Synthesis and SAR of 1,2-trans-(1-hydroxy-3-phenylprop-1-yl)cyclopentane carboxamide derivatives, a new class of sodium channel blockers

Article abstract of DOI:10.1016/j.bmcl.2005.11.051

Novel cyclopentane-based 3-phenyl-1-hydroxypropyl compounds were evaluated for inhibitory activity against the peripheral nerve sodium channel Na _V1.7 and off-target activity against the cardiac potassium channel hERG. The stereochemistry of the hydroxyl group and substitution on the phenyl rings with either fluorinated O-alkyl or alkyl groups were found to be critical for conferring potency against Na_V1.7. A benchmark compound from this series displayed efficacy in rat models of inflammatory and neuropathic pain.

Full text of DOI:10.1016/j.bmcl.2005.11.051

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1358–1361
Synthesis and SAR of 1,2-trans-(1-hydroxy-3-phenylprop-1-yl)
cyclopentane carboxamide derivatives, a new class
of sodium channel blockers
Dong Ok,a,* Chunshi Li,a Catherine Abbadie,c John P. Felix,b Michael H. Fisher,a
Maria L. Garcia,b Gregory J. Kaczorowski,b Kathryn A Lyons,a William J. Martin,c
Birgit T. Priest,b McHardy M. Smith,b Brande S. Williams,b Matthew J. Wyvratta
and William H. Parsonsa
aDepartment of Medicinal Chemistry, Merck Research Laboratories, Rahway, NJ 07065, USA
bDepartment of Ion Channels, Merck Research Laboratories, Rahway, NJ 07065, USA
cDepartment of Pharmacology, Merck Research Laboratories, Rahway, NJ 07065, USA
Received 29 September 2005; revised 13 November 2005; accepted 14 November 2005
Available online 5 December 2005
Abstract—Novel cyclopentane-based 3-phenyl-1-hydroxypropyl compounds were evaluated for inhibitory activity against the
peripheral nerve sodium channel NaV1.7 and off-target activity against the cardiac potassium channel hERG. The stereochemistry
of the hydroxyl group and substitution on the phenyl rings with either fluorinated O-alkyl or alkyl groups were found to be critical
for conferring potency against NaV1.7. A benchmark compound from this series displayed efficacy in rat models of inflammatory
and neuropathic pain.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Neuropathic pain is defined as chronic pain that arises
from injury to the peripheral or central nervous system
or from diseases that affect the nervous system, such
as diabetes mellitus and HIV. Neuropathic pain is
thought to be associated with hyperexcitability of senso-
ry afferents, leading to abnormal spontaneous firing.1
Voltage-gated sodium channels (VGSCs) have been
shown to accumulate in injured sensory nerves2, and
block of VGSCs inhibits injury-induced spontaneous fir-
ing.3 Several sodium channel blockers are used clinically
in the treatment of neuropathic pain. These include local
anesthetics (e.g., lidocaine), antiarrhythmics (e.g.,
mexiletine) and anticonvulsants (e.g., lamotrigine and
carbamazepine).4
a FRET-based membrane potential assay that mea-
sures the activity of the NaV1.7 channel subtype.
NaV1.7 channels are highly expressed in sensory neu-
rons and are thought to contribute to nociception.6
BPBTS (compound 1) has good in vitro potency,
blocking the inactivated state of the VGSC subtype
NaV1.7 with a Ki of 0.15 lM. When injected locally,
BPBTS dose-dependently inhibited pain behavior in
a rat model of tonic pain. Recently, we reported
the synthesis of CDA547 (compound 2), which has
improved pharmacokinetics compared to that of
BPBTS and is orally active in the spinal nerve liga-
tion (SNL) model, a rat model of neuropathic pain.8
Despite a reasonable pharmacokinetic profile in rats
(F = 44%, Clp = 14 mL/min/kg, T0.5 = 0.98 h), human
liver microsomal (HLM) stability studies with
CDA54 indicated rapid metabolic oxidation of the
N-Me amide side chain, leading to dealkylated
metabolites, and of the biphenylsulfonamide group.
The purpose of the present study was to further de-
fine the key structure–activity relationships (SARs)
of 2 and identify analogs with improved pharmaco-
kinetic properties.
