concentration of 1 ꢀM. All compounds, except for compound 21a with the
metabolically labile ester functionality, showed good microsomal stability in human and mouse
species (>70% remaining). Replacement of the –OCF group with the methylaminocarbonyl
3
group tended to increase the stability, especially in the case of the pair 17 and 19n (68.9% vs.
97.6%; 71.1% vs. 90.2%; and, 68.7% vs. 91.4%, for human, mouse and rat, respectively). In
addition, there was minimal interaction with the hERG channel according to the binding assay;
in fact, all assayed molecules showed hERG IC values >30 ꢀM, and some of them, compounds
5
0
1
6, 19d, 19e and 19n, even had IC values ≥100 ꢀM.
5
0
a
Table 5. ADME profile.
Cpd PAMPA
Pe
1A2b
(%)
2C9b
(%)
2C19 2D6
3A4
(%)
HLM
(%)
MLM
(%)
RLM
(%)
hERG
IC (ꢀM)
b
b
b
c
c
c
(%)
(%)
5
0
(nm/s)
13
16
17
19d
19e
0.009
0.007
N/D
0.010
N/D
6.1
0
19.4
8.4
30.5
0
18.9
40.2
11.1
6.3
0
0
0
0
87.7
92.1
68.9
91.1
90.2
97.6
1.8
83.8
87.2
71.1
97.5
90.6
90.2
0
52.3
37
100
4.9
13.6
0
0.4
25.6
7.4
8.8
0
0
12.3
0
10.6
19.9
5.7
28.8
37.2
5.1
95.6
68.7
99.8
92.6
91.4
3.3
50.3
>100
>100
>100
63.4
8.3
11.6
0
19n
N/D
0.002
2
1a
3.8
a
b
c
N/D: not determined; % inhibition at 10 ꢀM; % compound remaining after a 20ꢀmin incubation in human, mouse
or rat liver microsomes (HLM, MLM and RLM, respectively).
ACS Paragon Plus Environment
18

Page 19 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vivo efficacy in murine bleeding models. The most active compounds in the functional
antifibrinolytic assay (EC50 <1 ꢀM) with the appropriate ADME profiles, the molecules
described in Table 5, were advanced into a mouse model known as the hyperfibrinolytic tail
bleeding test. In this in vivo efficacy model, after administration of 0.5 mg/kg tissue
plasminogen activator (tPA) alone, bleeding time (defined as the interval between the initial
transection and the visual cessation of hemorrhaging) was significantly prolonged (29.28 ± 0.4
min), with 189 ± 26 ꢁL of blood loss observed (i.e., 10% of the total volume). It could be
demonstrated that compounds in Table 6, infused at 1 mg/kg/h, significantly reduced bleeding
time relative to vehicleꢀtreated controls and 2 (300 mg/kg/h), the current SoC. To assess toxicity
and the absence of thrombus formation, an anatomopathological analysis focused on the brain,
liver, lungs, and kidneys was carried out following treatment at 10 mg/kg/h, a 10ꢀfold higher
dose than the efficacious dose, to consider a wide safety window. Table 6 reports whether signs
of toxicity or thrombosis were observed (0) or not (1). Compounds 16 and 19d were among the
most efficient analogs at shortening bleeding time, with mean times of 0.99 ± 0.35 min and
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
0.85 ±0.45 min, respectively. However, mice treated with compound 16 presented signs of
thrombosis in the lungs (Figure 3); therefore, this safety issue prohibited any further
investigation with compound 16. However, as expected, compound 21a exhibited poor metabolic
stability (<5% for all three species, Table 5) and led to poor in vivo efficacy, which precluded its
further development. However, its reduction of bleeding time was still comparable to the current
SoC (2) at a 300ꢀfold lower dose. Unfortunately, 21a also led to thrombus formation. Further
studies are required to decipher the cause of thrombus formation; for example, comparison of
exhaustive offꢀtarget selectivity profilings, biodistribution, pharmacokinetic parameters, etc.
