Synthesis and dopaminergic properties of benzo-fused analogues of quinpirole and quinelorane

Article abstract of DOI:10.1021/jm9804533

In an analogy to the potent catechol dopamine D₁ agonists dihydrexidine (1) and dinapsoline (2), benzo rings were fused onto the structures of the dopamine D₂-selective agonists quinelorane (3) and quinpirole (4). Each of the phenyl

Full text of DOI:10.1021/jm9804533

J . Med. Chem. 1999, 42, 935-940
935
Brief Articles
Syn th esis a n d Dop a m in er gic P r op er ties of Ben zo-F u sed An a logu es of
Qu in p ir ole a n d Qu in elor a n e
Martin K.-H. Doll,† David E. Nichols,*,† J ason D. Kilts,‡ Cassandra Prioleau,‡ Cindy P. Lawler,‡
Mechelle M. Lewis,‡ and Richard B. Mailman‡
Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences,
Purdue University, West Lafayette, Indiana 47907, and Neuroscience Center and Departments of Pharmacology and
Psychiatry, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599-7250
Received August 3, 1998
In an analogy to the potent catechol dopamine D1 agonists dihydrexidine (1) and dinapsoline
(2), benzo rings were fused onto the structures of the dopamine D2-selective agonists quinelorane
(3) and quinpirole (4). Each of the phenyl ring-substituted derivatives had significant affinity
for D2 receptors, albeit somewhat lower than the two parent compounds, 3 and 4. Compounds
with N-propyl and N-allyl substituents (5b, 5c, 6c, and 6d ) had higher affinity for the D2
dopamine receptor than did their corresponding secondary amines (5a and 6a ). Slightly different
effects on affinity of an n-propyl and an n-allyl group in the new analogues of 3 and 4 suggest
that different binding orientations may be invoked at the receptor.
In tr od u ction
to selectivity for the D1 receptor. Indeed, we demon-
strated that this structural feature is necessary for high
affinity and agonist efficacy at the dopamine D1 recep-
tor.5 Surprisingly, we discovered that the pendant
phenyl ring does not exclude these ligands from D2-like
receptors. It does, however, appear to engender in these
molecules functional selectivity toward specific sub-
populations of dopamine D2 receptors. For example both
1 and its N-n-propyl derivative proved to have selectiv-
ity for postsynaptic D2 receptor function, with the latter
having much higher affinity.7 In addition, the 4-methyl-
N-n-propyl derivative of 1 proved to be a high-affinity
(K0.5 ) 2.1 nM) ligand for the dopamine D3 receptor,
having greater than 100-fold selectivity for the D3 versus
the D2L isoform.8
Dopamine-mediated neurotransmission plays a role
in several psychiatric and neurological disorders and
has been implicated in the action of many drugs of
abuse. The development of novel ligands for brain
dopamine receptors is therefore extremely important,
so that appropriate pharmacological tools will be avail-
able to assist in the elucidation of the functional roles
and importance of the various dopamine receptors.
Classification of dopamine receptors into D1 and D2
subtypes was based originally on pharmacological and
functional grounds, including their coupling to the
enzyme adenylate cyclase.1 Recent molecular cloning
studies have led to the identification of five genes that
code dopamine receptor proteins: two in the D1-like
family (D1 and D5 or D1B) and three in the D2-like family
(D2, D3, and D4).2-4 In this paper, the use of the terms
D1 and D2 generally will be to the respective families
unless otherwise noted.
At present, no validated three-dimensional orienta-
tion of the amino acid residues at or near the ligand
binding site has been elucidated. Thus, research still
relies heavily on structure-activity relationship (SAR)
information generated from molecules synthesized based
on a lead compound. Our most important findings to
date have arisen from our discovery of the high-affinity,
full D1 agonist activity of certain benzo[a]phenan-
thridines such as 15 and naphthoisoquinoline 2.6 Al-
though these molecules possess high potency at D1
receptors, they nevertheless also possess significant
activity at receptors of the D2-type. We had reasoned
initially that the pendant phenyl ring in 1 would lead
* Address correspondence to: Dr. David E. Nichols. Phone: (765)
494-1461. Fax: (765) 494-1414. E-mail: drdave@pharmacy.purdue.edu.
