C5-substituted derivatives of 5-OMe-BPAT: Synthesis and interactions with dopamine D2 and serotonin 5-HT(1A) receptors

Article abstract of DOI:10.1016/S0968-0896(99)00196-0

Eight new C5-substituted derivatives of the potential atypical antipsychotic agent 5-methoxy-2-[N-(2-benzamidoethyl)-N-n- propylamino]tetralin (5-OMe-BPAT, 1) have been prepared by chemical conversion of the 5-trifluoromethylsulfonyloxy (triflate) analogu

Full text of DOI:10.1016/S0968-0896(99)00196-0

Bioorganic & Medicinal Chemistry 7 (1999) 2541±2548
C5-Substituted Derivatives of 5-OMe-BPAT: Synthesis and
Interactions with Dopamine D2 and Serotonin 5-HT1A Receptors
Evert J. Homan, a,* Martin Th. M. Tulp, b Jonas E. Nilsson, c Hakan V. Wikstrom a
and Cor J. Grol a
aDepartment of Medicinal Chemistry, University Centre for Pharmacy, University of Groningen, Antonius Deusinglaan 1,
NL-9713 AV Groningen, The Netherlands
bDepartment of Pharmacology, Solvay Pharmaceuticals B.V., PO Box 900, NL-1380 DA Weesp, The Netherlands
cDepartment of Structural Chemistry, Pharmacia & Upjohn AB, S-751 82 Uppsala, Sweden
Received 3 December 1998; accepted 18 June 1999
AbstractÐEight new C5-substituted derivatives of the potential atypical antipsychotic agent 5-methoxy-2-[N-(2-benzamidoethyl)-N-n-
propylamino]tetralin (5-OMe-BPAT, 1) have been prepared by chemical conversion of the 5-tri¯uoromethylsulfonyloxy (tri¯ate) ana-
logue 4 via various Stille-type cross-couplings, a Heck reaction, and an amidation in moderate to good yields. The 5-acetyl, 5-cyano, 5-
methyl, 5-(2-furyl), 5-phenyl, methyl 5-carboxylate, and the 5-carboxamido analogues 5±11 thus obtained, the previously disclosed 5-
methoxy, 5-hydroxy, and 5-unsubstituted analogues 1±3, and the 5-tri¯ate analogue 4 were evaluated for their ability to compete for
[3H]-spiperone binding to rat striatal membranes containing dopamine D2 receptors, and their ability to compete for [3H]-8-OH-DPAT
binding to rat frontal cortex membranes containing serotonin 5-HT1A receptors in vitro. Compounds 1±11 displayed weak to high a-
nities for dopamine D2 receptors, with Ki-values ranging from 550 nM for the 5-carboxamido analogue to 4.9 nM for the 5-hydroxy
analogue. The relative anities of the 5-methoxy, 5-hydroxy, and 5-unsubstituted analogues suggested that these compounds may bind
to the same site and in a similar way as the 5-oxygenated DPATs, with the 5-methoxy substituent of 1 functioning as a hydrogen bond
acceptor. The serotonin 5-HT1A receptor tolerated more structural diversity at the C5-position of 1, as revealed by the higher Ki-values
of 1±11, which ranged from 60 nM for the 5-carboxamido analogue to 1.0 nM for the 5-unsubstituted analogue. Partial least-squares
(PLS) analysis of a set of 24 molecular descriptors, generated for each analogue, revealed no signi®cant correlation between the dopa-
mine D2 receptor anities of 1±11 and their molecular properties, supporting the view that they may have di€erent binding modes at this
receptor subtype. A PLS model with moderate predictability (Q2=0.49) could be derived for the serotonin 5-HT1A receptor anities of
1±11. According to the model, a relatively lipophilic, nonpolar C5-substituent should be optimal for a high anity at this receptor
subtype. # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
2-[N-2-(benzamidoethyl)-N-n-propylamino]tetralin (5-
OMe-BPAT, 1) were shown to possess di€erent pharma-
cological pro®les: whereas both enantiomers behave as
full serotonin 5-HT1A receptor agonists, (R)-1 appears to
block dopamine D2 receptors, while (S)-1 possesses
dopamine D2 receptor-stimulating properties.14 Using
molecular modeling studies, we were able to rationalise
these di€erences in pharmacological behaviour by show-
ing that (R)- and (S)-1 may have di€erent modes of bind-
ing to the dopamine D2 receptor.15 Speci®cally, di€erences
were observed in the interactions of the 5-methoxy sub-
stituents of both compounds and a serine residue (Ser193)
in transmembrane domain (TM) 5 of the dopamine D2
receptor, which may explain the di€erences in intrinsic
ecacy at this receptor subtype.
Preclinical1±4 and post-mortem5±8 studies suggest that
compounds, which simultaneously block dopamine D2
receptors and stimulate serotonin 5-HT1A receptors,
may possess atypical antipsychotic properties. Promis-
ing results from preclinical studies have recently been
reported for several novel compounds with the indicated
pharmacological pro®le.9±11 2-Aminotetralin-derived
benzamides comprise a novel class of compounds with
mixed dopamine D2, D3 and serotonin 5-HT1A receptor
binding properties.12,13 The enantiomers of 5-methoxy-
Key words: Dopamine; serotonin; atypical antipsychotic; tri¯ate;
QSAR.
* Corresponding author. Present address: Pharmacopeia, Inc., CN
5350, Princeton, NJ 08543-5350, USA. Tel.: +1-609-452-3753; fax:
+1-732-422-0156; e-mail: ehoman@pharmacop.com
In order to further explore the di€erences in structural
requirements of the dopamine D2 and serotonin 5-HT1A
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00196-0

2542
E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
receptor with respect to the C5-substituent of the 2-
aminotetralin-derived benzamides, the 5-tri¯ate, 5-ace-
tyl, 5-cyano, 5-methyl, 5-(2-furyl), 5-phenyl, methyl 5-
carboxylate, and the 5-carboxamido analogues of 1
were prepared and their anities for dopamine D2 and
serotonin 5-HT1A receptors were determined in vitro.
modi®cation of the procedure originally described by
Chambers and Widdowson.19 Thus, reaction of 4 with
potassium cyanide in the presence of zinc dust, triphe-
nylphosphine and bistriphenylphosphine-nickel(II)
chloride as a catalyst, employing DMF as a solvent,
gave complete conversion to 6 after 48 h at 120ꢀC.
Stille reactions, employing the modi®cations reported
by Saa et al. for hindered electron-rich aryl tri¯ates,20
were performed for the preparation of the 5-methyl (7),
5-(2-furyl) (8) and 5-phenyl (9) analogues. Thus, reac-
tion of 4 with the appropriate organostannane reagent
(tetramethylstannane, tributyl(2-furyl)stannane and tri-
butylphenylstannane, respectively) in the presence of
excess lithium chloride, triphenylphosphine, and cataly-
tic amounts of bistriphenylphosphine-palladium(II)
chloride and 2,6-di-tert-butyl-4-methylphenol (radical
scavenger), employing DMF as a solvent and elevated
reaction temperatures, gave the desired products 7, 8
and 9 in acceptable yields.
