Synthesis and evaluation of carbon-linked analogs of dual orexin receptor antagonist filorexant

Article abstract of DOI:10.1016/j.bmcl.2014.02.026

Analogs of the dual orexin receptor antagonist filorexant were prepared. Replacement of the ether linkage proved highly sensitive toward modification with an acetylene linkage providing compounds with the best in vitro and in vivo potency profiles.

Full text of DOI:10.1016/j.bmcl.2014.02.026

Bioorganic & Medicinal Chemistry Letters 24 (2014) 1784–1789
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of carbon-linked analogs of dual orexin
receptor antagonist filorexant
a
a
b
b
Scott D. Kuduk a, , Jason W. Skudlarek , Christina N. Di Marco , Joseph G. Bruno , Mark A. Pausch ,
Julie A. O’Brien b, Tamara D. Cabalu c, Joanne Stevens d, Joseph Brunner d, Pamela L. Tannenbaum d,
Anthony L. Gotter e, Christopher J. Winrow e, John J. Renger e, Paul J. Coleman a
⇑
a Department of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
b Department of In Vitro Pharmacology, Merck Research Laboratories, West Point, PA 19486, USA
c Department of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486, USA
d Department of In Vivo Pharmacology, Merck Research Laboratories, West Point, PA 19486, USA
e Department of Neuroscience, Merck Research Laboratories, West Point, PA 19486, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Analogs of the dual orexin receptor antagonist filorexant were prepared. Replacement of the ether linkage
proved highly sensitive toward modification with an acetylene linkage providing compounds with the
best in vitro and in vivo potency profiles.
Received 18 January 2014
Revised 7 February 2014
Accepted 10 February 2014
Available online 19 February 2014
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Orexin
Insomnia
Antagonist
The orexin neuropeptides and their receptors have a demon-
strated role in regulating sleep/wake state and recent research
has generated significant interest in targeting the orexin receptors
for related disorders. The characterization of a novel neuropeptide
secreted from the hypothalamus, naming the peptide hypocretin1
and orexin,2 was achieved via genetic and biochemical approaches,
with the orexin naming most frequently utilized.3 Orexin
neuropeptides (OX-A and OX-B) are derived from orexinergic
neurons localized in the lateral hypothalamus, and signal through
two G-protein coupled receptors (GPCR’s), Orexin Receptor 1
(OX1R) and Orexin Receptor 2 (OX2R).3
These receptors are disseminated in key brain regions responsi-
ble for governing wake, vigilance and reward seeking behaviors.
Orexin signaling is most active during the normal wake period
and falls silent during the sleep period.4–6 Several companies have
identified dual orexin receptor antagonists (DORAs) targeting both
OX1R and OX2R, or selective orexin receptor antagonists (SORAs),
and many have advanced through late stage clinical develop-
ment.7–10 Numerous reviews are available in this context.11–14
Two DORAs, suvorexant and filorexant, have been discovered
and advanced into clinical development at Merck (Fig. 1).
Filorexant was most recently described as a potent DORA that is
efficacious in promoting sleep in rats and dogs, and has been
reported to be in Phase II trials. Limited SAR data has been reported
for filorexant, mostly to the carboxamide and fluoro-pyridine
regions of the molecule.15 In this Letter, we describe SAR analysis
of the role of the ether linkage between the piperidine central ring
Cl
N
O
N
N
N
N
N
O
Me
Me
Suvorexant
Me
Me
O
N
O
N
N
N
F
Filorexant
⇑
Corresponding author. Tel.: +1 215 652 5147; fax: +1 215 652 3971.
E-mail address: scott_d_kuduk@merck.com (S.D. Kuduk).
Figure 1.
http://dx.doi.org/10.1016/j.bmcl.2014.02.026
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

S. D. Kuduk et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1784–1789
1785
and the fluoropyridine, with an emphasis of all carbon linkages
(ethyl) with varied degrees of flexibility (sp3, sp2, and sp
hybridization).
