Synthesis and SAR of novel 2-arylthiazolidinones as selective analgesic N-type calcium channel blockers

Article abstract of DOI:10.1016/j.bmcl.2006.10.098

A series of new N-type (Ca_v2.2) calcium channel blockers derived from the 'hit' structures 2-(3-bromo-4-fluorophenyl)-3-(2-pyridin-2-ylethyl)thiazolidin-4-one 9 and its 2-[4-(4-bromophenyl)pyridin-3-yl]-3-isobutyl analogue 10 is described. Exte

Full text of DOI:10.1016/j.bmcl.2006.10.098

Bioorganic & Medicinal Chemistry Letters 17 (2007) 662–667
Synthesis and SAR of novel 2-arylthiazolidinones as selective
analgesic N-type calcium channel blockers
Lars J. S. Knutsen,* Christopher J. Hobbs, Christopher G. Earnshaw, Andrea Fiumana,
Jenny Gilbert, Sarah L. Mellor, Fleur Radford, Nichola J. Smith, Philip J. Birch,
J. Russell Burley, Stuart D. C. Ward and Iain F. James
Ionix Pharmaceuticals Ltd,ꢀ 418 Cambridge Science Park, Milton Road, Cambridge CB4 0PA, UK
Received 23 June 2006; revised 30 October 2006; accepted 31 October 2006
Available online 6 November 2006
Abstract—A series of new N-type (Cav2.2) calcium channel blockers derived from the ‘hit’ structures 2-(3-bromo-4-fluorophenyl)-3-
(2-pyridin-2-ylethyl)thiazolidin-4-one 9 and its 2-[4-(4-bromophenyl)pyridin-3-yl]-3-isobutyl analogue 10 is described. Extensive
SAR studies using a range of synthetic approaches resulted in novel, patented compounds with IC50 values of up to 0.2 lM in
an in vitro IMR32 assay, and selectivities for N/L of up to 30-fold. The new compounds described have potential in treatment
of neuropathic pain.
Ó 2006 Elsevier Ltd. All rights reserved.
Ion channel blockers are attractive drug targets for an
expanding range of therapeutic indications1 and volt-
age-dependent Ca2+ channels play important roles in
critical biological processes.2 Based on their pharmaco-
logical and electrophysiological properties the channels
have been classified into several subtypes: N- (Cav2.2),
L- (Cav1.1-Cav1.4), P/Q- (Cav2.1), R- (Cav2.3) and T-
type (Cav3.1-Cav3.3). These targets for therapeutic inter-
vention play specialised roles in cellular function.3,4
of N-type activity is important in the management of
pain.6
The recently approved pain drug Ziconotide (PrialtTM) is
a potent blocker of N-type Ca2+ channels.7 Pre-clinical
and clinical studies of Ziconotide conclude that selective
blockade of the N-type channel is effective in reducing
inflammatory and neuropathic pain in humans.
The market for Ziconotide may be restricted by the need
for intrathecal delivery of this new peptide drug, and by
a range of neurological and cardiovascular side-effects.8
N-type channels are located primarily at pre-synaptic
nerve terminals and mediate spinal transmission of pain
signals from the periphery to the central nervous system
(CNS) by modulating release of nociceptive neuro-trans-
mitters and neuropeptides.4 Selective Ca2+ channel
blockers in particular are now emerging as prospective
therapeutics for the treatment of neuropathic and
inflammatory pain.5 Mice lacking N-type channels show
suppressed response to painful stimuli that induce
inflammation, and show reduced neuropathic pain
symptoms. This provides evidence that N-type channels
may be essential for development of neuropathic pain,
that is, pain associated with nerve injury, and control
Gabapentin 1 (Fig. 1) was the first drug approved for
post-herpetic neuralgia treatment. The cyclic amino acid
1 and Pregabalin 2 are ligands at the a2d domain of
Cav2.2, stimulating interest in Ca2+ channels as pain
drug targets.9,10 Flunarizine11 3 and Lomerizine12 4,
known as L-type ligands, exhibit N-type blockade
(Table 1). Given this promising target, we started a
search for novel drug-like analgesic N-type blockers.