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Table 6. In vivo efficacy in a mouse tPAꢀinduced tail bleeding model
Compound Bleeding time
SEM
P Value vs.
saline-treated
control
P Value vs. 2- Anatomopathological
treated control Analysis, at 10 mg/kg
(
min) at 1 mg/kg
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Saline
29.28
14.2
12.43
16.37
0.99
2.09
0.85
8.21
18.45
2.67
±0.4
±3.1
±4.3
1
1
1
N/D
0
0
1
1
N/D
0
b
b
2c
<0.0001
Aprotinin
4
0.001
0.0090
<0.0001
<0.0001
<0.0001
<0.0001
0.01
0.393
0.91
<0.0001
<0.0001
<0.0001
0.04
d
d
±3.88
±0.12
±0.70
±0.15
±3.66
±3.72
±1.27
±4.53
16
17
19d
19e
19m
19n
21a
0.44
0.001
0.44
<0.0001
0.01
13.43
0
a
b
n ≥10 mice/group; compound 2 was tested at 300 mg/kg, and its corresponding anatomopathological analysis was
also performed at this dose; aprotinin was tested at 10 mg/kg; N/D: not determined. Data were evaluated using
the Mann–Whitney U test, where bleeding time is reported as the mean value and SEM (standard error of the mean).
18 c
18 d
Figure 3. Anatomopathological images of mice treated with compounds 16 and 19d.
A)
B)
5
0 µm
33 µm
ACS Paragon Plus Environment
20

Page 21 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
C)
D)
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
00 µm
50 µm
Figures A) and B): Anatomopathological images of mice treated with 16 (10 mg/kg). A) Hematoxylin and eosin
staining of lung tissue vessels (magnification 400x) showing a microthrombus (arrows); and, B) kidney with slight
hypercellularity and small spheres of protein in the Bowman´s capsule (arrows) (magnification 600x). Figures C)
and D): Anatomopathological images of mice treated with 19d (10 mg/kg). Hematoxylin and eosin staining of lung
tissue vessels (magnification 100x) and kidney glomeruli and tubules (magnification 400x) with normal appearance
and no thrombus formation.
Compound 19d was a promising therapeutic agent in terms of in vitro ADME profile as well as
in vivo efficacy and safety; therefore, both enantiomers, Rꢀ (5) and Sꢀ (22), were resolved by
supercritical fluid chromatography (SFC). The absolute stereochemistry of these compounds was
confirmed by Xꢀray diffraction analysis of derivatives 23a and 23b (Figure 4). The synthesis of
18
2
3a has been previously described by our group, and we followed the same protocol for
compound 23b (details in the supplementary information).
Figure 4. Structures of compounds 23a and 23b obtained by Xꢀray crystallographic analysis.
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
A) 23a, as previously reported; b) 23b together with its corresponding colorꢀcoded stick representation. Details for
the Xꢀray crystallographic data of compound 23b and its corresponding coordinates are reported in the
Supplementary Information.
Compound 5 had IC s of 15 nM and 12 nM against MMPꢀ3 and MMPꢀ10, respectively. The
50
corresponding IC s for compound 22 were 18 nM (MMPꢀ3) and 17 nM (MMPꢀ10). For these
5
0
advanced compounds, the exact EC was determined (by testing a broad range of concentrations
5
0
below 400 nM) as 0.7 nM (5) and 17.6 nM (22). As shown in Table 7, both enantiomers
exhibited preliminary ADME properties comparable to the racemic compound 19d; except for
P450 inhibition. In this case, we performed a Cacoꢀ2 permeability assay instead of a PAMPA
assay, where only passive diffusion is assessed; although low permeability was confirmed for
both molecules, the data suggests that 5 likely acts as an efflux pump substrate (efflux ratio
>
2.5). Thus, this fact might favor the exposure of 5 to the bloodstream and enhance its efficacy
compared to that of 22. The solubility of these two enantiomers, 5 and 22, was very similar at
ACS Paragon Plus Environment
22

Page 23 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
03.25 ꢀg/mL and 97.96 ꢀg/mL, respectively, at pH=7.4; for both compounds, a patch clamp
electrophysiology assay gave IC values >30 ꢀM. Therefore, regarding cardiovascular safety,
5
0
these data suggest that there is a good correlation with the hERG binding values (Table 5), as
expected, and the assayed compounds do not raise any alerts from a cardiovascular safety
perspective. Notably, as reported below in Table 8, the in vivo efficacy of the R enantiomer 5
outperformed that of the S enantiomer 22 (mean bleeding time of 1.82 ± 0.42 min vs. 10.36 ±
4.31 min, respectively) as well as that of the current standards of care, 2 and aprotinin. Noteworthy,
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
18
pharmacokinetic study revealed a short halfꢀlife (1.4 h) for 5; thus, leading to the desirable profile
in acute treatment of bleeding: stable enough to achieve the goal and fast elimination.