The synthetic route used to access carbon-linked target
molecules is shown in Schemes 1–3. Coupling of triazole carboxylic
acid 6 and piperidinol 7 was mediated by EDC followed by
Me
Me
Me
Me
Me
Me
O
N
OH
N
b
a
OH
+
HN
OH
N
O
N
O
N
N
N
N
N
O
N
N
8
9
7
6
c
Me
Me
N
Me
O
O
O
P
Me
Me
N
Me
O
N
O
+
+
-
N
N
Me
N
O
N
N
N
N
11b
11a
10
2: 1 ratio
Scheme 1. (a) EDC, HOBt, Hunig’s base, DMF. (b) Diacetoxyiodobenzene, TEMPO, CH2Cl2. (c) Potassium carbonate, 10, MeOH.
Me
Me
Me
Me
c
b
a
PMBN
O
HN
OH
PMBN
OH
PMBN
15
14
13
7
d
Me
Me
N
Me
HN
e
f
PMBN
Me
O
N
N
F
N
N
N
F
N
16
17
5a (b-n)
F
Scheme 2. (a) p-Anisaldehyde, NaBH3CN, MeOH. (b) Oxalyl chloride, DMSO, Et3N, CH2Cl2. (c) Potassium carbonate, 10. (d) 2-Bromo-5-fluoropyridine, Pd(PPh3P)4, CuI, Et3N.
(e) (1) 1-Chloroethyl chloroformate, Et3N, CH2Cl2; (2) MeOH, reflux. (f) EDC, HOAt, Hunig’s base, DMF.
F
Me
F
Me
Me
N
a
BocN
PPh3Br
Me
O
BocN
BocN
+
N
N
20a
F
19
18
20b
b
1: 2.5
Me
Me
HN
N
c
N
O
N
F
N
N
N
F
21
3
Scheme 3. (a) NaH, THF, 0 °C-rt. (b) HCl, dioxane. (C) 6, EDC, HOAt, Hunig’s base.

1786
S. D. Kuduk et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1784–1789
oxidation of the alcohol to provide aldehyde 9.16 Alkynylation
using the Bestmann reagent 1017 provided alkynes 11a and 11b,
albeit in an undesirable 2:1 ratio of cis/trans isomers due to
epimerization of the aldehyde under the reaction conditions. The
aldehyde group in 9 is likely sensitive to epimerization due to
being in a pseudo-axial position due to the 1,3-allylic strain
imposed by the 2-methyl group and the arylcarboxamide.15
In order to circumvent this issue, an alternate protecting group
strategy was investigated that would avoid the 1,3-allylic strain
and bias both the 2- and 5-substituents in the equatorial position
to avoid epimerization. Toward this end, piperidinol 7 underwent
reductive amination with para-anisaldehyde to provide PMB
protected alcohol 13. Oxidation, followed by alkynylation provided
15 as a single trans isomer. Sonogashira cross-coupling with 2-
iodo-4-fluoropyridine provided 16. Subsequent removal of the
PMB group and amide coupling provided alkyne 5a. Analogs of
5a were prepared via this route using different aryl iodides or
bromides in with 15. Hydrogenation of 5a with Pd/C in MeOH
generated ethyl linked analog 2 while Lindlar reduction provided
cis-olefin 4 (not shown).
Lastly, Wittig olefination with phosphonium salt 18 and
aldehyde 19 provided 20a and 20b, as a separable mixture, in a ra-
tio of 1:2.5. Removal of the protecting group in 20a was followed
by coupling with acid 6 to deliver trans-olefin linker analog 3.
Having prepared the targeted new carbon-carbon linker modi-
fied analogs, evaluation in primary in vitro assays was conducted.
Compounds were evaluated for OX2R and OX1R potency through an
in vitro radioligand binding assay to membranes from human cells
overexpressing the orexin receptor and reported as Ki. In addition,
a cell-based functional FLIPR (fluorometric imaging plate reader)
assay in which Ca2+ flux is measured as a functional determinant
of orexin binding, was also conducted.18 Results for compounds
1–5a are shown in Figure 2.
