Substituted 2,2-bisaryl-bicycloheptanes as novel and potent inhibitors of 5-lipoxygenase activating protein

Article abstract of DOI:10.1016/j.bmcl.2008.01.105

The discovery and SAR of a novel series of substituted 2,2-bisaryl-bicycloheptane inhibitors of 5-lipoxygenase activating protein (FLAP) are herein described. SAR studies have shown that 2,5-substitution on the exo-aryl group is optimal for potency. The most potent compounds in this series have an ortho-nitrogen aryl linked with a methyleneoxy as the 5-substituent and a polar group such as a urethane as the 2-substituent. One of the most potent compounds identified is the 5-benzothiazolymethoxy-2-pyridinylcarbamate derivative 2 (FLAP IC₅₀ = 2.8 nM) which blocks 89% of ragweed induced urinary LTE₄ production in dogs (at an I.V. dose of 2.5 μg/kg/min). This compound inhibits calcium ionophore stimulated LTB₄ production in both human polymorphonuclear (PMN) leukocytes and human whole blood (IC₅₀ = 2.0 and 33 nM, respectively).

Full text of DOI:10.1016/j.bmcl.2008.01.105

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2023–2027
Substituted 2,2-bisaryl-bicycloheptanes as novel
and potent inhibitors of 5-lipoxygenase activating protein
Dwight Macdonald,* Christine Brideau, Chi Chung Chan, Jean-Pierre Falgueyret,
à
`
´
Richard Frenette, Jocelyne Guay, John H. Hutchinson, Helene Perrier,
à
§
´
Peptiboon Prasit, Denis Riendeau, Philip Tagari, Michel Therien,
Robert N. Young– and Yves Girard
Department of Medicinal Chemistry, Merck Frosst Centre for Therapeutic Research, 16711 Trans Canada Hwy.,
Kirkland, Que., Canada H9H 3L1
Received 22 November 2007; revised 28 January 2008; accepted 29 January 2008
Available online 2 February 2008
Abstract—The discovery and SAR of a novel series of substituted 2,2-bisaryl-bicycloheptane inhibitors of 5-lipoxygenase activating
protein (FLAP) are herein described. SAR studies have shown that 2,5-substitution on the exo-aryl group is optimal for potency.
The most potent compounds in this series have an ortho-nitrogen aryl linked with a methyleneoxy as the 5-substituent and a polar
group such as a urethane as the 2-substituent. One of the most potent compounds identified is the 5-benzothiazolymethoxy-2-pyrid-
inylcarbamate derivative 2 (FLAP IC50 = 2.8 nM) which blocks 89% of ragweed induced urinary LTE4 production in dogs (at an
I.V. dose of 2.5 lg/kg/min). This compound inhibits calcium ionophore stimulated LTB4 production in both human polymorpho-
nuclear (PMN) leukocytes and human whole blood (IC50 = 2.0 and 33 nM, respectively).
Ó 2008 Elsevier Ltd. All rights reserved.
Leucotrienes (LT’s) are potent inflammatory mediators
involved in the pathogenesis of diseases such as asthma
and atherosclerosis.1,2 The first step in the conversion of
arachidonic acid to the LT’s is catalyzed by the enzyme
5-lipoxygenase (5-LO).3 The enzyme 5-LO may be
inhibited either by the direct interaction with an inhibi-
tor or by the binding of an inhibitor to the 5-lipoxyge-
nase activating protein (FLAP).4 Clinical trials have
demonstrated that the FLAP inhibitor MK-0591 is an
effective treatment for asthma.5,6 FLAP inhibitors have
also been shown to be protective in animal models of
cardiovascular disease.1
By screening a large number of diverse chemical structures,
obtained from various sources, for their ability to prevent
the production of LTB4 by human polymorphonuclear
(PMN) leukocytes when stimulated with calcium iono-
phore,7 wediscovereda potent inhibitor, 1(IC50 = 60 nM).
This 2,2-bis(oxyaryl)-bicycloheptane was also found to be
a potentFLAPinhibitor8 (IC50 = 387 nM) whilenot inhib-
iting the purified 5-LO (IC50 > 8 lM).