Chemistry
The synthetic steps employed to access the di€erent 5-
substituted analogues of 1 are outlined in Schemes 1 and
2. The 5-tri¯ate analogue 4, which served as the starting
point for the preparation of the other derivatives, was
prepared from the previously reported hydroxy analogue
212 by reaction with N,N-bis(tri¯uoromethane-sulfonyl)
aniline (N-phenyltri¯imide) in dichloromethane, employ-
ing triethylamine as a base.16
A Heck reaction between 4 and vinyl butyl ether in the
presence of 1,3-bis(diphenylphosphino)propane (dppp)
and a catalytic amount of palladium(II) acetate, employ-
ing triethylamine as a base and DMF as a solvent, fol-
lowed by hydrolysis with 5% aqueous hydrochloric acid
solution, was used to prepare the 5-acetyl analogue 5.17
Since the carboxamide moiety has been shown to be an
interesting bioisostere for aromatic hydroxy and methoxy
substituents in structurally related serotonergic ligands,21
we wanted to prepare the 5-carboxamide analogue of 1.
However, di€erent attempts to convert the 5-cyano ana-
logue (6) directly to the 5-carboxamido analogue (11),
employing di€erent mild hydrolytic conditions, failed in
our hands. Thus, reaction with manganese dioxide in
CH2Cl2,22 mercury(II) acetate in glacial acetic acid,23
aqueous sodium perborate without24 or with25 methanol,
or hydrogen peroxide26 gave only starting material.
Application of stronger reaction conditions, such as
re¯uxing in aqueous sulfuric acid,18 resulted in the hydro-
lysis of both the nitrile group and the benzamide moiety.
As an alternative route for the preparation of 11, the
methyl 5-carboxylate analogue 10 was prepared from 4 by
a palladium-catalyzed crosscoupling with carbon mon-
oxide in the presence of excess methanol27 (Scheme 1).
The resulting methyl ester could be converted eciently to
the carboxamide 11 by reaction with formamide and 30%
methanolic sodium methoxide28 (Scheme 2).
The 5-cyano analogue 6 was prepared according to a
procedure reported by Hedberg et al.,18 which is a
Pharmacology
Scheme 1. Reagents and conditions: (a) N-phenyltri¯imide, Et3N,
CH2Cl2, rt; (b) butyl vinyl ether, Pd(OAc)2, dppp, Et3N, DMF, Á; 5%
HCl, Á; (c) KCN, (PPh3)2NiCl2, PPh3, Zn, DMF, Á; (d) tetra-
methylstannane, (PPh3)2PdCl2, PPh3, LiCl, 2,6-di-tert-butyl-4-methyl-
phenol, DMF, Á; (e) tributyl(2-furyl)stannane, (PPh3)2PdCl2, PPh3,
LiCl, 2,6-di-tert-butyl-4-methylphenol, DMF, Á; (f) tributylphenyl-
stannane, Pd(PPh3)4, LiCl, 2,6-di-tert-butyl-4-methylphenol, DMF, Á;
(g) CO(g), MeOH, Pd(OAc)2, dppp, Et3N, DMSO, Á.
The newly prepared C5-substituted analogues 4±11, as
well as the previously described 5-methoxy, 5-hydroxy,
and 5-hydrido analogues 1, 2 and 3 were evaluated for
their ability to compete for [3H]-spiperone binding to
rat striatal membranes containing dopamine D2 recep-
tors, and their ability to compete for [3H]-8-OH-DPAT
binding to rat frontal cortex membranes containing
serotonin 5-HT1A receptors in vitro. The results of these
binding studies are shown in Table 1. In order to char-
acterise the binding assay, the anities of spiperone and
8-OH-DPAT have been included as well.
Results and Discussion
In three previous reports we have elaborated on the
structure±anity relationships (SAFIRs) of a series of
Scheme 2. Reagents and conditions: (a) formamide, 30% NaOCH3,
DMF, Á.

E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
2543
Table 1. Receptor binding data of compounds 1±11, spiperone,
and 8-OH-DPAT
dopamine D2 receptor, the ranking is 2>1>3, while at
the serotonin 5-HT1A receptor, the ranking is 3>1>2.
Compound 3 has the highest anity for the serotonin 5-
HT1A receptor, and due to its low anity for the dopa-
mine D2 receptor, it is also the most selective serotonin 5-
HT1A receptor ligand of this series under these assay con-
ditions. The dopamine D2 receptor anity ranking, but
also the relative anities of 1±3 are consistent with what
may be expected for 5-substituted DPAT analogues: the
hydroxy-substituted congeners usually have the highest
anities, followed by the methoxy analogues, while
unsubstituted analogues have only weak to moderate
anities.31±35 Presumably, the hydroxy group at the C5-
position functions both as hydrogen bond donor and
acceptor, while a methoxy group can only act as a
hydrogen bond acceptor while interacting with the
dopamine D2 receptor.34 Unsubstituted DPATs are
incapable of forming hydrogen bonds, which would
explain their lower anity for the dopamine D2 receptor.
Ki (nM)a
5-HT1A
Compound
R
D2
D2/5-HT1A
1
2
3
4
5
6
OCH3
OH
H
OSO2CF3
COCH3
CN
6.3
4.9
160
16
120
25
2.4
6.5
1.0
6.3
9.3
36
2.6
0.8
163
2.5
13
0.7
7
8
9
10
CH3
35
30
1.3
12
8.3
8.3
60
27
2.5
14
13
9.2
0.002
>400
2-Furyl
Phenyl
COOCH3
CONH2
Ð
120
110
550
0.15
>1000
11
Spiperone
8-OH-DPAT
98
2.5
As no relationships between the nature of the C5-sub-
stituents of 1±11 and the anities of the compounds for
the two receptor subtypes were obvious at ®rst sight, we
applied partial least-squares projections to latent struc-
tures (PLS)36 in an attempt to derive quantitative struc-
ture±activity relationships (QSARs). Thus, 24 di€erent
physicochemical descriptors, describing either whole
molecule or C5-substituent properties, were generated
for each compound. PLS analysis of the descriptor
matrix (11Â24; 11 compounds and 24 descriptors) and
the corresponding receptor anities revealed no statis-
tically signi®cant correlation between the molecular
properties of 1±11 and their dopamine D2 receptor a-
nities, as re¯ected by a negative values of Q2. In con-
trast, for the serotonin 5-HT1A receptor anities a
statistically signi®cant model was obtained, with the
®rst two latent variables of the PLS model accounting
for 80% of the variance in the anities. Figure 1 shows
the correlation between the ®tted and the observed (i.e.
experimental) pKi values of 1±11 at the serotonin 5-
HT1A receptor. The predictability of the model was
evaluated by leave-one-out crossvalidation,37 resulting
in a Q2 value of 0.49 after two components (Fig. 1).
