1-Thia-4,7-diaza-spiro[4.4]nonane-3,6-dione: A Structural Motif for 5-hydroxytryptamine 6 receptor antagonism

Article abstract of DOI:10.1111/cbdd.12240

A series of potent 5-hydroxytryptamine 6 (5-HT₆) receptor antagonists based on 1-thia-4,7-diaza-spiro[4.4]nonane-3,6-dione motif has been disclosed. Enantiomers of potent racemate compound 8a (K_i = 26 nm) displayed difference in activity (K_i of 15 nm versus 855 nm) signaling the influence of the stereochemistry of the chiral center on potency. In addition, the potent enantiomer displayed significant selectivity in biological activities over several related family members. Based on HTS-hit compound 1a (Ki=5.73uM against h5-HT6),compound 8a (Ki=26nM) was developed. Enantiomers (Ki of 15nM vs,855nM)displayed effect of chirality on activity.

Full text of DOI:10.1111/cbdd.12240

Chem Biol Drug Des 2014; 83: 149–153
Research Letter
1-Thia-4,7-diaza-spiro[4.4]nonane-3,6-dione: A
Structural Motif for 5-hydroxytryptamine 6 Receptor
Antagonism
Compound 1a contained a unique motif for the receptor’s
antagonism, especially due to the absence of any sulfon-
amide or sulfone moiety in the framework, a frequent feature
of literature-reported potent antagonists from various
laboratories (compounds A–C, Fig. 2) (7–9). Subsequently,
a synthetic program was initiated around this scaffold to
expand the scope of the series. This research generated a
series of potent benzazepine-derived antagonists, exempli-
fied by generic structure 2 (Fig. 1). In this Letter, we disclose
a preliminary account from our ongoing effort in the area.
Greg Hostetler, Derek Dunn, Beth A. McKenna,
Karla Kopec and Sankar Chatterjee*
Cephalon, Inc., 145 Brandywine Parkway, West Chester,
PA 19380-4245, USA
*Corresponding author: Sankar Chatterjee,
Sankar.Chatterjee24@gmail.com
A series of potent 5-hydroxytryptamine 6 (5-HT6) recep-
tor antagonists based on 1-thia-4,7-diaza-spiro[4.4]
nonane-3,6-dione motif has been disclosed. Enantio-
mers of potent racemate compound 8a (Ki = 26 nM)
displayed difference in activity (Ki of 15 nM versus
855 nM) signaling the influence of the stereochemistry of
the chiral center on potency. In addition, the potent
enantiomer displayed significant selectivity in biological
activities over several related family members.
Scheme 1 displays a general synthetic procedure to gen-
erate various target compounds. N protected-benzazepine
3a–b was coupled with commercially available various 1H-
indole-2,3-dione 4 to generate intermediate imines of
general structure 5 that, without isolation, underwent a
spirocyclization reaction with thioglycolic acid to generate
compounds 6a–b. While deprotection of the t-Boc group
in compound 6a was carried out under acidic condition,
deprotection of the trifluoroacetyl group in compound 6b
was achieved under basic condition to generate the base
7. Reductive amination of compound 7 with formalin and
Key words: 5-HT6 receptor, antagonist, CNS disorders
Received 27 May 2013, revised 19 July 2013 and accepted
for publication 18 September 2013
various representative ketones generated
a series of
N-alkylated products 8a–k.
The 5-hydroxytryptamine 6 (5-HT6) receptor is one of the
prominent members of the serotonin receptor family (1).