† Purdue University.
‡ University of North Carolina School of Medicine.
10.1021/jm9804533 CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/09/1999

936 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 5
Brief Articles
Sch em e 1a
Sch em e 2a
a
Reagents and conditions: (a) allylamine/toluene, reflux; (b)
benzoyl chloride/CH2Cl2, 0-25 °C, 80% (for a and b); (c) 0.025 M
solution in MeOH, 24 °C, hν, 72%; (d) LAH/THF, reflux; (e) 1 N
aq HCl/THF, 25-80 °C, 85% (for d and e).
a
Based on these findings, it was of interest to modify
two known and highly potent ligands for D2-like recep-
tors, quinelorane (3) and quinpirole (4), by appending
a phenyl ring, leading to compounds 5 and 6. Both 3
and 4 have been previously characterized as dopamine
D2 ligands, with modest selectivity for the D3 isoform.9,10
It was anticipated that this strategy might lead to
ligands targeted to specific subpopulations of D2-like
receptors that differed from those targeted by the parent
molecules. This report describes the synthesis of mol-
ecules 5a -c and 6a -d , as well as preliminary pharma-
cological characterization using radioligand binding.
Reagents and conditions: (a) tris(dimethylamino)methane/
toluene, reflux, then guanidine carbonate/EtOH, reflux, 94%; (b)
Pd(dba)2/DPPB/THF, thiosalicylic acid, room temperature, 96% for
5a and 81% for 6a ; (c) Pd-C/EtOH, ambient pressure, 86% for 5c
and 68% for 6d ; (d) t-BuOK/THF, HCO2Et, 0 °C to room temper-
ature, then hydrazine, pH 9, 74%; (e) H2CO/NaBH3CN in MeOH/
H2O, room temperature, 77%.
subsequent acylation of the resulting enamine with
benzoyl chloride provided the required precursor 7,
which was then irradiated in a 25 mM solution in MeOH
to afford 8 in 72% yield. The trans ring fusion was
proved unambiguously by 1H NMR, showing a coupling
constant of 12 Hz for the H4a/H10b spin system.
Reduction of the amide carbonyl group using LAH,
followed by cleavage of the ketal, was carried out as a
“one-pot reaction” analogously to a procedure reported
in the literature.18 Thus, the key intermediate 9 was
obtained in 85% yield. The R-enamino ketone was
formed from 9 (Scheme 2) with tris(dimethylamino)-
methane as the reagent and then was treated with
guanidine carbonate to give the desired 5b. Catalytic
hydrogenation then afforded 5c.
In an analogous fashion, the regioselective formyl-
ation using t-BuOK/ethyl formate and subsequent con-
densation with hydrazine19 afforded the respective
N-allyl intermediate 6c in 74% yield, which was then
hydrogenated to provide the target compound 6d . The
regioselectivity of the heteroannelation reaction was
confirmed when it was observed in the EIMS spectra of
the N-allyl compounds 5b and 6c that the base peak at
m/z 170 was derived from a thermally promoted retro-
[4+2] reaction, cleaving the tetrahydropyridine ring
with generation of the N-allyl isoquinolinium cation
(after elimination of an H radical and rearomatization
of the corresponding parent ion, m/z 171). This retro-
Diels-Alder fragmentation was observed in the CIMS
spectra of all the compounds 5a -c and 6a -d .
Ch em istr y
For the synthesis of the two basic heterocyclic struc-
tures 5b and 6c, the trans-hexahydrophenanthridinone
9 was chosen as the common intermediate and was
efficiently prepared in a three-step sequence (overall
yield 49%; Scheme 1). Based on the pioneering work of
Ninomiya,11 the photocyclization of enamides proved to
be a highly useful tool in alkaloid synthesis and has
been extensively reviewed in the literature.12-14 Accord-
ingly, trans-4a,10b-hexahydrophenanthridines can be
synthesized in good yield and with excellent diastereo-
selectivity from the respective benzamides.15,16 There-
fore the photochemical cyclization of 7 was chosen as
the key step in the preparation of the desired interme-
diate 9.