Inspection of the regression coecients revealed that a
relatively lipophilic C5-substituent is bene®cial for a
high serotonin 5-HT1A receptor anity, while the a-
nity decreases with a larger dipole moment of the C5-
substituent. Size-related properties appeared to have
little e€ect on the serotonin 5-HT1A receptor anity.
Taken together, these ®ndings clearly demonstrate that
the serotonin 5-HT1A receptor tolerates considerable
structural diversity at the 5-position of 1 and that 1±11
may share a common mode of binding at this receptor
subtype, as opposed to the dopamine D2 receptor. The
results from a previously reported molecular modeling
study on the binding modes of the enantiomers of 1 to
the dopamine D2 and serotonin 5-HT1A receptor15 may
provide a plausible explanation for this phenomenon. It
was shown that both enantiomers of 1 could form spe-
ci®c interactions between their protonated tertiary
amine moieties, their amide carbonyl oxygen atom, their
amide hydrogen atom, and the phenyl ring of their
benzamide moieties, and certain amino acid residues in
Ð
a
Mean values of 1±3 determinations.
2-aminotetralin-derived benzamides and structurally
closely related analogues with mixed dopamine D2, D3
and serotonin 5-HT1A receptor binding properties.12
14
The SAFIRs suggested that the 2-aminotetralin moieties
of the compounds may share the same binding sites in
these receptor subtypes as the class of N,N-di-n-propyl-
aminotetralins (DPATs). In the present study the
e€ects of introduction of di€erent substituents at the
C5-position of the lead compound of the series, 5-OMe-
BPAT (1) on the anities for the dopamine D2 and
serotonin 5-HT1A receptor have been investigated.
Chemical diversity at this position was introduced by
chemical transformation of the 5-tri¯ate analogue 4,
using several Stille-type crosscouplings and a Heck
reaction. All reactions proceeded in moderate to good
yields, ranging from 52 to 82%. Aryl tri¯ates have been
shown to be eciently hydrogenated to their unsub-
stituted analogues in the presence of a palladium cata-
lyst, when tributylamine is employed as the hydride
source.29 This observation may explain why hydro-
genolysis is sometimes observed as a side-reaction of
palladium-catalysed crosscouplings of aryl tri¯ates, in
cases where triethylamine is added as a base.30 How-
ever, there was no indication of the formation of C5-
unsubstituted by-product during the preparation of 10.
Compounds 1, 2 and the 5-unsubstituted analogue 3
have been characterised previously under di€erent assay
conditions.12 The anities found in the present investi-
gation are somewhat lower than those previously
reported. Particularly, 3 shows a low anity for the
dopamine D2 receptor (Ki=160 nM), whereas at cloned
human dopamine D2 receptors, using [3H]-raclopride as
a radioligand, it had an anity of 10 nM. These di€er-
ences are likely to be the result of di€erences in the
assay conditions ([3H]-spiperone versus [3H]-raclopride
as a radioligand, tissue versus cloned receptors). Never-
theless, the anity ranking of 1±3 at both receptor sub-
types was the same under both assay conditions: at the

2544
E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
Figure 1. Plots showing (left) the observed versus ®tted and the (right) observed versus predicted pKi values of compounds 1±11 at the serotonin 5-
HT1A receptor after two PLS components.
the 7-transmembrane domains of the dopamine D2
receptor model. This multipoint interaction mode of the
benzamidoethyl side-chain, which is supported by
SAFIRs derived from relevant structural modi®ca-
tions,13 imposes certain restrictions on the way the 2-
aminotetralin moiety of 1 can occupy the agonist bind-
ing site, and hence is likely to decrease the tolerance of
the dopamine D2 receptor to structural variation on the
phenyl ring of the 2-aminotetralin moiety of 1. In con-
trast, the amide moieties of the enantiomers of 1 were
not involved in speci®c interactions in the serotonin 5-
HT1A receptor model, the only interaction points being
the protonated tertiary amines and the benzamide
phenyl rings. Presumably this allows for more con-
formational adaptation of the ligands, required for the
agonist binding site to be able to accommodate amino-
tetralin moieties with structurally diverse substituents.
used without further puri®cation. All basic amine pro-
ducts were converted to their corresponding hydrochlo-
ride or oxalate salts by adding an equimolar amount of
a 1 M ethereal HCl solution or an ethanolic solution of
oxalic acid to a solution of the free base in ether. All
chemical data, except for TLC analyses, were obtained
on the salt forms, unless otherwise stated. TLC analyses
were carried out on aluminium plates (Merck) pre-
coated with silica gel 60 F254 (0.2 mm), and spots were
visualised with UV light and I2. Gravity column chro-
matography was performed using silica gel (Merck 60).
Melting points were determined in open glass capillaries
on an electrothermal digital melting-point apparatus
and are uncorrected. IR spectra (KBr pellets) were
recorded on an ATI-Mattson Genesis Series FT-IR
spectrophotometer, and only the important absorptions
are indicated. Broad peaks (b) have been indicated as
such. 1H NMR spectra were recorded at 200 MHz
on a Varian Gemini-200 spectrometer or at 300 MHz
on a Varian VXR-300 spectrometer. 1H NMR che-
mical shifts are denoted in d units (ppm) relative to
CDCl3 (7.26) and converted to the TMS scale. The
following abbreviations are used to indicate spin
multiplicities: s (singlet), bs (broad singlet), d (doub-
let), dd (doublet of doublets), t (triplet), m (multi-
plet). 13C NMR spectra were recorded at 50 MHz on
a Varian Gemini-200 spectrometer or at 75 MHz on
a Varian VXR-300 spectrometer. 13C NMR chemical
shifts are denoted in d units (ppm) relative to CDCl3
(76.91) and converted to the TMS scale. All chemical
ionisation mass spectra were recorded on a NER-
MAG R 3010 triple quadrupole mass spectrometer
equipped with a home-built atmospheric pressure
ionisation source and ionspray interface. Elemental
analyses (C, H and N) for target compounds were
performed at the Micro Analytical Department, Uni-
versity of Groningen.
Conclusions
The 5-methoxy substituent of 1 plays a role in the
binding of the compound to the dopamine D2 receptor,
since replacement of this substituent by other moieties
a€ects the anity for this receptor subtype. However,
no clear structure±anity relationship could be derived
from the limited set of C5-substituted analogues. The 5-
methoxy, 5-hydroxy and 5-unsubstituted analogues
probably share their binding sites and binding modes
with the 5-oxygenated DPATs, but the other analogues
may bind to the dopamine D2 receptor in a completely
di€erent way. The serotonin 5-HT1A receptor tolerates
more structural diversity at the C5-position of 1. For
high anity at this receptor subtype, the C5-substituent
should be relatively lipophilic and nonpolar.
