A σ1 receptor pharmacophore derived from a series of N-substituted 4-azahexacyclo[5.4.1.02,6.03,10.0 5,9.08,11]dodecan-3-ols (AHDs)

Article abstract of DOI:10.1016/j.bmcl.2012.08.046

A library of N-substituted 4-azahexacyclo[5.4.1.0^2,6.0 ^3,10.0^5,9.0^8,11]dodecan-3-ols (AHDs) was synthesized and subjected to competition binding assays at σ₁ and σ₂ receptors, as well as off-target screening of representative members at 44 other common central nervous system (CNS) receptors, transporters, and ion channels. Excluding 3 low affinity analogs, 31 ligands demonstrated nanomolar K_i values for either σ receptor subtype. Several selective σ₁ and σ₂ ligands were discovered, with selectivities of up to 29.6 times for σ₁ and 52.4 times for σ₂, as well as several high affinity, subtype non-selective ligands. The diversity of structures and σ₁ affinities of the ligands allowed the generation of a σ₁ receptor pharmacophore that will enable the rational design of increasingly selective and potent σ₁ ligands for probing σ₁ receptor function.

Full text of DOI:10.1016/j.bmcl.2012.08.046

Bioorganic & Medicinal Chemistry Letters 22 (2012) 6053–6058
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A
r1 receptor pharmacophore derived from a series of N-substituted
4-azahexacyclo[5.4.1.02,6.03,10.05,9.08,11]dodecan-3-ols (AHDs)
Samuel D. Banister a,b, Miral Manoli b,c, Munikumar Reddy Doddareddy d, David E. Hibbs d,
Michael Kassiou a,b,c,
⇑
a School of Chemistry, The University of Sydney, Sydney, NSW 2006, Australia
b Brain and Mind Research Institute, Sydney, NSW 2050, Australia
c Discipline of Medical Radiation Sciences, The University of Sydney, Sydney, NSW 2006, Australia
d Faculty of Pharmacy, The University of Sydney, Sydney, NSW 2006, Australia
a r t i c l e i n f o
a b s t r a c t
A library of N-substituted 4-azahexacyclo[5.4.1.02,6.03,10.05,9.08,11]dodecan-3-ols (AHDs) was synthesized
Article history:
Received 27 June 2012
Revised 7 August 2012
Accepted 13 August 2012
Available online 21 August 2012
and subjected to competition binding assays at
resentative members at 44 other common central nervous system (CNS) receptors, transporters, and ion
channels. Excluding 3 low affinity analogs, 31 ligands demonstrated nanomolar Ki values for either
receptor subtype. Several selective r1 and r2 ligands were discovered, with selectivities of up to 29.6
times for r1 and 52.4 times for 2, as well as several high affinity, subtype non-selective ligands. The
diversity of structures and 1 affinities of the ligands allowed the generation of a 1 receptor pharmaco-
r1 and r2 receptors, as well as off-target screening of rep-
r
r
Keywords:
r
r
Trishomocubanes
Sigma receptors
CNS
phore that will enable the rational design of increasingly selective and potent r1 ligands for probing r1
receptor function.
Ó 2012 Elsevier Ltd. All rights reserved.
Structure–activity relationships
Sigma (
pressed in the central nervous system (CNS) and certain peripheral
organs.1 The two currently defined subtypes, r1 and
2, differ in
apparent molecular size, function, and ligand discrimination.2
The 1 receptor has been cloned from numerous sources, including
r
) receptors are a neuromodulatory protein, widely ex-
memory deficits,18–21 and motor dysfunction.22 Indeed, many clini-
cally used antidepressants and antipsychotics from disparate
r
mechanistic classes are known interact with
therapeutically relevant concentrations.23–27
Receptors are also
involved in the physiological processes underlying addiction,
and many drugs of abuse have been shown to interact with recep-
r receptors at
r
r
human brain tissue, and shows no sequence homology to any other
mammalian protein.3 The r1 receptor resides primarily at the
mitochondria-associated endoplasmic reticulum membrane
(MAM) where it acts as a molecular chaperone for type 3 inosi-
tol-1,4,5-triphosphate receptors to maintain correct interorganelle
r
tors, including cocaine, methamphetamine, and phencyclidine
(PCP).28–31
While
multiple disorders of the CNS, the development of truly selective
receptor ligands remains problematic. Many of the earliest r
r receptors remain a promising therapeutic target for
signalling and cytosolic Ca2+ concentrations.4,5 However,
r
1 recep-
r
tors are also known to undergo translocation to the nuclear enve-
lope and plasma membrane, accounting for their modulation of
various plasma membrane-bound proteins and maintenance of
Ca2+ homeostasis by multiple mechanisms.6–9 Historically, the
elucidation of r2 receptor structure and function has proven more
receptor ligands were discovered serendipitously and were multi-
functional ‘dirty drugs’, such as haloperidol. The structural hetero-
geneity of current
remains an impediment to understanding
r
ligands is extreme, and ligand promiscuity
receptor pharmacol-
r
ogy. Several selective ligands have been identified for both r1
and r2 subtypes, however, such compounds comprise relatively
few distinct structural classes. The identification of new chemo-
difficult, however, it was very recently proposed that the r2 recep-
tor is actually progesterone receptor membrane component 1
(PGMRC1).10
types with
r subtype selectivity and truly negligible off-target
The diverse, neuromodulatory pharmacology exhibited by
r
activity remains a goal for the development of pharmacological
tools and potential therapeutic agents targeting these sites.