Due to its exclusive distribution in the central nervous
system (CNS), this receptor has emerged as a promising
target for the pharmacological intervention for the
treatment for several central nervous system (CNS)
related disorders, for example, cognitive function in Alz-
heimer’s disease and schizophrenia, anxiety, obesity,
depression, and sleep–wake activity (2–4). Thus, the
discovery of novel and potent antagonists of this receptor
has become a recent area of research for the pharma-
ceutical industries (5). It had been revealed that
SB-742457, a potent 5-HT6 antagonist, demonstrated
significant improvement in global function in the treatment
for dementia in Alzheimer’s disease in a phase IIb pla-
cebo-controlled study (6). In search of novel and potent
5-HT6 receptor antagonists, our team initially profiled our
corporate chemical library in a high-throughput screening
(HTS) platform. From this exercise, the team encountered
the 1-thia-4,7-diaza-spiro[4.4]nonane-3,6-dione-derived ‘hit’
compound 1a [Ki of 5.73 lM against human 5-HT6 (h5-HT6)
receptor, Fig. 1].
In Scheme 2, commercially available indole-carboxylic
acids 9a–b underwent an internal oxidative cyclization pro-
cess to generate lactones 10a–b (10). Compound 10a
underwent a ring-opening reaction in the presence of
methanol to generate the hydroxyl esters 11a; however,
compound 10b needed treatment with TMSN2 in methanol
to generate compound 11b (11). Compounds 11a–b were
converted to corresponding chloro derivatives 12a–b that
were coupled with compound 3a to generate compounds
13a–b. Basic hydrolysis of the methyl ester group in 13a–
b generated carboxylic acids 14a–b that underwent an
acid-catalyzed internal spirocyclization reaction to generate
compounds 15a–b. Deprotection of the t-boc group
in 15a–b generated 16a–b that on further alkylation-
generated compounds 17a–b. Finally, selective reduction
in the 3o-amide group in ring B over the 2o-amide group
in ring C in compound 16a produced compound 18 (12).
Following a literature-reported procedure, binding proper-
ties of the target compounds were assessed against
ª 2013 John Wiley & Sons A/S. doi: 10.1111/cbdd.12240
149

Hostetler et al.
X
nonane-3,6-dione derivatives offering a more potent ana-
log 1b (Fig. 1 and Table 1, Ki of 88 nM). A cursory analysis
of the molecules 1a–b revealed that the incorporation of a
lipophilic moiety, for example, methoxy (or chlorine, data
not shown) into 3, 5 positions of the phenyl ring (ring A),
attached to the thiazolidinone ring of the parent compound
1a generated a 65-fold more potent compound 1b. This
indicated that this area of the antagonist might be interact-
ing with a lipophilic site on the receptor. However, it was
also realized that both these antagonists lacked a basic
nitrogen atom (a common feature to various literature-
reported potent 5-HT6 antagonists) that interacts with a
critical aspartic acid moiety on the receptor. In the
absence of any knowledge of bound conformations of
compounds 1a–b to the receptor, a decision was made to
fuse an azepine system (ring E) with the above-mentioned
phenyl ring (ring A) maintaining a region of lipophilicity in
that part of the molecule with a basic nitrogen embedded
in the area. As shown in Table 1, while an N-acylated
3
2
O
B
4
A
D
O
B
1
A
E
Y
N
X
N
5
N
S
S
C
D
C
N
z
O
N
H
O
1 a X = H
b X = OMe
2
Figure 1: Structures of compounds 1a–b and 2.
recombinant human 5-HT6 (h5-HT6) receptors by displace-
ment of [3H] LSD and reported as Ki values from an aver-
age of two experiments, done in triplicate in Table 1 (13).
Discovery of HTS-hit 1a (Ki of 5.73 lM, Table 1) prompted
us to profile additional available 1-thia-4,7-diaza-spiro[4.4]
Me
N
N
H
NH
O
O
N
S
O
N
O
S
F
SO2
Br
O
N
O
Me
CF3
A
B
C
Figure 2: Structures of compounds A, B, and C.
