Small molecule and macrocyclic pyrazole derived inhibitors of myeloperoxidase (MPO)

Article abstract of DOI:10.1016/j.bmcl.2021.128010

Myeloperoxidase (MPO), a critical enzyme in antimicrobial host-defense, has been implicated in chronic inflammatory diseases such as coronary artery disease. The design and evaluation of MPO inhibitors for the treatment of cardiovascular disease are reported herein. Starting with the MPO and triazolopyridine 3 crystal structure, novel inhibitors were designed incorporating a substituted pyrazole, which allowed for substituents to interact with hydrophobic and hydrophilic patches in the active site. SAR exploration of the substituted pyrazoles led to piperidine 17, which inhibited HOCl production from activated neutrophils with an IC₅₀ value of 2.4 μM and had selectivity against thyroid peroxidase (TPO). Optimization of alkylation chemistry on the pyrazole nitrogen facilitated the preparation of many analogs, including macrocycles designed to bridge two hydrophobic regions of the active site. Multiple macrocyclization strategies were pursued to prepare analogs that optimally bound to the active site, leading to potent macrocyclic MPO inhibitors with TPO selectivity, such as compound 30.

Full text of DOI:10.1016/j.bmcl.2021.128010

Bioorg. Med. Chem. Lett. 42 (2021) 128010
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Small molecule and macrocyclic pyrazole derived inhibitors of
myeloperoxidase (MPO)
Carol H. Hu*, Meriah W. Neissel Valente*, O. Scott Halpern, Sutjano Jusuf, Javed A. Khan,
Gregory A. Locke, Gerald J. Duke, Xiaoqin Liu, Franck J. Duclos, Ruth R. Wexler, Ellen K. Kick,
Joanne M. Smallheer
Research and Development, Bristol-Myers Squibb Company, P. O. Box 5400, Princeton, NJ 08543, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
Myeloperoxidase (MPO), a critical enzyme in antimicrobial host-defense, has been implicated in chronic in-
flammatory diseases such as coronary artery disease. The design and evaluation of MPO inhibitors for the
treatment of cardiovascular disease are reported herein. Starting with the MPO and triazolopyridine 3 crystal
structure, novel inhibitors were designed incorporating a substituted pyrazole, which allowed for substituents to
interact with hydrophobic and hydrophilic patches in the active site. SAR exploration of the substituted pyrazoles
Atherosclerosis
Myeloperoxidase
Triazolopyridine
Macrocycle
MPO inhibitor
MPO
led to piperidine 17, which inhibited HOCl production from activated neutrophils with an IC50 value of 2.4 μM
and had selectivity against thyroid peroxidase (TPO). Optimization of alkylation chemistry on the pyrazole ni-
trogen facilitated the preparation of many analogs, including macrocycles designed to bridge two hydrophobic
regions of the active site. Multiple macrocyclization strategies were pursued to prepare analogs that optimally
bound to the active site, leading to potent macrocyclic MPO inhibitors with TPO selectivity, such as compound
30.
Myeloperoxidase (MPO) is a human heme-containing peroxidase.
Other members of this family include thyroid peroxidase (TPO), eosin-
ophil peroxidase (EPX), lactoperoxidase (LPO) and vascular peroxidase
(VPO). MPO is the major protein component of the primary granules in
neutrophils and is present in monocytes and macrophages. Neutrophil
MPO combats infection by producing strong oxidants to destroy mi-
croorganisms.1 A unique feature of this enzyme is its ability to generate
HOCl, a powerful oxidant. Other downstream products of the MPO ac-
tivity include a variety of reactive molecules such as HOBr and HOSCN,
which can participate in modification of amino acids, proteins and
lipids. TPO catalyzes both the iodination of tyrosyl residues in thyro-
globulin and the coupling of these residues to form key thyroid hor-
mones, so it is critical that MPO inhibitors display excellent selectivity
for MPO versus TPO.2 Like MPO, EPX oxidizes halides to hypohalous
acids such as HOBr, and has been implicated in the release of reactive
oxidants and other radical species in the development of inflammatory
conditions such as asthma.3
between plasma MPO levels and risk of cardiovascular disease in
humans.5–8 MPO is found in human atherosclerotic lesions and con-
tributes to atherosclerosis through oxidative modifications to both low
density and high density lipoproteins (LDL and HDL).9 Furthermore, a
potential role for MPO as a downstream modulator of atrial fibrosis in
heart failure has been proposed. In a study reported by Rudolph and
coworkers,10 MPO-deficient mice pretreated with Ang-II (an effector of
atrial remodeling and fibrosis) showed lower atrial tissue levels of the
MPO biomarker 3-chlorotyrosine, reduced matrix metalloprotease ac-
tivity and reduced atrial fibrosis compared to wild type mice. Based on
studies such as these, inhibition of MPO in plasma and the artery wall
may have an overall benefit in human cardiovascular disease.
