Modulation of selective serotonin reuptake inhibitor and 5-HT1A antagonist activity in 8-aza-bicyclo[3.2.1]octane derivatives of 2,3-dihydro-1,4-benzodioxane

Article abstract of DOI:10.1016/j.bmcl.2003.10.024

2,3-Dihydro-1,4-benzodioxanes with aryl 8-aza-bicyclo[3.2.1]oct-3-ene attachments 2 produce compounds with potent 5-HT-T affinity, and weak 5-HT _1A affinity and α₁ affinity. This compares with 2,3-dihydro-1,4-benzodioxanes containing 8-aza-bicyclo[3.2.1] octan-3-ol attachments 4 which possess potent 5-HT_1A affinity, moderate to good selectivity over α₁ and little 5-HT-T affinity. A 3-benzothiophene analogue of 4 (30) was synthesized which possesses potent 5-HT_1A affinity and especially good selectivity over both α₁ and 5-HT-T.

Full text of DOI:10.1016/j.bmcl.2003.10.024

Bioorganic & Medicinal Chemistry Letters 14 (2004) 515–518
Modulation of selective serotonin reuptake inhibitor and 5-HT1A
antagonist activity in 8-aza-bicyclo[3.2.1]octane derivatives of
2,3-dihydro-1,4-benzodioxane
Adam M. Gilbert,a,* Gary P. Stack,b Ramaswamy Nilakantan,a Jason Kodah,a
Megan Tran,b Rosemary Scerni,c Xiaojie Shi,c Deborah L. Smithc
and Terrance H. Andreec
aChemical and Screening Sciences, Wyeth Research, 401 N. Middletown Road, Pearl River, NY 10945, USA
bChemical and Screening Sciences, Wyeth Research, CN-8000, Princeton, NJ 08852, USA
cNeuroscience Discovery Research, Wyeth Research, CN-8000, Princeton, NJ 08852, USA
Received 1 August 2003; accepted 3 October 2003
Abstract—2,3-Dihydro-1,4-benzodioxanes with aryl 8-aza-bicyclo[3.2.1]oct-3-ene attachments 2 produce compounds with potent 5-HT-
T affinity, and weak 5-HT1A affinity and a1 affinity. This compares with 2,3-dihydro-1,4-benzodioxanes containing 8-aza-bicyclo[3.2.1]
octan-3-ol attachments 4 which possess potent 5-HT1A affinity, moderate to good selectivity over a1 and little 5-HT-T affinity. A
3-benzothiophene analogue of 4 (30) was synthesized which possesses potent 5-HT1A affinity and especially good selectivity over
both a1 and 5-HT-T.
# 2003 Elsevier Ltd. All rights reserved.
Inhibition of serotonin (5-HT) receptors, namely the
serotonin transporter (5-HT-T), has led to the develop-
ment of many clinically approved agents for depression
and anxiety.1 However, these drugs have noticeable side
effects as well as a delayed onset of action. Selective
antagonists for the 5-HT1A receptor have been postu-
lated to be useful in the treatment of CNS disorders
such as anxiety,2 and Alzheimer’s disease.3 In addition,
5-HT1A antagonists may have an important role in
treating depression in that they could facilitate the onset
of action of selective serotonin receptor inhibitors
(SSRIs) by blocking 5-HT1A autoreceptors which are
believed to decrease the firing of serotonergic neurons in
the presence of 5-HT.4 Thus, an agent with both activ-
ities (SSRI and 5-HT1A antagonism) would be a sig-
nificant improvement on currently approved therapies.
As proof of concept, co-administration of 5-HT1A
antagonists/partial antagonists and SSRIs has been
clinically shown to induce faster antidepressant action
than administration of SSRIs alone.5,6
began synthesizing molecules containing the aryl 8-aza-
bicyclo[3.2.1]oct-3-ene moiety 1. Benzodioxane deriva-
tives of 1 (2) were found to possess potent affinity and
inhibitory activity for the 5-HT-T (Fig. 1). Interestingly
changing the aryl 8-aza-bicyclo[3.2.1]oct-3-ene moiety
in 1 to an 8-aza-bicyclo[3.2.1] octan-3-ol 3 produces 4, a
series of molecules with potent 5-HT1A affinity and
relatively little 5-HT-T affinity. In addition, many of the
members of series 4 were found to possess potent 5-
In the course of our study of novel ligands that would
possess both SSRI and 5-HT1A antagonist activities, we
* Corresponding author. Fax: +1-845-602-5561; e-mail: gilbera@
wyeth.com
Figure 1. Aryl 8-aza-bicyclo[3.2.1] octanes of interest.
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.10.024