O
X
N
a
X
N
+
]
O
[
N
NH2
N
z
H
O
3 a X = tBoc
b X = COCF3
4
N
z
H
5
b
O
O
R
N
X
N
e
N
N
*
*
S
S
O
O
N
H
N
z
z
6 a X = tBoc
H
b X = COCF3
8a - k
c
d
7
X = H
Scheme 1: Reagents and conditions: (a) Toluene, reflux under Dean-Stark condition, overnight; (B) HS-CH2-COOH, catalytic ZnCl2,
reflux, 50–60% over two steps; (c) 4 N HCl in dioxane, room temperature, 2 h, quantitative; (d) 7 N NH3 in MeOH, room temp., overnight,
quantitative; (e) for R = Me: formalin, methanol, sodium triacetoxyborohydride, 0 °C to room temp., 60%; for R = 2o alkyl or cyclo-alkyl:
corresponding ketone, methanol, sodium triacetoxyborohydride, 0 °C to room temp., 50–60%.
150
Chem Biol Drug Des 2014; 83: 149–153

5-HT6 Antagonists
O
O
X
OH
()n
OMe
()n
O
a
O
()n
O
b
N
H
O
N
H
N
H
11 a - b X = OH (n = 1, 2)
12 a - b X = Cl (n = 1, 2)
c
10 a - b (n = 1, 2)
9 a - b (n = 1, 2)
d
Y
N
O
OX
O
O
O
f
N
N
O
()n
()n
O
NH
N
H
N
H
15 a - b Y = t-Boc (n = 1, 2)
g
h
16a Y = H, n = 1
16b Y = H, n = 2
13 a - b X = Me (n = 1, 2)
14 a - b X = H (n = 1, 2)
e
17a Y = Me, n = 1
17b Y = Me, n = 2
HN
i
16a
N
N
H
O
18
Scheme 2: Reagents and conditions: (a) t-butyl bromide, DMSO, 50 °C, overnight, 60–65%; (b) (i) for n = 1, MeOH, 4-DMAP, room
temp., overnight, 70%; (ii) for n = 2, TMSN2, C6H6, MeOH, 60%; (c) SOCl2, Et3N, CH2Cl2, 0 °C to room temp., quantitative; (d) 3a, Et3N,
K2CO3, CH3CN, 60 °C, overnight, 50–55%; (e) LiOH.H2O, MeOH, H2O, room temp., quantitative; (f) (i) for n = 1, ZnCl2, THF, reflux,
overnight, 60%; (ii) for n = 2, HOAc, 110 °C, 50% (g) 4 N HCl in dioxane, room temperature, 2 h, quantitative; (h) (i) formalin, methanol,
sodium triacetoxyborohydride, 0 °C to room temp., 60%; (ii) LiBH3NMe2, THF, 0 °C to room temp., 55%.
analog 6b was ineffective, the corresponding deprotected
analog 7 containing a basic amine moiety was ca. 38-fold
more potent than compound 1a. Methylating the azepine
nitrogen atom (ring E) in compound 7 generated com-
pound 8a that was ca. 220-fold more potent than com-
pound 1a. On the other hand, methylation of the indolone
nitrogen atom of (ring C) of the spiro system in 7 gener-
ated 8b that was ca. 25-fold more potent than the parent
compound 1a. Methylation of both nitrogen atoms of com-
pound 7 generated 8c that was ca. 37-fold more potent
than the parent compound 1a. Substitution on the azepine
nitrogen atom (ring E) with bulkier isopropyl as well as
cycloalkyl groups was detrimental for the series (cf. 8d–g
versus 7) indicating steric limitation at the site. Interest-
ingly, a benzyl group attached to either nitrogen atom of
azepine ring (ring E, compound 8h) or indolone ring (ring
C, compound 8i) maintained the potency of compound 7.
It is possible that the planar aromatic moiety from either
compound might be interacting in a beneficial manner with
a distant site on the receptor. Incorporation of a methyl
group in the phenyl segment (ring D) of the indolone sys-
tem (compounds 8j–k) was not detrimental to activity. In
expanding the SAR, the requirement of the sulfur atom of
the thiazolidinone ring (ring B) for activity was also probed.