The catalytic pathway by which MPO catalyzes the 2-electron
oxidation of Cl- to form HOCl is shown in Fig 1.11 When MPO reacts
with H2O2, a key redox intermediate is formed by oxidation of the native
state Fe(III) of the heme to an iron IV oxo species, Fe(=O)(IV), known as
MPO(I) that is the primary catalytic complex. MPO(I) reacts rapidly with
chloride ion to form HOCl, returning MPO back to its native state. An
alternate peroxidation pathway also exists in which the MPO(I) complex
is converted back to native MPO via two sequential one electron
Numerous studies have demonstrated that high plasma levels of MPO
and MPO oxidation products are associated with inflammatory diseases,
including atherosclerosis.4 In addition, correlations have been reported
* Corresponding authors.
E-mail addresses: carol.hu@bms.com (C.H. Hu), meriah.valente@bms.com (M.W. Neissel Valente).
https://doi.org/10.1016/j.bmcl.2021.128010
Received 7 January 2021; Received in revised form 22 March 2021; Accepted 28 March 2021
Available online 1 April 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
IC50 of 190 nM.15 Compound 1 also had good selectivity over TPO, but
was a known irreversible inhibitor of methylguanine methyltransferase
(MGMT),18 a key DNA repair enzyme, and had poor chemical stability at
low pH. Both of these issues were resolved with triazolopyridine ana-
logs, 216 and 3,17 with improved potency compared to 1. However, both
2 and 3 suffer from time-dependent cytochrome P450 3A4 (TD-CYP3A4)
inhibition, as well as inadequate selectivity over TPO. Furthermore, it
was found that only small substituents were tolerated on the phenyl ring
of 3, which limited exploration of larger lipophilic groups that would
project into a hydrophobic pocket observed above the heme in the MPO
crystal structures. Therefore, 5-membered heterocyclic replacements
were explored for the phenyl ring of 3, resulting in the identification of a
potent pyrazole lead 4 (Fig. 2). The design and synthesis of linear and
macrocyclic pyrazole analogs with improved MPO potency and selec-
tivity for TPO are described herein.
An overlay of 4 bound in MPO with the MPO crystal structure of 3
(Fig. 3) shows a tight overlap of the triazolopyridine cores, but a marked
difference in the position of the pyrazole ring, which is rotated
compared to the phenyl ring of 3. This places one edge of the pyrazole
Fig. 1. MPO catalytic pathways.
reductions involving a second intermediate, Compound (II), which re-
tains the iron IV heme. Compound (II) reacts slowly with reducing
agents to convert this complex back to the native enzyme, consuming
H2O2, but not generating HOCl. Peroxidase substrates for MPO include
endogenous species such as tyrosine, melatonin and ascorbate. The wide
variety of compounds that MPO can oxidize via this pathway make it
challenging to find reversible MPO inhibitors that are not substrates.
Multiple irreversible inhibitors of MPO have been reported. A pub-
lished X-ray crystal structure of compound TX2 (Fig. 2) bound to MPO
shows that it forms a covalent bond to MPO through the exocyclic sulfur
to a methyl group on the heme moiety.12 Other irreversible inhibitors of
MPO that act by a similar mechanism, PF-0628299913 and AZD-324114
have also been reported (Fig. 2).
We focused on identifying reversible inhibitors that bind above the
heme of native MPO, have little reactivity with the enzyme and block the
interaction of MPO with H2O2, such as described for the benchmark
triazolopyrimidines and triazolopyridines 1, 2 and 3 shown in
Fig. 2.15–17 Compound 1, the initial lead in the program, was identified
from high throughput screening and had modest potency in MPO-
mediated chlorination assay (AminoPhenyl Fluorescein – APF) with an
Fig. 3. Overlay of crystal structures of compounds 317 (in yellow, PDB 6WYD)
and 4 (in green, PDB 7LAE) bound in the active site of MPO. Select amino acids
are shown in the active site.