Experimental
Chemistry
5-{[(Tri¯uoromethyl)sulfonyl]oxy}-2-[N-(2-benzamido-
ethyl)-N-n-propylamino]tetralin oxalate (4). Triethyl-
amine (2.87 g, 28.4 mmol) was added to a stirred solution
of N-phenyltri¯imide (6.09 g, 17.0 mmol) and 213 (5.00 g,
14.2 mmol) in CH2Cl2 (100 mL). The reaction mixture
General remarks. Unless otherwise indicated, all mate-
rials were purchased from commercial suppliers and

E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
2545
was stirred overnight at room temperature under a nitro-
gen atmosphere. H2O (50 mL) was added to the reaction
mixture, the phases were separated and the organic layer
was subsequently washed with 10% aq Na2CO3 solution
(3Â50 mL), H2O (50 mL) and brine (50 mL). After drying
(Na2SO4) and ®ltration of the organic layer, the solvent
was evaporated, which gave the crude tri¯ate as a brown
oil. Chromatographic puri®cation [eluent: EtOAc:pe-
troleum ether (bp 40-±60), 1:2 (v/v)] gave 5.68 g (11.7
mmol, 83%) of the pure base of 4 as a light yellow oil: mp
147±149ꢀC (acetone); IR: cm 1 3377, 2973, 2880, 2651 (b),
atmosphere at room temperature for 15 min. Then the
reaction vessel was sealed and stirring was continued at
120ꢀC for 48 h. After cooling to room temperature, the
reaction mixture was poured into H2O (25 mL) and the
mixture was extracted with Et2O (3Â25 mL). The
organic layers were collected and washed with brine (25
mL). After drying (Na2SO4), the organic solution was
concentrated in vacuo which gave the crude nitrile as an
orange oil. Puri®cation by column chromatography (elu-
ent: EtOAc) yielded 0.260 g (0.72 mmol, 70%) of the pure
base of 6 as a colourless oil: mp 117±120ꢀC; IR: cm 1 3421
(b), 3246 (b), 3058, 2937, 2880, 2475 (b), 2224, 1654, 1578,
1534; 1H NMR (base, 200 MHz, CDCl3): d 0.86 (t, J=7.3
Hz, 3H), 1.39±1.70 (m, 3H), 1.96±2.05 (m, 1H), 2.45±2.52
(m, 2H), 2.68±3.15 (m, 7H), 3.33±3.47 (m, 2H), 7.05±7.43
(m, 7H), 7.75 (d, J=7.1 Hz, 2H); 13C NMR (base, 50
MHz, CDCl3): d 11.6, 21.8, 24.8, 28.2, 31.8, 38.0, 48.5,
52.1, 55.3, 111.9, 117.8, 126.0, 126.7, 128.4, 130.2, 131.2,
1
2504 (b), 1718, 1660, 1579, 1525, 1417, 1210; H NMR
(base, 300 MHz, CDCl3): d 0.85 (t, J=7.3 Hz, 3H), 1.40±
1.55 (m, 3H), 1.97±2.01 (m, 1H), 2.45±2.50 (dd, J=7.0 Hz,
7.7 Hz, 2H), 2.60±3.02 (m, 7H), 3.37±3.43 (m, 2H), 6.97±
7.10 (m, 3H), 7.25±7.41 (m, 4H), 7.77 (dd, J=8.3 Hz, 1.3
Hz, 2H); 13C NMR (base, 75 MHz, CDCl3): d 11.7, 22.0,
24.1, 24.8, 32.1, 38.5, 48.8, 52.4, 55.7, 118.3, 118.6 (q,
J=320 Hz), 126.9, 128.4, 129.4, 131.2, 134.6, 140.0, 148.1,
167.3; MS (CI with AcOH): m/z 486 (M+1). Anal. calcd
133.9, 134.5, 137.8, 139.6, 167.2; MS (CI with AcOH): m/z
1
.
.
362 (M+1). Anal. calcd for C23H27N3O HCl 4H2O: C
68.63, H 7.15, N 10.44; obsd C 68.55, H 7.14, N 9.84.
.
for C25H27N2O4SF3 C2H2O4: C 52.25, H 5.10, N 4.88;
obsd C 51.87, H 5.04, N 5.36.
5-Methyl-2-[N-(2-benzamidoethyl)-N-n-propylamino]-
tetralin hydrochloride (7). A solution of the free base of 4
(250 mg, 0.5 mmol) in dry DMF (3 mL) was added to a
mixture of (PPh3)2PdCl2 (42 mg, 0.06 mmol), LiCl (181
mg, 4.27 mmol) and PPh3 (82 mg, 0.31 mmol) in dry DMF
(7 mL). The reaction mixture was stirred at room tem-
perature for 10 min, and then (CH3)4Sn (0.29 mL, 2.09
mmol) and a few crystals of 2,6-di-tert-butyl-4-methyl-
phenol were added. The reaction vessel was sealed and the
reaction mixture was stirred overnight at 120ꢀC. After
cooling, the reaction mixture was partitioned between
CHCl3 and 10% aqueous NaHCO3 solution. The com-
bined organic layers were subsequently washed with 10%
aqueous KF solution, H2O and brine. After drying
(Na2SO4) and ®ltering the organic layer, the solvent was
evaporated, which gave the crude product as a brown
oil. Puri®cation by column chromatography [eluent:
MeOH:CH2Cl2, 1:20 (v/v)] yielded 150 mg (0.43 mmol,
82%) of the pure base of 7 as a colourless oil: mp 122±
5-Acetyl-2-[N-(2-benzamidoethyl)-N-n-propylamino]tetra-
lin hydrochloride (5). Pd(OAc)2 (6 mg, 0.03 mmol), dppp
(11 mg, 0.03 mmol), Et3N (0.20 g, 1.98 mmol), and
butyl vinyl ether (0.50 g, 5.0 mmol) were added to a
solution of the free base of 4 (0.24 g, 0.50 mmol) in dry
DMF (2 mL). The reaction mixture was stirred under a
nitrogen atmosphere at room temperature for 15 min.