The objective of Ionix’s N-type drug discovery pro-
gramme was to identify orally active selective small mol-
ecule N-type blockers with an acceptable therapeutic
index for the treatment of chronic pain. Several patented
compounds are claimed to be selective N-type blockers.
Examples include NMED 39-45-3 5 and cinnarizine
analogue MC 34D 7, from a Neuromed piperazine
Keywords: N-type calcium channel; Analgesia; Thiazolidinone.
*
Corresponding author. Tel.: +1 610 738 6505; fax: +1 610 738 6558;
e-mail: LKnutsen@cephalon.com
ꢀ
Vernalis PLC, 613 Reading Road, Winnersh, RG41 5UA, UK
(www.vernalis.com) acquired Ionix Pharmaceuticals, July 2005.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.098

L. J. S. Knutsen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 662–667
663
F
CO2H
CO2H
NH2
(S)-
NH2
N
H3C
N
F
R
CH3
Gabapentin 1
Pregabalin 2 Flunarizine 3, R =
O
O
CH3
CH3
Lomerizine 4, R =
O
CH3
Figure 1. Structures of reference calcium channel modulators.
Table 1. IC50 values for 3-(2-pyridin-2-ylethyl)thiazolidine-4-ones in the IMR32 assay23
R2
O
5
4
N
X 1
3
2
N
R1
Compound
R1
R2
X
N-type IC50 (lM)
L-type IC50 (lM)
Ratio L-type/N-type
9
26
23
27
28
29
3
3-Br, 4-F–Ph
3-CF3–Ph
3-Br, 4-F–Ph
3-CF3–Ph
3-Br, 4-F–Ph
3-Br, 4-F–Ph
—
H
S
S
S
S
1.68
2.41
1.34
1.29
8.13
29.8
1.67
2.43
9.51
11.5
2.92
2.08
>30
>30
0.78
1.48
5.6
4.8
2.2
1.6
—
H
N-iBu-acetamide
N-iBu-acetamide
H
SO2
S@O
—
H
—
—
—
0.5
0.6
4
—
—
series,13 as well as the amide 6 with in vivo activity in
pain models.14 The quinoline 8 has reported activity in
the late stage formalin paw pain model.15 We noted
the structural diversity in Figure 2, and sought differing
starting points as leads in our search for novel blockers.
The carbinol 14, derived from 13, was converted into the
nicotinaldehyde 15 utilising a Vilsmeier reagent, fol-
lowed by ammonium acetate-induced cyclisation.21
Aldehyde 15 was cyclocondensed with mercaptoacetic
acid and isobutylamine to provide target 10 (Scheme 1).
We utilised a high-throughput in vitro fluorescence-
based assay to assess IC50 values for blockade of both
N- and L-type channels in a human neuroblastoma cell
line, IMR32, essentially in a single assay.16 Evaluation
of the Ionix corporate compound library identified
2-arylthiazolidinones, including 9 and 10, as promising
N-type blockers, and both were selected as promising
screening ‘hit’ structures for elaboration of SAR
(Fig. 3).
The intermediate 18 was prepared from aldehyde 17
available from formylation of 16.21 Suzuki coupling of
18 with 4-CF3-phenylboronic acid provided 19 (Scheme
2). A higher yielding method for synthesis of 19 is via
aldehyde 21; Suzuki coupling22 of 21 afforded 22 and
cyclocondensation provided 19. Syntheses of related
thiazolidinones are illustrated in Scheme 3 and 4.
Thiazolidinone 9 proved to be one of the more selective
examples with a selectivity ratio for the N-/L-type of >5,
with promising solubility and physical properties. How-
ever, N-type potency of >1 lM was not improved in any
of the analogues of 9 in Table 1, and dropped away
markedly upon oxidation of the sulfur. Selectivity could
not be readily improved, even by probing neighbouring
areas of the binding site with bulky R2 groups at the
thiazolidinone 5-position, for example, the acetamide
23 (Scheme 3).