Fig. 2. Literature MPO inhibitors and benchmark triazolopyrimidine and triazolopyridine inhibitors of MPO. Inhibition of MPO chlorination was measured using an
AminoPhenyl Fluorescein (APF) assay. IC50 values are reported as the mean of at least two individual determinations. See references 15–17 and Appendix A.
Supplementary data for assay details.
2

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
closer to the methylenes of the side chain of Arg239 and orients the
pyrazole ring to make an edge to face interaction with Phe366. It also
projects the fluorophenyl substituent into a small lipophilic pocket,
leaving the larger lipophilic pocket bordered by residues Phe407 and
Met411 unoccupied. Synthetically the pyrazole served as a handle to
introduce lipophilic groups with various substitutions to reach into the
lipophilic regions.
Table 1
SAR of pyrazole analogs.a
The assay results for 4 and further elaborated pyrazole analogs are
shown in Table 1. Compounds were evaluated for inhibition of MPO as
well as in assays to measure inhibition of EPX and TPO as previously
described.15 TD-CYP3A4 inhibition was monitored. The unsubstituted
phenyl analog 5 was similar in potency in all assays to 4, and pyridyl
analog 6 showed a modest 3-fold improvement in selectivity for MPO
relative to TPO. A scan of increasing size substituents around the phenyl
ring showed that groups larger than F or Me in the ortho or para positions
provided less potent analogs (data not shown). Although it appeared
from molecular modeling that larger groups in the meta position may
make contact with the lipophilic pocket, an improvement in MPO po-
tency wasn’t achieved, and increased TD-CYP3A4 inhibition was
observed as compounds became more lipophilic, as is exemplified with
m-phenyl analog 7. This indicated a different trajectory was needed to
effectively extend into this space. Molecular modeling suggested that
introduction of a methylene between the pyrazole and the phenyl sub-
stituent would extend the aryl substituent into an alternate region. The
benzyl compound 8 showed promising potency as a starting point for
further substitution. A wide range of different groups were tolerated in
combination with the methylene linker including 2-naphthyl 9, 3-
biphenyl 10, quinoline 11 and tetrahydroisoquinoline 12. However, in
general the introduction of lipophilic groups provided potent CYP3A4
inhibitors with no improvement in TPO selectivity. Notably, the intro-
duction of the basic nitrogen in 12 provided an encouraging off-target
profile with 120-fold selectivity for TPO and reduced CYP3A4 activity.
Modeling suggested that insertion of an additional one atom linker be-
tween the two phenyls of 10 could provide the desired trajectory to
project the second phenyl into the lipophilic pocket between Phe407
and Met411; however, the resulting compounds 13 and 14 were of
similar potency in the APF assay to 4 and both were very potent TD-
CYP3A4 inhibitors. Additional heterocyclic analogs were prepared
with basic nitrogens to investigate the impacts on TPO and TD-CYP3A4
activities. The isoindoline ether 15 resulted in a 13-fold decrease in the
TD-CYP3A4 IC50 compared to phenyl ether 14. Further reducing the
lipophilicity with the pyrrolidine ether 16 or piperidine 17, provided
similar inhibition of MPO in the APF assay, while greatly reducing TD-
CYP3A4 inhibition and improving TPO selectivity. The trend toward
increased selectivity for MPO over TPO with basic amines in this region
of the active site is consistent with recent reports.17 The analogs in
Table 1 consistently had potent EPX inhibition (IC50 3–30 nM) and may
be useful inhibiting EPX in lung, although this was outside the scope of
this work.
Cmpd.