Relative to potent DORA comparator 1 containing the ether
linkage found in filorexant, the direct oxygen for methylene
replacement in the form of ethyl analog 2 lost considerable affinity
in the binding assay (>10Â) and potency in the functional assay,
highlighting a key role for the oxygen heteroatom. The effect was
more pronounced for OX1R in both assays. The trans-olefin 3
retained considerably more potency at OX2R in the functional
assay with an IC50 = 34 nM, but with an ꢀ8-fold decrease at
OX1R, similar to that seen with ethyl 2. The corresponding cis-olefin
linkage 4 was less effective than the trans providing binding and
FLIPR data similar to that for ethyl compound 2. Lastly, alkyne 5a
provided very high binding affinity at both sub-types (Ki = 1.1 nM)
and potent functional activity at OX2R (IC50 = 24 nM) and OX1R
(IC50 = 31 nM). Based on this promising potency and dual activity
profile, additional SAR evaluation was conducted with the alkyne
moiety in place.
In order to more fully scope out the SAR using the new alkynyl
modification, examination of the role of the pyridine ring and sub-
stituents was investigated as shown in Table 1. Removal of the
fluorine atom (5b) led to a 2-3-fold loss in functional activity,
but a more pronounced effect (ꢀ10Â) was noted in the binding
assay. Pyridine isomer 5c lost considerable binding affinity relative
to 5b, while 4-pyridyl 5d proved similar.
Alkyne 5a was highly lipophilic (HPLC logD = 3.7)19 so addi-
tional polar substituents on the pyridine such as hydroxyl groups
were investigated. Hydroxy pyridine 5e provided balanced binding
affinity at both receptor sub-types (Ki = 24 nM), while isomer 5f
showed preferred binding at OX2R over OX1R (Ki’s = 7.5 and
866 nM, respectively), and this preference was noted in the
Me
Me
Me
Me
Me
O
Me
O
N
O
N
N
N
N
N
O
N
N
N
F
F
N
N
N
N
F
N
N
3
1
2
hOX2R Ki = 3.0 nM
hOX1R Ki = 17 nM
hOX2R IC50 = 34 nM
hOX1R IC50 = 140 nM
hOX2R Ki = 0.2 nM
hOX1R Ki = 1 nM
hOX2R IC50 = 15 nM
hOX1R IC50 = 16 nM
hOX2R Ki = 26 nM
hOX1R Ki = 470 nM
hOX2R IC50 = 145 nM
hOX1R IC50 = 861 nM
F
Me
Me
N
Me
Me
O
N
N
N
N
O
N
N
N
N
N
F
4
5a
hOX2R Ki = 28 nM
hOX2R Ki = 1.1 nM
hOX1R Ki = 1.1 nM
hOX2R IC50 = 24 nM
hOX1R IC50 = 31 nM
hOX1R Ki = 135 nM
hOX2R IC50 = 417 nM
hOX1R IC50 = 706 nM
Figure 2.

S. D. Kuduk et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1784–1789
1787
Table 1
Me
Me
N
N
O
R
N
N
L#
5a
R
OX2R BINDa
1.1
OX1R BINDa
1.1
OX2R FLIPRa
24
OX1R FLIPRa
31
N
N
F
5b
5c
5d
5e
11
41
7.0
24
27
42
28
nd
nd
100
83
107
26
N
nd
nd
N
24
N
HO
HO
N
5f
7.5
866
211
52
nd
3878
nd
5g
130
N
N
OH
OH
5h
5.6
18
26
59
OH
5i
5j
42
76
nd
nd
nd
nd
N
ꢀ150
339
OH
OH
N
N
5k
5l
32
108
nd
nd
OH
OH
25
75
100
6.3
298
6.3
N
5m
0.53
0.56
a
Values represent the average of n P 2 experiments.
functional assay as well. Adding a hydroxy methyl group on the
2-pyridyl was examined (5g–i) and SAR indicated the position
of the group significantly affected the binding affinity. Analog
5h with the hydroxymethyl at the 3-position was most tolerant
in terms of binding with a ꢀ5–15 fold decrease, while functional
potency (OX2R IC50 = 26 nM, OX1R IC50 = 59 nM) was similar to
parent alkyne 5a. The 3-hydroxymethyl analogs in the context
of the 3-pyridine (5j,k) showed weaker binding affinity than the
2-pyridyl counterpart 5i. Also, the 4-pyridyl-3-hydroxymethyl 5l
was substantially weaker in binding and FLIPR. Lastly, a 3-
hydroxylmethyl analog with the pyridine nitrogen absent in the
form of phenyl analog 5m was also evaluated. This compound
proved to be a very potent DORA in the FLIPR assay (OXR
IC50 = 6.3 nM) with highest affinity (OX2R Ki = 0.53 nM, OX1R
Ki = 0.56 nM) of any analog prepared, but was also an inhibitor
of CYP3A4 (IC50 ꢀ4 uM).