S
O
OH
O
O
N
Keywords: 5-LO; FLAP; LT; Asthma; Inflammatory bowel disease;
Rheumatoid arthritis; Psoriasis; Atherosclerosis.
H
N
N
*
Corresponding author. Tel.: +1 514 428 3089; fax: +1 514 428
4900; e-mail: dwight_macdonald@merck.com
O
O
OH
2
Present address: Neurochem Inc., 275 Armand-Frappier Blvd.,
Laval, Que., Canada H7V 4A7.
1
à
§
Present address: Amira Pharmaceuticals, 9535 Waples St., Suite 100,
San Diego, CA 92121, USA.
S
O
CO2H
Cl
N
Present address: Chemistry Research and Discovery, Amgen, Inc.,
One Amgen Center Drive, Thousand Oaks, CA 91320, USA.
Present address: Department of Chemistry, Simon Fraser University,
Burnaby, BC, Canada V5A 1S6.
N
–
MK-0591
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.105

2024
D. Macdonald et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2023–2027
In order to more closely mimic the position of the quin-
OH
olinyl and acid groups of MK-0591, these two groups
were also placed on the same aromatic ring of the bis-
arylbicycloalkane. The synthesis of the 2-phenyl-2-(2,5-
alkoxyphenyl) bicycloalkanes followed the procedure
shown in Scheme 2. Phenylmagnesium bromide was
added to norbornanone 8, which resulted in production
of only the exo-phenyl tertiary alcohol 9. Friedel-Crafts
alkylation of hydroquinone with the tertiary alcohol 9
gave the required diphenol 10 in modest yield (27%).
Only the product from exo addition to the bicyclo sys-
tem was observed. The 2,5-diphenol 10 was deproto-
nated with 1 equivalent of t-BuOK followed by
alkylation with a halomethyl-aryl to produce the bis
(not shown) and mono-alkylated product 11. Further
functionalization of monophenol 11 with Cs2CO3 and
various halides in DMF or with an aryl isocyanate in
toluene supplied the compounds shown in Scheme 2.
Similar chemistry was used to prepare the correspond-
ing 2,4-diphenol products (Scheme 3) and monocyclic 5
and 6-member compounds (Scheme 4). The Friedel-
Crafts alkylation gave good yields for the bicyclo 2,4-
diphenol but very poor yields for the monocyclic com-
pounds. The monocyclic compounds were best pre-
pared from the corresponding alkenes as shown in
Scheme 4. The methyl benzoate derivative 12r was pre-
pared by converting phenol 11a to a triflate, followed
by Pd catalyzed carbonylation (Scheme 2).
Cs2CO3
+
Cl
DMF, RT.
N
3
OH
O
O
N
N
+
5
4
O
OH
N
BrCH2CO2Me, Cs2CO3
DMF, RT.
O
N
O
N
LiOH (aq)
THF, RT.
O
CO2Me
O
CO2H
The in vitro evaluation of the compounds was as fol-
lows: (a) Each compound was evaluated in a 5-lipoxyge-
nase activating protein (FLAP) binding assay. This
assay determines the ability of the compound to displace
the radio-labeled ligand [125I]-L-691,831.8
6
7
10
Scheme 1.
Synthesis of compounds 3–7.
(b) The compounds were screened in a calcium-iono-
phore-activated human polymorphonuclear (PMN) leu-
kocyte and a human whole blood assay for their ability
to inhibit LTB4 production.7
Structure–activity relationship (SAR) study of this
FLAP inhibitor has led to the identification of the
5-benzothiazolymethoxy-2-pyridinylcarbamate deriva-
tive 2 which has shown excellent potency in an in vivo
model for blocking the production of systemic LT’s.
The SAR studies and synthesis of this novel series of
FLAP inhibitors is herein described.