Then the reaction vessel was sealed and the reaction
mixture was heated at 120ꢀC for 24 h. After cooling to
ꢁ90ꢀC, 5% aq HCl solution (3 mL) was added to the
reaction mixture and stirring was continued for 15 min
at ambient temperature. The reaction mixture was then
cooled to room temperature, poured into H2O (25 mL)
and solid NaHCO3 was added until basic. The mixture
was extracted with CH2Cl2 (3Â25 mL), the organic layers
were collected and subsequently washed with H2O (3Â25
mL) and brine (25 mL). After drying (Na2SO4) and ®lter-
ing, the organic layer was concentrated under reduced
pressure to give the crude product as a brown oil. Pur-
i®cation by column chromatography (eluent: EtOAc)
yielded 0.154 g (0.4 mmol, 82%) of the pure base of 5 as a
colourless oil: mp 81±83ꢀC; IR: cm 1 3256 (b), 2963, 2494
124ꢀC; IR: cm 3249 (b), 2965, 2467 (b), 1655, 1601,
1
1578, 1533; 1H NMR (base, 200 MHz, CDCl3): d 0.92 (t,
J=7.3 Hz, 3H), 1.43±1.79 (m, 3H), 2.04±2.10 (m, 1H),
2.21 (s, 3H), 2.57 (dd, J=7.3 Hz, 7.3 Hz, 2H), 2.65±3.08
(m, 7H), 3.46±3.56 (m, 2H), 6.92±7.09 (m, 4H), 7.41±7.56
(m, 3H), 7.80 (dd, J=7.7 Hz, 1.6 Hz, 2H); 13C NMR
(base, 50 MHz, CDCl3): d 11.6, 19.4, 21.6, 25.8, 27.0, 32.3,
37.5, 48.2, 51.9, 55.5, 125.5, 126.7, 127.2, 128.4, 131.2,
1
(b), 1676, 1654, 1601, 1578, 1533; H NMR (base, 200
MHz, CDCl3): d 0.88 (t, J=7.3 Hz, 3H), 1.46±1.64 (m,
3H), 1.95±2.10 (m, 1H), 2.51 (s, 3H), 2.58±2.66 (m, 2H),
2.81±3.13 (m, 7H), 3.51±3.55 (m, 2H), 7.15 (d, J=5.4 Hz,
2H), 7.35±7.51 (m, 5H), 7.80 (dd, J=7.7 Hz, 1.8 Hz, 2H);
13C NMR (base, 50 MHz, CDCl3): d 11.5, 20.8, 25.1, 27.7,
29.7, 32.1, 37.4, 48.8, 52.4, 56.0, 125.4, 126.8, 127.0, 128.4,
131.3, 132.8, 134.2, 135.7, 137.0, 137.7, 167.2; MS (CI with
134.5, 134.6, 135.8, 136.2, 167.1; MS (CI with AcOH): m/z
3
.
.
351 (M+1). Anal. calcd for C23H30N2O HCl 4H2O: C
68.97, H 8.20, N 7.00; obsd C 69.16, H 8.03, N 7.07.
AcOH): m/z 379 (M+1). Anal. calcd for C24H30N2
3
5-(2-Furyl)-2-[N-(2-benzamidoethyl)-N-n-propylamino]-
tetralin hydrochloride (8). A mixture of (PPh3)2PdCl2 (43
mg, 0.1 mmol), LiCl (175 mg, 4.1 mmol), PPh3 (80 mg,
0.3 mmol) and the free base of 4 (220 mg, 0.45 mmol) in
dry DMF (4 mL) was stirred at room temperature
under a nitrogen atmosphere. After 5 min a solution of
tributyl(2-furyl)stannane (0.325 mL, 1.03 mmol) in dry
DMF (6 mL) was added, followed by a few crystals of
2,6-di-tert-butyl-4-methylphenol. After stirring for 45
min at 120ꢀC under a nitrogen atmosphere, the reaction
.
.
O2 HCl 4H2O: C 67.26, H 7.66, N 6.54; obsd C 67.03, H
7.32, N 6.44.
5-Cyano-2-[N-(2-benzamidoethyl)-N-n-propylamino]-
tetralin hydrochloride (6). Zn (dust, 103 mg, 1.6 mmol),
(PPh3)2NiCl2 (172 mg, 0.3 mmol), PPh3 (140 mg, 0.5
mmol) and KCN (83 mg, 1.27 mmol) were added to a
solution of the free base of 4 (0.50 g, 1.03 mmol) in dry
DMF (5 mL). The mixture was stirred under a nitrogen

2546
E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
mixture was cooled and partitioned between CH2Cl2 and
10% aqueous NaHCO3 solution. The organic solution
was subsequently washed with H2O and brine, dried
(Na2SO4) and concentrated under reduced pressure. The
resulting crude product was puri®ed by column chroma-
tography (eluent: EtOAc), which yielded 140 mg (0.35
mmol, 77%) of the pure base of 8 as a colourless oil: mp
127±130ꢀC; IR: cm±1 3416 (b), 3246 (b), 3057, 2939, 2880,
reaction mixture was cooled to room temperature,
resaturated with CO and extra MeOH (1.0 mL) was
added in order to complete the reaction. After 72 h of
heating at 70ꢀC TLC analysis revealed absence of start-
ing material. The reaction mixture was cooled to room
temperature and poured into H2O (50 mL). The aqueous
phase was extracted with CH2Cl2 (5Â15 mL), and the
collected organic layers were washed with H2O (5Â15 mL)
and brine (15 mL). After drying (Na2SO4) and ®ltering
of the organic layer, the solvent was evaporated, yielding
the crude ester as a brown oil. Puri®cation by column
chromatography [eluent: MeOH:CH2Cl2, 1:20 (v/v)]
yielded 0.50 g (1.3 mmol, 61%) of the pure base of 10 as a
1
2611, 2475, 1654, 1578, 1534; H NMR (base, 200 MHz,
CDCl3): d 0.92 (t, J=7.3 Hz, 3H), 1.46±1.67 (m, 3H),
2.03±2.09 (m, 1H), 2.53±2.60 (m, 2H), 2.75±3.11 (m, 7H),
3.46±3.56 (m, 2H), 6.48 (s, 2H), 7.03±7.06 (m, 2H), 7.18 (t,
J=7.7 Hz, 1H), 7.39±7.54 (m, 5H), 7.80 (dd, J=7.7 Hz,
1.8 Hz, 2H); 13C NMR (base, 50 MHz, CDCl3): d 11.6,
21.7, 25.9, 28.6, 32.6, 37.6, 48.2, 51.9, 55.3, 108.8, 111.0,
125.4, 125.6, 126.7, 128.4, 129.1, 130.1, 131.2, 133.3, 134.6,
colourless oil: mp 90±91ꢀC; IR: cm 3400 (b), 3247 (b),
1
3057, 2965, 2880, 2626 (b), 2492 (b), 1718, 1654, 1578,
1534; 1H NMR (base, 200 MHz, CDCl3): d 0.89 (t, J=7.3
Hz, 3H), 1.43±1.62 (m, 3H), 1.99±2.05 (m, 1H), 2.53 (dd,
J=7.3 Hz, 7.3 Hz, 2H), 2.72±3.10 (m, 6H), 3.28±3.49 (m,
3H), 3.84 (s, 3H), 7.04 (bs, 1H), 7.12±7.21 (m, 2H), 7.36±
7.47 (m, 3H), 7.66±7.79 (m, 3H); 13C NMR (base, 50
MHz, CDCl3): d 11.6, 21.7, 25.6, 28.0, 32.6, 37.7, 48.3,
51.6, 51.9, 55.2, 125.2, 126.7, 128.1, 128.4, 129.6, 131.1,
133.3, 134.6, 137.4, 137.7, 161.1, 168.2; MS (CI with
136.8, 141.6, 153.3, 167.1; MS (CI with AcOH): m/z 403
1
.