Although only a few different substituents at the
amide nitrogen have been examined in the photo-
cyclization reaction11,17 and N-benzyl derivatives are
usually employed, N-allylation in benzamide 7 seemed
to be most suitable in our case. N-Allyl derivatives can
be readily converted into both types of target com-
pounds, the desired N-propyl congeners (catalytic hy-
drogenation) as well as into the corresponding secondary
amines by deallylation procedures. In addition, in the
event that the N-n-propyl compounds should prove to
have pharmacological utility, the allyl moiety could be
easily reduced catalytically with tritium to provide a
radioligand for receptor binding studies.
Two major methods have generally been employed for
N-deallylation reactions, based on isomerization of the
allyl to a 2-propenyl group that readily undergoes in
situ hydrolysis to the secondary amine and propanal.20,21
A more recent procedure22 relying on transfer of the allyl
cation to the mercapto group of thiosalicylic acid by
means of Pd(0) catalysis proved to be most successful
Condensation of commercially available cyclohexane-
dione monoethylene glycol ketal with allylamine and

Brief Articles
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 5 937
Ta ble 1. Dopamine Receptor Affinities of Compounds 5 and 6 and Their Analogues
K0.5 (nM)
D1 affinity
D2 affinity
D2 long
D3 affinity
compound no.
rat striatum
rat striatum
C-6 glioma cells
C-6 glioma cells
SCH23390
spiperone
0.4 ( 0.02
0.21 ( 0.03
58.1 ( 7.8
31.3 ( 4.4
4.78 ( 0.2
28.8 ( 9.2
467 ( 115
102 ( 37
45 ( 8.4
0.16 ( 0.02
1490 ( 270
174 ( 26
710 ( 143
2250 ( 620
nd
0.37 ( 0.22
170 ( 23
nd
1 (DHX)
2 (dinapsoline)
3 (quinelorane)
4 (quinpirole)
5a
5b
5c
6a
6b
6c
6d
6.22 ( 1.07
5.93 ( 0.45
>5 µM
0.67 ( 0.30
4.5 ( 1.6
324 ( 109
15.8 ( 4.3
20.2 ( 4.6
955 ( 297
187 ( 0.75
73.7 ( 23.5
162 ( 32
>5 µM
>5 µM
>5 µM
>5 µM
>5 µM
>5 µM
>5 µM
641 ( 187
189 ( 115
100 (32
>5 µM
>5 µM
3090 ( 150
1880 ( 850
4300 ( 1410
>5 µM
>5 µM
209 ( 54
for providing the secondary amines 5a and 6a in yields
of 96% and 81%, respectively. The N-methyl derivative
6b was additionally prepared in 77% yield using form-
aldehyde/NaCNBH3.
the receptor in different orientations. These observa-
tions imply that future attempts to carry out molecular
modeling and/or receptor docking studies should care-
fully consider alignment of the various molecules and
that the correct alignments in apparently structurally
similar molecules may not be intuitively obvious.
P h a r m a cology
Selectivity of Com p ou n d s for Str ia ta l D1 vs D2
Dop a m in e Recep tor s. Binding affinities were as-
sessed via competition binding assays using rat striatal
membranes and with [3H]SCH23390 and [3H]spiperone
as D1 and D2 radioligands, respectively (see Table 1).
None of the derivatives showed significant D1 affinity
(K0.5 > 5 µM). In addition, each of the phenyl ring-
substituted derivatives displayed some loss of D2 recep-
tor affinity when compared to the two parent com-
pounds, quinelorane (3) and quinpirole (4). Significant
affinity at D2 receptors was observed, however. Com-
pounds with N-propyl and N-allyl substituents (5b, 5c,
6c, and 6d ) had higher affinity for the D2 dopamine
receptor than did their corresponding secondary amines
(5a and 6a ), a fact consistent with the known SAR of
dopamine D2 agonists.