(c) Direct inhibition of recombinant 5-LO of selected
compounds was evaluated using a spectrophotometric
assay monitoring 5-hydroperoxyeicosatetraenoic acid
production.12
The initial direction of our SAR studies was led by the
observation that other FLAP inhibitors such as MK-
0591 have two structural features which are important
for their high potency.9 These two functional groups,
namely the quinolinylmethoxy and a carboxylic acid,
were then introduced at the most convenient locations
on the core of the lead compound 1 (Scheme 1). The
commercially available 2,2-bis(hydroxyphenyl)-bicyclo-
heptane 3 was alkylated with one equivalent of 2-chlo-
romethylquinoline in DMF using cesium carbonate as
base, to yield both the mono-quinolinyl and the bis-
quinolinyl compounds 4 and 5. The mono quinolinyl
phenol was then coupled to methyl bromoacetate under
the same conditions. Hydrolysis of the resulting methyl
ester 6 then afforded the required carboxylic acid 7.
All the compounds described in this paper were isolated
as mixtures of stereoisomers except where noted
otherwise.
Introduction of a quinoline and a carboxylic acid func-
tion to the core of the lead structure at the two para
positions of the phenyl rings 7, produced an encourag-
ing fourfold enhancement of the FLAP potency over
that of the lead compound 1 (Scheme 1, Table 1). The
bis-quinolinyl 5, quinolinyl-phenol 4, and the quinoli-
nyl-ester 6 were also more potent than the lead com-
pound. This enhancement in potency by the
introduction of the quinolinyl, especially if an acid is
added 7, is analogous to the effects of these groups in
other FLAP inhibitors such as MK-0591.9
Placement of the quinolinyl and acid groups on the
same phenyl ring of the lead compound core structure
is a better mimic of the position of these groups on more
potent known inhibitors, such as MK-0591 (Scheme 2,
Table 1). A 30-fold enhancement in FLAP binding po-
tency over 1 is observed when the quinolinyl and the

D. Macdonald et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2023–2027
2025
R1
R2
Compound Reagents
N
11a
12a
12b
12c
12d
12e
12f
c
N
O
CO Me
2
c,d
d,c
c,d,e
d,c,e
c,f
N
O
CO Me
2
N
O
CO H
2
b
a
N
O
CO H
2
OH
O
N
N
O
O
CO Et
2
8
9
CO H
2
c,f,e
c,g
c,h
c,i
N
N
R1
12g
12h
12i
O
O
OH
O
Me
n-Bu
reagents
N
OH
OH
O
N
N
NMe
2
12j
c,j
O
10
11
SO Me
2
O
12k
12l
c,k
c,l
N
N
N
O
N
R1
O
12m
12n
12o
12p
12q
2
m,l
n,l
O
N
N
N
O
O
reagents
H
N
N
O
R2
o,l
O
N
N
N
O
O
p,l
N
12
q,l
S
S
N
H
N
O
q,o
N
O
N
CO Me
2
12r
12s
c,r,s
c,r,s,t
N
CO H
2
10
Scheme 2.
Synthesis of compounds 10–12. Reagents and conditions: (a) PhMgBr, THF, À60 to 25 °C; (b) Hydroquinone, p-TSA monohydrate,
toluene, reflux, 3 h, 27%; (c) t-BuOK, DMF, 2-(chloromethyl)quinoline, 25 °C, 18 h, 32% mono, 17% bis; (d) Cesium carbonate, DMF, methyl
bromoacetate, 25 °C, 18 h; (e) LiOH(aq), THF, 25 °C, 18 h; (f) Cesium carbonate, DMF, ethyl-4-bromobutyrate, 25 °C, 18 h; (g) Cesium carbonate,
DMF, MeI, 25 °C, 18 h; (h) Cesium carbonate, DMF, bromobutane, 25 °C, 18 h; (i) Cesium carbonate, DMF, benzyl bromide, 25 °C, 18 h; (j)
Cesium carbonate, DMF, N,N-dimethylamino ethylchloride, 25 °C, 18 h; (k) Triton-B, THF, methyl vinyl sulfone, 25 °C, 18 h. (l) Cesium carbonate,
DMF, 2-picolyl chloride, 25 °C, 18 h; (m) t-BuOK, DMF, 2-picolyl chloride, 25 °C, 18 h; (n) t-BuOK, DMF, 3-picolyl chloride, 25 °C, 18 h; (o)
3-pyridyl isocyanate (in situ from refluxing nicotinyl azide in toluene 30 m), DIPEA, toluene, reflux, 3 h, 63%; (p) t-BuOK, DMF,
5-chloromethylfuro[3,2-B]pyridine, 25 °C, 18 h; (q) t-BuOK, DMF, 2-(chloromethyl)-1,3-benzothiazole, 25 °C, 18 h, 24% mono, 10% bis; (r)
Tf2O, pyridine, 0–25 °C, 18 h; (s) Palladium (II) acetate, dppf, TEA, CO(1 atm), DMF, MeOH, 80 °C, 18 h, 26%; (t) KOH(aq), THF,
1,2-propanediol, 110 °C, 18 h, 59%.