R
MPO
APF
TPO
IC50
TD-CYP3A4 IC50
(nM) at 30 min
(ratio compared to
5 min)
IC50
(nM)
(nM)
4
5
53
700
640
110 (3.3)
55
210 (2.7)
6
7
54
85
2000
710
210 (3.2)
7.1 (2.4)
8
9
120
52
2300
1100
1100
1800
310 (3.2)
40 (2.4)
17 (2.8)
210 (4.1)
10
11
55
140
12
13
58
39
6700
1900
7000 (2.8)
37 (2.8)
14
15
16
44
26
46
1100
1700
11 (2.9)
140 (3.3)
14,000
13,000 (1.6)
Both analogs 12 and 17 showed little binding to human plasma
proteins with 50 and 80% free, which translated to good potency in a
human neutrophil assay15 with IC50 values of 5.0 and 2.4
μM, respec-
tively. Both 12 and 17 were metabolically stable with HLM t1/2 > 120
min. Good pharmacokinetic properties (PK) were observed with 17 in
mice after intravenous dosing (t1/2 = 1.2 h, Vss = 0.2 L/kg, CL = 2.6 mL/
min/kg, at a 1 mpk dose), but oral bioavailability was lacking, consistent
with poor permeability observed in a PAMPA assay.
17
100
21,000
3300 (4.3)
Development of macrocyclic drugs, such as cyclic peptides, has
demonstrated the possibility of improving the drug-like properties of
molecules.19 In some cases oral bioavailability has been achieved,20
despite a high molecular weight and polar backbone. Macrocyclic
compounds can be designed to allow for additional substituents to
maximize lipophilic or H-bonding opportunities as needed to optimize
potency and off-target liabilities. Overlay of the MPO crystal structures
of 14 with the amino indane 1817 reported by Shaw and coworkers
inspired the design of macrocyclic MPO inhibitors (Fig. 4). The
a
The AminoPhenyl Fluorescein (APF) assay measures inhibition of MPO
chlorination activity in the presence of chloride. IC50 values are reported as the
mean of at least two individual determinations. See references 15–17 and Ap-
pendix A. Supplementary data for assay details.
3

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
Fig. 4. Overlay of crystal structures of compounds 14 (in green, PDB 7LAG) and 1817 (in brown, PDB 7LAL) bound in the active site of MPO. Only key residues in the
active site are shown.
potency was observed (APF IC50 = 8 nM). An 18-membered macrocycle
24 was modestly less potent than its 17-membered ring analog 23.
Additional macrocycles with other variations in the linker were explored
(compounds 25–28, Table 2), with 20–30 nM IC50 values. This data
suggests that these compounds are probably not making close contact
with specific residues, but are just extending into the open center of the
targeted lipophilic pocket.
Following the SAR of the previously reported branched inhibitors,17
macrocycles 29 and 30 were designed to place a pendant phenyl or
benzyl group that extended into a hydrophobic pocket to interact with
Val410, as reported for linear compound 31 (Fig. 6). Macrocycle 29 was
designed to include the same phenyl-2,2,2-system of 31.17 As a 1:1
mixture of diastereomers, compound 29 had an APF IC50 of 7 nM. The
Fig. 5. Proposed generic macrocyclic scaffold. L represents a linker group that
benzyl substituted pentane analog 30 was predicted by modeling to have
the most favorable interactions with the hydrophobic pocket, so the
material was separated into enantiomer pairs. Pentane 30 had an APF
IC50 of 5 nM with good selectivity over TPO (TPO IC50 = 7.9 µM, 1580-
fold selective). The stereochemical (R)-configuration was determined by
X-ray crystallography (Fig. 6). The (S)-isomer of 30 was approximately
3-fold less potent (data not show). The added lipophilicity resulted in
potent TD-CYP3A4 inhibition for both molecules (29, IC50 = 1.6 µM at
30 min, a 2.4-fold shift from 5 min; 30, IC50 = 0.58 µM after 30 min, a
4.3-fold shift from 5 min) with no added permeability benefit over
previous compounds. An overlay of the MPO crystal structures with 30
and 3117 (Fig. 6) shows that the pendant phenyl group in both molecules
extends into the hydrophobic pocket formed by Pro220, Val410, and
Asp218, confirming the binding mode of 30 predicted by molecular
modeling.
can accommodate additional substitution.
macrocycles were designed to incorporate features of both 14 and 18 to
optimize binding and improve potency and physicochemical properties.
A generic macrocyclic scaffold was envisioned as shown in Fig. 5, where
the linker region L could be modified with lipophilic substitutions to
optimally fill the hydrophobic pocket formed by residues Phe366,
Phe407, Pro220 and Met411. Table 2 shows a variety of linkers that
were successfully incorporated to provide potent macrocyclic inhibitors.
