Imidazopyridyl compounds as aldosterone synthase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.12.003

The inhibition of aldosterone synthase (CYP11B2) may be an effective treatment of hypertension and heart failure, among other ailments. Previously reported benzimidazole CYP11B2 inhibitors led the way for bioisosteric imidazopyridines that are both potent

Full text of DOI:10.1016/j.bmcl.2016.12.003

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Imidazopyridyl compounds as aldosterone synthase inhibitors
a
a
a
a
a
Brent R. Whitehead a, , Michael M.-C. Lo , Amjad Ali , Min K. Park , Scott B. Hoyt , Yusheng Xiong ,
Jiaqiang Cai b, Emma Carswell b, Andrew Cooke b, John MacLean b, Paul Ratcliffe b, John Robinson b,
D. Jonathan Bennett b, Joseph A. Clemas a, Tom Wisniewski a, Mary Struthers a, Doris Cully a,
Douglas J. MacNeil a
⇑
a MRL, Rahway, NJ 07065, USA
b MRL, Newhouse, Lanarkshire ML1 5SH, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
The inhibition of aldosterone synthase (CYP11B2) may be an effective treatment of hypertension and
heart failure, among other ailments. Previously reported benzimidazole CYP11B2 inhibitors led the
way for bioisosteric imidazopyridines that are both potent and selective over CYP11B1.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 4 November 2016
Revised 30 November 2016
Accepted 1 December 2016
Available online xxxx
Keywords:
Aldosterone
CYP11B2
Hypertension
Imidazopyridine
Heart failure
Aldosterone is a principal mineralocorticoid that promotes
increased blood pressure, inflammation, fibrosis, and organ
damage.1–4 The final steps of its biosynthesis are catalyzed by a
single mitochondrial cytochrome P450 enzyme aldosterone syn-
thase (CYP11B2), found predominantly in zona glomerulosa cells
of the adrenal gland.5,6 Inhibition of CYP11B2 should lower plasma
aldosterone levels, potentially providing an effective treatment for
a variety of ailments, including hypertension, heart failure, and
chronic kidney disease.
A great design challenge to advance a CYP11B2 inhibitor in the
clinic is to maintain its high selectivity against other CYPs, in
particular cortisol synthase CYP11B1, which share >93% homology
with CYP11B2. CYP11B1 catalyzes the biosynthesis of cortisol, and
inhibition of CYP11B1 can result in impaired stress response and
altered glucose metabolism. Selectivity against other steroidogenic
CYPs, such as CYPs 17, 19, and 21, is crucial as well, since they reg-
ulate the production of androgens and estrogens.
undesired dose-limiting impairment of cortisol response was also
observed, presumably due to the inhibition of CYP11B1.9–11
The results from LCI-699 highlight the need to discover
CYP11B2 inhibitors that are both potent and selective against
CYP11B1 for clinical development. Our discovery of the benzimida-
zole series13 has led us to further explore the isosteric equivalents
of the heterocycle. We are pleased to find that the imiadzopyridine
analog 2 offers good potency on CYP11B2 while maintaining simi-
lar selectivity against CYP11B1 as the initial benzimidazole lead 1
(Fig. 1).14
The preparation of the imidazopyridines is illustrated with the
synthesis of compound 2 (Scheme 1).14 5-Fluoropyridin-2-amine
was treated with neat ethyl bromoacetate, and the resultant
intermediate 2a was heated in phosphorus oxychloride to afford
the corresponding 2-chloro-6-fluoroimidazo[1,2-a]pyridine 2b
after careful quenching and isolation. The intermediate 2b was
halogenated at the 3-position with N-iodosuccinimide to yield
2-chloro-6-fluoro-3-iodoimidazo[1,2-a]pyridine 2c. Two iterative
Suzuki couplings of 2c, first with cyclopropylboronic acid and the
corresponding pyridineboronic acid pinacol ester, afforded the
desired compound 2.
A
host of small molecule CYP11B2 inhibitors have been
reported.7,8,12 Among them, LCI-699 has been shown to lower
aldosterone levels and blood pressure in the clinic, providing the
proof-of-concept for the therapeutic target. At the same time,
As of the compounds reported in Table 1, compound 3 was
prepared by employing methylboronic acid instead of cyclopropy-
lboronic acid in step d. Compound 4 was prepared directly from 2b
and the corresponding boronate ester. Chlorination of compound 4
⇑
Corresponding author.