Table 7. ADME profile for the enantiomers of 19d: 5 (R) and 22 (S)
Cpd Caco-2
Caco-2
Efflux ratio
1A2a
(%)
2C9
(%)
2C19 2D6
3A4
(%)
HLM MLM RLM hERG
b b b
a
a
a
a
Papp
(%)
(%)
(%)
(%)
(%)
IC50
(
nm/s)
0.0016
0.0059
(ꢀM)
1
8
5
2
%
2.60
1.12
28.7
28.7
9.4
8.2
28.6
22.2
43.7
45.6
17.6
15.9
93.5
97.0
98.0
89.6
98.1
97.4
>100
>100
2
a
b
inhibition at 10 ꢀM; % compound remaining after a 20ꢀmin incubation in human, mouse or rat liver microsomes
(
HLM, MLM and RLM, respectively).
a
Table 8. In-vivo efficacy in a mouse tPAꢀinduced tail bleeding model
Compound Bleeding time
SEM
P Value vs
saline-treated
control
P Value vs 2- P Value vs Aprotinin-
treated control treated control
(
min) at 1 mg/kg
18
5
1.82
10.36
±0.42
±4.31
<0.0001
0.001
0.001
0.12
0.079
0.529
22
a
n ≥10 mice/group. Data were evaluated using the Mann–Whitney U test, where bleeding time is reported as the
mean value and SEM (standard error of the mean).
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Chemistry. As a general strategy, all of the analogs in this series were first prepared as racemic
mixtures. For selected compounds, the racemic mixture was resolved by chiral HPLC to give
both enantiomers. The preparation of αꢀspiropiperidine hydroxamates with different linkers is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
summarized in the following schemes. Racemic αꢀspiropiperidine hydroxamate compounds 11ꢀ
13, 16, 19aꢀc and 19fꢀi were synthesized as depicted in Scheme 1. The methodology is closely
21
related to that previously reported.
Commercially available 8ꢀtertꢀbutylꢀ2ꢀmethylꢀ8ꢀ
azaspiro[4.5]decaneꢀ2,8ꢀdicarboxylate 24 was deprotonated with LDA and sulfonylated with 4ꢀ
fluorobenzenesulfonyl fluoride. Then, saponification of methyl ester 25 gave carboxylic acid 26,
which was converted into the THPꢀprotected hydroxamic acid 27 by reaction with Oꢀ(tetrahydroꢀ
2Hꢀpyranꢀ2ꢀyl)hydroxylamine using EDC/HOBt as the coupling system. Subsequent acidic
deprotection of the hydroxamate and BOC moieties of compound 27 afforded the 4ꢀfluorophenyl
intermediate 11. Alternatively, reaction of 27 with aqueous NaOMe gave the protected precursor
of methoxy compound 12, obtained after removal of the BOC and THP groups. Nucleophilic
aromatic substitution of the aryl fluoride 27 with the appropriate aryl alcohols in the presence of
cesium carbonate afforded the corresponding aryl ethers 29aꢀi. Acidic removal of both BOC and
THP groups with HCl afforded the target compounds 13, 16, 19aꢀc, and 19fꢀi.
a
Scheme 1. Synthesis of compounds 11ꢀ13, 16, 19aꢀc and 19fꢀi
ACS Paragon Plus Environment
24

Page 25 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a)
b)
c)
24
26
25
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
d)
e)
2
7
e)
12
28
f)
1
1
13, X = C, R = H, R = H, R = OCH
1
2
3
3
16, X = C, R = H, R = H, R = OCF
1
2
3
3
1
1
1
1
1
1
1
9a, X = C, R = H, R = H, R = CH
1
2
3
3
9b, X = C, R = H, R = H, R = C(CH )
3 3
1
2
3
e)
9c, X = C, R = H, R = H, R = OCHF
2
1
2
3
9f, X = C, R = H, R = H, R = CH N(CH )
3 2
1
2
3
2
9g, X = N, R = H, R = H, R = OCH
3
1
2
3
9h, X = C, R = H, R = OCH , R = H
1
2
3
3
29
9i, R = OCH , R = H, R = H
1
3
2
3
a
Reagents and conditions: (a) LDA, THF, ꢀ78 ºC, 1 h, then 4ꢀfluorobenzenesulfonyl fluoride, ꢀ78 ºC, 1 h to RT
overnight; (b) LiOH·H O, THF/MeOH/H O (3:3:2), RT, 4 h; (c) EDC·HCl, HOBt, THPONH , NMM, DMF, RT,
2
2
2
overnight; (d) NaOMe, overnight; (e) HCl/dioxane (2N), RT, 1 h; (f) corresponding aryl alcohol, Cs CO , DMF, 80ꢀ
2
3
9
0 ºC, MW or conventional heating.