acid are in the 5 and 2 positions of the aryl, respectively,
12c. An even more interesting observation is that ester
12a is more potent than the acid, with its FLAP potency
80-fold more than the lead compound. If the position of
the acid/ester and the quinoline are exchanged, the po-
tency is much less, 20-fold with ester 12b and 200-fold
with acid 12d. The relatively small differences in the
FLAP potency between the ester and acid pairs reported
in this paper (one to threefold) are in agreement with
compounds related to MK-0591, where a neutral group
in place of an acid led to a potency within threefold of
the acid compound.9
The corresponding 2,4-substituted compounds are sig-
nificantly less potent than the 2,5-substituted com-
pounds (Scheme 3, Table 1). Using a longer 3-carbon

2026
D. Macdonald et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2023–2027
chain to connect the acid/ester to the phenol gave com-
pounds 12e and 12f that have similar FLAP potency to
the 1-carbon linked ester 12a. The most interesting
contrast between the long and short chain compounds
is that the long chain ester and acid have only a two-
fold difference in potency in the human whole blood
assay (WBA) while the short chain compounds have
more than a 10-fold difference. If the acid/ester is di-
rectly attached to the aryl (12r and 12s) then the po-
tency profile is similar to the 2-atom linked
compounds. We have thus obtained several compounds
(12a, 12e, 12f, and 12r) with a similar in vitro potency
profile to that of the clinically effective MK-0591.
O
N
c
a, b
9
OH
13
O
N
14a R = Me
14b R = H
O
CO2R
d
Further SAR study on the 2-oxyaryl substituent
(Scheme 2, Table 1) shows that a polar group is impor-
tant for potency in the whole blood assay. The amine
12j, sulfone 12k, and 2-pyridyl 12l are quite potent in
the WBA (100–300 nM) while the methyloxy, butyloxy
and benzyloxy analogs (12g, 12h, and 12i, respectively)
are >1000 nM. One of the best polar groups is the 3-
pyridinylcarbamate, as in 2 (WBA IC50 = 33 nM).
Replacement of the quinoline with other heterocycles
is possible, provided that an ortho-nitrogen is present
(compare 12m to 12n, Scheme 2, Table 1). This obser-
vation is analogous to compounds related to MK-
0591, where a naphthyl in place of the quinoline gave
a much less potent compound.9 The benzothiazole
proved to be equivalent to the quinoline, and when
combined with a 3-pyridinylcarbamate derivative, pro-
vided the extremely potent and orally bioavailable 2.
When dosed orally in rats at 20 mg/kg as a solution
in PEG400, the bioavailability was 19% with a Cmax
of 1.1 lM at 6 h. These compounds were resolved on
ChiralpakÒ AD (eluted with 4:1 hexane/2-propanol)
and the enantiomers (+)2 and (À)2 were found to have
similar potency. This small difference in potency be-
tween the enantiomers is similar to that of a chiral com-
pound related to MK-0591(L-674,636), where there was
only a twofold potency difference.9 Compound 2 was
also very potent for inhibiting ragweed induced urinary
LTE4 production in dogs (89% inhibition at an I.V.
dose of 2.5 lg/kg/min).13 Both the PK and the in vivo
efficacy studies were preformed on the racemate of
compound 2.