E-mail address: brent.whitehead@merck.com (B.R. Whitehead).
http://dx.doi.org/10.1016/j.bmcl.2016.12.003
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Whitehead B.R., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.003

2
B.R. Whitehead et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
N
N
N
N
N
Prior work in the benzimidazole series has established that sub-
stitution at the 2- or the 6-position of the pyridine is not tolerated
by CYP11B2.13 With the cyclopropyl group confirmed as the opti-
mal one balancing potency at CYP11B2 and selectivity against
CYP11B1, we proceeded to examine the effect of different sub-
stituents R2 on the pyridine (Table 2). Compounds 7–17 were syn-
thesized similarly as compound 2, with the appropriate boronate
ester used in the second Suzuki coupling.
N
F
F
OH
OH
Me
Me
Me
Me
2
1
CYP11B2 IC50 = 13 nM
CYP11B1 IC50 = 830 nM
CYP11B1/B2 = 63
CYP11B2 IC50 = 1.4 nM
CYP11B1 IC50 = 100 nM
CYP11B1/B2 = 75
Compounds 7 and 8, bearing a tertiary alcohol with one of the
methyl groups replaced by a trifluoromethyl group, show similar
activity on CYP11B2 with similar levels of selectivity against
CYP11B1 as compound 2. The carboxylate ester group improves
the CYP11B2 activity while decreasing the CYP11B1 activity (com-
pound 9), resulting in a highly selective compound, but it does not
offer enough metabolic stability to go forward. Electron-withdraw-
ing groups such as fluoro or trifluoromethyl (compounds 10, 12)
have little effect on either the CYP11B2 activity or CYP11B1 selec-
tivity, while the cyano group (compound 11) leads to some erosion
of CYP11B2 activity. Groups such as methoxy, methyl or phenyl
improve the CYP11B2 activity and CYP11B1 selectivity slightly
(compounds 13–15). Finally, replacing the entire pyridine with iso-
quinoline gives a potent CYP11B2 inhibitor with similar selectivity
as compound 2 against CYP11B1 (compound 17).
Fig. 1. Comparison of the benzimidazole lead 1 and the imidazopyridine lead 2.
NH2
N
NH
a
c
b
d
O
N
N
N
F
F
F
F
O
Me
2a
N
Cl
Cl
Cl
N
I
2b
2c
N
N
e
N
N
N
F
F
OH
Me
Me
2d
2
We then sought to investigate the effect of different R4 and R5
on the imidazopyridine. We investigated two series, one with the
pyridine bearing a tertiary alcohol at the 5-position of the pyridine
(Table 3) and another with the pyridine bearing a 5-fluoro group
(Table 4). In the 5-tertiary alcohol series, when the fluorine is
moved from the 6-position to the 7-position (compound 18),
reduction in potency at CYP11B2 and selectivity against CYP11B1
was observed. Replacing the fluoro with a cyano group at the 6-
position results in little change in potency and selectivity (com-
pound 19).
In the 5-fluoro series, moving the halogen from the 6-position
to the 7-position has a similar effect in loss of potency at CYP11B2
(compound 20). Replacing the fluoro with a chloro group generates
a compound with a similar profile (compound 21). By introducing
an extra halogen at the 7-position (R5), we obtained a compound
with better potency at CYP11B2 while maintaining similar selec-
tivity at CYP11B1 (compound 22). Incorporating additional alkyl
Scheme 1. Reagents and conditions for preparation of compound 2: (a) ethyl
bromoacetate, neat, room temperature; (b) phosphorous oxychloride, neat, 110 °C;
(c) N-iodosuccinimide, MeCN, room temperature; (d) c-PrB(OH)2, Pd(PPh3)2(OAc)2
(0.1 equiv), Cs2CO3, THF, 100 °C; (e) 5-(1-hydroxy-1-methylethyl) pyridine-3-
boronic acid pinacol ester, Pd(OAc)2 (0.1 equiv), S-Phos (0.2 equiv), K3PO4, THF,
100 °C.
by N-chlorosuccinimide afforded compound 5 The precursor for
compound 6 (6d) was prepared through lithiation of intermediate
2c (Scheme 2).
Our investigation focused on improving selectivity over
CYP11B1 and to bring forward a suitable candidate for in vivo stud-
ies. In the benzimidazole series, N-substitution was found to exert
a substantial effect on CYP11B2 inhibition, with optimal activity
limited to a small cyclopropyl-containing group.13 Here in the imi-
dazopyridine series, we observed a similar SAR pattern (Table 1).
Poor activity was observed when the corresponding C-3 posi-
tion of the imidazopyridine is unsubstituted (compound 4). Both
the methyl group (compound 3) and the chloro group (compound
5) give compounds that are more potent than the corresponding
cyclopropyl compound at CYP11B2 but less selective against
CYP11B1. The pseudohalide cyano group leads to a less potent
and less selective compound (compound 6).