516
A. M. Gilbert et al. / Bioorg. Med. Chem. Lett. 14 (2004) 515–518
HT1A antagonist activity. We now wish to report the
synthesis and initial structure–activity relationships
(SARs) of these two new classes of 5-HT antagonist
compounds.
equivalent of 1 or 3 with 1 equivalent of 16 in warm
DMSO (Scheme 4).
All aryl 8-aza-bicyclo[3.2.1] oct-3-ene benzodioxanes 2
and aryl 8-aza-bicyclo[3.2.1] octan-3-ol benzodioxanes 4
were tested in vitro to determine affinity for the 5-HT1A,
Schemes 1–4 show the synthesis of the target molecules.
The aryl 8-aza-bicyclo[3.2.1] oct-3-ene derivatives 1,
were prepared by demethylation of tropinone 5 with 1-
chloroethyl chloroformate,7 protection of the secondary
amine with (BOC)2O to form 6,8 addition of an aryl
lithium, followed by dehydration/deprotection with
TFA (Scheme 1).
5-HT transporter (5-HT-T) and a1 receptors. Human 5-
HT1A (HC-5-HT1A) receptor binding was determined
via the displacement of [3H]-8-OH-DPAT from human
5-HT1A transfected CHO cells according to the method
of Dunlop et al.12 Assessment of compound agonism/
antagonism on the HC 5-HT1A receptor was determined
using a [35S]-GTPgS13 and/or a forskolin stimulated
cyclic AMP assay (c-AMP). A protocol similar to that
of Cheetham et al. was used to determine 5-HT trans-
porter affinity (RB5-HT-T).14 IC50 values (HC-5-HT-T)
were calculated from the Ki values according to the
method of Cheng et al.15 Selectivity over the a1 receptor
was determined by displacement of [3H]-prazosin (a1)16
since a1 receptor affinity could lead to unwanted side
effects for a selective 5-HT agent. An optimal com-
pound would have potent 5-HT-T/5-HT1A inhibitory
activity and weak or no (a1) affinity.
The aryl 8-aza-bicyclo[3.2.1] octan-3-ol derivatives 3
were prepared by demethylating tropinone 5 with 1-
chloroethyl chloroformate,7 protection of the resulting
secondary amine with BnBr to give 8, addition of an
aryl lithium to exclusively give the endo-alcohol product
9,9 followed by removal of the benzyl group with Pd/C,
HCO2NH4 (Scheme 2).10
The 8-OMe and 8-OEt benzodioxane headpieces were
synthesized as tosylates 16 (R=Me and R=Et), as
shown in Scheme 3. The sequence features a Baeyer–
Villiger oxidation/basic hydrolysis conversion of alde-
hyde 12 to phenol 13, the attachment of (2R)-(À)-glyci-
dyl tosylate, debenzylation with concomitant ring
closure/epoxide opening to produce 15, and tosylation
with TsCl to produce 16 (Scheme 3).
Biological data for all newly prepared aryl 8-aza-bicy-
clo[3.2.1]oct-3-enes 2 are presented in Table 1. In all
The target aryl 8-aza-bicyclo[3.2.1] oct-3-ene benzo-
dioxanes 2 and aryl 8-aza-bicyclo[3.2.1] octan-3-ol ben-
zodioxanes 411 were prepared by heating 1 or 3 with 16
in the presence of K2CO3 in MeCN or by heating 2
Scheme 3. Reagents and conditions: (a) NaH, BnBr, THF, 23 ꢀC; (b)
NaH, MeI or EtI, DMF, 23 ꢀC; (c) (1) m-CPBA, CH2Cl2, 23 ꢀC; (2)
Al2O3, MeOH, 23 ꢀC; (d) NaH, (2R)-(À)-glycidyl tosylate, solvent,
XꢀC; (e) 5% Pd/C, 1,4-cyclohexadiene, NaHCO3, EtOH, 80 ꢀC; (f)
TsCl, TEA, DMAP, CH2Cl2, 23 ꢀC.
Scheme 1. Reagents and conditions: (a) (1) 1-chloroethyl chlor-
oformate, DCE, 80 ꢀC; (2) MeOH, 50 ꢀC; (3) (BOC)2O, i-PrOH, H2O,
NaOH, 23 ꢀC; (b) ArLi, THF, À78 to 23 ꢀC; (c) TFA, CH2Cl2, 23 ꢀC.
Scheme 2. Reagents and conditions: (a) (1) 1-chloroethyl chlor-
oformate, DCE, 80 ꢀC; (2) MeOH, 50 ꢀC; (3) BnBr, Et3N, THF, 23 ꢀC;
(b) ArLi, THF, À78 to 23 ꢀC; (c) 10% Pd/C, HCO2NH4, MeOH,
50 ꢀC.
Scheme 4. Reagents and conditions: (a) K2CO3, MeCN, 80 ꢀC or
1 equiv 16, 2 equiv 1 or 9, DMSO, 80 ꢀC.