Scheme 3. 10 Synthesis of compounds 13–14. Reagents and conditions:
(a) Resorcinol, 3-mercaptopropionic acid, HCl (conc), 1 h, 41%; (b) t-
BuOK, DMF, 2-(chloromethyl)quinoline, 25 °C, 18 h; (c) Cesium
carbonate, DMF, methyl bromoacetate, 25 °C, 18 h; (d) LiOH(aq),
THF, 25 °C, 18 h.
OH
reagents
a
n
OH
n
16
15
N
N
O
O
reagents
n
n
O
OH
R
18
17
Compound Reagents
n
R
17a
18a
18b
18c
18d
18e
18f
b
1
1
1
1
1
2
2
CO Me
2
b,c
b
The corresponding monocyclic 5 and 6-member ring
compounds have similar potency and SAR to the bicy-
clo series (Scheme 4, Table 1). The similar potency of
the monocyclic compounds to the corresponding bicy-
clic compounds, in addition to the similar potency of
the enantiomers (+)2r and (À)2r, suggests that the
bicyclic and monocyclic moiety is either solvent ex-
posed, lipid exposed, or in a large hydrophobic pocket
when these molecules are bound to FLAP.
N
N
b,d
b,e
b,c
b
H
N
N
O
CO Me
2
N
In conclusion, a highly potent series of FLAP inhibitors
(2,2-bisaryl-bicycloheptanes) has been discovered. A
representative compound 2 has a FLAP, HPMN, and
human whole blood IC50 of 2.8, 2.0 and 33 nM, respec-
tively. This compound also blocks 89% of ragweed in-
duced urinary LTE4 production in dogs at an I.V.
dose of 2.5 lg/kg/min.
Scheme 4. 11 Synthesis of compounds 16–18. Reagents and conditions:
(a) Hydroquinone, p-TSA monohydrate, toluene, reflux, 3 h, ꢀ7%; (b)
t-BuOK, DMF, 2-(chloromethyl)quinoline, 25 °C, 18 h; (c) Cesium
carbonate, DMF, methyl bromoacetate, 25 °C, 18 h; (d) t-BuOK,
DMF, 3-picolyl chloride, 25 °C, 18 h; (e) 3-picolyl isocyanate (in situ
from refluxing nicotinyl azide in toluene, 30 min), DIPEA, toluene,
reflux, 3 h.

D. Macdonald et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2023–2027
Table 1. Summary of in vitro assay data
2027
a
FLAP IC50 (nM)
a
HPMN IC50 (nM)
a
5-LO IC50 (nM)
a
HWBA IC50 (nM)
Compound
MK-0591
1
1.6 0.3
387 50
2.8 0.1
2.5 0.2
2.4 0.6
145 50
127 15
182 15
98 50
3.1 0.5
60 20
—
510 50
—
>8000
—
—
2( )
2(+)
2(À)
4
2.0 1.0
1.8 0.5
3.0 0.5
186 50
>1800
33 10
86 10
37 10
—
—
—
5
—
—
—
—
6
169 50
44
7
10
2
>6300
—
—
—
2070 700
13.2 0.1
5.0 1.0
418 250
19 14
11a
12a
12b
12c
12d
12e
12f
12g
12h
12i
5000
>6000
—
580 50
290 160
—
5.0 05
5.9 3.0
94
14
1
1
16
282 70
2
—
—
>3000
—
3155 1000
3.7 0.1
3.5 0.5
8.6 1.0
42
8
8
4
—
—
150 20
300 30
1100 100
>3000
>3000
335 50
114 60
225 100
275 75
—
9.0 1.5
80 40
6
2
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
8.6 0.4
6.5 1.5
5.5 1.2
4.1 0.2
18.3 6.0
217 100
8.7 4.0
4.2 1.0
7.0 2.5
3.9 0.7
12j
12k
12l
12m
12n
12o
12p
12q
12r
12s
13
134
9
24
—
1
181 23
4.9 0.5
7.6 0.2
5.6 1.0
12.4 1.0
980 200
776 50
3595 275
>3000
80 10
220 50
140 20
—
2.2 0.8
7.0 4.0
2.6 0.5
9.5 1.0
>2372
—
—
14a
14b
17a
18a
18b
18c
18d
18e
18f
47 16
152 50
5.0 0.5
1.7 0.5
—
22
3
1920 200
890 100
1080 100
200 20
9.3 1.5
4.5 0.6
2.5 0.1
2.4 0.3
3.7 0.3
4.2 0.5
10.2
4.1
3
1
<1.3
1.8 0.5
6.1
52
200 200
4
1
ꢀ3000
a Values are means of at least two experiments, standard error.