Selectivity of Com p ou n d s for Molecu la r Iso-
for m s of D2-lik e Recep tor s. To determine the utility
of these compounds as ligands to probe the D2-like
receptor isoforms, their affinity for the D2L or D3
dopamine receptor was determined in C-6 glioma cells
transfected with these receptors. As shown in Table 1,
5a -6d did not demonstrate high affinity for the D2L
receptor but exhibited some selectivity for the D3 vs D2
receptor. Those quinelorane derivatives with N-propyl
and -allyl substituents (5b and 5c) had good affinity
and a high degree of selectivity for the D3 vs D2 receptor,
much like quinpirole (4) and quinelorane (3) themselves.
The appended phenyl ring clearly lacks the effect that
is seen in catechol-containing agonists such as 1 and 2.
D1 receptor potency is not enhanced, and affinity for D2-
like receptors is attenuated significantly. Comparing
quinelorane (3) with its benzo-fused analogue 5c, the
latter has some 30-fold lower affinity at the D3 isoform.
This divergence in SAR found with catechol versus non-
catechol dopamine agonists suggests that different
binding orientations may be invoked at the receptor.
Interestingly, for the quinpirole series, the N-allyl 6c
had an approximately 2-fold higher affinity than did
N-propyl 6d , while no similar effect occurred in the
quinelorane analogues. This fact would seem to suggest
that perhaps even quinpirole and quinelorane bind to
Exp er im en ta l Section
Ch em ica l P r oced u r es. Melting points were determined
on a Thomas-Hoover Meltemp apparatus and are uncorrected.
Unless otherwise noted, 1H NMR spectra (300 MHz) were
recorded on a Bruker ARX 300 instrument and the solvent
(CHCl3) was used as internal standard; 500-MHz spectra were
performed on a Varian VXR 500 S spectrometer. Chemical
shifts are reported in δ (ppm) using the common abbreviations.
Elemental analysis was carried out at Purdue Microanalysis
Laboratory using a Perkin-Elmer 2400 apparatus, and all of
the results were within (0.40% from the calculated values.
Mass spectra were recorded on a Finnigan 4000 quadrupole
mass spectrometer using isobutane as the reactant gas for
chemical ionization (CI) and a -70 eV potential for the electron
impact spectra (EI); all ion peaks are listed as m/z (relative
intensity) values. THF was freshly distilled from Na/benzo-
phenone ketyl. All other chemicals were used as purchased.
Pd(dba)2 (dba ) dibenzylideneacetone) was obtained from
Aldrich and DPPB (1,4-bis(diphenylphosphino)butane) from
Acros Organics (Fischer Scientific). Salts were prepared either
from the analytically pure bases by use of the stoichiometric
amount of diluted ethanolic HCl or HBr, respectively, or from
the purified bases (column chromatography) followed by re-
crystallization.
N -Allyl-N -(1,4-d ioxa sp ir o[5.4]d e ca -7-e n -8-yl)b e n z-
a m id e (7). A solution of cyclohexanedione monoethylene glycol
ketal (25.00 g, 160.1 mmol), and allylamine (14.5 mL, 196.2
mmol) in toluene was held at reflux for 4 h, while water was
removed continuously by means of a Dean-Stark trap. An-
other 10 mL of allylamine was added, and reflux was continued
for 2 h. The mixture then was concentrated under reduced
pressure, the residue taken up into 160 mL of CH2Cl2, and
Et3N (24.4 mL, 176.1 mmol) added. The solution was cooled
in an ice bath, and benzoyl chloride (18.6 mL, 160.1 mmol) in
CH2Cl2 (120 mL) was added dropwise with cooling. The
mixture was stirred overnight at room temperature. After
extraction with H2O and aqueous NaHCO3 (5%), the organic
layer was dried (Na2SO4) and evaporated to leave a yellow
solid. Recrystallization from hexane provided 7 (38.11 g, 80%)
as thick yellow crystals: mp 71 °C; CIMS m/z 300 ([M + H]+,
21), 105 ([C6H6CO]+, 100); 1H NMR δ 1.65-1.69 (m, 2H), 2.13
(br s, 2H), 2.25-2.27 (br m, 2H), 3.91 (s, 4H), 4.22-4.24 (d, J
) 6.0 Hz, 2H), 5.17-5.28 (m, 3H), 5.86-5.99 (m, 1H), 7.32-
7.36 (m, 3H), 7.50-7.53 (m, 2H). Anal. (C18H21NO3) C, H, N.