Macrocyclic compounds in Table 2 were racemic unless otherwise
indicated.
Racemic 16-membered macrocycle 19 had an IC50 value of 36 nM in
the APF assay, similar to 14 (APF IC50 = 44 nM) and 18 (APF IC50 = 58
nM), thus confirming that potency could be achieved. In order to probe
the SAR of macrocycle size, the 17-membered macrocycle 20 was pre-
pared, adding an extra carbon in the hydrophobic pocket. Racemic 20
was modestly more potent than 19 with an APF IC50 of 14 nM. Chiral
separation of 20 provided enantiomer 21, a 4 nM inhibitor in the APF
assay (37-fold more potent than the less active enantiomer, not shown)
which maintained the EPX potency (EPX IC50 = 37 nM) of linear com-
pounds. The stereochemical assignment was not determined. Compound
21 showed improved selectivity over TPO (TPO IC50 = 28 µM, >7000-
fold selective) with minimal TD-CYP3A4 inhibition (IC50 = 11 µM after
30 min, a 1.8-fold shift from 5 min). Permeability as measured by
PAMPA or Caco-2 was not improved. The alternate 17-membered
macrocycle with a carbon inserted in the aryl ether linkage 22 was a
25 nM inhibitor in the APF assay. Interestingly, when the benzyl ether
linkage between the aryls was reversed to give 23, a 3-fold boost in
The optimized synthesis that enabled late-stage introduction of
pyrazole substituents (Table 1) is exemplified in Scheme 1 with the
synthesis of compound 17 (described in the Supporting Information).
The pyrazole 34 was prepared in a one-pot synthesis from the bis-trityl
7-bromotriazolopyridine 32.16 Conversion of 32 to the neo-
pentylboronate followed by a subsequent Suzuki coupling with the
bromide 33, and selective removal of the Boc protecting group afforded
34. Alkylation of the pyrazole nitrogen with the alcohol 35 under Mit-
sunobu conditions followed by global deprotection provided 17. Ana-
logs in Table 1 were prepared by similar procedures or by direct
arylation of the pyrazole nitrogen (data not reported).
In macrocycle synthesis, generally the most critical and challenging
synthetic step is the ring closure. Only a small number of reactions are
readily applicable to macrocyclization, limiting both the range of
4

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
Table 2
Examples of macrocyclic MPO inhibitors.a
Cmpd.
L
Ring Size
16
MPO APF IC50 (nM)
36b
Cmpd.
L
Ring Size
15
MPO APF IC50 (nM)
29
19
25
20
21c
22
23
17
17
17
17
14
4
26
27
28
29
17
15
16
15
31
26
21
7
25
8
24
18
17
30c
15
5
a
The AminoPhenyl Fluorescein (APF) assay measures inhibition of MPO chlorination activity in the presence of chloride, IC50 values are reported as the mean of at
least two individual determinations, unless otherwise indicated. See references 15–17 and Appendix A. Supplementary data for assay details. Compounds were racemic
unless otherwise indicated.
b
N = 1.
c
Single enantiomer.
potential targets and the flexibility of the synthesis.21,22 Various cycli-
zation strategies were therefore explored to prepare the desired
macrocyclic inhibitors. The first MPO macrocyclic inhibitor 19 was
successfully obtained via an intramolecular Ullmann coupling between
an aryl iodide and a phenol, followed by trityl deprotection as shown in
Scheme 2 (described in the Supporting Information). The macrocyclic
precursor was prepared in seven steps from 32, which was coupled to
prop-2-yn-1-ol, followed by conversion to the key vinyl iodide
Fig. 6. MPO crystal structure overlay of macrocyclic inhibitor 30 (in yellow, PDB 7LAN) and branched amine series lead 3117 (in purple, PDB 6WY7). Only key
residues in the active site are shown.