A. M. Gilbert et al. / Bioorg. Med. Chem. Lett. 14 (2004) 515–518
517
cases, the 2(R) enantiomers were prepared since is
known that the 2(R) enantiomer of (2,3-dihydro-ben-
zo[1,4]dioxin-2-yl)-methylamines show stronger 5-HT
receptor affinity/function compared to the 2(S) enan-
tiomer.17 2-Naphthyl analogue 16 was initially prepared
since 2-naphthyl 8-aza-bicyclo[3.2.1] oct-3-ene is known
to have potent 5-HT-T affinity.18 This compound shows
potent 5-HT-T affinity/inhibition, good selectivity over
the a1 receptor and some affinity for the 5-HT1A receptor.
The 1-napthyl compound 17 comparably shows weaker 5-
HT-T affinity but more 5-HT1A affinity. Indole 18 shows
the best balance of 5-HT-T/5-HT1A affinity, but the
potency at the a1 receptor increases as well. The 8-quino-
line analogue 19 compares with the 1-naphtyl compound
17 except that it shows more a1 affinity, and the 3,4-
dichlorophenyl molecule 20 shows a comparable activity
profile to 16. In general, the RB5-HT-T affinities are
single to double digit nM while the HC5-HT-T func-
tional IC50s are much less potent (ꢁ2 orders of magni-
tude). The origin of this difference is not known.
At this point we were interested in exploring the biolo-
gical profile of compounds where the 8-aza-bicy-
clo[3.2.1]oct-3-ene moiety 1 would be changed to an 8-
aza-bicyclo[3.2.1] octan-3-ol 3. This structural change
produces compounds 4 with a dramatic activity profile
shift compared to 2. 2-Napthyl analogue 21 shows potent
5-HT1A affinity and only moderate 5-HT-T affinity
(Table 2). This is a reverse of the activity profile seen for
2-naphtyl-8-aza-bicyclo[3.2.1]oct-3-ene compound 16.
Moreover, 21 shows potent 5-HT1A antagonist activity
in both the [35S]GTPgS and c-AMP assays. In an effort
to reduce the a1 affinity of 21, compound 22 (R=Et)
was prepared. This compound shows reduced a1 affinity
compared to 21, good 5-HT1A affinity and is devoid of
5-HT-T affinity at 100 nM.
Hoping to optimize 21 and 22 to produce selective 5-
HT1A antagonists, different aryl groups were attached
to the 8-aza-bicyclo[3.2.1] octan-3-ol benzodioxane
core. Phenyl analogues 23 and 24 showed potent 5-
Table 1. 5-HT1A affinity, 5-HT1A functional activity, 5-HT-T affinity, 5-HT-T functional activity and a1 affinities for 8-aza-bicyclo[3.2.1]oct-3-ene
2,3-dihydro-1,4-benzodioxanes 2
Compd
Ar
HC5-HT1A
Ki (nM)a
RB5-HT-T
Ki (nM)
HC5-HT-T
IC50 (nM)b
a1
Ki (nM)
16
17
18
19
20
2-Naph
1-Naph
3-(5-F-1H-indole)
8-Quinoline
3,4-Dichlorophenyl
43% @ 1000 nM
357
127.7
307.1
35% @ 1000 nM
1.4
26.0
8.5
33.0
4.69
76.7
951
138
109.0
260.0
229
26% @ 100 nM
71
46.2
>1000
a Ki values are the mean of two experiments run at six different concentrations. 95% confidence limits were generally Æ10% of the mean value.
b IC50, Emax and Imax values are reported from one experimental run at six different concentrations.
Table 2. 5-HT1A affinity, 5-HT1A functional activity, 5-HT-T affinity, 5-HT-T functional activity and a1 affinities for aryl 8-aza-bicyclo[3.2.1]octan-
3-ol 2,3-dihydro-1,4-benzodioxanes 4
Compd
R
Ar
HC5-HT1A
Ki (nM)a
[
35S]GTPgS: Imax (%)
cAMP: Emax (%)
IC50 (nM)b
RB5-HT-T
Ki (nM)
a1
Ki (nM)
IC50 (nM)b
21
22
23
24
25
26
27
28
29
30
31
Me
Et
Me
Et
Me
Me
Et
Me
Et
2-Naph
2-Naph
Ph
0.81
5.87
1.53
3.05
39.0
0.33
2.92
46.0
61.6
1.6
100, 9.85
NA
0, 21
NA
0, 21
NA
NA
NA
NA
NA
NA
216.5
42
0% @ 100 nM
0% @ 100 nM
0% @ 100 nM
12% @ 100 nM
9% @ 100 nM
0% @ 100 nM
0% @ 100 nM
6% @ 100 nM
0% @ 100 nM
4% @ 100 nM
103.5
NA
32.7
NA
9.9
12.9
NA
147
100, 25
NA
Ph
2-OMe-Ph
3-CF3-Ph
3-CF3-Ph
2-Pyr
E40c
E90d
NA
100, 1484
NA
92, 36
NA
2-Pyr
3-Benzothiophene
3-Benzothiophene
Me
Et
0, 130
NA
106
123.5
16.7
a Ki values are the mean of two experiments run at six different concentrations. 95% confidence limits were generally Æ10% of the mean value.
b IC50, Emax and Imax values are reported from one experimental run at six different concentrations.
c Emax=40%.
d Emax=90%. NA, not available.