While hydroxymethylpyridine 5h did have the desirable effect
of decreasing the lipophilicity (HPLC logD = 2.1), and maintained
good functional potency relative to 5a, it proved to be a substrate
for the CNS efflux transporter P-glycoprotein (P-gp).20 Fluoropyri-
dine 5a gave the best overall profile in terms of binding, functional
potency, and P-gp properties (Fig. 3). Accordingly, more advanced
characterization was conducted to see if advantages could be
realized over 5-ether linked DORAs.

1788
S. D. Kuduk et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1784–1789
To evaluate its in vivo pharmacodynamics and selectivity, DORA
Me
N
5a was examined in a mouse sleep model (Fig. 4). Both wild type
and OX2R knock-out mice surgically implanted with radio teleme-
try transmitters measuring electroencephalography (EEG) and
electromyogram (EMG) activity to determine vigilance state over
time as previously described.22 In wildtype mice compound 5a
dosed orally at 100 mg/kg promoted sleep with sleep stage archi-
tecture consistent with other orexin receptor antagonists across
species.21,22 Significant active wake reduction was observed for
2 h following treatment accompanied by corresponding increases
in delta and REM sleep. These responses are qualitatively similar
to that induced by DORA-22, an analog of filorexant evaluated in
mice, where active wake reductions are accompanied by propor-
tionally similar slow wave sleep and REM increases.22 While
compound 5a, promotes sleep to a similar magnitude as DORA-
22, its effects are more rapid and brief; it attenuates active wake
and increases REM and delta sleep to a greater extent during the
first 30 min after treatment relative to DORA-22 used at the same
dose (100 mg/kg), and the duration of sleep promoting effects are
shorter, lasting for 2 h relative to 4 h induced by DORA-22.22
Notably, the sleep promoting effects of compound 5a were largely
Me
O
N
N
N
N
F
5a
HPLC logD = 3.7
HPLC sol (pH 7) = 124 μM
PPB = 99.1 (h), 96.1 (d)
Dog IV PK: Cl = 1.6 mL/min/kg,
Vdss = 0.26, t1/2 = 2.0 h
hERG = 23 μΜ
CYP inhibition: 3A4 > 50 μM,
2C9: 32 μM, CYP2D6 > 50 μM
Figure 3.
Figure 4. Sleep promoting responses to compound 5a are largely due to activity at OX2R. Wildtype (left, n = 5) and OX2R knockout mice (right, n = 6) were treated orally with
vehicle (closed symbols, 20% Vit. E TPGS) and 100 mg/kg compound 5a (open symbols) in a balanced cross-over paradigm. Dose time (arrow) occurred during the dark, or
active phase 4 h prior to lights on (grey, white horizontal bars, bottom). Mean time in vigilance state during 30 min analysis intervals following vehicle and compound
treatment conditions were compared using a linear mixed effects model for repeated measures t-test where significant responses are indicated with grey lines where tick
marks indicate significance level (short, medium, long, p <0.05, 0.01, 0.001).

S. D. Kuduk et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1784–1789
1789
7. Bettica, P.; Squassante, L.; Zamuner, S.; Nucci, G.; Danker-Hopfe, H.; Ratti, E.
Sleep 2012, 35, 1097.
8. Hoever, P.; de Haas, S. L.; Dorffner, G.; Chiossi, E.; van Gerven, J. M.;
Dingemanse, J. J. Psychopharmacol. 2012, 26, 1071.
9. Herring, W. J.; Snyder, E.; Budd, K.; Hutzelmann, J.; Snavely, D.; Liu, K.; Lines, C.;
Roth, T.; Michelson, D. Neurology 2012, 79, 2265.