The 4-azahexacyclo[5.4.1.02,6.03,10.05,9.08,11]dodecan-3-ol
receptors has implicated these proteins in virtually all major CNS
diseases,11,12 with compelling evidence that
r receptors play a cen-
tral or ancillary role in anxiety and depression,13–15 psychosis, 16,17
(AHD) scaffold confers affinity for
pended at the nitrogen atom, as in 1 (Fig. 1), and analogous hemia-
minals have demonstrated selectivity for receptors over 44 other
major CNS receptors, transporters, and ion channels. Members of
r receptors when judiciously ap-
r
⇑
Corresponding author.
E-mail address: michael.kassiou@sydney.edu.au (M. Kassiou).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.08.046

6054
S. D. Banister et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6053–6058
F
F
OH
OH
R
OH
N
N
N
O
1: R = H
2: R = F
3
4
O
O
OH
H
N
OH
N
N
O
O
5
6
N
O
N
N
N
N
O
8
7 (SA-4503)
Figure 1. Selected
r
receptor ligands.
this class, including 2 and 3, have displayed promising anti-cocaine
effects in mice,32 but few structure-affinity relationships (SAfiRs)
have been established for this class due to the limited structural
variability of reported members.33–35 Preliminary trends suggest
that distance between the polycyclic hemiaminal and the aryl
The synthesis of target AHDs is depicted in Scheme 1. The Diels-
Alder reaction of cyclopentadiene (9) and 1,4-benzoquinone (10)
gave the well-precedented adduct 11,42 which underwent [2+2]
photocyclization to give pentacyclo[5.4.0.02,6.03,10.05,9]undecane-
8,11-dione (Cookson’s diketone, 12).43 Although cage diketone 12
is commercially available, the quantities required and the ease of
synthesis deemed in-house production more economical, with up
to 12 g produced in a single run from inexpensive precursors. Pro-
tection of a single ketone functionality of 12 as its ethylene acetal
gave the racemic ketal 13. Condensation of the remaining ketone of
13 with appropriate primary amines at high temperature (sealed
tube) gave the corresponding imines, which were subsequently re-
duced by sodium borohydride to stereoselectively afford corre-
sponding endo-amines of type (14). Acetal hydrolysis in aqueous
acid with acetone as co-solvent and donor ketone, followed by
basic work-up, furnished the transannularly-cyclized products
15–40. The structures and yields of AHDs 15–39 are presented in
Table 1.