Very few thiazolidinones modulating ion channels are
known. Compound 11 is one of a series of Na+ channel
blockers17 and CP-060S 12 is an anti-ischaemic with
Ca2+ overload inhibition and antioxidant activity.18
Thiazolidinones with K+ modulation are known.19 Giv-
en promising in vitro activity and selectivity of 9 and 10
in the IMR32 assay (Tables 1 and 2), we embarked on a
detailed SAR study of 2-arylthiazolidinones. Tables 1–4
summarise SAR studies, focusing on structural features
aiming to provide selectivity for N-type over L-type cal-
cium channels, and represent a major chemistry effort.
Experimental detail has already been published.20
We opted therefore to focus on an SAR study around
2-[4-(4-bromo-phenyl)pyridin-3-yl]-3-isobutylthiazolidin-
4-one 10, which proved to be robustly selective with an

664
L. J. S. Knutsen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 662–667
F
F
CH3
CH3
O
O
CH3
N
H
N
N
NH
O
CH3
O
CH3
H3C
F
F
Neuromed Technologies
Warner Lambert/Elan 6
NMED 39-45-3 5
CH3
HN
H
N
N
Ph
N
N
O
F
Astra Zeneca 8
NMED MC 34D 7
Figure 2. Structures of reference N-type blockers.
O
O
CH3
S
N
Br
N
S
H3C
N
N
Br
F
9
10
O
O
O
CH3
S
N
N
N
S
O
O
N
But
tBu
O
CF3
OH
Purdue Pharma 11
Chugai CP-060S 12
Figure 3. Structures of reference and ‘hit’ thiazolidin-4-ones.
channel in this whole series (IC50 210 nM); and cyclo-
pentyl 32 retained good selectivity (>16) while the selec-
tivity of cyclohexyl derivative 33 fell away.
Br
Br
b
a
H3C
HO
O
CH3
CH3
14
13
We next replaced 4-bromophenyl with alternative aryl
functions while retaining an N-isobutyl; potency and
selectivity was retained in 4-toluyl, 4-ethenylphenyl
and 4-CF3-phenyl derivatives. When sulfur was replaced
by sulfinyl or oxygen, the potency fell away.
O
N
CH3
Br
S
O
N
H
Br
H3C
c
2-[3-(4-Bromophenyl)pyridin-4-yl]-3-isobutylthiazolidin-
4-one 41, the pyridine-4-yl isomer of 10, showed similar
properties to of its parent, so SAR of these analogues
was elucidated (Table 3). In contrast to the 3-pyridyl
series, simple replacement of bromine with chlorine
did not decrease selectivity, and beneficial changes in
terms of selectivity were 4-methoxy-, 4-methyl- and
4-ethenylphenyl.
N
10
15
Scheme 1. Reagents and conditions: (a) MeMgI, C6H6, 0 °C, 80%; (b)
i—POCl3, DMF, 0–60 °C, 18 h; ii—NH4OAc, 60 °C, 39%; (c)
HSCH3CO2H, isobutylamine, C6H6, 70 °C, 18 h, 99%.
N/L ratio of 12 (Table 2). It was possible to retain selec-
tivity by altering substituents at the 3-position, replacing
the isobutyl group in 10 progressively by 2-butyl, cyclo-
propyl, cyclopentyl and cyclohexyl. The 2-butyl deriva-
tive 30 exhibited the highest potency at the N-type
Unlike the 3-pyridyl series, N-cyclopentyl in place of
isobutyl did not confer any improvement, and was dele-
terious to activity when a basic substituent such as
4-dimethylamino was introduced in place of bromine.

