M4 agonists/5HT7 antagonists with potential as antischizophrenic drugs: Serominic compounds

Article abstract of DOI:10.1016/j.bmcl.2007.01.093

Chronic low-dose treatment of rats with the psychomimetic drug, phencyclidine, induces regionally specific metabolic and neurochemical changes in the CNS that mirror those observed in the brains of schizophrenic patients. Recent evidence suggests that drugs targeting serotoninergic and muscarinic receptors, and in particular 5-HT₇ antagonists and M₄ agonists, exert beneficial effects in this model of schizophrenia. Compounds that display this combined pattern of activity we refer to as serominic compounds. Based upon leads from natural product screening, we have designed and synthesised such serominic compounds, which are principally arylamidine derivatives of tetrahydroisoquinolines, and shown that they have the required serominic profile in ligand binding assays and show potential antipsychotic activity in functional assays.

Full text of DOI:10.1016/j.bmcl.2007.01.093

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2649–2655
M4 agonists/5HT7 antagonists with potential as
antischizophrenic drugs: Serominic compounds
Colin J. Suckling,a,* John A. Murphy,a Abedawn I. Khalaf,a Sheng-ze Zhou,a
Dimitris E. Lizos,a Albert Nguyen van Nhien,a Hiroshi Yasumatsu,b,c Allan McVie,b
Louise C. Young,e Corinna McCraw,e Peter G. Waterman,e Brian J. Morris,b,d
Judith A. Prattb,e and Alan L. Harveye
aDepartment of Pure & Applied Chemistry, University of Strathclyde, 295 Cathedral Street, Glasgow G1 1XL, Scotland, UK
bYoshitomi Research Institute for Neuroscience in Glasgow (YRING), University of Glasgow, Glasgow G12 8QQ, Scotland, UK
cMitsubishi Pharma Corporation, Yokohama, Japan
dInstitute of Biomedical and Life Sciences, University of Glasgow, Glasgow G12 8QQ, Scotland, UK
eStrathclyde Institute for Pharmaceutical & Biomedical Sciences, University of Strathclyde, Glasgow G4 ONR, Scotland, UK
Received 30 November 2006; revised 29 January 2007; accepted 30 January 2007
Available online 2 February 2007
Abstract—Chronic low-dose treatment of rats with the psychomimetic drug, phencyclidine, induces regionally specific metabolic and
neurochemical changes in the CNS that mirror those observed in the brains of schizophrenic patients. Recent evidence suggests that
drugs targeting serotoninergic and muscarinic receptors, and in particular 5-HT7 antagonists and M4 agonists, exert beneficial effects
in this model of schizophrenia. Compounds that display this combined pattern of activity we refer to as serominic compounds. Based
upon leads from natural product screening, we have designed and synthesised such serominic compounds, which are principally ary-
lamidine derivatives of tetrahydroisoquinolines, and shown that they have the required serominic profile in ligand binding assays
and show potential antipsychotic activity in functional assays.
Ó 2007 Elsevier Ltd. All rights reserved.
Schizophrenia is a widespread disorder that affects
approximately 1% of the population worldwide. Current-
ly available treatments for psychotic diseases including
schizophrenia have a limited response from patients but
also havesignificant side effects.1 The first generation anti-
psychotic drugs including haloperidol (1) are effective to
some extent against the so-called positive symptoms of
schizophrenia, which include hallucinations and delu-
sions. However such compounds are ineffective against
the so-called negative symptoms, which include loss of
emotional responsiveness, lack of motivation and social
withdrawal, and also in the remediation of cognitive de-
fects in working memory, attention and executive func-
tion. It is generally accepted that conventional
antipsychotic drugs are dopamine D2 antagonists, a prop-
erty that has been associated with their activity against
positive symptoms but also with side effects such as motor
defects and hyperprolactinemia.2 The introduction of clo-
zapine (2) offered an improved clinical profile against the
cognitive deficits and negative symptoms. However cloza-
pine is a weak D2 antagonist at clinical doses.3 This clearly
indicates that antipsychotic activity is associated with
much more than D2 antagonist activity. Several mecha-
nismshavebeenproposed toexplain the atypicality ofclo-
zapine. These include relatively strongers 5-HT2A
receptor affinity compared with dopamine D2 receptor
affinity3 and ‘fast dissociation’ from the D2 receptor.4
However, there is no general agreement on the mecha-
nisms underlying the atypical antipsychotic profile of
OH
O
Cl
N
F
N
1
N
N
Cl
Keywords: Anti-schizophrenic; M4 receptor agonist; 5HT7 receptor
antagonist; Tetrahydroisoquinolines; Amidines.