t r a n s-4a ,10b -5-Allyl-1,2,3,4,4a ,5,6,10b -oct a h yd r o-6-
p h en a n th r id on e-2-sp ir o-2′-d ioxola n e (8). A 0.025 M solu-
tion of 7 (4.64 g, 15.5 mmol) in absolute MeOH (620 mL) was

938 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 5
Brief Articles
placed into an Ace glass 500-mL photochemical reactor and
irradiated with a 450-W Hanovia medium pressure, quartz,
mercury-vapor lamp seated in a cold tap water-cooled, quartz
immersion well. Throughout the reaction time, the lamp and
the solution were flushed continuously with argon and stirring
was maintained using a Teflon-coated magnetic stirring bar.
The reaction temperature was not allowed to exceed 24 °C
by stirring a mixture of Pd(dba)2 (9.9 mg) and DPPB (7.3 mg)
in dry THF (0.5 mL) for 20 min under argon. To a solution of
5b (100.0 mg, 0.34 mmol) stirred under argon in dry THF (6
mL) was added thiosalicylic acid (62.9 mg, 0.41 mmol). The
solution of the preformed catalyst then was added dropwise.
The mixture was stirred for 1 h at ambient temperature, then
diluted with water, and concentrated to ca. one-third of its
volume. Water was added; the solution was basified (1 N
aqueous NaOH) and extracted (CHCl3). The combined organic
layers were washed with brine, dried (Na2SO4), and evapo-
rated. The residue was purified on silica gel, eluting with
CHCl3-MeOH (8:2.5) to give 5a (83.1 mg, 96%) as an off-white
solid: mp 229 °C (sharp, decomposition; 2-propanol); mp
(monohydrochloride salt) >260 °C; CIMS m/z 253 ([M + H]+,
100), 131 (6), 130 (8); 1H NMR δ 2.54-2.72 (m, 2H), 2.91-
2.95 (m, 3H), 3.46 (dd, J ) 18.6, 4.0 Hz, 1H), 4.09 (d, J ) 6.0
Hz, 1H), 4.23 (d, J ) 6.0 Hz, 1H), 4.87 (br s, 2H), 7.06 (d, J )
7.3 Hz, 1H), 7.15-7.24 (m, 2H), 7.32 (d, J ) 7.3 Hz, 1H), 8.09
(s, 1H). Anal. (C15H16N4) C, H, N.
tr a n s-6a ,12b-3-Am in o-7-p r op yl-1,6,6a ,7,8,12b-h exa h y-
d r op yr im id in o[4,5-b]p h en a n th r id in e (5c). A solution of 5b
(210 mg, 0.72 mmol) in absolute EtOH (80 mL) containing 10%
Pd/C (30 mg) was stirred vigorously overnight under an H2
atmosphere at ambient pressure. The mixture was filtered
though Celite, and the residue from evaporation was chro-
matographed on silica gel, eluting with CHCl3-CH3CN-
MeOH-NH4OH (180:20:4:1) to yield 5c (181 mg, 86%) as a
yellowish resin that solidified upon standing: mp 132-136 °C
dec; mp (monohydrochloride salt) 238-241 °C dec; CIMS m/z
295 ([M + H]+, 100), 173 (17), 172 (32); 1H NMR δ 0.94 (t, J )
7.3 Hz, 3H), 1.56-1.64 (m, 3H), 2.47-2.81 (m, 5H), 3.05-3.09
(m, 1H), 3.43 (dd, J ) 18.3, 5.2 Hz, 1H), 3.68 (d, J ) 15.4 Hz,
1H), 3.99 (d, J ) 15.3 Hz, 1H), 4.86 (br s, 2H), 7.06 (d, J ) 7.3
Hz, 1H), 7.14-7.24 (m, 2H), 7.29 (d, J ) 7.3 Hz, 1H), 8.11 (s,
1H). Anal. (C18H22N4) C, H, N.