Recently, the structurally novel sodium channel
blocker N-{[20-(aminosulfonyl)biphenyl-4-yl]methyl}-
N0-(2,20-bithien-5-ylmethyl)succinamide (BPBTS)5 was
identified through high-throughput screening, using
*
Corresponding author. Tel.: +1 732 594 7364; fax: +1 732 594
5350; e-mail: dong_ok@merck.com
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.051

D. Ok et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1358–1361
1359
Table 1. Potencies in functional NaV1.7 and hERG binding assays
O
OH
O
O
H
N
SO2
NH2
N
R2
N
H
R1
H
S
O
O
1
2
S
Compound R1
R2
NaV1.7 IC50 MK-0499
(lM)
binding
(% inhibition
at 10 lM)
N
H
N
N
OCF3
2
3
0.29
1.49
>3
54
ndc
ndc
92
89
52
68
60
91
73
87
78
93
70
ndc
SO2NH2
10
11
H
H
2-OCF3
3-OCF3
4-OCF3
4-OCF3
4-OCF3
H
H
0.32
0.23
0.64
0.52
0.36
0.98
1.15
0.44
0.44
0.87
1.21
O
O
O
12
H
13
H
N
H
13Aa
13Bb
14
H
OCF3
H
3
4
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
2-CF3
3,5-di-Cl
3,5-di-Cl
15
16
2-OCF3
3-OCF3
3-OCF3
4-OCF3
4-OCH3
OH
17
18
N
H
R1
R2
19
20
4-SO2CH3 4-OCF3 >3
a 13A, enantiomer A of 13.
b 13B, enantiomer B of 13.
c nd, not determined.
Initial SAR studies began with the removal of the met-
abolically labile biphenylsulfonamide group leading to
the N-Me benzylamide 39, which showed a 5-fold loss
in potency in the functional NaV1.7 assay10, when com-
pared to that of 2 (IC50 ꢀ 1.5 vs 0.29 lM for 2). We next
sought to decrease the peptidyl nature of the compound
by replacing an amide bond in 3 with various surrogates.
Ether, amine, sulfonamide, reverse amide, and hydroxy-
propyl tethers led to a further decrease in potency.
However, phenyl substitution, and modifications of the
right-hand benzyl amide region in molecules containing
a 1-hydroxy-3-phenylprop-1-yl tether produced several
compounds with NaV1.7 potency similar to that of 2.
O
O
c
a, b
X
R1
5a X = MeO
5b X = NMeOMe
6
O
O
CN
CO2H
d
R1
R1
7 (+/-)
8 (+/-)
The synthesis of the compounds in Table 1 is described
in Scheme 1. Commercially available methyl 1-cyclopen-
tene-1-carboxylate (5a) was treated with N,O-di-
methylhydroxylamine hydrochloride and isopropyl
magnesium chloride to give Weinreb amide 5b. This
was reacted with excess Grignard reagent to give ke-
tones 6 (range of yields 53–65% from 5a).11 In a key
step, a high selectivity of the trans stereochemistry of
nitriles 7 (>88% dec) was obtained in P90% yield via
the reaction of 6 with potassium cyanide in refluxing
methanol. The nitriles 7 were then hydrolyzed to the
corresponding acids 8 (concd HCl, 80 ꢁC, 5 h, dioxane)
in quantitative yield.12 The acids 8 were converted to
the benzyl amides 9 using the requisite benzylamine
through BOP-mediated amide bond formation. The car-
bonyl group in 9 was finally reduced with NaBH4 to
provide a set of racemic, diastereomeric alcohols 10–
20. The two diastereomers were readily separated by
standard silica gel chromatography.13 The faster moving
diastereomers were typically more potent against
NaV1.7, and all compounds shown in Tables 1 and 2 re-
fer to the faster moving diastereomer, except for 10
which is a mixture of two diastereomers.
O
O
e
f
N
R2
10 - 20 (+/-)
H
R1
9 (+/-)
Scheme 1. Reagents and conditions: (a) Me(MeO)NH Æ HCl, i-
PrMgCl, THF, 93%; (b) R1-Ph(CH2)2MgBr, ether; (c) KCN, CH3OH,
reflux; (d) concd HCl, dioxane, 80 ꢁC, 5 h, quant.; (e) R2-PhCH2NH2,
HOBT, BOP, i-Pr2NEt; (f) NaBH4, CH3OH, 95%.