N
H
*
Corresponding author. Tel.: +44 141 548 2271; fax: +44 141 548
5743; e-mail: c.j.suckling@strath.ac.uk
2
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.01.093

2650
C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
clozapine and other recently introduced drugs such as
olanzapine.
Further consideration of these structures, those of the
natural ligands and those of known synthetic ligands
led to the following definitions of structural require-
ments anticipated for serominic activity (Fig. 1).
In order to establish a new basis for the discovery of
antischizophrenic compounds, we demonstrated that
chronic intermittent exposure to phencyclidine induces
schizophrenia-like patterns of activity in the rat brain
and distinguishes between the behaviour of haloperidol
and clozapine.5,6 Moreover metabolic activity mea-
sured by 2-deoxyglucose autoradiography identified
hypoactivity in the prefrontal cortex (hypofrontality),
thalamus and temporal lobes. The regionally specific
changes together with the associated cognitive deficits
mirror those observed in schizophrenic patients.5,6 M4
muscarinic acetylcholine receptors located in the pre-
frontal cortex have been implicated in the pathology
of schizophrenia.7,8 Some of the more effective atypical
antipsychotic drugs have significant 5-HT7 affinity in
their pharmacological profile and 5-HT7 receptors are
highly localized in the thalamic nuclei9 and prefrontal
cortex where their level of expression may be altered
in schizophrenia. Based upon this information, we
hypothesised that a favourable primary profile for an
antischizophrenic drug would be 5-HT7 antagonist
activity, M4 agonist activity and low affinity for the
D2 receptor. We call this the serominic concept. Impor-
tantly, neither 5-HT7 antagonists alone nor M4 antag-
onists alone have shown activity in animal models
predictive of the negative symptoms of schizophrenia
although some activity has been claimed against posi-
tive symptoms.10,11 The selective targeting of muscarin-
ic receptor subtypes as an approach to novel therapies
for psychotic disorders has been highlighted.12 and the
difficulty in obtaining selective compounds has also
been argued.13 The importance of the dual acting com-
pound expressed by the serominic concept therefore
appears strong.
1. a framework that contains an N+.
2. a 5-HT7 responsive group, which would typically be
an aromatic system possibly with alkoxy substituents.
3. an M4 responsive group, which would typically be a
hydrogen bond acceptor such as methylenedioxy,
thiadiazole, or alkoxy.
4. the three components should be joined in such a way
as to provide an approximately planar or slightly
puckered molecule with some but limited conforma-
tional flexibility.
Of the known semi-selective M4 agonists, PTAC (6),10
and xanomeline (7)15 can adopt two primary conforma-
tions (Fig. 2) but only conformation 1 is available for
the M4 agonist 8 introduced by Lilly.16 In PTAC and
xanomeline, it is also possible to identify the same nom-
inal separation between the positively charged nitrogen
atom and a hydrogen bond acceptor as that noted in
berberine (5) above. Interestingly, xanomeline and
PTAC have been proposed as candidate antipsychotic
drugs.15,17 The Lilly M4 agonist 8 does not conform to
the same nominal pattern but, in view of its activity
proven for a required component of a serominic com-
pound, substructures from 8 were included in the design
of compounds (see below). These structural concepts,
although loosely drawn from screening and published
information, were sufficient to stimulate the design and
synthesis of compounds to evaluate the serominic con-
cept as a novel approach to antipyschotic drugs.