.
(M+1). Anal. calcd for C26H30N2O2 HCl 2H2O: C 69.69,
H 7.21, N 6.25; obsd C 69.64, H 7.18, N 6.25.
5-Phenyl-2-[N-(2-benzamidoethyl)-N-n-propylamino]-
tetralin hydrochloride (9). A solution of tributylphenyl-
stannane (228 mg, 0.62 mmol) in 1,4-dioxane (9 mL) was
added to a mixture of the free base of 4 (250 mg, 0.52
mmol), LiCl (68 mg, 1.60 mmol) and Pd(PPh3)4 (30 mg,
0.03 mmol) in dry DMF (3 mL). Then a few crystals of 2,6-
di-tert-butyl-4-methylphenol were added and the reaction
mixture was stirred overnight at 120ꢀC under a nitrogen
atmosphere. After cooling, the solids were removed from
the reaction mixture by ®ltration (Celite1) and the ®ltrate
was poured into H2O (25 mL). The aqueous solution was
extracted with CHCl3 (3Â25 mL) and the extracts were
combined. The organic solution was subsequently washed
with aqueous NaHCO3 solution (3Â25 mL), H2O (25 mL)
and brine. After drying (Na2SO4) and ®ltering, the organic
solution was concentrated under reduced pressure, which
gave the crude product as a brown oil. Puri®cation by silica
column chromatography [eluent: MeOH:CH2Cl2, 1:20
(v/v)] yielded 110 mg (0.27 mmol, 52%) of the pure base of
AcOH): m/z (relative intensity) 337 (8), 395 (100, M+1).
1
.
.
Anal. calcd for C24H30N2O3 HCl 2H2O: C 65.51, H 7.35,
N 6.37; obsd C 65.35, H 7.32, N 6.48.
5-Carboxamido-2-[N-(2-benzamidoethyl)-N-n-propylami-
no]tetralin hydrochloride (11). Formamide (69 mg, 1.6
mmol) was added to a solution of the free base of 10 (150
mg, 0.4 mmol) in dry DMF (20 mL). The reaction mixture
was heated at 100ꢀC under a nitrogen atmosphere and then
30% NaOCH3 solution in MeOH (0.1 mL) was added
dropwise via a syringe. Heating was continued at 100ꢀC for
1 h and then the reaction mixture was allowed to cool to
room temperature. After concentrating the reaction mix-
ture under reduced pressure, the residue was dissolved in
CH2Cl2 (25 mL), and the resulting solution was subse-
quently washed with 10% aqueous NaHCO3 solution
(3Â25 mL), H2O (25 mL) and brine (25 mL). After drying
(Na2SO4) and ®ltering, the solvent was evaporated, which
gave the crude carboxamide as a yellow oil. Puri®cation by
silica column chromatography [eluent: MeOH:CH2Cl2,
1:10 (v/v)] yielded 110 mg (0.3 mmol, 76%) of the pure base
of 11 as a colourless oil: mp 190±192ꢀC; IR: cm 1 3297 (b),
9 as a colourless oil: mp 103±105ꢀC; IR: cm 3256 (b),
1
3057, 2965, 2479 (b), 1654, 1601, 1578, 1533; 1H NMR (300
MHz, CDCl3): d 1.00 (as, 3H), 1.77±1.80 (m, 1H), 1.99±
2.06 (m, 2H), 2.32±2.47 (m, 1H), 2.69±2.79 (m, 2H), 3.05±
3.22 (m, 4H), 3.31±3.44 (m, 1H), 3.51±3.60 (m, 1H), 3.64±
3.74 (m, 1H), 3.79±3.87 (m, 1H), 4.02±4.11 (m, 1H), 6.93±
7.54 (m, 12H), 8.06±8.10 (m, 2H); 13C NMR (75 MHz,
CDCl3): d 11.6, 21.3, 25.6, 28.2, 32.0, 37.4, 48.6, 52.2, 56.2,
125.6, 126.8, 127.4, 127.9, 128.5, 128.6, 129.0, 131.2, 133.6,
1
3169 (b), 2965, 2611 (b), 2520 (b), 1655, 1578, 1533; H
NMR (base, 200 MHz, CHCl3): d 0.88 (t, J=7.3 Hz, 3H),
1.39±1.67 (m, 3H), 1.97±2.02 (m, 1H), 2.52 (dd, J=7.4 Hz,
7.4 Hz, 2H), 2.71±3.16 (m, 7H), 3.42±3.47 (m, 2H), 6.05 (bs,
1H), 6.26 (bs, 1H), 7.07±7.12 (m, 3H), 7.19±7.24 (m, 1H),
7.36±7.52 (m, 3H), 7.75 (dd, J=7.9 Hz, 1.6 Hz, 2H); 13C
NMR (base, 50 MHz, CDCl3): d 11.6, 21.6, 25.3, 27.0, 32.3,
37.6, 48.3, 52.0, 55.4, 124.4, 125.5, 126.7, 128.4, 131.2,
131.4, 134.2, 134.4, 135.1, 137.3, 167.2, 172.4; MS (CI with
134.4, 135.9, 141.6, 141.8, 167.1; MS (CI with AcOH): m/z
1
.
.
413 (M+1). Anal. calcd for C28H32N2O HCl 2H2O: C
73.41, H 7.50, N 6.12; obsd C 73.14, H 7.70, N 6.28.
Methyl 2-[N-(2-benzamidoethyl)-N-n-propylamino]tetra-
lin - 5 - carboxylate hydrochloride (10). Methanol (3.32
mL, 131 mmol) was added to a mixture of the free base
of 4 (1.00 g, 2.1 mmol), Pd(OAc)2 (14 mg, 0.1 mmol),
dppp (26 mg, 0.1 mmol), Et3N (0.42 g, 4.2 mmol) and
DMSO (15 mL). The mixture was ¯ushed with nitrogen
until all reagents were dissolved (1 h) and then the reac-
tion mixture was saturated with CO gas (caution: highly
toxic!) by ¯ushing it at room temperature during 1 h.
Then the reaction mixture was heated at 70ꢀC under an
atmosphere of CO. After 24 and 48 h of heating, the
AcOH): m/z (rel. intensity) 288 (3), 380 (100, M+1). Anal.
1
.
.
calcd for C23H29N3O2 HCl 4H2O: C 65.69, H 7.33, N 9.99;
obsd C 65.37, H 7.33, N 9.93.