The compounds listed in Tables 1 and 2 were screened at
1 lM in the functional NaV1.7 assay. IC50 determina-
tions were obtained on compounds exhibiting >50%
inhibition at 1 lM. In parallel, compounds were exam-
ined for activity on the cardiac potassium channel
hERG using a 35S-labeled MK-0499 binding assay
(MK-0499 binding)14, since blockade of hERG can lead
to delayed cardiac repolarization and cause a potentially
lethal arrhythmia termed torsades de pointes.15
Extensive SAR studies were carried out to assess opti-
mal substitution patterns of the phenyl groups of 4 in
order to maximize NaV1.7 potency, while limiting

1360
D. Ok et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1358–1361
Table 2. Potencies in the NaV1.7 functional assay and the hERG
binding assay
To characterize the individual enantiomers of 13, the race-
mic parent was resolved by chiral HPLC16 to yield 13A
and 13B, as the faster and slower eluting enantiomers,
respectively. Enantiomer B (13B) displayed slightly better
properties with respect to NaV1.7 and hERG activity.
OH
R
To examine the other benzylic amide moiety, a series of
compounds containing modifications on the amide chain
of 14 were synthesized (Schemes 2–4) and tested (Table 2).
The 3,5-di-Cl compound 22, a reverse amide analog of 14,
was practically inactive. However, the OCF3analog of 22
(compound 30) exhibited a 3-fold increase in activity as
compared to 14. In fact, compound 30 is one of the most
potent NaV1.7 blockers prepared in the alcohol series, but
inhibited MK-0499 binding to hERG by 45% and 84% at
1 lM and 10 lM, respectively. The interesting effect of the
OCF3 group is not confined to the reverse amide analog
30, but is also found in the corresponding sulfonamide
analogs. While 3,5-di-Cl sulfonamide analog 29 was de-
void of activity, the OCF3 analog 33 exhibited similar
activity to 14. The SAR also shows that replacement of
the amide with an ether linkage was well tolerated (24
and 31), while the isomeric ether analogs (28 and 32)
showed a greatly diminished activity.
Compound
R
NaV1.7
IC50 (lM)
MK-0499
binding
(% inhibition
at 10 lM)
O
Cl
N
H
14
0.98
91
Cl
O
Cl
Cl
N
H
22
24
28
>3
79
59
40
Cl
Cl
O
1.11
>3
O
Cl
Cl
Cl
Cl
O
O
S
The data in Tables 1 and 2 show that many of the com-
pounds containing a hydroxypropyl tether have similar
potency on NaV1.7 as 2. Apparently, the amide moieties
are not critical determinants of potency, and can be re-
placed with no appreciable loss of activity. However,
N
29
30
31
32
33
H
>3
90
84
65
70
90
0.33
0.82
>3
OTIPS
d, e
NH2
a-c
22(+/-)
7
21
Scheme 2. Reagents and conditions: (a) NaBH4, CH3OH (97%); (b)
TIPSOTf, 2,6-lutidine, CH2Cl2 (90%); (c) H2, Raney Ni, NH3 (50 atm),
CH3OH, 50 ꢁC, 3 h (95%); (d) 3,5-Cl2C6H3COOH, HOBT, BOP,
i-Pr2NEt, CH2Cl2; (e) n-Bu4NF, THF.
0.63
hERG activity (Table 1). Block of NaV1.7 occurred with
different substituents at various positions in the aromat-
ic rings, but most compounds also had substantial
hERG binding activity.
OTIPS
e, f
a-d
24(+/-)
OH
8
23
In general, potency of the unsubstituted lead 10 was in-
creased by substitution at various positions on the phen-
yl rings with an OCF3 group, as illustrated by analogs
11–13, which all exhibited improved NaV1.7 activity
compared to that of 10. The 3,5-di-Cl analog (14) also
resulted in an increase in potency over 10, suggesting
that these substituents may contribute to hydrophobic
interaction surfaces. However, simultaneous substitu-
tions on both phenyl rings generally led to a loss of
activity as compared to substitutions with a single
OCF3 group (for example, compare 16 and 12), pointing
to potential steric effects. In contrast to –CF3 and
–OCF3 substituents, 4-methoxy 19 or sulfone 20 substi-
tutions were not well tolerated.