MeO
Lead identification. Radioligand 5-HT7, M4 and D2
receptor binding assays were established using tritiated
5-CT (5-carboxytryptamine), N-methylscopolamine
and spiperone, respectively.14 About 2000 plant extracts
from the natural products library of SIDR, University
of Strathclyde, were screened in these ligand binding as-
says. Extracts displaying significant activity in both the
5-HT7 and M4 receptor binding assays were fractionated
by solvent partitioning and purified by HPLC before
being reassayed. Several active compounds were identi-
fied by NMR spectroscopy. Significant activity in the
5-HT7 screen was identified in aporphines of which lili-
otulipiferine 3 was one of the strongest binding (Ki
80 nM). In the M4 assay, whilst aporphines themselves
were inactive, the introduction of an oxygen atom in
ring C to give oxoaporphines, exemplified by liriodenine
4, gave compounds with activity in the micromolar
range. The common structural elements in 3 and 4 asso-
ciated with the isoquinoline suggested that it might be
possible to obtain serominic compounds designed by a
conceptual fusion of the two structures. In support of
this concept, we found that berberine 5 showed both
measurable 5-HT7 (Ki ꢀ 5lM) and M4 (Ki ꢀ 2lM)
activity.
NMe
HO
HO
3
OMe
O
NMe
O
O
4
O
Me
N
O
OMe
OMe
5
Figure 1. Alkaloids from natural product screening that contributed to
the design of serominic compounds. The atoms in bold italics indicate
the conceptual binding determinants for the relevant receptors: 5-HT7
in 3 and 5; M4 in 4 and 5.

C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
2651
binding affinities for both 5-HT7 and M4 receptors and
(ii) at least 50-fold selectivity for 5-HT7 over D2 recep-
tors (Scheme 2).
S
N
N
S
N
N
C4H9S
SC4H9
NH
6 conformation 1
NH
The synthesis of series A compounds followed two
routes. For the aryl carbazoles (9–14), 5-methoxytrypta-
mine was acylated with the appropriate carboxylic acid
(>95%) and the resulting amide cyclised with phospho-
rus oxychloride to give dihydrocarbazoles (53–84%).20
The tetrahydrocarbazoles were obtained by reduction
with sodium borohydride. The benzyl carbazoles (15–
22) were prepared by condensation of 5-methoxytrypta-
mine (9–96%) with a series of arylidene oxazolones
themselves prepared from the corresponding aryl car-
boxaldehyde and N-acetylglycine (9–95%).21 The com-
plete series of compounds, 9–22, is shown in Table 1.
6 conformation 2
S
N
OH
N
Me
N
N
N
H
NH
O
O
C6H11
8
7
Cl
Series B is a more extensive collection of compounds
that takes the tetrahydroisoquinoline structure as its
Cl
Figure 2. Structural and conformational relationships in M4 agonists.
The several binding determinants corresponding to those of berberine
(Fig. 1, 5) are shown in bold italics.
NH2
OMe,H
These arguments suggest that the requirements for ser-
ominic compounds (Fig. 3) can be satisfied by a combi-
nation of two subunits, merged at N+ with geometries
consistent with the patterns shown in Fig. 2. In the 5-
HT7 component, X represents a hydrogen bond accep-
tor, in particular an alkoxy group. In the M4 binding
component, Y represents a hydrogen bond acceptor
or, based upon PTAC, a thioether.
N
i. ArCOCl
ii. POCl3
H
+
ArCHO +
AcNHCH2CO2H
Ac2O
N
O
N
OMe,H
O
Ar
Ar
N
H
OMe,H
N
NH
Ar
Compound design and synthesis. We have studied two
series of compounds that embody the serominic concept
and avoid structural relationships with haloperidol or
clozapine. Series A includes indolylethylamine and tetra-
hydrocarboline derivatives that bear hydrogen bond
acceptors or sulfur substituent in the side chain. This
series was designed to investigate new types of molecular
scaffold differing significantly from structures that had
so far shown activity. As such, this represents the more
speculative series (Scheme 1). Series B contains tetrahy-
droisoquinoline and benzylamine analogues of the Lilly
M4 agonist (8). Muscarinic antagonists from polycyclic
alkaloid analogues have been described18 and aporphine
derivatives that bind strongly to the 5-HT7 receptor
also19 but serominic dual activity is novel. We propose
that acceptable properties of a serominic are (i) similar
OMe,H
Ar
N
N
H
H
Scheme 1. Generalised synthetic scheme for series A.