5

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
Scheme 1. Reagents and Conditions: (a) 5,5,5′,5′-tetramethyl-2,2′-bi(1,3,2-dioxaborinane), KOAc, Pd2(dba)3, tricyclohexylphosphonium tetrafluoroborate, dioxane,
100 ◦C; (b) 33, PdCl2(dppf), 2 M Na2CO3, THF, 70 ◦C; (c) 1 M NaOH, MeOH, 65 ◦C; 43% over 3 steps; (d) TMAD, nBu3P, 35, toluene, 88%; (e) TFA, DCM then
Et3SiH, 57%.
Scheme 2. Ullmann Coupling as Ring Closure Step. Reagents and Conditions: (a) PdCl2(PPh3)2, prop-2-yn-1-ol, CuI, Et3N, DMF, 85 ◦C, 70%; (b) (i) LAH, NaOMe,
THF, 0 ◦C (ii) (MeO)2CO, 0–78 ◦C, I2, then RT, 45%; (c) 37, PdCl2(dppf)-DCM adduct, K2CO3, THF/H2O, 80 ◦C, 92%; (d) MnO2, DCM, 92%; (e) 3-iodobenzylamine,
Et3N, EtOH, 60 ◦C, then NaBH4, 24%; (ii) TBAF, THF, 70%; (iii) NaBH4, NiCl2, MeOH/THF, rt, 100%; (f) CuI, Cs2CO3, 1,10-phenanthroline, toluene (5 mM), 145 ◦C,
29%; (g) TFA, DCM then Et3SiH, 26%.
Scheme 3. Mitsunobu Reaction as Ring Closure Step with Secondary Alcohols. Reagents and Conditions: (a) 41, TEA, EtOH, reflux; then NaBH4, 72%; (b) Boc2O, 1 M
NaOH, THF, 87%; (c) CMBP, toluene (5 mM), 100 ◦C, 85%; (d) TFA, DCM then Et3SiH; (e) PtO2, MeOH, H2, 50 psi (28% over 2 steps).
6

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
Fig. 7. Proposed macrocyclization via Suzuki-Miyaura condensation.
intermediate 36 in moderate yields. Suzuki-Miyaura coupling of this
iodide with pyrazole boronate 37 provided allylic alcohol 38. Oxidation
of the alcohol to the aldehyde 39 with MnO2, followed by reductive
amination with 3-iodobenzylamine, reduction of the double bond and
deprotection of the phenol afforded 40, setting the stage for the key
cyclization step. An Ullmann reaction run under dilute conditions (CuI,
Cs2CO3 and 1,10-phenanthroline in 5 mM toluene) provided the desired
16-member macrocycle in good yields, and deprotection of the tri-
azolopyridine gave 19. Ring expanded analog, 20, was similarly pre-
pared by replacing the 3-iodobenzyl amine with 3-iodophenethyl amine
in step e. The single enantiomer 21 was obtained from the racemate by
chiral SFC.
the 15-membered macrocycle 29 from the sterically demanding sec-
ondary alcohol intermediate 42 in good yield (Scheme 3).
In order to facilitate further SAR in this region of the macrocycle, a
universal macrocyclization strategy was desired. To this end, cyclization
via an intramolecular Suzuki-Miyaura coupling between a pyrazole
boronate and a vinyl iodide was explored (Fig. 7). Literature precedent
existed for macrocyclization via intramolecular Suzuki-Miyaura
condensation of a vinyl iodide and a vinyl boronate in the synthesis of
Rhizopodin,24 and we had demonstrated the intermolecular condensa-
tion between the vinyl iodide and a pyrazole boronate (Scheme 2). We
were therefore pleased to find that not only was this an effective method
for macrocyclization, it was by far the most versatile ring closure
method for a wide variety of target macrocycles, regardless of the
complexity of “L” as illustrated by compounds 25–28 and 30 in Table 2.