´
8. Charleson, S.; Prasit, P.; Leger, S.; Gillard, J. W.; Vickers,
P. J.; Mancini, J. A.; Charleson, P.; Guay, J.; Ford-
Hutchinson, A. W.; Evans, J. F. Mol. Pharmacol. 1992,
41, 873.
References and notes
1. Funk, C. D. Nat. Rev. Drug Disc. 2005, 4, 664.
2. (a) Friesen, R. W.; Riendeau, D. Ann. Rep. Med. Chem.
2005, 40, 199; (b) Osher, E.; Weisinger, G.; Limor, R.;
Tordjman, K.; Stern, N. Mol. Cell. Endocrinol. 2006, 252,
201.
3. Radmark, O.; Samuelsson, B. Biochem. Biophys. Res.
Commun. 2005, 338, 102.
4. Ferguson, A. D.; McKeever, B. M.; Shihua, X.; Wisniew-
ski, D.; Miller, D. K.; Yamin, T.-T.; Spencer, R. H.; Chu,
L.; Ujjainwalla, F.; Cunningham, B. R.; Evans, J. F.;
Becker, J. W. Science 2007, 317, 510.
5. Diamant, Z. et al. J. Allergy Clin. Immunol. 1995, 95, 42.
6. Young, R. Eur. J. Med. Chem. 1999, 34, 671.
7. Brideau, C.; Chan, C.; Charleson, S.; Denis, D.; Evans, J.
F.; Ford-Hutchinson, A. W.; Fortin, R.; Gillard, J. W.;
9. Prasit, P.; Belley, M.; Blouin, M.; Brideau, C.; Chan, C.;
Charleson, S.; Evans, J. F.; Frenette, R.; Gauthier, J. Y.;
Guay, J.; Gillard, J. W.; Grimm, E.; Ford-Hutchinson, A.
W.; Hutchinson, J. H.; Fortin, R.; Jones, T. R.; Mancini,
J.; Leger, S.; McFarlane, C. S.; Piechuta, H.; Riendeau,
D.; Roy, P.; Tagari, P.; Vickers, P. J.; Young, R. N.;
Zamboni, R. J. Lipid Mediators 1993, 6, 239.
10. Chu, L.; Goulet, M. T.; Ujjainwalla, F.; Frenette, R.;
Girard, Y.; Therien, M.; Macdonald, D.; Hutchinson, J.
H.; Chang, L. Patent: WO 2005/009951 A2, 2005.
11. Armstrong, H. M.; Chang, L. L.; Frenette, R Macdonald,
D.; Ok, H. O.; Therien, M.; Ujjainwalla, F. Patent: WO
2006/098912 A1, 2006.
12. Falgueyret, J.-P.; Denis, D.; Macdonald, D.; Hutchinson,
J. H.; Riendeau, D. Biochemistry 1995, 34, 13603.
13. Tagari, P.; Becker, A.; Brideau, C.; Frenette, R.; Sadl, V.;
Thomas, E.; Vickers, P.; Ford-Hutchinson, A. J. Phar-
macol. Exp. Ther. 1993, 2651, 416.
´
Guay, D.; Guevremond, D.; Hutchinson, J. H.; Jones, T.
R.; Leger, S.; Mancini, J. A.; McFarlane, C. S.; Pickett,
C.; Piechuta, H.; Prasit, P.; Riendeau, D.; Rouzer, C. A.;
Tagari, P.; Vickers, P. J.; Young, R. N.; Abraham, W. M.
Can. J. Physiol. Pharmacol. 1992, 70, 799.
´