tr a n s-5a ,11b-6-Allyl-1,5,5a ,6,7,11b-h exa h yd r op yr a zolo-
[3,4-b]p h en a n th r id in e (6c). A suspension of t-KOBu (0.93
g, 8.3 mmol) in THF (10 mL) was stirred under argon at 0 °C
while a solution of ethyl formate (1.34 mL, 16.6 mmol) and 9
(1.00 g, 4.2 mmol) in THF (4 mL) was added via syringe over
a 10-min period. The ice bath was removed, and stirring was
continued for 90 min at room temperature. To this mixture
was slowly added a 1 M hydrazine solution in THF (12.5 mL,
12.5 mmol) followed by a few drops of 1 N aqueous HCI to
adjust the pH to ca. 9. After stirring for 20 h, the mixture was
basified, diluted with twice the volume of water, and extracted
(CH2Cl2). The combined organic layers were washed with
brine, dried (Na2SO4), and evaporated to give an amorphous
solid. Purification on silica gel, eluting with CHCl3-MeOH (10:
0.5), afforded 6c (0.81 g, 74%) as a colorless foam: mp
(monohydrochloride salt) >205-215 °C continuous dec; CIMS
m/z 266 ([M + H]+, 100), 170 (15); EIMS m/z 266 ([M + H]+,
21), 265 (M+., 28), 264 ([M - H]+, 32), 171 (67), 170 (100), 129
(33); 1H NMR δ 2.59-2.74 (m, 3H), 3.14-3.25 (m, 3H), 3.47-
3.70 (m, 3H), 3.96 (d, J ) 15.3 Hz, 1H), 5.21-5.30 (m, 2H),
5.92-6.06 (m, 1H), 7.07 (d, J ) 7.1 Hz, 1H), 7.17-7.33 (m,
3H), 7,36 (s, 1H). Anal. (C17H19N3) C, H, N.
tr a n s-5a ,11b-1,5,5a ,6,7,11b-Hexa h yd r op yr a zolo[3,4-b]-
p h en a n th r id in e (6a ). The reaction was carried out as
described for 5a . For chromatography, CHCl3-MeOH-NH4OH
(95:5:0.5) was used as eluent, resulting in 81% of 6a as a white
solid that could be recrystallized from CHCl3 as its mono-
CHCl3 solvate: mp 164-165 °C (light yellow needles); mp
(monohydrochloride salt) >260 °C; CIMS m/z 226 ([M + H]+,
100), 131 (6), 130 (9); 1H NMR (500 MHz) δ 2.45-2.60 (m, 2H),
2.94-3.00 (m, 3H), 3.53-3.57 (m, 1H), 4.08 (d, J ) 16 Hz, 1H),
4.22 (d, J ) 15.7 Hz, 1H), 7.06 (d, J ) 7.3 Hz, 1H), 7.17 (t, J
) 7.1 Hz, 1H), 7.21-7.24 (m, 1H), 7.34-7.36 (m, 2H). Anal.
(CHCl3 solvate; C15H16N3Cl3) C, H, N, Cl.
1
while the progress of the reaction was monitored by H NMR.
After 5 h the irradiation was stopped, the slightly yellow
solution was evaporated, and the residue was chromato-
graphed on silica gel, using CH2Cl2-Et2O (100:6) as the eluent
to give 8 (3.35 g, 72%) as a colorless highly viscous oil that
crystallized upon standing in the cold: mp 76-77 °C; CIMS
m/z 300 ([M + H]+, 100), 284 (29), 101 (21); 1H NMR (500 MHz)
δ 1.56-1.71 (m, 2H), 1.82-1.92 (m, 2H), 2.20-2.25 (m, 1H),
2.43-2.48 (m, 1H), 3.13 (ddd, J ) 12.3, 12.3, 3.9 Hz, 1H), 3.39
(ddd, J ) 11.8, 12.0, 3.8 Hz, 1H), 3.90-4.05 (m, 5H), 4.73-
4.78 (m, 1H), 5.10-5.14 (m, 2H), 5.79-5.86 (m, 1H), 7.20 (br
d, J ) 7.7 Hz, 1H), 7,32-7.35 (br t-like m, 1H), 7.45 (ddd, J )
7.6, 3.0, 1.5 Hz, 1H), 8.10 (dd, J ) 7.7, 1.4 Hz, 1H); 13C NMR
(500 MHz) δ 27.34, 33.06, 36.18, 38.99, 43.85, 57.59, 64.57,
64.64, 107.74, 116.02, 122.67, 126.99, 128.73, 128.88, 131.93,
133.88, 140.39, 165.23. Anal. (C18H21NO3) C, H, N.