group is the primary determinant of
r subtype selectivity, with
benzyl AHD derivatives generally preferring the r2 receptor and
phenethyl AHD derivatives typically demonstrating selectivity for
the r1 receptor, allowing the generation of ligands for either r sub-
type from a common scaffold. Aromatic substitution patterns have
been explored to a limited extent, and small, electronegative atoms
in the 3-position appear to confer the greatest enhancement of
affinity and subtype selectivity. A molecular hybridization strategy
utilizing AHD produced 4, 5, and 6, which demonstrated profound
alterations of
r subtype selectivity and attenuated off-target
interaction when compared to their respective parent compounds;
haloperidol, NE-100, and RHM-2.36
Having recently explored the effect of hemiaminal isomeriza-
tion37,38 and expansion of the trishomocubane cage39 of the AHD
The requisite amines for step (d) of Scheme 1 were generally
commercially available, with the exception of those comprising
chemotype on
would expand the SAfiRs for this class, potentially allowing the
generation of subtype pharmacophores. The present study aimed
r binding, exploration of the N-substituent region
r
O
O
O
to systematically explore the importance of (i) alkyl chain length,
(ii) alkoxyaromatic substitution patterns, (iii) heteroaromatic
incorporation, and (iv) aliphatic N-substitution. To address the first
goal, analogs of 1 and 2 containing two to four methylene unit-
spacers were envisaged. The second aim sought to explore the
importance of substitution of the aromatic ring, given the frequent
H
a
b
O
H
O
O
9
10
11
12
recurrence of the (poly)methoxyphenyl motif among
r ligands,
such as SA-4503 (7, r1 Ki = 4.63 nM, r2
/
r
1 = 13.6).40 Such com-
c
pounds may be amenable to the incorporation of carbon-11, there-
by providing potential positron emission tomography (PET) tracers
for imaging
r receptors in living systems. The incorporation of het-
e
d
eroaromatic rings has not been explored within the AHD class,
however, the inclusion of certain pyridine regioisomers was shown
O
O
OH
O
O
N(CH2)nR
to confer r2 selectivity to
r
ligands such as 8 (r2 Ki = 4.91 nM, r1
/
O
N(CH2)nR
r
2 = 16.9).41 Finally, molecular modeling of this class had indicated
15-40
14
13
that
crease
p
density on the aromatic ring of AHD analogs 2 and 3 may de-
receptor affinity.37 Therefore, a similarly sized N-substitu-
r
Scheme 1. Reagents and conditions: (a) PhMe,À10 °C–rt, 80%; (b) h
m, hexane-
ent that is alicyclic rather than aromatic may possess a more
optimal highest occupied molecular orbital (HOMO) for
interaction.
Me2CO (90:10), rt, 14 h, 93%; (c) HOCH2CH2OH, p-TsOH (cat.), PhMe, reflux, Dean-
Stark, 5 h, 93%; (d)(i) R(CH2)nNH2, EtOH, 100 °C, sealed tube, 18 h; (ii) NaBH4, EtOH,
0 °C–rt, 8 h; (e) 4 M aq. HCl, Me2CO, rt, 12 h, basic work-up, 22–65% (over 3 steps).
r receptor

S. D. Banister et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6053–6058
6055
4, 5, 6, 38, and 40. The synthesis of 4, 5, and 6 has been previously
described, but 38 and 40 required 3-(3-fluorophenyl)propylamine
(41, Scheme 2) and 4-(3-fluorophenyl)butylamine (42, Scheme 3)
respectively. The former was synthesized from 3-fluorocinnamic
acid (43) via 3-fluorocinnamamide (44), with complete reduction
Consistent with the previously established SAfiRs, the benzylic
derivatives 1 and 2 were both r2 selective ligands (Table 1). How-
ever, contrasting previous findings, the simple benzylic derivative
1 displayed the greatest r2 selectivity (Ki = 12.6 nM, r1
with a 3-fluoro substituent (2) reducing r2 affinity (Ki = 31.0 nM)
and diminished selectivity over r1 2 = 4.9). The phenethyl
derivative 27 showed a preference for r1 receptors (Ki = 26 nM,
1 = 2.6) and a 3-fluoro substituent doubled 1 affinity, but pro-
duced little net improvement in selectivity over 2 (Ki = 12 nM, r2
1 = 4.0). Increasing the distance between the trishomocubyl
hemiaminal and aromatic ring led to improved binding at both
r1 and r2 receptors, effectively abolishing subtype selectivity.
/r2 = 26.7),
of the
a,b-unsaturated amide achieved by lithium aluminum hy-
(r1/r
dride, to furnish 41 in 54% yield over 3 steps. The synthesis of
the latter started with N-(3-bromopropyl)phthalimide (45), which
was reacted with triphenylphosphine to give phosphonium bro-
mide 46. 3-Fluorobenzaldehyde was subjected to a Wittig reaction
with 46 and the obtained distaereomeric mixture of alkenes (47)
was hydrogenated over palladium on carbon to give saturated
amine 48. Removal of the phthalimide group was achieved by
hydrazinolysis to give 42 in 57% overall yield.