Thus, it was replaced with a methylene (ÀCH2) group in
compounds 7 and 8a to generate 16a and 17a, respec-
tively. As shown, such maneuver was detrimental to
potency. Subsequently, it was argued that a relatively lar-
ger ethylene group (–CH2CH2–) might be a better bio-isos-
teric replacement of sterically bulky sulfur atom. However,
homologation of 16a and 17a generating 16b and 17b,
respectively, diminished the activity. Interestingly, selective
reduction in the 3o–carbonyl group in ring B over 2o–
carbonyl group in ring C of 16a generated compound 18
that was ca. 33-fold more potent than its precursor and
equipotent to compound 8a. A detailed knowledge of the
three-dimensional structures of the receptor and antago-
nist complexes will be needed to explain the equipotency
of compounds 18 and 8a. Finally, to probe the effect of
the chirality on potency, compound 8a was separated into
respective enantiomers (>98% ee) via chiral HPLC. The
enantiomers displayed ca. 57-fold difference in binding
potency (Ki of 15 nM versus 855 nM). The activity of the
potent enantiomer compares favorably with that of the
compounds A (Ki of 1.43 nM), B (Ki of 2.40 nM), and C
(Ki of 6.50 nM) depicted in Fig. 2. The determination of the
absolute configuration of either isomer has been an
ongoing research effort. The potent enantiomer from
Chem Biol Drug Des 2014; 83: 149–153
151

Hostetler et al.
Table 1: Biologic data of target compoundsa,b
Acknowledgements
P
B
P
Authors wish to thank Drs. Ed Bacon and John P. Mallamo
for their encouragement during the course of this research.
E
A
X
N
()n
R
N
D
C
N
Y
Conflict of Interest
O
z
Authors, as the employees of Cephalon, Inc., were
engaged in discovering potent 5–HT6 antagonists.
c
Compound
X
Y
Z
P, P
R
n
Ki nM
1a
1b
6b
7
–
–
–
–
–
–
–
–
–
–
S
S
S
S
S
S
S
S
S
S
S
S
S
–
–
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
5730
88
>30 000
150
References
COCF3
H
Me
H
Me
CHMe2
cyclo-butyl
cyclo-pentyl
cyclo-hexyl
CH2Ph
H
H
Me
H
Me
H
H
H
H
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Me
H
H
H
H
H
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
1. Marazziti D., Baroni S., Borsini F., Picchetti M., Falaschi
V., Catena-Dell M.O. (2013) Serotonin receptors of type
6 (5-HT6): from neuroscience to clinical pharmacology.
Curr Med Chem;20:371–377.
2. Ramirez M.J. (2013) 5-HT6 receptors and Alzheimer’s dis-
ease. Alzheimers Res Ther;5:15. (epub ahead of print).
3. Wesołowska A. (2010) Potential role of the 5-HT6
receptor in depression and anxiety: an overview of
clinical data. Pharmacol Rep;62:564–577.
4. Ly S., Pishadari B., Lok L.L., Hajos M., Kocsis B.
(2013) Activation of 5-HT6 receptors modulates sleep-
wake activity and hippocampal theta oscillation. ACS
Chem Neurosci;4:191–199.
8a
8b
8c
8d
8e
8f
26
227
155
954
>10 000
1400
>10 000
92
8g
8h
8i
CH2Ph
99
45
39
723
8j
H
H
H
H
H
H
H
8k
16a
17a
16b
17b
18
CH2
CH2
CH2
CH2
337
>30 000
>30 000
22
ꢀ
5. Rosse G., Schaffhauser H. (2010) 5-HT6 receptor
Me
H
H, H CH2
antagonists as potential therapeutics for cognitive
impairment. Curr Top Med Chem;10:207–221.
aEnzyme assay against recombinant h5-HT6; internal ligand [3H]
LSD (binding of this ligand to the h5-HT6 membranes was satura-
ble with Bmax = 6.2 pmol/mg protein and Kd = 2.3 nM). Ki value of
a test compound was then calculated according to the Cheng–
Prusoff method.
bAll compounds were tested as racemates.
cAverage of three experiments.