Aryl fluoride 27 was also coupled with other alcohols in the presence of NaH to yield
intermediates 30aꢀb and 31, which, after acidic deprotection, afforded compounds 17, 19m and
1
9o (Scheme 2).
a
Scheme 2. Synthesis of compounds 17, 19m and 19o
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
a)
(b)
27
(
a)
3
0a, R = OCF3
0b, R = H
17, R = OCF3
19m, R = H
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
(b)
31
19o
a
Reagents and conditions: (a) corresponding alcohol, NaH, RT, 30 min, then 27, 80ꢀ90 ºC, overnight; (b)
HCl/dioxane (2N), RT, 1 h.
Similar chemistry could be used to arrive at amides 19d, 19e, 19j and 19l through the
corresponding esterꢀsubstituted aryl ether precursors 32aꢀc (Scheme 3), which could be
converted into carboxylic acids 33aꢀc with LiOH. These intermediates were coupled with the
appropriate amines. Finally, the protected amide analogs 34aꢀd were deprotected to obtain the
desired hydroxamic acids 19d, 19e, 19j and 19l.
a
Scheme 3. Synthesis of compounds 19d, 19e, 19j and 19l
ACS Paragon Plus Environment
26

Page 27 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a)
b)
27
32a, R = H, R = H, R = COOCH
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
33a, R = H, R = H, R = COOH
1 2 3
1
2
3
3
32b, R = H, R = COOCH , R = H
1
2
3
3
33b, R = H, R = COOH, R = H
1 2 3
32c, R = H, R = OCH , R = COOCH
1
2
3
3
3
33c, R = H, R = OCH , R = COOH
1 2 3 3
d)
1
c) or 2c)
3
4a, R = H, R = H, R = CONHCH
19d, R = H, R = H, R = CONHCH
1 2 3 3
1
2
3
3
3
4b, R = H, R = H, R =
19e, R = H, R = H, R =
1 2 3
1
2
3
34c, R = H, R = CONHCH , R = H
19j, R = H, R = CONHCH , R = H
1 2 3 3
1
2
3
3
34d, R = H, R = OCH , R = CONHCH
19l, R = H, R = OCH , R = CONHCH
1 2 3 3 3
1
2
3
3
3
a
Reagents and conditions: (a) corresponding alcohol, Cs CO , DMF, 80ꢀ90 ºC, overnight or DMA, 120 ºC, MW, 2 h;
2
3
(
b) LiOH·H O, THF/H O (3:1 or 3:2) or THF/MeOH/H O (3:3:2), RT, 4 h; (1c) EDC·HCl, HOBt, CH NH·HCl or
2 2 2 3
pyrrolidine, NMM, DMF, RT, overnight; (2c) Et N, HATU, CH NH·HCl, CH CN, RT, overnight; (d) HCl/dioxane
3
3
3
(2N), RT, 1ꢀ2 h.
Preparation of compounds 14, 15 and 19p is shown in Scheme 4. In this case, compound 24 was
coupled with the appropriate benzenesulfonyl fluoride in each case to afford intermediates 35aꢀc.
Then, reaction with an aryl alcohol led to the corresponding esters, which were hydrolyzed to
carboxylic acids 37aꢀc. These products were converted into hydroxamic acids 14, 15 and 19p, as
previously described.