r2/r
r
r
/
r
r
Phenylpropyl derivative 37 (r1 Ki = 17 nM, r2 Ki = 26 nM) and its
3-fluoro analog, 38 (r1 Ki = 15 nM, r2 Ki = 6.3 nM), displayed al-
most equivalent affinity for the r1 and r2 receptor subtypes. Phen-
ylbutyl derivative 39 displayed modest selectivity for the r1
receptor (r1 Ki = 10 nM, r2 Ki = 54 nM), but its 3-fluoro analog
The N-substituted 4-azahexacyclo[5.4.1.02,6.03,10.05,9.08,11]dod-
ecan-3-ols 15–40 were subjected to protein binding assays against
a panel of CNS receptors. The Ki values for 15–40 at r1 and r2
receptors are shown in Table 1. Rat brain homogenates were used
(40) was a high affinity, dual
r receptor ligand (r1 Ki = 5.5 nM, r2
as the source of r1 receptors, whilst PC12 cells were used as the r2
Ki = 8.9 nM).
receptor source. The radioligands [3H](+)-pentazocine and [3H]DTG
were used in the r1 and r2 receptor assays, respectively. Selected
AHDs were comprehensively screened against major CNS recep-
tors, transporters, and ion channels (see Table S1 of Supplementary
data for full binding profiles). In general, compounds 15–40
showed negligible affinity for adrenergic, dopamine, GABA, hista-
mine, 5-HT, muscarine, and opioid receptors, as well as mono-
Substitution of the benzyl group of 1 with a single methoxy unit
was equally well tolerated by 2 receptors at the 2-, 3-, and 4-posi-
tions (15, 16, and 17 respectively). Like 1 itself, the methoxybenzyl
regioisomers were selective for r2 receptors, and this was most
pronounced for 3-methoxy analogue 16 (r1/r2 = 12.2). A 3,4-dime-
r
thoxy substitution pattern was optimal for r2 binding, with 19
possessing both the greatest affinity and subtype selectivity for
amine transporters (DAT, NET, SERT), and ion channels (Ca2+
NMDA/PCP), confirming the utility of AHD for the development
,
the r2 receptor (Ki = 12.6 nM, r1
a 2,4-dimethoxy substitution pattern (18) was tolerated at both
r subtypes, but 3,5-dimethoxy analog 20 showed markedly
/r2 = 18.1) within this series, and
of selective
r
ligands.
Table 1
Yields and binding affinities of compounds 1–6 and 15–39 for r1 and r2 receptors.
Cmpd
n
R
Yielda (%)
Ki (nM SEM)b
Selectivitiy
r1
r2
r1
r2
1
1
1
2
—
—
—
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
4
4
Ph
3-FPh
3-FPh
—
—
—
63
34
52
42
50
65
22
53
48
61
41
45
38
61
38
36
37
39
61
33
37
49
39
53
41
56
48
33
63
55
58
49
337 21
153 17
12
31
48 10
55
93
225 18
27.3 4.1
26.8 3.7
27.3 4.0
77
12.6 2.2
125
259 38
>10000
2.2 0.3
20.0 1.0
3234 397
514 46
68 14
56.9 8.9
100 14
56.9 9.0
2
6
28
4.9
2c
3c
12
27
20
1
2
1
4.0
2.0
4.7
4d
4
5
5d
6d
7.6 1.0
190 22
327 26
200 26
29.6
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
2-(OCH3)Ph
3-(OCH3)Ph
4-(OCH3)Ph
2,4-(OCH3)2Ph
3,4-(OCH3)2Ph
3,5-(OCH3)2Ph
3,4-(OCH2O)Ph
3,4,5-(OMe)3Ph
Cyclohexane
2-Pyridine
3-Pyridine
4-Pyridine
Ph
2-(OCH3)Ph
3-(OCH3)Ph
4-(OCH3)Ph
2,3-(OCH3)2Ph
3,4-(OCH3)2Ph
3,4-(OCH2O)Ph
2-Pyridine
3-Pyridine
4-Pyridine
Ph
7.0
12.2
7.3
66
2
7
1.2
2.3
228 20
610 34
435 51
4289 395
6.7 0.8
1048 30
>10000
1502 148
18.1
4.9
1.7
5
3.0
52.4
3.1
2.9
26
2
2.6
0.8
1.9
3.8
7.1
2.6
3.7
68.7 6.0
52.3 4.0
15.1 1.1
15.0 0.7
101 13
107
259 38
97 10
26.8 4.1
100 14
404 32
5
26
2
1170 186
815 126
68.0 6.0
43.7
8.2
5.9
1.5
17
15
10
1
1
1
26
6
3-FPh
Ph
3-FPh
6.3 0.7
54 13
8.9 1.0
2.4
5.4
1.6
5.5 1.0
a
b
c
Un optimized yield over three steps.