tr a n s-4a ,10b-5-Allyl-3,4,4a ,5,6,10b-h exa h yd r o-1H-p h en -
a n th r id in -2-on e (9). A two-necked flask (500 mL) equipped
with a dropping funnel with drying tube and argon inlet was
charged with a suspension of LiAlH4 (3.40 g, 89.6 mmol) in
THF (300 mL) and precooled to 0 °C. Over a 1-h period, a
solution of 8 (5.70 g, 19.1 mmol) in THF (120 mL) was added
dropwise with cooling and stirring. The dropping funnel was
replaced by a condenser, and the mixture was heated at reflux
for 2 h. After cooling, the flask was immersed in an ice bath,
3 N aqueous NaOH (25 mL) was added dropwise, and the
resulting mixture was filtered. The filtered alumina salts were
washed repeatedly with THF, the combined organic layers
evaporated, and the resulting oil was dissolved in THF (100
mL). To this solution was slowly added 500 mL of 1 N aqueous
HCl, and stirring was continued overnight at room tempera-
ture under argon and then at 80 °C for 1 h. The acidic solution
was extracted with CH2Cl2, then basified (3 N aqueous NaOH),
and extracted with CHCl3. The CHCl3 extract was dried
(Na2SO4) and evaporated to provide 4.48 g of crude product
that was chromatographed on silica gel, eluting with EtOAc-
hexane-Et3N (150:100:1) affording 9 as a viscous oil (3.90 g,
85%). After standing in the cold, the pure product solidified
and was recrystallized from hexane: mp 54-56 °C; CIMS m/z
242 ([M + H]+, 100); 1H NMR δ 1.72-1.89 (m, 1H), 2.31-2.61
(m, 4H), 2.73-2.81 (m, 1H), 3.04-3.11 (m, 2H), 3.16-3.23 (m,
1H), 3.39-3.45 (m, 1H), 3.78 (d, J ) 15.7 Hz, 1H), 3.97 (d, J
) 15.7 Hz, 1H), 5.19-5.27 (m, 2H), 5.87-6.00 (m, 1H), 7.04-
7.20 (m, 4H). Anal. (C16H19NO) C, H, N.
tr a n s-6a ,12b-7-Allyl-3-a m in o-1,6,6a ,7,8,12b-h exa h yd r o-
p yr im id in o[4,5-b]p h en a n th r id in e (5b). To a solution of 9
(2.50 g, 10.4 mmol) in dry toluene (22 mL) was added tris-
(dimethylamino)methane (6.5 mL, 37.6 mmol), and the mix-
ture was stirred under argon at 90 °C for 4 h. The solvents
were removed, the residue was dissolved in EtOH (45 mL),
guanidine carbonate (4.00 g, 22.2 mmol) was added, and the
mixture was heated at reflux for 17 h. The solvents were
evaporated, the residue was partitioned between CHCl3 and
H2O, and the aqueous layer was repeatedly extracted (CHCl3).