10. Kuduk, S. D.; Winrow, C. J.; Coleman, P. J. Annu. Rep. Med. Chem. 2013, 48, 73.
11. Coleman, P. J.; Cox, C. D.; Roecker, A. J. Curr. Top. Med. Chem. 2011, 11, 696.
12. Rufoff, C.; Cao, M.; Guilleminault, C. Curr. Pharm. Des. 2011, 17, 1476.
13. Gotter, A. L.; Roecker, A. J.; Hargreaves, R.; Coleman, P. J.; Winrow, C. J.; Renger,
J. J. Prog. Brain Res. 2012, 198, 163.
absent in mice lacking OX2 receptors which not only demonstrates
that these effects of compound 5a are mediated through orexin
receptors, but is also consistent with OX2R being the primary
mediator of orexin-induced arousal.23,24 Average plasma exposure
in satellite animals was 60
9 lM-h.
While acetylene 5a was not an inhibitor of CYP3A4 as noted for
5m, it proved to be a potent time-dependent 3A4 inhibitor (with
an IC50 shift ratio >6.4). In fact, it was found that all acetylene
containing compounds prepared in Table 1 are time dependent
inhibitors highlighting a potential for drug–drug interactions.
Accordingly, further activities involving acetylene linked orexin
antagonists described herein were ceased.
In summary, the SAR of carbon replacements of the piperidine
ether region of the DORA filorexant was reported. This region could
be replaced with acetylene linkages as represented by 5a, which
was a potent DORA that showed pharmacological activity in a
mouse sleep EEG model. While this structural change also led to
undesirable time-dependent inhibition of CYP3A4 leading to a no
go for this series, additional SAR evaluation of this region of the
filorexant remains and will be reported in due course.
14. Mieda, M.; Sakurai, T. CNS Drugs 2013, 27, 83.
15. Coleman, P. J.; Schreier, J. D.; Cox, C. D.; Breslin, M. J.; Whitman, D. B.; Bogusky,
M. J.; McGaughey, G. B.; Bednar, R. A.; Lemaire, W.; Doran, S. M.; Fox, S. V.;
Garson, S. L.; Gotter, A. L.; Harrell, C. M.; Reiss, D. R.; Cabalu, T. D.; Cui, D.;
Prueksaritanont, T.; Stevens, J.; Tannenbaum, P. L.; Ball, R. G.; Stellabott, J.;
Young, S. D.; Hartman, G. D.; Winrow, C. J.; Renger, J. J. Chem. Med. Chem. 2012,
7, 415.
16. Cox, C. D.; Breslin, M. J.; McGaughey, G. B.; Whitman, D. B.; Schreier, J. D.;
Bogusky, M. J.; Roecker, A. J.; Mercer, S. P.; Bednar, R. A.; Lemaire, W.; Bruno, J.
G.; Reiss, D. R.; Harrell, C. M.; Murphy, K. L.; Garson, S. L.; Doran, S. M.; Koblan,
K. S.; Prueksaritanont, T.; Anderson, W. B.; Tang, C.; Roller, S.; Cabalu, T. D.; Cui,
D.; Ball, R. G.; Hartman, G. D.; Winrow, C. J.; Renger, J. J.; Coleman, P. J. J. Med.
Chem. 2010, 53, 5320.
17. Müller, S.; Liepold, B.; Roth, G.; Bestmann, H. J. Synlett 1996, 521.
18. Bergman, J. M.; Roecker, A. J.; Mercer, S. P.; Bednar, R. A.; Reiss, D. R.; Ransom,
R. W.; Harrell, C. M.; Pettibone, D. J.; Lemaire, W.; Murphy, K. L.; Li, C.;
Preuksaritanont, T.; Winrow, C. J.; Renger, J. J.; Koblan, K. S.; Hartman, G. D.;
Coleman, P. J. Bioorg. Med. Chem. Lett. 2008, 18, 1425.