Ki values represent the mean SEM of four experiments.
Data extracted from Ref.37.
d
Data extracted from Ref.36.

6056
S. D. Banister et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6053–6058
O
O
F
a, b
F
c
F
HO
H2N
H2N
43
44
41
Scheme 2. Reagents and conditions: (a) SOCl2, DMF (cat.), CHCl3, reflux, 4 h; (b) 28% aq. NH4OH, rt, 18 h, 66% over 2 steps; (c) LiAlH4, Et2O, reflux, 14 h, 81%.
O
O
O
N
F
b
d
a
N
Br
N
PPh3Br
O
O
O
45
46
47
48
c
O
H2N
F
N
F
O
42
Scheme 3. Reagents and conditions: (a) p-xylene, reflux, 45 h, 92%; (b) 3-FPhCHO, NaOMe, THF, 0 °C to rt, 41 h, 67%; (c) H2, 10% Pd/C (cat.), THF-MeOH (50:50), rt, 18 h, 98%;
(d) NH2NH2ÁH2O, EtOH, reflux, 2 h, 95%.
reduced affinity for both
enedioxy bridge (21) was detrimental to
r
receptors. Installation of a 3,4-methyl-
binding, as was further
With the exception of 3,4,5-trimethoxybenzyl derivative 22,
and 3- and 4-pyridine analogs 25 and 26 respectively, all N-substi-
r
elaboration to 3,4,5-trimethoxybenzyl analogue 22, with these
analogs showing submicromolar and micromolar binding affinities
tuted AHDs were found to interact with
cromolar affinity. The distance between the hemiaminal and the
aromatic ring is the greatest determinant of subtype selectivity
r receptors with submi-
respectively. The difference in
r
affinity between 19 and 22 sug-
r
gests that subtle steric or electronic alterations have a profound
influence on r2 selectivity within this series.
Alkoxy-substituted phenethyl AHDs 28–33, with the exception
of 28, showed a slight preference for r1 binding, with Ki values rang-
within this series; benzyl derivatives are r2 selective, phenethyl
derivatives are r1 preferring. However, fewer high affinity, highly
subtype selective r1 ligands were identified compared to r2 li-
gands. To understand the origin of subtype selectivity in N-substi-
tuted AHDs, and guide the development of increasingly potent and
ing from 15–101 nM, and generally low levels of subtype selectivity.
Methoxy substitution in the 2- or 3-position (28 and 29 respec-
selective r1 ligands, a r1 pharmacophore model for the series of 32
tively), imparted moderate affinity for both
4-methoxy derivative 30 displaying the greatest preference for r1
(Ki = 15.1 nM, r2 1 = 3.8). Unlike its benzylic counterpart, the 3,4-
dimethoxyphenethyl analog 32 displayed only low affinity for both
receptors, with regioisomeric 2,3-dimethoxy 31 proving superior
in terms of r1 affinity and subtype selectivity (Ki = 15.0 nM, r2
1 = 7.1). Incorporation of a 3,4-methylenedioxy bridge (33) gave
a modestly selective 1 ligand (Ki = 26 nM, r2 1 = 3.7).
r
subtypes, with
N-substituted AHDs shown in Table 1 was developed using the
Phase program provided in Maestro.44,45
/r
The generated five feature pharmacophore represented as dis-
tances and angles is shown in Figures 2a and 2b respectively,
and contained two hydrophobic regions (H6 and H11), one aro-
matic ring (R15), one positive ionizable feature (P14), and one
hydrogen bond acceptor (A1). The distances between the central
positive ionizable group and the two hydrophobic regions are in
agreement with the few published r1 pharmacophores.46–49 The
distance between the positive ionizable N and hydrophobic region
(H6) near the aromatic ring, which Glennon et al.46 described as
the primary hydrophobic region, is about 6.3 Å, whereas the dis-
tance between N and hydrophobic region (H11), near the hydrogen
bond acceptor group, is 3.6 Å (secondary hydrophobic region).