Pharmacology
[3H]-Spiperone binding to dopamine D2 receptors. Male
Wistar rats, weighing 200±300 g (TNO, The Netherlands),
were decapitated and the striata were dissected. After

E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
2547
weighing, the striata were taken up in 20 volume
amounts of 50 mM Tris±HCl bu€er (pH 7.7) and
homogenised using a Potter-Elvehjem homogenizer
(10Â600 rpm). The resulting homogenate was cen-
trifuged at 4ꢀC during 10 min at 50,000Âg in a Sorvall
RC-5 centrifuge. The supernatant was decanted and the
pellet was resuspended in 20 volume amounts of 50 mM
Tris±HCl bu€er, using a Vortex mixer. The suspension
was again homogenised (5Â600 rpm) and centrifuged at
4ꢀC during 10 min at 50,000Âg. After having repeated
this washing procedure two more times, the tissue was
taken up in 20 volume amounts of incubation medium
(47 mM Tris±HCl bu€er, 120 mM NaCl, 5 mM KCl, 2
mM CaCl2, 1 mM MgCl2, 0.1% ascorbic acid, 5 mM
pargyline, pH 7.7) and homogenised (5Â600 rpm). The
membrane fraction was then taken up in 100 volume
amounts of incubation medium, to a ®nal concentration
of 10 mg original wet weight per mL. The homogenate
was incubated batchwise at 37ꢀC during 10 min (acti-
vation of pargyline), distributed in 96-well plates (0.5
mL per well) by a Tecan dilution robot, and stored at
4ꢀC until further use. Half an hour prior to the incuba-
tion the tissue was allowed to warm to room tempera-
ture. About 15 min prior to the incubation 50 mL of test
compound, dissolved in 0.1% ascorbic acid solution, or
0.1% ascorbic acid solution (determination of total
binding) was added to the tissue. The incubation was
then initiated by adding 50 mL of [3H]-spiperone (New
England Nuclear, Boston, MA; speci®c activity 20±40
Ci/mmol, Kd=0.12 nM), dissolved in 0.1% ascorbic
acid solution, resulting in a concentration of 0.5 nM
during incubation. The incubation mixture was incu-
bated at 25ꢀC on a water bath during 20 min. Binding in
the presence of haloperidol (0.1 mM) was de®ned as
non-speci®c. The incubation was terminated by ®ltra-
tion of the plates through Whatman GF/B glass®ber
®lters and subsequent washing with 50 mM Tris±HCl
bu€er (2Â5 mL) of 4ꢀC, using Tomtek cell harvesters.
The ®lters, each containing 96 samples, were dried and
transferred in polyurethane envelopes, together with
solid scintillation sheets. The envelops were heat-sealed
and counted in a Wallac betaplate counter.
and homogenised (5Â600 rpm). The membrane fraction
was then taken up in 50 volume amounts of incubation
medium, to a ®nal concentration of 20 mg original wet
weight per mL. The homogenate was incubated batch-
wise at 37ꢀC during 15 min (activation of pargyline), dis-
tributed in 96-well plates (0.5 mL per well) by a Tecan
dilution robot, and stored at 4ꢀC until further use. Half an
hour prior to the incubation the tissue was allowed to
warm to room temperature. About 15 min prior to the
incubation 50 mL of test compound, dissolved in 0.1%
ascorbic acid solution, or 0.1% ascorbic acid solution
(determination of total binding) was added to the tissue.
The incubation was then initiated by adding 50 mL of [3H]-
8-OH-DPAT (New England Nuclear, Boston, MA; spe-
ci®c activity 158 Ci/mmol, Kd=2.0 nM), dissolved in
0.1% ascorbic acid solution, resulting in a concentration
of 1.0 nM during incubation. The incubation mixture (0.6
mL) was incubated at 37ꢀC on a water bath during 10
min. Binding in the presence of serotonin (10 mM) was
de®ned as non-speci®c. The incubations were terminated
and the radioactivity was determined as described for the
dopamine D2 receptor binding assay.
Data analysis. Concentrations of unlabeled test com-
pound causing 50% displacement of the speci®c binding
of a labelled compound (IC50 values) were obtained by
computerised log-probit linear regression analysis of
data obtained in experiments in which four to six dif-
ferent concentrations of the test compound were used.
Inhibition constants (Ki) were calculated using the
Cheng±Pruso€ equation.38 Mean Ki values were calcu-
lated from at least three values obtained from indepen-
dent experiments, that is, in experiments performed on
di€erent days with di€erent membrane preparations.
All incubations were done in triplicate.
Quantitative structure±activity relationships. All calcula-
tions were performed on an Octane R10,000 Silicon
Graphics workstation running IRIX 6.4. The program
Tsar39 was used to generate 24 descriptors for com-
pounds 1±-11. The following whole molecule descriptors
were generated: molecular mass; molecular surface area;
molecular volume; total dipole moment; number of
heteroatoms; number of H-bond donors; number of H-
bond acceptors; number of N atoms; number of O
atoms; number of S atoms. In addition, the following
descriptors were generated for the C5-substituents only:
molecular mass; molecular surface area; molecular
volume; Verloop L; Verloop B1; Verloop B2; Verloop
B3; Verloop B4; Verloop B5; total dipole moment; log
P; molecular refractivity; number of H-bond donors;
number of H-bond acceptors. A matrix was formed by
putting the descriptors in the columns and the com-
pounds in the rows. The ®rst two columns were formed
by the dopamine D2 and serotonin 5-HT1A pKi values
of the compounds. After mean-centring and autoscaling
of the matrix, PLS analysis was performed using the
PLS_Toolbox40 for MatLab.41
[3H]-8-OH-DPAT binding to serotonin 5-HT1A receptors.
Male Wistar rats, weighing 200±-300 g (TNO, The
Netherlands), were decapitated and the frontal cortices
were dissected. After weighing, the frontal cortices were
taken up in 10 volume amounts of 0.32 M sucrose
solution and homogenised using a Potter-Elvehjem
homogeniser (10Â600 rpm). The resulting homogenate
was centrifuged at 4ꢀC during 10 min at 700Âg in a
Sorvall RC-5 centrifuge. The supernatant was decanted
and again centrifuged at 4ꢀC during 10 min at
50,000Âg. The pellet was resuspended in 10 volume
amounts of 50 mM Tris±HCl bu€er (pH 7.7), using a
Vortex mixer and homogenised during 10 s using a
Polytron (level 100). The resulting homogenate was
incubated during 10 min at 37ꢀC in order to remove
endogenous serotonin. Again the homogenate was cen-
trifuged at 4ꢀC during 10 min at 50,000Âg. The result-
ing pellet was resuspended in 10 volume amounts of
incubation medium (50 mM Tris±HCl bu€er, 4 mM
CaCl2, 0.1% ascorbic acid, 10 mM pargyline, pH 7.7)
References
1. Wadenberg, M. L.; Ahlenius, S. J. Neural Transm. Gen.
Sect. 1991, 83, 43.

2548
E. J. Homan et al. / Bioorg. Med. Chem. 7 (1999) 2541±2548
2. Rollema, H.; Lu, Y.; Schmidt, A. W.; Zorn, S. H. Eur. J.
Pharmacol. 1997, 338, 3.