L. J. S. Knutsen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 662–667
Table 2. IC50 values for (4-phenylpyridin-3-yl)-thiazolidine-4-ones
665
O
X
N
N
R4
R3
Compound
R3
R4
X
N-type IC50 (lM)
L-type IC50 (lM)
Ratio L-type/N-type
10
30
31
32
33
34
35
36
37
38
39
19
40
24
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
4-Br–Ph
Ph
Isobutyl
2-Butyl
S
S
S
S
S
S
S
S
S
S
S
S
0.66
0.21
1.30
0.63
1.32
3.98
0.60
1.29
0.49
0.88
0.51
1.07
3.97
4.42
8.41
2.33
4.69
10.3
2.01
16.4
2.18
6.66
8.20
5.54
9.99
10.3
15.0
29.3
12.8
11.2
3.6
Cyclopropyl
Cyclopentyl
Cyclohexyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
16.3
1.5
4.1
3.6
4-Cl–Ph
4-F–Ph
5.2
4-CH3–Ph
4-Ethyl-Ph
4-Ethenyl-Ph
4-CF3–Ph
4-Br–Ph
4-Br–Ph
16.7
6.3
19.5
9.6
S@O
O
3.8
6.6
Table 3. IC50 values for (3-phenylpyridin-4-yl)thiazolidinones
O
X
N
R4
R3
N
Compound
R3
R4
N-type IC50 (lM)
L-type IC50 (lM)
Ratio L-type/N-type
41
42
43
44
45
46
47
48
4-Br–Ph
4-Cl–Ph
4-F–Ph
Isobutyl
Isobutyl
0.50
0.58
4.15
1.40
0.59
0.53
10.7
1.37
3.08
4.58
11.7
13.7
8.03
4.16
18.2
4.25
6.1
7.9
2.8
9.5
13.6
7.8
1.7
3.1
Isobutyl
Isobutyl
4-CH3O–Ph
4-CH3–Ph
4-Ethenyl-Ph
4-(CH3)N–Ph
4-(CF3)–Ph
Isobutyl
Isobutyl
Cyclopentyl
Cyclopentyl
Table 4. IC50 values for (2-phenylpyridin-3-yl)thiazolidinones
O
X
N
R4
R3
N
Compound
R3
R4
N-type IC50 (lM)
L-type IC50 (lM)
Ratio L-type/ N-type
49
50
51
52
53
54
55
4-Br–Ph
4-Cl–Ph
4-F–Ph
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
Isobutyl
0.70
1.26
4.92
8.93
1.96
0.56
1.33
6.20
7.49
12.9
20.6
12.5
5.44
4.23
8.9
5.9
2.6
2.3
6.4
9.7
3.2
4-MeO–Ph
4-Me–Ph
4-(CF3)–Ph
4-Ethenyl-Ph

666
L. J. S. Knutsen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 662–667
O
CH3
O
N
H
CF3
S
N
O
N
H
H3C
O
N
H
Br
Cl
Br
e
h
N
i
f
22
18
21
17
16
O
CH3
g
d
S
N
CF3
H3C
Br
Cl
N
N
N
20
19
Scheme 2. Reagents and conditions: (d) i—LDA, THF, À78 °C, 30 min; ii—DMF, À78 °C, 15 min, to rt 25%; (e) HSCH3CO2H, isobutylamine,
C6H6, 70 °C, 18 h, 41%; (f) 4-CF3PhB(OH)2, Pd(PPh3)4, K2CO3 aq, DME, 85 °C, 18 h, 29%; (g) i—LDA, THF, À78 °C, 1.5 h; ii—N-CHO-
piperidine, À78 °C, 1 h, 63%; (h) 4-CF3PhB(OH)2, Pd(OAc)2, PPh3, K2CO3 aq, DME, 85 °C, 18 h, 81%; (i) HSCH3CO2H, isobutylamine, (EtO)3CH,
C6H6, 80 °C, 75%.
OH
In the series featuring 2-[2-(4-bromophenyl)pyridin-3-
O
O
CH3
B
CH3
OH
yl]-3-isobutylthiazolidin-4-ones (Table 4), no real
improvement in potency was recorded despite changing
the aromatic functionality at the pyridyl 2-position.
S
R5
S
N
N
H3C
H3C
R5
H
H
Br
K2CO3
N
N
Thiazolidinones prepared by the methods outlined in
Schemes 1–4 are racemic mixtures. In order to compare
the N-type selectivity and potency between enantiomers,
racemic intermediate 18 was separated by chiral LC24 to
provide the enantiomers 56 and 57, which were convert-
ed into enantiomerically pure compounds 58–61
(Scheme 5).