Figure 3 shows the mapping of the generated pharmacophore
with two of the best fitting r1 active compounds; 31 and 3. The
aromatic ring region is occupied by a phenyl ring in both com-
pounds, with primary hydrophobic region (H6) occupied by a
methoxy group in 31 and a fluorine atom in 3. The other three
pharmacophoric features, namely positive ionizable, hydrogen
bond acceptor, and secondary hydrophobic region, are present in
most of the dataset and are occupied by a basic nitrogen atom, hy-
droxy group, and the hexacyclo[5.4.1.02,6.03,10.05,9.08,11]dodecane
cage respectively. Figure 2 also includes yellow excluded volumes
generated using the sterically unfavored regions of inactive com-
pounds. Both depicted active compounds clearly do not overlap
r
/
r
r
/r
The 2-pyridylmethyl derivative 24 displayed high affinity for r2
receptors (Ki = 20.0 nM), and more than 50 times selectivity over
the
r1 receptor. The regioisomeric 3-pyridylmethyl (25) and 4-pyr-
idylmethyl (26) analogs were much poorer
r
2 ligands, with submi-
cromolar and micromolar binding affinities respectively. The same
general trend was observed for the pyridylethyl regioisomers, with
a 2-pyridyl group imaprting high r2 affinity and selectivity to 34
(Ki = 26.8 nM, r1
sessed moderate affinity and selectivity for
idylethyl derivative 36 showed moderate affinity for the r1
receptor (Ki = 68.0 nM, r2 1 = 5.9).
/r2 = 43.7). The 3-pyridylethyl derivative 35 pos-
r2, however, the 4-pyr-
/r
AHD 23, containing a cyclohexane ring in place of the phenyl
ring of 1, was prepared to investigate the hypothesis that HOMO
density from aromatic
receptor affinity. Confirming this hypothesis, 23 was found to
possess high affinity for both receptor subtypes (r1 Ki = 6.7 nM,
p orbitals may be inversely correlated with
r
r
r1 Ki = 2.2 nM), making it the most potent
r receptor ligand iden-
tified within the AHD series to date.

S. D. Banister et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6053–6058
6057
Figure 2. Best ranked five feature pharmacophoric hypothesis with (a) distances and (b) angles. Spheres represent favored hydrophobic regions (H6 and H11), positive
ionizable region (P14), hydrogen bond acceptor (A1), and aromatic ring (R15).
Figure 3. Five feature pharmacophore mapping to active compounds (a) 31 and (b) 3. Yellow spheres represent sterically unfavored excluded volumes.
with these regions, indicating no unfavorable steric interactions
with the receptor.
The AHD scaffold represents an excellent platform for the devel-
opment of highly selective, high affinity receptor ligands. Judi-
r
Figure 4 shows the mapping of the pharmacophore with the
two least r1 active compounds of the dataset; 25 and 22. Com-
pound 25 (Fig. 4a) does not occupy the primary hydrophobic re-
gion (H6), and contains an electron deficient pyridine in place of
the phenyl ring of 1. Even though compound 22 (Fig. 4b) contains
all the pharmacophoric features and maps well to the hypothesis,
the two yellow excluded volumes are occupied by two methoxy
groups, causing unfavorable steric interactions with the receptor.
One of the limitations of the hypothesis is that compound 23,
one of the most active of the dataset, does not map properly due
to the absence of an aromatic ring. However, its high affinity indi-
cates that alicyclic hydrophobic groups are accepted in place of
substituted phenyl rings.
cious selection of a suitable N-arylalkyl substituent has permitted
the discovery of several selective r2 ligands (1, 19, 24, and 34),
and r1 ligands (6, 31), as well many highly
r selective, dual sub-
type ligands (e.g. 23, 38, and 40). Excluding 22, 25, and 26, Ki values
for N-substituted AHDs binding at the r1 receptor subtype ranged
from 5.5–1170 nM, and from 6.3–404 nM at the r2 receptor. The
proposed SAfiRs indicate that highly r2 selective ligands can be
generated by appropriate ring substitution of the N-benzylic region
of 1, or substitution of this group with a 2-pyridinealkyl unit. Few-
er r1 selective ligands have been generated from AHD, prompting
the generation of a five feature pharmacophore to suggest a direc-
tion for the rational design of such agents in the absence of any
structural information describing the r1 binding site. Ongoing
Figure 4. Five feature pharmacophore mapping to inactive compounds (a) 25 and (b) 22. Yellow spheres represent sterically unfavored excluded volumes.