(CH2)n
O
KNO3 / H2SO4
OMe,H
(CH2)n
NH
O2N
OMe,H
O
Functional group
interconversions
t
BuOCOCHO
Et3N
X
H
OMe,H
OMe,H
n = 1 - 3
N
X = OH, OAc, OPr,
SBu, S(CH2)2S,
O2CC6H3Cl2
X
+
R
(CH2)n
X
N
H
N
H2N
O
5HT7 - binding component
POCl3
Y
R
Y
N
R
OMe,H
OMe,H
X
N
(CH2)n
M4 - binding component
Figure 3. Proposed structural components of serominic compounds.
Scheme 2. Generalised synthetic scheme for series B.

2652
C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
Table 1. Carboline derivativesa
Compound
R1
R2 R3
..=..
X
5-HT7 Ki (lM)
M4 Ki (lM)
D2 Ki (lM)
D2/5-HT7 ratio
39
9
H
3-OMe
0.8
42
31
R1
N
R2
R3
N
H
9 - 13
X
10
11
12
13
14
H
H H
..=..
H H
H H
..=..
3-OH
3-OMe
H
1.1
3.5
>30
78
88
25
80
7
OMe
H
97
>300
>300
70
>300
nt
200
>3
—
26
H
OMe
3-OMe
Nitropyrrole
1.7
7.8
R1
NH
N
H
X
15 - 22
15
16
17
18
19
20
21
22
H
3-OMe
3-OMe
0.026
0.045
0.009
0.036
0.02
35
18
1.6
1.5
4.6
1.9
nt
62
33
511
53
—
—
—
—
OMe
H
3,4-OCH2O
3,4-OCH2O
2-SPr
24
12
OMe
H
94
95
H
2-SMe
4-SMe
2-OH,3-F
0.035
0.17
0.046
nt
H
68
>300
nt
H
nt
N
MeO
Me
N
N
H
14
NO2
a 5-HT7, M4 and D2 receptor binding assays were performed using tritiated 5-CT, N-methylscopolamine and spiperone, respectively (see Supple-
mentary material). n = 1–5, variation <20%.
basis in place of the tetrahydrocarboline of series A
compounds. The structural variations include the
substituents on the tetrahydroquinoline, the ring size
of the benzofused system connected through the ami-
dine, and the substitution pattern of the alicyclic ring.
A variety of benzofused alicyclic compounds were re-
quired; indanes, tetralins and benzocycloheptanes and
the corresponding ketones were obtained by adaptations
of standard methods.22 The amines required for amidine
preparation were obtained by nitration23 of the appro-
priate benzofused alicyclic compounds, functional
group modification of the nitro compounds to afford
the required alicyclic ring substituents (acetates, dic-
hlorobenzoates, ethers, thioethers and thioacetals).
Hydrogenation of the functionalised nitro compounds
over palladium on carbon afforded the corresponding
aromatic amines, which were converted into the re-
quired amidines by condensation with the appropriate
formamide in the presence of phosphorus oxychloride.24
Typical yields for the amidine formation were 40–60%.
Formylations of tetrahydroquinolines were facilitated
with the use of trimethylacetic formic anhydride.25 It
should be pointed out that alcohols like 29, 30 cannot
be used as precursors to make different esters as the ami-
dine group in the structure is very sensitive to acetyl
chloride, benzoyl chloride and acetic anhydride.
Evaluation of compounds. Although series A (Table 1)
did not contain a compound that satisfied all of the
requirements for a serominic described above, it has
nevertheless provided important pointers towards the
required properties. The aryl dihydrocarbolines 9, 11
and 14 show measurable binding activity in each assay
but the aryl tetrahydrocarbolines 10 and 13 have no
measured M4 activity. On the other hand, substituted
benzyl tetrahydrocarbolines have binding activity in
both assays. Tetrahydro-b-carbolines bind strongly to
5-HT7 receptors [e.g., 15, 16 and 18]; they also indicate

C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
2653
that a single nitrogen atom can act as the cationic site to
elicit both 5-HT7 and M4 activity. 17 and 18 were pre-
pared with the methylenedioxy ring which is an isostere
of the thiadiazole ring of PTAC. These compounds
demonstrate very effective binding to 5-HT7 receptor
compared with D2 binding activity; 17, for example, is
more than 500-fold selective for the 5-HT7 receptor
compared with the D2 receptor but, like others in series
A showed insufficient M4 binding activity for further
investigation. For comparison of affinity and selectivity,
the Kis of PTAC at 5-HT7 and M4 receptors were deter-
mined in our assays to be 3.8 and 0.002 lM, respective-
ly, whilst those of the 5-HT7 selective ligand,
SB2598741, were 0.02 and 4.3 lM, respectively.