3. Newman-Tancredi, A.; Gadauvan, S.; Conte, C.; Chaput,
C.; Touzard, M.; Verriele, L.; Audinot, V.; Millan, M. J. Eur.
J. Pharmacol. 1998, 355, 245.
4. Christo€ersen, C. L.; Meltzer, L. T. Neuropsychopharma-
cology 1998, 18, 399.
5. Hashimoto, T.; Nishino, N.; Nakai, H.; Tanaka, C. Life
Sci. 1991, 48, 355.
6. Joyce, J. N. Psychopharmacology Berl. Suppl. 1993, 112, 16.
7. Hashimoto, T.; Kitamura, N.; Kajimoto, Y.; Shirai, Y.;
Shirakawa, O.; Mita, T.; Nishino, N.; Tanaka, C. Psycho-
pharmacology Berl. Suppl. 1993, 112, 35.
8. Sumiyoshi, T.; Stockmeier, C. A.; Overholser, J. C.; Dilley,
G. E.; Meltzer, H. Y. Brain Res. 1996, 708, 209.
9. Reitz, A. B.; Bennet, D. J.; Blum, P. S.; Codd, E. E.; Mar-
yano€, C. A.; Ortegon, M. E.; Renzi, M. J.; Scott, M. K.;
Shank, R. P.; Vaught, J. L. J. Med. Chem. 1994, 37, 1060.
10. Bartoszyk, G. D.; Greiner, H. D.; Seyfried, C. A. Soc.
Neurosci. Abstr. 1997, 23, 530.
11. Wustrow, D.; Bellitotti, T.; Glase, S.; Ross Kesten, S.;
Johnson, D.; Rubin, R.; Blackburn, A.; Akunne, H.; Corbin,
A.; Davis, M. D.; Georgic, L.; Whetzel, S.; Zoski, K.; He€ner,
T.; Pugsley, T.; Wise, L. J. Med. Chem. 1998, 41, 760.
12. Homan, E. J.; Copinga, S.; Elfstrom, L.; Van der Veen, T.;
Hallema, J.-P.; Mohell, N.; Unelius, L.; Johansson, R.; Wik-
strom, H. V.; Grol, C. J. Bioorg. Med. Chem. 1998, 6, 2111.
13. Homan, E. J.; Kroodsma, E.; Copinga, S.; Unelius, L.;
Wikstrom, H. V.; Grol, C. J. Bioorg. Med. Chem. 1999, 7,
1111.
21. Liu, Y.; Yu, H.; Svensson, B. E.; Cortizo, L.; Lewander,
T.; Hacksell, U. J. Med. Chem. 1993, 36, 4221.
22. Cook, M. J.; Forbes, E. J.; Khan, G. M. Chem. Commun.
1966, 121.
23. Plummer, B. F.; Menendez, M.; Songster, M. J. Org. Chem
1989, 54, 718.
24. Reed, K. L.; Gupton, J. T.; Solarz, T. L. Synth. Commun.
1990, 20, 563.
25. McKillop, A.; Kemp, D. Tetrahedron 1989, 45, 3299.
26. Cacchi, S.; Misiti, D. Synthesis 1980, 243.
27. Dolle, R. E.; Schmidt, S. J.; Kruse, L. I J. Chem. Soc.
Chem. Commun 1987, 904.
28. Jagdmann, E. G.; Munson, R. H.; Gero, T. W. Synth.
Commun. 1990, 20, 1203.
29. Saa, J. M.; Dopico, M.; Martorell, G.; Garcia-Raso, A. J.
Org. Chem. 1990, 55, 991.
30. Sonesson, C.; Lin, C.-H.; Hansson, L.; Waters, N.; Svens-
son, K.; Carlsson, A.; Smith, M. W.; Wikstrom, H. J. Med.
Chem. 1994, 37, 2735.
31. Sonesson, C.; Boije, M.; Svensson, K.; Ekman, A.; Carls-
son, A.; Romero, A. G.; Martin, I. J.; Duncan, J. N.; King, L.
J.; Wikstrom, H. J. Med. Chem. 1993, 36, 3409.
32. Hacksell, U.; Svensson, U.; Nilsson, J. L.; Hjorth, S.;
Carlsson, A.; Wikstrom, H.; Lindberg, P.; Sanchez, D. J. Med.
Chem. 1979, 22, 1469.
33. Yu, H.; Liu, Y.; Malmberg, A.; Mohell, N.; Hacksell, U.;
Lewander, T. Eur. J. Pharmacol. 1996, 303, 151.
34. Malmberg, A.; Nordvall, G.; Johansson, A. M.; Mohell,
N.; Hacksell, U. Mol. Pharmacol. 1994, 46, 299.
35. Van Vliet, L. A.; Tepper, P. G.; Dijkstra, D.; Damsma, G.;
Wikstrom, H.; Puglsey, T. A.; Akunne, H. C.; He€ner, T. G.;
Glase, S. A.; Wise, L. D. J. Med. Chem. 1996, 39, 4233.
36. Wold, S.; Johansson, E.; Cocchi, M. In 3D QSAR in Drug
DesignÐTheory, Methods and Applications; Kubinyi, H., Ed.;
ESCOM: Leiden, 1993; pp 523±550.
37. Geladi, P.; Kowalski, B. R. Anal. Chem. Acta 1986, 185, 1.
38. Cheng, Y.-C.; Pruso€, W. H. Biochem. Pharmacol 1973,
22, 3099.
39. Tsar 3.0, Oxford Molecular Group, Oxford OX4 4GA,
UK.
14. Homan, E. J.; Copinga, S.; Unelius, L.; Jackson, D. M.;
Wikstrom, H. V.; Grol, C. J. Bioorg. Med. Chem. 1999, 7, 1263.
15. Homan, E. J.; Wikstrom, H. V.; Grol, C. J. Bioorg. Med.
Chem. 1999, 7, 1805.
16. Hendrickson, J. B.; Bergeron, R. Tet. Lett. 1973, 46, 4607.
17. Cabri, W.; Candiana, I.; Bedeschi, A. J. Org. Chem. 1990,
55, 3654.
18. Hedberg, M. H.; Linnanen, T.; Jansen, J. M.; Norvall, G.;
Hjorth, S.; Unelius, L.; Johansson, A. M. J. Med. Chem. 1996,
39, 3503.
19. Chambers, M. R. I.; Widdowson, D. A. J. Chem. Soc.
Perk. Trans. 1 1989, 1365.
40. PLS_Toolbox version 1.5 for Matlab, Eigenvector
Research, Inc., Manson,WA 98831, USA.
20. Saa, J. M.; Martorell, G.; Garcia-Raso, A. J. Org. Chem.
1992, 57, 678.
41. MatLab 5.2, The MathWorks, Inc., Natick, MA 01760-
1500, USA.