Scheme 3. Reagents and conditions: (a) HCl (g), CH3OH (92%); (b)
NaBH4, CH3OH; (c) TIPSOTf, 2,6-lutidine, CH2Cl2; (d) LiAlH4, ether
(87%); (e) NaH, 3,5-Cl2C6H5CH2Br, DMF; (f) n-Bu4NF, THF.
O
OH
O
O
f-l
a, b
TIPSO
c-e
O
TIPSO
28 (+/-)
O
25
26
27
Scheme 4. Reagents and conditions: (a) LiAlH4, ether (87%); (b)
TIPSCl, imidazole, CH2Cl2 (95%) and separation of cis and trans
isomers (approximately 1:1); (c) NaH, CH2@CHCH2Br, DMF (70%);
(d) 4-Methylmorpholine N-oxide, OsO4, CH2Cl2; (e) Pb(OAc)4,
pyridine, CH3OH; (f) 3,5-Cl2C6H3MgBr, ether; (g) H2, Pd/C, CH3OH
(92%); (h) n-Bu4NF, THF; (i) Jones reagent, acetone (80%);
(j) Me(MeO)NHHCl, HOBT, BOP, i-Pr2NEt, CH2Cl2; (k)
PhCH2CH2MgCl, THF; (l) NaBH4, CH3OH.
Of the compounds in Table 1, 13 displayed the greatest
selectivity for block of NaV1.7 over binding to hERG.

D. Ok et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1358–1361
Table 3. Pharmacokinetic properties and anti-allodynic efficacy
1361
References and notes
2
13
13
40
13B
1. Waxman, S. G.; Cummins, T. R.; Dib-Hajj, S.; Fjell, J.;
Black, J. A. Muscle Nerve 1999, 22, 1177.
2. England, J. D.; Happel, L. T.; Kline, D. G.; Gamboni, F.;
Thouron, C. L.; Liu, Z. P.; Levinson, S. R. Neurology
1996, 47, 272.
3. Liu, X.; Zhou, J. L.; Chung, K.; Chung, J. M. Brain Res.
2001, 900, 119.
4. (a) Anger, T.; Madge, D. J.; Mullar, M.; Riddall, D.
J. Med. Chem. 2001, 44, 115; (b) Mao, J.; Chen, L. L. Pain
2000, 87, 7.
5. Priest, B. T.; Garcia, M. L.; Middleton, R. E.; Brochu, R.
M.; Clark, S.; Dai, G.; Dick, I. E.; Felix, J. P.; Liu, C. J.;
Reiseter, B. S.; Schmalhofer, W. A.; Shao, P. P.; Tang, Y.
S.; Chou, M. Z.; Kohler, M. G.; Smith, M. M.; Warren, V.
A.; Williams, B. S.; Cohen, C. J.; Martin, W. J.; Meinke,
P. T.; Parsons, W. H.; Wafford, K. A.; Kaczorowski, G. J.
Biochemistry 2004, 43, 9866.
HLM stabilitya
% remaining after 1 h
4
49
Pharmacokinetics (1 mg/kg iv)
Oral bioavailability (%)
Clp (mL/min/kg)
44
65
14.0
1.0
1.0
1.8
9.0
9.0
1.6
2.3
9.0
T0.5 (h)
b
1.6
1.9
5.3
Cmax (lM)
AUCb (lM/h)
Anti-allodynic efficacy (3 mg/kg po)
Maximal reversal (%) 43
25
31
a Human liver microsomal preparation.
bCmax and bAUC determined after oral dosing with 2 mg/kg for 2 or
3 mg/kg for 13 and 13B.
6. Djouhri, L.; Newton, R.; Levinson, S. R.; Berry, C. M.;
Carruthers, B.; Lawson, S. N. J. Physiol. 2003, 546, 565.
7. Shao, P. P.; Ok, D.; Fisher, M. H.; Garcia, M. L.;
Kaczorowski, G. J.; Li, C.; Lyons, K. A.; Martin, W. J.;
Meinke, P. T.; Priest, B. T.; Smith, M. M.; Wyvratt, M. J.;
Ye, F.; Parsons, W. H. Bioorg. Med. Chem. Lett. 2005, 15,
1901.