5-HT7 and M4 receptors (27 and 34). The compound
with the most promising profile was 29, which was great-
er than 750-fold selective for the 5-HT7 receptor; it was
taken further, together with 26, into functional assays
(see below).
Bearing in mind the likely significance of sulfur in
the binding of PTAC to the M4 receptor, compounds
35–38 were prepared to explore the role of sulfur in
binding in series B. Table 2 shows that compounds
35 and 38 have similar binding affinities to the oxygen
containing compounds described above but 36 and 37
are markedly less effective in binding activity. The vinyl
sulfide 36 might have been predicted to be closer in
Lewis basicity to the sulfur in PTAC, but clearly in
our compounds, sulfur bound to a tetrahedral carbon
provides more effective binding to M4 receptors. The
bulkier dithiane groups in 35 and 39 might have caused
steric problems for binding but this appears not to be
the case. However, the most intriguing change seen
with 36 and 37 is the low binding affinities for 5-HT7
receptors.
For the series B compounds, the coupling of a tetrahy-
droisoquinoline with a substituted benzocycloalkane
led to a number of dual active compounds (serominics).
Thus several compounds (23, 26, 29, 30, 33, 34, 38 and
39) all incorporate the required units and shown signif-
icant binding activity to both 5-HT7 and M4 receptors.
The tetrahydroisoquinolines may be unsubstituted or
substituted with alkoxy groups and a variety of hydroxy
and alkoxy substituents are acceptable in the cycloal-
kane ring. Encouragingly, in some compounds D2 bind-
ing activity was significantly lower (23, 25, 29 and 30)
but in others, D2 activity remained close to that at the
The results of a ligand binding assay do not show
whether a compound is behaving as an agonist or antag-
onist. The most promising compounds from the ligand
binding assays were therefore evaluated in a functional
Table 2. Amidine derivativesa
Compound
R1
R2
n
X
5-HT7 Ki (lM)
M4 Ki (lM)
D2 Ki (lM)
D2/5-HT7 ratio
(CH2)n
R1
R1
X
N
N
R2
23 - 39
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
H
OAc
2
1
2
3
2
1
1
1
1
1
1
1
2
CH
N
0.2
2.2
1.4
50.0
0.2
2.4
12
24
t
OMe
OMe
OMe
OMe
OMe
OMe
H
CO2 Bu
OAc
OAc
54.0
11.0
14.0
42.0
14.0
>300
>300
nt
CH
CH
CH
N
0.4
0.5
28
28
0.2
OPr
5.3
12.0
2.0
8
0.6
CO2Me
OH
OH
25.0
0.4
CH
CH
CH
CH
CH
CH
C
0.3
2.8
>750
>111
—
2.7
H
O2CC6H3Cl2
O2CC6H3Cl2
OAc
>30
>30
1.0
160.0
5.3
0.2
OMe
OMe
H
>300
12.0
1.7
—
12
OAc
0.1
0.3
1.6
0.7
17
OMe
OMe
OMe
OMe
OMe
S(CH2)2S
SBu
SBu
>100
8.2
3.1
>333
<0.03
0.09
2.1
(alkene)
C
>300
36.0
0.8
11.0
4.5
1
3
1
CH
CH
C
OH
S(CH2)2S
0.3
0.3
1.7
>100
0.2
>500
N
N
R2
Me
40, 41
40
41
OAc
OH
0.1
3.6
4.5
19.0
1.6
59.0
16
16
a 5-HT7, M4 and D2 receptor binding assays were performed using tritiated 5-CT, N-methylscopolamine and spiperone, respectively (see Supple-
mentary material). n = 1–5, variation <20%.