N
N
f
Cl
Cl
N
N
F
F
I
CN
2c
6d
Scheme 2. Reagents and conditions for preparation of intermediate 6d:
(f) n-butyllithium, THF, À78 °C, 5 min; p-toluenesulfonyl cyanide.
Table 1
Effect of R1 substitution on CYP11B2/CYP11B1 inhibition.a
N
N
N
F
R1
OH
Me
Me
Compound
R1
hCYP11B2 (IC50, nM)
hCYP11B1 (IC50, nM)
B1/B2b
2
3
4
5
6
Cyclopropyl
Me
H
Cl
CN
13
1.2
52
1.2
23
830
37
880
33
63
31
17
27
27
600
a
All IC50’s reported in this table correspond to n P 2, reported as their geometric mean.
b
Ratio of hCYP11B1 IC50/hCYP11B2 IC50
.
Please cite this article in press as: Whitehead B.R., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.003

B.R. Whitehead et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
3
Table 2
Effect of R2 substitution on CYP11B2/CYP11B1 inhibition.
N
N
N
N
N
N
F
F
R2
7-16
17
(S)
(R)
OH
OH
CF3
a:
b:
Me
Me
CF3
Compound
R2
hCYP11B2 (IC50, nM)
hCYP11B1 (IC50, nM)
B1/B2a
2
7
8
9
10
11
12
13
14
15
16
17
CMe2(OH)
a
b
CO2Me
F
CN
CF3
OMe
Me
Ph
13b
830b
460c
250c
640b
1800b
>8700b
590b
530b
280b
300b
490c
100b
63
69
53
>260
71
>55
60
130
85
46
15
6.6c
4.7c
2.5b
25b
160b
10b
4.0b
3.3b
6.6b
33b
CH2NHSO2Et
0.6b
190
a
b
c
Ratio of hCYP11B1 IC50/hCYP11B2 IC50
IC50’s determined as n P 2, reported as their geometric mean.
IC50’s determined as n = 1.
.
Table 3
Effect of R4 and R5 substitution on CYP11B2/ CYP11B1 inhibition in the 5-tertiary alcohol series.a
R5
N
N
N
R4
OH
CF3
Me
Compound
R4
R5
hCYP11B2 (IC50, nM)
hCYP11B1 (IC50, nM)
B1/B2b
7
18
19
F
H
CN
H
F
H
6.6
26
5.8
460
630
220
69
25
37
a
All IC50’s reported in this table correspond to n P 2, reported as their geometric mean.
b
Ratio of hCYP11B1 IC50/hCYP11B2 IC50
.
Table 4
Effect of R3, R4 and R5 substitution on CYP11B2 /CYP11B1 inhibition in the 5-fluoro pyridine series.a
R5
N
N
N
R4
R3
F
Compound
R4
R5
R3
hCYP11B2 (IC50, nM)
hCYP11B1 (IC50, nM)
B1/B2b
10
20
21
22
23
24
F
H
F
H
F
H
H
H
H
H
H
Me
Et
25
98
23
7.3
7.9
6.6
1800
4600
1100
760
930
800
71
47
47
100
120
120
H
Cl
Cl
F
F
a
All IC50’s reported in this table correspond to n P 2, reported as their geometric mean.
b
Ratio of hCYP11B1 IC50/hCYP11B2 IC50
.
Please cite this article in press as: Whitehead B.R., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.003

4
B.R. Whitehead et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Table 5
Table 7
Activity of selected compounds at related CYP targets (IC50 in nM).
Rat pharmacokinetic profile for compound 7a (1 mg kgÀ1 IVb/2 mg kgÀ1 POc).
Compound
CYP17a
CYP19a
CYP21a
CYP3A4b
Parameters
7
>10,000
ND
ND
>10,000
>10,000
5700
7700
2200
790
2300
420
>10,000
>10,000
>10,000
ND
>10,000
>10,000
>50,000
>50,000
3500
>25,000
11,000
2800
Clp (mL minÀ1 kgÀ1
)
23
0.82
0.58
0.61
0.92 0.54
0.7 0.5
47%
10
17
22
23
24
Vdss (L kgÀ1
MRT (h)
t1/2 (h)
)
PO AUCN (
l
M h kg mgÀ1
)
120
Cmax
F%
(lM)
ND not determined.
a
a
b
c
IC50’s determined as n P 2, reported as their geometric mean.
IC50’s determined as n = 1.