8. Kim, S. H.; Chung, J. M. Pain 1992, 50, 355.
9. Compound 3 was prepared according to a method for the
synthesis of 2 described in Ref. 7.
NaV1.7 activity is strongly influenced by the nature of
the substituents on both phenyl rings. The alcohols in
Tables 1 and 2 are a structurally simplified design and
lack one of the metabolic liabilities of compound 2.
Compounds 13 and 13B were chosen to examine the
effect of the hydroxypropyl substitution on the rat
pharmacokinetic profile (Table 3).
Compounds 13 displayed a slightly greater stability than
compound 2 in human liver microsome incubation stud-
ies and stability was greatly improved for 13B. Oral bio-
availability of 13 was comparable to that of 2, whereas
the single enantiomer 13B displayed improved bioavail-
ability, Cmax, and AUC. Compounds 13 and 13B
showed reduced clearance rates and increased half-lives
compared to 2, suggesting that removal of the N-Me
amide did indeed eliminate a site for metabolism. Com-
pounds 2, 13, and 13B were evaluated for anti-allodynic
efficacy in a rat model of chronic pain. In the CFA
model (intradermal injection of complete Freundꢀs adju-
vant)17, 13 and 13B significantly reversed CFA-induced
allodynia but failed to show an improvement in efficacy
over 2, possibly owing to their somewhat lower in vitro
potency against NaV1.7, although other factors such as
local tissue concentration or protein binding cannot be
ruled out.
10. Felix, J. P.; Williams, B. S.; Priest, B. T.; Brochu, R. M.;
Dick, I. E.; Warren, V. A.; Yan, L.; Slaughter, R. S.;
Kaczorowski, G. J.; Smith, M. M.; Garcia, M. L. Assay
Drug Dev. Technol. 2004, 2, 260.
11. Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini,
G.; Dolling, U.-H.; Grabowski, E. J. J. Tetrahedron Lett.
1995, 31, 5461.
12. The stereochemistry of the cyclopentane has been
unequivocally confirmed to be trans by an independent
synthesis of 34, which was prepared in a 4-step procedure
from commercially available trans-DL-1,2-cyclopentane
dicarboxylic acid.
O
O
CO2H
a, b
HO2C
N
N
O
H
OCF3
OCF3
O
OH
c, d
N
H
In summary, replacement of the N-Me amide group of
the diamide 2 with an alcohol moiety has afforded a
number of potent amide–alcohol hybrids with in vitro
activity comparable to that of diamide 2 and with an
improved pharmacokinetic profile. Two benchmark
amide–alcohols are efficacious in a rat model of chronic
pain, and additional experiments will determine the
therapeutic potential of these compounds for the treat-
ment of chronic pain.
34
Synthesis of 34: (a) CDI, C6H5CH2NH2, THF; (b)
Me(MeO)NH Æ HCl, HOBT, BOP, i-Pr2NEt; (c)
PhCH2CH2MgBr, ether; (d) NaBH4, CH3OH.
13. Spectral (1H NMR, LC–MS) data on all intermediates and
final compounds were consistent with the proposed
structures.
14. Wang, J.; Della-Penna, K.; Wang, H.; Karczewski, J.;
Conolly, T. M.; Koblan, K. S.; Bennett, P. B.; Salata, J. J.
Am. J. Physiol. Heart Circ. Physiol. 2003, 284, H256.
15. Tamargo, J. Jpn. J. Pharmacol. 2000, 83, 1.
16. Column used: ChiralCel OD 4.6 · 250 mm, 10 lm. UV
detection at 300 nm. Elution (.75 ml/min) with heptane:
isopropyl alcohol (90:10) afforded successively the enan-
tiomers 13A (7.5 min) and 13B (8.9 min).
Acknowledgments
The authors thank Dr. Joseph L. Duffy for critically
reading the manuscript. We further thank Erin McGo-
wan, Xiaohua Li, Feng Ye, Pengcheng P. Shao, William
A. Schmalhofer, and Vivien A. Warren for their out-
standing technical contributions to this work.
17. Gould, H. J., 3rd; Gould, T. N.; Paul, D.; England, J. D.;
Liu, Z. P.; Reeb, S. C.; Levinson, S. R. Brain Res. 1999,
824, 296.