2654
C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
assay using mouse N1E-115 cells, which produce a pure
population of M4 receptors.25 Known muscarinic ago-
nists such as oxotremorine and acetylcholine act on
muscarinic M4 receptors coupled to an inhibition of
adenylyl cyclase, leading to reduced cAMP levels.26
While full dose–response curves were not obtained for
each agonist, the maximum response observed with 26
and 29 was similar to that found for PTAC suggesting
that both 26 and 29 behave as full agonists. Because it
is unclear to what extent neutral or intrinsic 5-HT7
activity is central to our serominic hypothesis, full func-
tional characterization of 26 and 29 at this receptor has
not been carried out.
As shown in Fig. 4, both 26 and 29 at nM concentra-
tions caused a substantial reduction of cAMP levels to
less than 50% of control; this decrease was blocked by
the muscarinic antagonist, atropine. To demonstrate
in vivo antipsychotic activity, three compounds, 25,
29 and 33, were examined in the amphetamine induced
hyperactivity test in rats; this test is widely used for
assessing antipsychotic activity.27 The selection of these
compounds was in part due to the higher affinity of 25
and 33 at the D2 receptor compared with 29. Com-
pounds 25, 29 and 33 were tested at a concentration
of 10 mg/kg ip. Although very similar in structure,
the small differences between these three compounds
were found to have major effects on their activity in
this test in a manner that was not predictable
(Fig. 5a). It would not be expected that all compounds
that showed an appropriate profile in receptor binding
assays would be active in vivo because such assays take
no account of bioavailability and metabolism. Howev-
er, the most active compound, 29, was found to sup-
Figure 4. Functional assay of 26 and 29. Modulation of PACAP-
induced stimulation of cAMP by low concentrations (10À9 and
10À8 M). Antagonism by atropine (10À4 M) is clearly indicated.
affinity for the D2 receptor (Ki > 300lM) in contrast
to compounds 25 and 33, which were both measurably
active.
Beginning from studies of brain metabolism, through
lead generation by screening of natural products, li-
gand binding and functional assays, we have demon-
strated in vivo the potential of the serominic concept
as a valid basis for the discovery of new antipsychotic
compounds. Future studies will address whether com-
pounds acting at these targets display activity in our
chronic PCP model of metabolic hypofunction thereby
assessing our ability to predict an improved therapeutic
profile over existing antipsychotic drugs. The antipsy-
chotic activity of 29 without appreciable D2 antagonist
activity is a striking finding that demonstrates the po-
tential of serominic compounds as novel antipsychotic
agents.
press hyperactivity in
a dose dependent manner
(Fig. 5b) with an ED50 of around 8 mg/kg ip (95% con-
fidence limits). Interestingly 29 had the most pro-
nounced effect in the amphetamine test, whereas 25
and 33 appeared ineffective. The activity of compound
29 is unlikely to be related to antagonism of dopamine
receptors since this compound has minimal binding
Figure 5. Demonstration of serominic activity by the amphetamine induced hyperactivity test. (a) Left: effect of compounds 23, 33 and 25 at
10 mg kgÀ1. (b) Right: dose–response of compound 29.

C. J. Suckling et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2649–2655
2655
15. Shannon, H. E.; Rasmussen, K.; Bymaster, P.; Hart, J. C.;
Peters, S. C.; Swedberg, M. D. B.; Jeppesen, L.; Shear-
down, M. J.; Sauerberg, P.; Fink-Jensen, A. Schizophr.
Res. 2000, 42, 249.
16. Hollinshead, S. P.; Huff, B. E.; Hughes, P. F.; Mendoza, J.
S.; Mitch, C. S.; Staszak, M. A.; Ward, J. S.; Wilson, J. W.
WO 99/04778.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.bmcl.
2007.01.093.
17. Sauerberg, P.; Jeppesen, L.; Olesen, P. H.; Rasmussen,
T.; Swedberg, M. D. B.; Sheardown, M. J.; Fink-
Jensen, A.; Thomsen, T.; Thoegersen, H.; Rimvall, K.;
Ward, J. S.; Calligaro, D. O.; DeLapp, N. W.;
Bymaster, F. P.; Shannon, H. E. J. Med. Chem.
1998, 41, 4378.