PPh3, DME, H2O
R5 = CF3, CH3
56
OH
O
O
CH3
CH3
B
OH
S
R5
S
N
N
H3C
H3C
R5
H
H
Br
K2CO3
N
N
PPh3, DME, H2O
58-61
57
It is clear from Table 5 that there are large differences in
in vitro activity between enantiomers, illustrated by the
Scheme 5. Synthesis of single enantiomer sets of 19 and 37.
CH3
H
Table 5. IC50 values for single enantiomer 3-(4-phenylpyridin-3-yl)-
thiazolidine-4-ones
O
N
CH3
CH3
O
H3C
N
S
Compound R5
N-type
L-type
Ratio
O
H
IC50 (lM) IC50 (lM) L-type/N-type
j, k
58
59
60
61
4-CF3
4-CF3
3.32
0.77
12.9
22.9
12.9
23.2
3.9
29.1
2.9
Br
Br
4-CH3 4.44
4-CH3 0.73
31.8
F
F
23
24
selectivity of 59 and 61, both of which are derived from
the same precursor single enantiomer of 18. Compounds
59 and 61 exhibit a selectivity ratio for N-/L-type of
Scheme 3. Reagents and conditions: (j) mercaptosuccinic acid, 2-(2-
aminoethyl)pyridine, C6H6, 80 °C, 18 h, 82%; (k) i—isobutylamine,
EDC, HOBT, NEM, THF, 40 °C, 18 h; ii—HCl, Et2O, rt, 56%.
O
O
CH3
Br
O
N
H
O
O
HN
Br
Br
N
H3C
l
m
N
N
15
24
25
Scheme 4. Reagents and conditions: (l) glycolamide, BF3ÆOEt2, THF, 65 °C, 48 h, 85%; (m) NaH, 1-Br-2-CH3-propane, DMF, 70 °C, 18 h, 43%.

L. J. S. Knutsen et al. / Bioorg. Med. Chem. Lett. 17 (2007) 662–667
667
13. Snutch, T. P. PCT Int. Appl. WO 01/45709, 2001; Chem.
Abstr. 2001, 135, 61354.
ꢀ30, and represent the most selective examples in this
series of thiazolidinone N-type Ca2+ channel blockers.
14. Song, Y.; Bowersox, S. S.; Connor, D. T.; Dooley, D. J.;
Lotarski, S. M.; Malone, T.; Miljanich, G.; Millerman, E.;
Rafferty, M. F.; Rock, D.; Roth, B. D.; Schmidt, J.;
Stoehr, S.; Szoke, B. G.; Taylor, C.; Vartanian, M.; Wang,
Y.-X. J. Med. Chem. 2000, 43, 3474.
15. Chapdelaine, M.; Kemp, L.; McCauley, J. PCT Int.
Appl. WO 02/36567, 2002; Chem. Abstr. 2002, 136,
369614.
In conclusion, we have identified a series of novel 3-aryl-
thiazolidin-4-ones derived from the initial HTS ‘hits’ 9
and 10. By analysis of SAR, we have identified com-
pound 30 (Table 2) as the most potent, and single enan-
tiomer 61 (Table 5) as the most selective compounds in
this series. These compounds show promise as lead
structures in the quest for clinically effective N-type
blockers in the treatment of pain.
16. Takahara, A.; Fujita, S.; Moki, K.; Ono, Y.; Koganei, H.;
Iwayama, S.; Yamamoto, H. Hypertens. Res. 2003, 26,
743.
17. Sun, Q.; Tafesse, L.; Limberis, J. T.; Islam, K.; Kyle, D. J.
Comb. Chem. High Throughput Screening 2003, 6, 481;
Sun, Q.; Kyle, D. J. PCT Int. Appl. WO 03/008398, 2003;
Chem. Abstr. 2003, 138, 131144.
18. Kato, T.; Ozaki, T.; Tamura, K.; Suzuki, Y.; Akima, M.;
Ohi, N. J. Med. Chem. 1999, 42, 3134.