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 4. Synthesis of compounds 14, 15 and 19p
a)
1b) or 2b)
c)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24
3
3
3
5a, R = H, R = Br, R = H
36a, R1 = H, R = -O-(4-OCH -C H ), R = H
1
2
3
2
3
6
4
3
5b, R = F, R = H, R3 = H
36b, R1 = -O-(4-OCH -C H ), R = H, R = H
3 6 4 2 3
1
2
5c, R1 = H, R2 = Cl, R3 = F
36c, R = H, R = Cl, R = -O-(4-CF -C H )
1 2 3 3 6 4
d)
e)
37a, R = H, R2 = -O-(4-OCH -C H ), R = H
38a, R1 = H, R = -O-(4-OCH -C H ), R = H
14, R = H, R = -O-(4-OCH -C H ), R = H
1 2 3 6 4 3
1
3
6
4
3
2
3
6
4
3
3
7b, R = -O-(4-OCH -C H ), R = H, R = H
38b, R1 = -O-(4-OCH -C H ), R = H, R3 = H
15, R = -O-(4-OCH -C H ), R = H, R = H
1 3 6 4 2 3
1
3
6
4
2
3
3
6
4
2
3
7c, R1 = H, R = Cl, R = -O-(4-CF -C H )
38c, R = H, R = Cl, R = -O-(4-CF -C H )
19p, R1 = H, R = Cl, R = -O-(4-CF -C H )
2 3 3 6 4
2
3
3
6
4
1
2
3
3
6
4
a
Reagents and conditions: (a) LDA, THF, ꢀ78 ºC, 1 h, then corresponding sulfonyl fluoride, ꢀ78 ºC, 1 h and RT
overnight; (1b) corresponding alcohol, Cs CO , DMF, 80ꢀ100 ºC, overnight; (2b) corresponding alcohol, Cs CO ,
2
3
2
3
CuI, N,N´ꢀdimethylglycine, dioxane, reflux, overnight; (c) LiOH·H O, THF/MeOH/H O (3:3:2), RT, 4 h; (d)
2
2
EDC·HCl, HOBt, THPONH , NMM, DMF, RT, overnight; (e) HCl/dioxane (2N), RT, 2ꢀ3 h.
2
In the case of methyl amides 19k and 19n, phenol 40 was first prepared from intermediate 27
(
Scheme 5). Subsequent reaction with the corresponding benzoate afforded 41aꢀb. Ester
hydrolysis gave the corresponding carboxylic acids 42aꢀb, which were then converted to the
corresponding amides 43aꢀb. Final deprotection under acidic conditions led to the desired
products 19k and 19n.
ACS Paragon Plus Environment
28

Page 29 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Scheme 5. Synthesis of compounds 19k and 19n
O
O
O
O
O
O
N
THPO
S
THPO
S
1c) or 2c)
b)
a)
N
H
N
H
2
7
O
O H
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
N
Boc
3
9
40
Boc
O
O
O
N
O
O
O
N
O
O
O
N
THPO
S
THPO
S
THPO
S
N
H
d)
N
H
e)
N
H
R
R
R
O
O
O
Boc
Boc
Boc
41a, R = 2-(COOCH CH )C H
42a, R = 2-(COOH)-C H4
43a, R = 2-(CONHCH )C H
3 6 4
2
3
6
4
6
41b, R = -CH -4-(COOCH )C H
42b, R = -CH -4-(COOH)-C H
43b, R = -CH -4-(CONHCH )C H
2
3
6
4
2
6
4
2
3
6
4
f)
O
O
O
HO
S
N
H
9k, R = 2-(CONHCH )C H
3 6 4
R
1
1
O
9n, R = -CH -4-(CONHCH )C H
2
3
6
4
N
H
a
Reagents and conditions: (a) NaH, benzyl alcohol, THF, RT, 30 min, then 27, 80 ºC, overnight; (b) Pd/C, H (1
2
atm), MeOH, RT, 1 h; (1c) ethyl 2ꢀfluorobenzoate, Cs CO , DMF, 80 ºC, overnight; (2c) methyl 4ꢀ
2
3
(
bromomethyl)benzoate, K CO , DMF, 0 ºC to RT, overnight; (d) LiOH·H O, THF/H O (3:1), RT, overnight; (e)
2 3 2 2
EDC·HCl, HOBt, CH NH·HCl, NMM, DMF, RT, overnight; (f) HCl/dioxane (2N), RT, 1 h.
3
The synthetic route for the preparation of compound 18 with an amino linker is outlined in
Scheme 6. Starting from intermediate 27, [4ꢀ(trifluoromethoxy)phenyl]methanamine was used
for the nucleophilic aromatic substitution to obtain precursor 44, which was finally deprotected
in acidic media.
ACS Paragon Plus Environment
29