18. Augelli-Szafran, C. E.; Jaen, J. C.; Moreland, D. W.;
Nelson, C. B.; Penvose-Yi, J. R.; Schwartz, R. D. Bioorg.
Med. Chem. Lett. 1998, 8, 1991.
19. Linnanen, T.; Brisander, M.; Uenelius, L.; Sundholm, G.;
Hacksell, U.; Johansson, A. M. J. Med. Chem. 2000, 43,
1339.
20. Kanaoka, Y.; Sato, E.; Ban, Y. Chem. Pharm. Bull. Jpn.
1967, 101.
21. Desai, N. C.; Bhatt, J. J.; Shah, B. R.; Undavia, N. K.;
Trivedi, P. B.; Narayanan, V. Farmaco 1996, 51, 361;
Audia, E.; Evrard, D. A.; Murdoch, G. R.; Droste, J.
J.; Nissen, J. S.; Schenck, K. W.; Fludzinski, P.;
Lucaites, V. L.; Nelson, D. L.; Cohen, M. L. J. Med.
Chem. 1996, 39, 2773–2780; Dakin, H. D. J. Biol. Chem.
1929, 82, 439.
22. Wan, P.; Davis, M. J.; Teo, M. A. J. Org. Chem. 1989, 54,
1354.
23. Beccalli, E. M.; Marchesini, A.; Pilati, T. Synthesis 1991,
127.
References and notes
1. Rowley, M.; Bristow, L. J.; Hutson, P. H. J. Med. Chem.
2001, 44, 477.
2. Seeman, P. Clin. Neurosci. Res. 2001, 1, 53.
3. Meltzer, H. Y. Curr. Opin. Pharmacol. 2005, 5, 101.
4. Kapur, S.; Remington, G. Biol. Psychiatry 2001, 50, 883.
5. Morris, B. J.; Cochran, S. M.; Pratt, J. A. Curr. Opin.
Pharmacol. 2005, 5, 101.
6. Cochran, S. M.; Kennedy, M.; McKerchar, C. E.;
Steward, L. J.; Pratt, J. A.; Morris, B. J. Neuropsycho-
pharmacology 2003, 28, 265.
7. Birdsall, N. J. M.; Nathanson, N. M.; Schwarz, R. D.
Trends Pharmacol. Sci. 2001, 22, 215.
8. Crook, J. M.; Tomaskovic-Crook, E.; Copolov, D. L.;
Dean, B. Am. J. Psychiatry 2001, 158, 918.
9. Dean, B.; Pavey, G.; Thomas, D.; Scarr, E. Schizophr.
Res. 2006, 88, 265.
10. Bymaster, E. P.; Shannon, H. E.; Rasmussen, K.; Delapp,
N. W.; Mitch, C. H.; Ward, J. S.; Calligaro, D. O.;
Ludvigsen, T. S.; Sheardown, M. J.; Olesen, P. H.;
Swedberg, M. D. B.; Sauerberg, P.; Fink-Jensen, A. Eur.
J. Pharmacol. 1998, 356, 109.
11. Pouzet, B.; Didricksen, M.; Arnt, J. Pharmacol. Biochem.
Behav. 2002, 71, 655.
12. Sur, S.; Kinney, G. K. Curr. Neuropharmacol. 2005, 3, 63.
13. Jensen, A. A.; Frolund, B.; Liljefors, T.; Krogsgaard-
Larsen, P. J. Med. Chem. 2005, 48, 4705.
24. Yusybov, M. S.; Filimonov, V. D. Recl. Trav. Chim. Pays-
Bas 1982, 101, 460.
25. McKinney, M.; Miller, J.; Gibson, V.; Nickelson, L.;
Aksoy, S. Mol. Pharmacol. 1991, 40, 1014.
26. Olianas, M. C.; Adem, A.; Kalsson, E.; Onali, P. Eur.
J. Pharmacol. 1998, 357, 235.
14. Thomas, D. R.; Atkinson, P. J.; Ho, M.; Bromidge, S. M.;
Lovell, P. J.; Villani, A. J.; Hagan, J. J.; Middlemiss, D.
N.; Price, G. W. Br. J. Pharmacol. 2000, 130, 409.
27. Arnt, J. Eur. J. Pharmacol. 1995, 283, 55.