Acknowledgements
19. The high throughput (HT) HPLC logD (pH 7) value was determined on and
Agilent 1200 HPLC/DAD system and ChemStation software, both from Agilent
Technologies, USA. Separations are carried out on a Poroshell 120 EC-C18,
We thank Mr. Yunfu Luo (WuXi AppTec) for assistance in the
preparation of intermediates 15 and 20.
30 mm  3.0 mm I.D., 2.7
lm, (Agilent Technologies, USA). The mobile phase
consists of phosphate buffered saline at pH 7 (mobile phase A) and acetonitrile
(mobile phase B). The chromatographic system is calibrated with a set of
standards with published shake-flask logD values. Linear regression is used to
determine the calibration line relating the retention time to logD for the
calibration standards. This line is then used to determine the HT HPLC logD
(pH 7) value of API from the measured retention time by the HPLC/DAD
analysis of the API solution.
References and notes
1. de Lecea, L.; Kilduff, T. S.; Peyron, C.; Gao, X. B.; Foye, P. E.; Danielson, P. E.;
Fukuhara, C.; Battenberg, E. L. F.; Gautvik, V. T.; Bartlett, F. S.; Frankel, W. N.;
van den Pol, A. N.; Bloom, F. E.; Gautvik, K. M.; Sutcliffe, J. G. PNAS 1998, 95, 322.
2. Sakurai, T.; Amemiya, A.; Ishii, M.; Matsuzaki, I.; Chemelli, R. M.; Tanaka, H.;
Williams, S. C.; Richardson, J. A.; Kozlowski, G. P.; Wilson, S.; Arch, J. R. S.;
Buckingham, R. E.; Haynes, A. C.; Carr, S. A.; Annan, R. S.; McNulty, D. E.; Liu, W.
S.; Terrett, J. A.; Elshourbagy, N. A.; Bergsma, D. J.; Yanagisawa, M. Cell 1998, 92,
573.
3. Gotter, A. L.; Webber, A. L.; Coleman, P. J.; Renger, J. J.; Winrow, C. J. Pharmacol.
Rev. 2012, 64, 389.
4. Grady, S. P.; Nishino, S.; Czeisler, C. A.; Hepner, D.; Scammell, T. E. Sleep 2006,
29, 295.
5. Taheri, S.; Sunter, D.; Dakin, C.; Moyes, S.; Seal, L.; Gardiner, J.; Rossi, M.;
Ghatei, M.; Bloom, S. Neurosci. Lett. 2000, 279, 109.
6. Xie, X. M.; Wisor, J. P.; Hara, J.; Crowder, T. L.; LeWinter, R.; Khroyan, T. V.;
Yamanaka, A.; Diano, S.; Horvath, T. L.; Sakurai, T.; Toll, L.; Kilduff, T. S. J. Clin.
Invest. 2008, 118, 2471.
20. Hitchcock, S. A. J. Med. Chem. 2012, 55, 4877.
21. Brisbare-Roch, C.; Dingemanse, J.; Koberstein, R.; Hoever, P.; Aissaoui, H.;
Flores, S.; Mueller, C.; Nayler, O.; van Gerven, J.; deHaas, S.; Hess, P.; Qiu, C.;
Buchmann, S.; Scherz, M.; Weller, T.; Fischli, W.; Clozel, M.; Jenck, F. Nat. Med.
2007, 13, 150.
22. Winrow, C. J.; Gotter, A. L.; Cox, C. D.; Tannenbaum, P. L.; Garson, S. L.; Doran, S.
M.; Breslin, M. J.; Schreier, J. D.; Fox, S. V.; Harrell, C. M.; Stevens, J.; Reiss, D. R.;
Cui, D.; Coleman, P. J.; Renger, J. J. Neuropharmacology 2012, 62, 978.
23. Dugovic, C. l.; Shelton, J.; Sutton, S.; Yun, S.; Li, X.; Dvorak, C.; Carruthers, N.;
Atack, J.; Lovenberg, T. Sleep 2008, 31, A33.
24. Morairty, S. R.; Revel, F. G.; Malherbe, P.; Moreau, J. L.; Valladao, D.; Wettstein,
J. G.; Kilduff, T. S.; Borroni, E. PLoS ONE 2012, 7, 39131.