6. Maher-Edwards G., Zvartau-Hind M., Hunter A.J., Gold
M., Hopton G., Jacobs G., Davy M., Williams P. (2010)
Double-blind, controlled phase II study of a 5-HT6
receptor antagonist, SB-742457, in Alzheimer’s dis-
ease. Curr Alzheimer Res;7:374–385.
7. Nirogi R.V., Badange R., Kambhampati R., Chindhe A.,
Deshpande A.D., Tiriveedhi V., Kandikere V., Muddana
N., Abraham R., Khagga M. (2012) Design, synthesis
and pharmacological evaluation of 4-(piperazin-1-yl
methyl)-N1-arylsulfonyl indole derivatives as 5-HT6 recep-
tor ligands. Bioorg Med Chem Lett;22:7431–7435.
8. Henderson A.J., Guzzo P.R., Ghosh A., Kaur J., Koo
J.M., Nacro K., Panduga S., Pathak R., Shimpukade B.,
Tan V., Xiang K., Wierschke J.D., Isherwood M.L., M. L.
(2012) Epiminocyclohepta[b]indole analogs as 5-HT6
antagonists. Bioorg Med Chem Lett;22:1494–1498.
9. Sundar B.G., Bailey T.R., Dunn D.D., Bacon E.R., Salv-
ino J.M., Morton G.C., Aimone L.D. et al. (2012) Novel
brainpenetrant benzofuropiperidine 5-HT6 receptor
antagonists. Bioorg Med Chem Lett;22:120–123.
10. Labroo R.B., Labroo V.M., King M.M., Cohen L.A.
(1991) An improved synthesis of dioxindole-3-propionic
acid and some transformations of the C-3 hydroxyl
group. J Org Chem;56:3637–3642.
above-mentioned pair was further profiled against
various members of the serotonin family displaying
>200-fold selectivity against each of the following subtype:
5HT1a, 5HT2b, 5HT2C, 5HT3, 5HT5a, and 5HT7, respec-
tively.
In conclusion, in this Letter, we disclosed 1-thia-4,7-
diaza-spiro[4.4]nonane-3,6-dione as a structural motif for
5HT6 antagonism. The reported research revealed that the
presence of a sulfonamide or a sulfonyl moiety, a frequent
feature of literature-reported potent antagonists at the time
of research, is not an absolute necessity for a potent
antagonist. From the series, enantiomers of potent race-
mate compound 8a (Ki = 26 nM) displayed influence of
chirality on potency (Ki of 15 nM versus 855 nM). The
potent enantiomer also displayed selectivity for 5HT6
receptor over various other members of the serotonin fam-
ily. Additional research is continuing with the series that
will be the subject matter of future publications.
11. Myers A.G., Charest M.G., Lerner C.D., Brubacker
J.D., Siegel D.R. (2005) Patent application WO 2005/
112945:86.
152
Chem Biol Drug Des 2014; 83: 149–153

5-HT6 Antagonists
12. Flaniken J.M., Collins C.J., Lanz M., Singaram B.
(1999) Aminoborohydrides. 11. Facile Reduction of
N-Alkyl Lactam to the Corresponding Amines Using
Lithium Aminoborohydrides. Org Lett;1:799–801.
13. Bacon E.R., Bailey T.R., Dunn D.D., Hostetler G.A.,
ꢀ
McHugh R.J., Morton G.C., Rosse G.C., Salvino J.M.,
Sundar B.G., Tripathy R. (2011) Patent application WO
2011/087712:1–222.
Chem Biol Drug Des 2014; 83: 149–153
153