The synthesis of 30 using this strategy is shown in Scheme 4
(described in the Supporting Information). Amine 43 was alkylated with
the allylic bromide 44 in the presence of CsOH and KI to provide the
allylic amine, which was protected as a t-butyl carbamate to provide the
macrocycle precursor 45. Intramolecular Suzuki-Miyaura coupling of
the pyrazole boronate with the vinyl iodide gave 15-membered
While macrocycles 22–24 were prepared easily from a standard
Mitsunobu ring closure using DIAD/PPh3, this method was not suitable
for cyclizing secondary alcohols. In order to prepare the macrocycle 29
containing the 4-phenylbicyclo[2.2.2]octan-1-amine17 that was intro-
duced in intermediate 42 via a reductive amination of intermediate 39
and 41 (described in the Supporting Information), alternative Mitsu-
nobu conditions were explored. It was found that CMBP (cyano-
methylene tri-n-butylphosphorane)23 at elevated temperature provided
Scheme 4. Suzuki-Miyaura Coupling as Ring Closure Step. Reagents and Conditions: (a) CsOH, KI, DMF, rt; (b) Boc2O; 51% over 2 steps; (c) PdCl2(dppf)DCM adduct,
K2CO3, dioxane/H2O (1 mM), 85 ◦C, 45%; (d) TFA, DCM then Et3SiH; (e) Pt2O, MeOH, H2, 50 psi; 21% isolated yield of 30 over 2 steps after prep HPLC.
7

C.H. Hu et al.
Bioorganic & Medicinal Chemistry Letters 42 (2021) 128010
macrocycle 46 in 45% yield. Global deprotection, followed by reduction
of the double bond and separation of the resulting two diastereomers via
prep HPLC, gave macrocycle 30. The absolute configuration was
assigned based on the crystal structure with the molecule bound to MPO
(Fig. 6). The intramolecular Suzuki-Miyaura ring closure conditions
provided yields of 20–60% for diverse systems including smaller or more
constrained rings. It was also found that the Suzuki-Miyaura macro-
cyclization can be run at room temperature using Ba(OH)2 as base in
DMF at higher concentration in comparable yields.
References
1 Nauseef WM. Myeloperoxidase in human neutrophil host defense. Cell Microbiol.
2014;16(8):1146–1155.
2 Taurog A, Dorris ML, Doerge DR. Mechanism of simultaneous iodination and
coupling catalyzed by thyroid peroxidase. Arch Biochem Biophys. 1996;330(1):24–32.
3 Mitra SN, Slungaard A, Hazen SL. Role of eosinophil peroxidase in the origins of
protein oxidation in asthma. Redox Rep. 2000;5(4):215–224.
4 Nicholls SJ, Hazen SL. Myeloperoxidase, modified lipoproteins, and atherogenesis.
J Lipid Res. 2009;50(Suppl):S346–S351.
5 Baseri M, Heidari R, Mahaki B, Hajizadeh Y, Momenizadeh A, Sadeghi M.
Myeloperoxidase levels predicts angiographic severity of coronary artery disease in
patients with chronic stable angina. Adv Biomed Res. 2014;3:139.
6 Baldus S, Heeschen C, Meinertz T, et al. Myeloperoxidase serum levels predict risk in
patients with acute coronary syndromes. Circulation. 2003;108(12):1440–1445.
7 Brennan M-L, Hazen SL. Myeloperoxidase: a mechanistically linked biomarker for
cardiovascular disease. Curr Cardiovascular Risk Rep. 2007;1(1):58–65.
8 Chaikijurajai T, Tang WHW. Myeloperoxidase: a potential therapeutic target for
coronary artery disease. Expert Opin Ther Targets. 2020;24:695–705.
9 Shao B, Heinecke JW. HDL, lipid peroxidation, and atherosclerosis. J Lipid Res. 2009;
50(4):599–601.
In summary, potent inhibitors of MPO were prepared by designing
pyrazole analogs to facilitate extension of lipophilic groups into a hy-
drophobic pocket in the MPO protein. Inhibitors 12 and 17 demon-
strated potent inhibition in human neutrophils while maintaining
excellent selectivity for MPO versus TPO by incorporating an amine
substituent. In an attempt to improve the potency and drug-like prop-
erties, MPO crystal structures and molecular modeling were used to
design macrocyclic MPO inhibitors with varied ring sizes and substitu-
tion patterns. Multiple macrocyclization strategies were employed to
prepare the diverse structures. Intramolecular ring-closure via a Suzuki-
Miyaura condensation between a pyrazole boronate and a vinyl iodide
was a versatile method for making diverse macrocycles. Although the
macrocycle strategy did not improve the oral bioavailability of the tri-
azolopyridine series, the macrocycles 21, 29 and 30 provided increased
in vitro potency compared to linear analogs 12 or 17 while maintaining
good TPO selectivity.