6058
S. D. Banister et al. / Bioorg. Med. Chem. Lett. 22 (2012) 6053–6058
15. Hayashi, T.; Stahl, S. M. Drugs Future 2009, 34, 137.
16. Deutsch, S. I.; Weizman, A.; Goldman, M. E.; Morihisa, J. M. Clin.
Neuropharmacol. 1988, 11, 105.
17. Snyder, S. H.; Largent, B. L. J. Neuropsychiatry Clin. Neurosci. 1989, 1, 7.
18. Maurice, T.; Lockhart, B. P. Prog. Neuropsychopharmacol. Biol. Psychiatry 1997,
21, 69.
efforts to further elaborate the substituted AHDs will utilize the
accumulated SAfiRs and five feature pharmacophore, ultimately
providing increasingly selective ligands for either
r subtype from
a common scaffold. These pharmacological tools demonstrably
possess favourable in vivo properties, and will be used to delineate
the functions of r1 and r2 receptors in CNS disease.
19. Maurice, T.; Phan, V. L.; Privat, A. Brain Res. 2001, 898, 113.
20. Maurice, T. Drug News Perspect. 2002, 15, 617.
21. Phan, V. L.; Urani, A.; Sandillon, F.; Privat, A.; Maurice, T. Neurobiol. Aging 2003,
24, 865.
Acknowledgments
22. Walker, M. J. Eur. Neuropsychopharmacol. 1996, 6, 159.
23. Tam, S. W.; Cook, L. Proc. Nat. Acad. Sci. U.S.A. 1984, 81, 5618.
24. Itzhak, Y.; Kassim, C. O. Eur. J. Pharmacol. 1990, 176, 107.
25. Rao, T. S.; Cler, J. A.; Mick, S. J.; Dilworth, V. M.; Contreras, P. C.; Iyengar, S.;
Wood, P. L. Neuropharmacology 1990, 29, 1191.
26. Narita, N.; Hashimoto, K.; Tomitaka, S.; Minabe, Y. Eur. J. Pharmacol. 1996, 307,
117.
27. Hashimoto, K. Cent. Nerv. Syst. Agents Med. Chem. 2009, 9, 197.
28. Witkin, J. M. Neurosci. Biobehav. Rev. 1994, 18, 121.
29. Nguyen, E. C.; McCracken, K. A.; Liu, Y.; Pouw, B.; Matsumoto, R. R.
Neuropharmacology 2005, 49, 638.
Ki determinations for targets included in Table 1 and Table S1
were generously provided by the National Institute of Mental
Health’s Psychoactive Drug Screening Program, Contract
#NO1MH32004 (NIMH PDSP). The NIMH PDSP is directed by Bryan
L. Roth MD, PhD at the University of North Carolina at Chapel Hill
and Project Officer Jamie Driscol at NIMH, Bethesda MD, USA. For
experimental details please refer to the PDSP web site http://
www.pdsp.med.unc.edu/.
30. Brammer, M. K.; Gilmore, D. L.; Matsumoto, R. R. Eur. J. Pharmacol. 2006, 553,
141.
31. Matsumoto, R. R. Expert Rev. Clin. Pharmacol. 2009, 2, 351.
32. Liu, X.; Banister, S. D.; Christie, M. J.; Banati, R.; Meikle, S.; Coster, M. J.; Kassiou,
M. Eur. J. Pharmacol. 2007, 555, 37.
Supplementary data
33. Nguyen, V. H.; Kassiou, M.; Johnston, G. A.; Christie, M. J. Eur. J. Pharmacol.
1996, 311, 233.
34. Liu, X.; Kassiou, M.; Christie, M. J. Aust. J. Chem. 1999, 52, 653.
35. Liu, X.; Kassiou, M.; Christie, M. J.; Hambley, T. W. Aust. J. Chem. 2001, 54, 31.
36. Banister, S. D.; Moussa, I. A.; Jorgensen, W. T.; Chua, S. W.; Kassiou, M. Bioorg.
Med. Chem. Lett. 2011, 21, 3622.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.
08.046.
References and notes
37. Banister, S. D.; Moussa, I. A.; Jordan, M. J. T.; Coster, M. J.; Kassiou, M. Bioorg.
Med. Chem. Lett. 2010, 20, 145.