19. Castle, N. A.; Gross, M.; Mendoza, J. S. PCT Int. Appl.
WO 99/62891, 1999; Chem. Abstr. 1999, 132, 12317.
20. Knutsen, L. J. S.; Hobbs, C. J.; Radford, F.; Fiumana, A.
PCT Int. Appl. WO 04/17965, 2004; Chem. Abstr. 2004,
140, 217639; Knutsen, L. J. S.; Hobbs, C. J.; Radford, F.;
Fiumana, A.; Earnshaw, C. G.; Mellor, S. L.; Smith, N. J.
PCT Int. Appl. WO 04/17966, 2004; Chem. Abstr. 2004,
140, 217640.
References and notes
1. Gribkoff, V. K.; Starrett, J. E. Expert Opin. Pharmacother.
1999, 1, 61.
2. Cox, B.; Denyer, J. C. Expert Opin. Ther. Patents 1998, 8,
1237; Elmslie, K. S. J. Neurosci. Res. 2004, 75, 733.
3. Spafford, J. D.; Zamponi, G. W. Curr. Opin. Neurobiol.
2003, 13, 308.
4. Catterall, W. A. Cell Calcium 1998, 24, 307.
5. Dickenson, A. H.; Matthews, E. A.; Diaz, A. M. Pain
2003, 619; Sabido-David, C.; Faravelli, L.; Salvati, P.
Expert Opin. Investig. Drugs 2004, 13, 1249; McGivern, J.
G.; McDonough, S. I. Curr. Drug Targets CNS Neurol.
Disord. 2004, 3, 457.
6. Saegusa, H.; Kurihara, T.; Zong, S.; Kazuno, A.; Matsu-
da, Y.; Nonaka, T.; Han, W.; Toriyama, H.; Tanabe, T.
EMBO J. 2001, 20, 2349.
7. Miljanich, G. P. Curr. Med. Chem. 2004, 11, 3029;
Matthews, E. A.; Dickenson, A. H. Pain 2001, 92, 235.
8. Atanassoff, P. G.; Hartmannsgruber, M. W. B.; Thrasher,
J.; Wermeling, D.; Pharm, D.; Longton, W.; Gaeta, R.;
Singh, T.; Mayo, M.; McGuire, D.; Luther, R. R. Reg.
Anesth. Pain. Med. 2000, 25, 274.
9. Sills, G. J. Curr. Opin. Pharmacol. 2006, 6, 108;; Canti, C.;
Davies, A.; Dolphin, A. C. Curr. Neuropharmacol. 2003,
1, 209.
10. Stahl, S. M. J. Clin. Psychiatry 2004, 65, 1033.
11. Pauwels, P. J.; Leysen, J. E.; Janssen, P. A. J. Life Sci.
1991, 48, 1881.
12. Hara, H.; Morita, T.; Sukamoto, T.; Cutrer, F. M. CNS
Drug Rev. 1995, 1, 204.
21. Thomas, A. D.; Asokan, C. V. J. Chem. Soc., Perkin
Trans. 1 2001, 20, 2583.
22. Lohse, O.; Thevenin, P.; Waldvogel, E. Synlett 1999, 1, 45.
23. Reference compound data in the IMR32 assay were
obtained as the following pIC50 values: x-Conotoxin
GVIA: 8.42 0.22 at N-type, no measurable block at L-
type. Nimodipine: 4.97 0.26 at N-type and 7.87 0.27 at
L-type. See Ward, S. D. C.; Hick, C. A.; Thomas, D.
Online Abstract Viewer/Itinerary Planner, Society for
Neuro-science Annual Meeting, New Orleans, Louisiana,
November 8–12, 2003; Society for Neuroscience, Wash-
ington, DC; Program No. 165.7.
24. The HPLC separation method utilised was ChiralpakÒ
50801 (20 lm) as chiral stationary phase. Mobile phase
CH3OH; UV detection at 280 nm. Analysis conditions:
ChiralcelÒ OD-H; mobile phase—95% heptane/5% iso-
propanol; UV detection at 225 nm. Flow rate—1 ml/min,
at 25 °C.