´
10 Rudolph V, Andrie RP, Rudolph TK, et al. Myeloperoxidase acts as a profibrotic
mediator of atrial fibrillation. Nat Med. 2010;16(4):470–474.
11 Soubhye J, Aldib I, Elfving B, et al. Design, synthesis, and structure-activity
relationship studies of novel 3-alkylindole derivatives as selective and highly potent
myeloperoxidase inhibitors. J Med Chem. 2013;56(10):3943–3958.
´
¨
12 Tiden A-K, Sjogren T, Svensson M, et al. 2-Thioxanthines are mechanism-based
inactivators of myeloperoxidase that block oxidative stress during inflammation.
J Biol Chem. 2011;286(43):37578–37589.
13 Ruggeri RB, Buckbinder L, Bagley SW, et al. Discovery of 2-(6-(5-chloro-2-
methoxyphenyl)-4-oxo-2-thioxo-3,4-dihydropyrimidin-1(2H)-yl)acet amide (PF-
06282999): a highly selective mechanism-based myeloperoxidase inhibitor for the
treatment of cardiovascular diseases. J Med Chem. 2015;58(21):8513–8528.
14 Tong X, Zhou D, Savage A, et al. Population pharmacokinetic modeling with
enterohepatic circulation for AZD3241 in healthy subjects and patients with multiple
system atrophy. J Clin Pharmacol. 2018;58(11):1452–1460.
Declaration of Competing Interest
15 Duclos F, Abell LM, Harden DG, et al. Triazolopyrimidines identified as reversible
myeloperoxidase inhibitors. MedChemComm. 2017;8(11):2093–2099.
16 Wurtz NR, Viet A, Shaw SA, et al. Potent triazolopyridine myeloperoxidase
inhibitors. ACS Med Chem Lett. 2018;9(12):1175–1180.
All co-authors were employees of Bristol-Myers Squibb (Princeton,
NJ, USA) at the time the studies were conducted. The authors declare
that they have no known competing financial interests or personal re-
lationships that could have appeared to influence the work reported in
this paper.
17 Shaw SA, Vokits BP, Dilger AK, et al. Discovery and structure activity relationships of
7-benzyl triazolopyridines as stable, selective, and reversible inhibitors of
myeloperoxidase. Bioorg Med Chem. 2020;28(22):115723.
18 Chae M-Y, Swenn K, Kanugula S, Dolan ME, Pegg AE, Moschel RC. 8-Substituted O6-
benzylguanine, substituted 6(4)-(benzyloxy)pyrimidine, and related derivatives as
inactivators of human O6-alkylguanine-DNA alkyltransferase. J Med Chem. 1995;38
(2):359–365.
Acknowledgements
This research was conducted at Bristol-Myers Squibb and did not
receive any specific grant from funding agencies in the public, com-
mercial, or not-for-profit sectors. The authors wish to thank Richard
Rampulla, Arvind Mathur and the scientists at the Biocon Bristol-Myers
Squibb Research Center (BBRC) for the synthesis of key chemical in-
termediates, as well as Dr. Lynn M. Abell for her support and input to
this study.
19 Augustijns PF, Brown SC, Willard DH, et al. Hydration changes implicated in the
remarkable temperature-dependent membrane permeation of cyclosporin A.
Biochemistry. 2000;39(25):7621–7630.
20 Bockus AT, McEwen CM, Lokey RS. Form and function in cyclic peptide natural
products: a pharmacokinetic perspective. Curr Top Med Chem. 2013;13(7):821–836.
21 Collins JC, Farley KA, Limberakis C, Liras S, Price D, James K. Macrocyclizations for
medicinal chemistry: synthesis of druglike macrocycles by high-concentration
Ullmann coupling. J Org Chem. 2012;77(24):11079–11090.
22 Yu X, Sun D. Macrocyclic drugs and synthetic methodologies toward macrocycles.
Molecules. 2013;18(6):6230–6268.
ˆ
23 Tsunoda T, Kaku H, Ito S. New Mitsunobu reagents. TCIMail. 2005;123.
Appendix A. Supplementary data
24 Dieckmann M, Kretschmer M, Li P, Rudolph S, Herkommer D, Menche D. Total
synthesis of rhizopodin. Angew Chem Int Ed. 2012;51(23):5667–5670.
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.128010.
8