38. Banister, S. D.; Moussa, I. A.; Beinat, C.; Reynolds, A. J.; Schiavini, P.; Jorgensen,
W. T.; Kassiou, M. Bioorg. Med. Chem. Lett. 2011, 21, 38.
39. Banister, S. D.; Yoo, D. T.; Chua, S. W.; Cui, J.; Mach, R. H.; Kassiou, M. Bioorg.
Med. Chem. Lett. 2011, 21, 5289.
40. Lever, J. R.; Gustafson, J. L.; Xu, R.; Allmon, R. L.; Lever, S. Z. Synapse 2006, 59,
350.
41. Maeda, D. Y.; Williams, W.; Kim, W. E.; Thatcher, L. N.; Bowen, W. D.; Coop, A.
Bioorg. Med. Chem. Lett. 2002, 12, 497.
42. Yates, P.; Switlak, K. Can. J. Chem. 1894, 1990, 68.
43. Marchand, A. P.; Allen, R. W. J. Org. Chem. 1974, 39, 1596.
44. Dixon, S.; Smondyrev, A.; Knoll, E.; Rao, S.; Shaw, D.; Friesner, R. J. Comput.
Aided. Mol. Des. 2006, 20, 647.
45. Dixon, S. L.; Smondyrev, A. M.; Rao, S. N. Chem. Biol. Drug Des. 2006, 67, 370.
46. Glennon, R. A.; Ablordeppey, S. Y.; Ismaiel, A. M.; el-Ashmawy, M. B.; Fischer, J.
B.; Howie, K. B. J. Med. Chem. 1994, 37, 1214.
47. Gund, T. M.; Floyd, J.; Jung, D. J. Mol. Graph. Model. 2004, 22, 221.
48. Laggner, C.; Schieferer, C.; Fiechtner, B.; Poles, G.; Hoffmann, R. D.; Glossmann,
H.; Langer, T.; Moebius, F. F. J. Med. Chem. 2005, 48, 4754.
49. Zampieri, D.; Mamolo, M. G.; Laurini, E.; Florio, C.; Zanette, C.; Fermeglia, M.;
Posocco, P.; Paneni, M. S.; Pricl, S.; Vio, L. J. Med. Chem. 2009, 52, 5380.
1. Walker, J. M.; Bowen, W. D.; Walker, F. O.; Matsumoto, R. R.; De Costa, B.; Rice,
K. C. Pharmacol. Rev. 1990, 42, 355.
2. Quirion, R.; Bowen, W. D.; Itzhak, Y.; Junien, J. L.; Musacchio, J. M.; Rothman, R.
B.; Su, T. P.; Tam, S. W.; Taylor, D. P. Trends Pharmacol. Sci. 1992, 13, 85.
3. Kekuda, R.; Prasad, P. D.; Fei, Y. J.; Leibach, F. H.; Ganapathy, V. Biochem.
Biophys. Res. Commun. 1996, 229, 553.
4. Hayashi, T.; Su, T.-P. Cell 2007, 131, 596.
5. Tsai, S.-Y.; Hayashi, T.; Mori, T.; Su, T.-P. Cent. Nerv. Syst. Agents Med. Chem.
2009, 9, 184.
6. Morin-Surun, M. P.; Collin, T.; Denavit-Saubie, M.; Baulieu, E. E.; Monnet, F. P.
Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 8196.
7. Hayashi, T.; Maurice, T.; Su, T. P. J. Pharmacol. Exp. Ther. 2000, 293, 788.
8. Aydar, E.; Palmer, C. P.; Klyachko, V. A.; Jackson, M. B. Neuron 2002, 34, 399.
9. Hayashi, T.; Su, T. P. J. Pharmacol. Exp. Ther. 2003, 306, 726.
10. Xu, J.; Zeng, C.; Chu, W.; Pan, F.; Rothfuss, J. M.; Zhang, F.; Tu, Z.; Zhou, D.; Zeng,
D.; Vangveravong, S.; Johnston, F.; Spitzer, D.; Chang, K. C.; Hotchkiss, R. S.;
Hawkins, W. G.; Wheeler, K. T.; Mach, R. H. Nat. Commun. 2011, 2, 380.
11. Maurice, T.; Su, T.-P. Pharmacol. Ther. 2009, 124, 195.
12. Banister, S.; Kassiou, M. Curr. Pharm. Des. 2012, 18, 884.
13. Bermack, J. E.; Debonnel, G. J. Pharmacol. Sci. 2005, 97, 317.
14. Kulkarni, S. K.; Dhir, A. Expert Rev. Neurother. 2009, 1021, 9.