Male Wistar Han rats.
n = 2, fasted rats, compound dosed in ethanol/PEG300/ water (10/60/30).
n = 3, fasted rats, compound dosed in 0.5% methylcellulose.
b
Table 6
Metabolic stability of selected compounds in liver microsomes.a
Compound
Human
Rat
Rhesus
(compounds 23, 24) is detrimental to metabolic stability across
multiple species.
7
8
9
10
16
21
22
23
24
88%
96%
ND
93%
92%
90%
93%
88%
37%
38%
33%
1%
70%
31%
77%
80%
56%
2%
31%
31%
0%
29%
45%
37%
87%
46%
0%
To confirm that the liver microsome stability screen can identify
compounds with good pharmacokinetic profiles, we have submit-
ted compound 7 for rat PK determination (Table 7). As expected,
compound 7 displays moderate plasma clearance and good oral
exposure, which would allow the compound to be further profiled
in vivo.
a
In summary, we have demonstrated that imidazopyridines are a
suitable bioisostere of benzimidazoles for inhibition of aldosterone
synthase (CYP11B2). Further optimization of the series improved
activity on CYP11B2 with the necessary selectivity on remaining
CYPs. By studying the SAR on the series, we brought forth a lead
candidate for in vivo studies.
Reported as % parent remaining 30 min after incubating a 1
g/mL of liver microsomes and NADPH.
lM solution with
250
l
groups at the pyridine (compounds 23, 24) has little impact on the
potency at CYP11B1.
It is important to confirm that the CYP11B2 inhibitors progress-
ing into development do not interfere against CYPs that play a role
in androgen and estrogen biosynthesis (CYPs 17, 19, and 21) as
well as hepatic CYPs such as CYP3A4. As shown in Table 5, most
of the compounds in the chemical series are selective against the
related CYP targets. They do not inhibit CYP17 or CYP 21, and for
exemplified compounds such as 7 and 22, there is a good window
(>1000-fold) for the inhibition of CYP19. The erosion of CYP19
selectivity is evident when alkyl groups are introduced into the
4-position of the pyridine (compounds 23, 24), or when the pyri-
dine is replaced by isoquinoline (compound 17). With regard to
the hepatic CYPs, compounds 7 and 22 are selective against CYP
Acknowledgment
The authors thank M.V.S. Rao for conducting the pharmacoki-
netic experiment.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.12.
003.
3A4 (IC50 > 25
lM), CYP 2D6 (IC50 > 25 lM), and CYP 2C9
References
(IC50 = 6, >25 M respectively).
l
To identify development candidates that display reasonable oral
bioavailability, we have screened some of the compounds for their
metabolic stability against liver microsomes (Table 6). Most of the
compounds have good metabolic stability against human micro-
somes. Stereochemistry at the tertiary alcohol has little impact
on the metabolic stability of compound 7 vs 8. Compound 9 has
an excellent in-vitro profile but unfortunately displayed poor
metabolic stability across human, rat and rhesus. Metabolic
stability against rhesus microsomes is moderate for the selected
compounds, which appeared to be improved by extra halogens
on the imidazopyridine ring (compound 10 vs compound 22).
The introduction of alkyl groups in the 4-position of the pyridine
1. Shavit L, Lifschitz MD, Epstein M. Kidney Int. 2012;81:955.
2. Carey RM. Curr Opin Endocrinol Diabetes Obes. 2010;17:194.
3. Tomaschitz A, Pilz S, Ritz E, Obermayer-Pietsch B, Pieber TR. Nat Rev Endocrinol.
2010;6:83.
4. Gilbert KC, Brown NJ. Curr Opin Endocrinol Diabetes Obes. 2010;17:199.
5. Rainey WE. Mol Cell Endocrinol Metab. 1999;151:151.
6. Bureik M, Lisurek M, Bernhardt R. Biol Chem. 2002;383:1537.
7. Hu Q, Yin L, Hartmann RW. J Med Chem. 2014;57:5011.
8. Cerny MA. Curr Top Med Chem. 2013;13:1385.
9. Calhoun DC, White WB, Krum H, et al. Circulation. 2011;124:1945.
10. Azizi M, Amar L, Menard. J Neprhol Dial Transplant. 2013;28:36.
11. Meredith EL, Ksander G, Monovich LG, et al. ACS Med Chem Lett. 2013;4:1203.
12. Papillon JPN, Lou C, Singh AK, et al. J Med Chem. 2015;58:9382.
13. Hoyt SB, Park MK, London C, et al. ACS Med Chem Lett. 2015;6:573.
14. Ali A, Bennett DJ, Cai J, et al. WO201343518, Mar 28; 2013.
Please cite this article in press as: Whitehead B.R., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.12.003