518
A. M. Gilbert et al. / Bioorg. Med. Chem. Lett. 14 (2004) 515–518
HT1A affinity, no 5-HT-T affinity, but the R=Et analo-
gue still possessed a1 affinity. Analogues 26 and 27 also
show good 5-HT1A affinity, practically no 5-HT-T affi-
nity, but these compounds behave as partial agonists/
agonists at the 5-HT1A receptor. The 2-pyridyl com-
pounds 28 and 29 show reduced 5-HT1A affinity
compared to the corresponding phenyl analogues and the
3-benzothiophene molecules 30 and 31 compare favor-
ably with 21 and 22. Compound 30 is the most selective 5-
HT1A antagonist that we have prepared (5-HT1A: Ki:
1.6 nM [35S]GTPgS: Imax: 92%, IC50: 36 nM, a1: 106 nM).
2. Fletcher, A.; Cliffe, I. A.; Dourish, C. T. Trends. Phar-
macol. Sci. 1993, 14, 441.
3. Bowen, D. M.; Francis, P. T.; Pangulus, M. N.; Stephens,
P. H.; Procter, A. W. Lancet 1992, 11, 132.
4. Blier, P.; Bergeron, R. J. J. Clin. Psychiatry 1998, 59, 16.
5. Argitas, F.; Perez, V.; Alvarez, E. Arch. Gen. Psychiatry
1994, 51, 248.
6. Blier, P.; Bereron, R. J. Clin. Psychopharmacol. 1995, 15,
217.
7. Olofson, R. A.; Martz, J. T.; Semet, J. P.; Diteau, M.;
Malfroot, T. J. Org. Chem. 1984, 49, 2081.
8. Patchett, A. A.; Tata, J. R.; Lu, Z. British U.K. Patent
Appl. GB 2298647, 1996; Chem. Abstr. 1997, 126, 60192.
9. Cope, A. C.; D’Addieco, A. J. Am. Chem. Soc. 1951, 73,
3419.
Thus we have disclosed a series of aryl 8-aza-bicy-
clo[3.2.1] oct-3-ene benzodioxanes 2 and aryl 8-aza-
bicyclo[3.2.1] octan-3-ol benzodioxanes 4 that have
potent affinity for the 5-HT-T and 5-HT1A receptors
respectively. In addition, several of the 8-aza-bicy-
clo[3.2.1] octan-3-ol analogues 4 show good 5-HT1A
antagonist activity. A 3-benzothiophene 8-aza-bicy-
clo[3.2.1] octan-3-ol analogue 30 shows potent 5-HT1A
affinity and antagonism, no 5-HT-T affinity and is >50
fold selective for the 5-HT1A receptor over a1. Further
studies concerning agents that target the 5-HT1A and 5-
HT-T receptors will be reported in due course.
10. Audia, J. E.; Hibschman, D. J.; Krushinski, J. H.; Mabry,
T. E.; Nissen, J. S.; Rasmussen, K.; Rocco, V. P.; Schaus,
J. M.; Thompson, D. C.; Wong, D. T. EP 722941.
1
11. All new compounds gave satisfactory H NMR, IR and
MS data in accord with the assigned structure.
12. Dunlop, T.; Zhang, Y.; Smith, D. L.; Schechter, L. E. J.
Pharmacol. Toxicol. Methods 1998, 40, 47.
13. Lazareno, S.; Birdsall, N. J. M. Br. J. Pharmacol. 1993,
109, 1120.
14. Cheetham, S. C.; Viggers, J. A.; Slater, N. A.; Heal, D. J.;
Buckett, W. R. Neuropharmacology 1993, 8, 737.
15. Cheng, Y.-C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
23, 3099.
16. Morrow, A. L.; Creese, I. Mol. Pharmacol. 1986, 29, 321.
17. Hibert, M. F.; Gittos, M. W.; Middlemiss, D. N.; Mir,
A. K.; Fozard, J. R. J. Med. Chem. 1988, 31, 1087.
18. Moldt, P.; Scheel, K.; Joergen, O. G. M.; Nielsen, E. O.
WO 97/13770.
References and notes
1. Romero, L.; Casanovas, J. M.; Hervas, I.; Cortes, R.;
Artigas, F. Antidepressants 1997, 1.