Drying (Na2SO4) and evaporation of the organic layer left a
residue that was chromatographed over silica gel, eluting with
CHCl3-MeOH (25:1) to yield 5b (2.86 g, 94%) as a yellowish
amorphous foam: mp (monohydrochloride salt; H2O) 274-276
°C; CIMS m/z 293 ([M + H]+, 100), 170 (14); EIMS m/z 292
([M - H]+, 24), 171 (45), 170 (100), 130 (11), 129 (28); 1H NMR
δ 2.65-2.75 (m, 3H), 3.12-3.21 (m, 3H), 3.45 (dd, J ) 17.7,
5.0 Hz, 1H), 3.63-3.70 (m, 2H), 3.98 (d, J ) 15.4 Hz, 1H), 4.98
(br s, 2H), 5.23-5.32 (m, 2H), 5.92-6.05 (m, 1H), 7.07 (d, J )
7.3 Hz, 1H), 7.15-7.32 (m, 3H), 8.12 (s, 1H). Anal. (monohy-
drochloride salt; C18H21N4Cl) C, H, N, Cl.
tr a n s-5a ,11b -6-Met h yl-1,5,5a ,6,7,11b -h exa h yd r op yr a -
zolo[3,4-b]p h en a n th r id in e (6b). To a solution of 6a (100 mg,
0.4 mmol) in MeOH (4 mL) and H2O (1.5 mL) were added 37%
aqueous formaldehyde solution (0.18 mL, 2.4 mmol) and
NaBH3CN (92.9 mg, 1.5 mmol). The mixture was stirred at
tr a n s-6a ,12b-3-Am in o-1,6,6a ,7,8,12b-h exa h yd r op yr im i-
d in o[4,5-b]p h en a n th r id in e (5a ). The catalyst was prepared

Brief Articles
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 5 939
room temperature overnight under argon, then concentrated
to ca. one-third, and diluted with H2O. After basifying and
extraction (CHCl3), the organic layers were dried (Na2SO4) and
evaporated. The residue was purified on silica gel, eluting with
CHCl3-MeOH-NH4OH (400:20:1) to give 6b (70 mg, 77%) as
a colorless glass: mp (monohydrobromide salt, acetone/H2O)
>225 °C dec; CIMS m/z 240 ([M + H]+, 100), 145 (22), 144
binding of the radioligand to serotonin sites. Total binding was
defined as radioligand binding in the absence of any competing
drug. Nonspecific binding of [3H]SCH23390 was defined by
adding 1 µM unlabeled SCH23390, whereas 1 µM chlorprom-
azine was used to define nonspecific binding of [3H]spiperone.
Triplicate determinations were performed in each assay. After
incubation at 37 °C for 15 min, in some experiments tubes
were filtered rapidly through Skatron glass fiber filter mats
(model number 11734) and rinsed with 5 mL of ice-cold wash
buffer (50 mM Hepes, 4 mM MgCl2; pH 7.4) using a Skatron
microcell harvester (Skatron Instruments Inc., Sterling, VA).
Filters were allowed to dry and then punched into scintillation
vials (Skatron Instruments Inc., Sterling, VA). OptiPhase
HiSafe II scintillation cocktail (1 mL) was added to each vial.
After vials were shaken for 30 min, radioactivity in each
sample was determined on an LKB Wallac 1219 Rackbeta
liquid scintillation counter (Wallac Inc., Gaithersburg, MD).
In other experiments, 96-tube plates were filtered rapidly
through Packard 96 GF/C filters and rinsed four times with 5
mL of ice-cold wash buffer (10 mM Tris, 0.9% NaCl; pH 7.4)
using a Packard microcell harvester (Packard Instruments,
Downers Grove, IL). Filters were allowed to dry, and then 25
µL of Microscint-20 scintillation cocktail was added to each
filter well. Each plate was sealed using a microplate heat
sealing film on a microplate sealer and then counted on a
Packard topcount microplate scintillation counter (Packard,
Downers Grove, IL). Tissue-protein levels were estimated as
described for C-6 cell membranes.
Clon ed r ecep tor a ssa ys: Cell membranes were suspended
in assay buffer (50 mM Tris, 120 mM NaCl; pH 7.4) at a
concentration of 12.5 µg of wet weight/mL, in a final assay
volume of 1.0 mL for experiments using the Skatron harvester
(Skatron, Inc., Sterling, VA) and 0.5 mL for experiments using
the Packard harvester (Meridan, CT). For competition curves,
D2L and D3 receptors were labeled with [3H]spiperone (0.07
and 0.14 nM, respectively) and several concentrations of test
compound. Total binding was defined as radioligand bound in
the absence of any competing drug, and nonspecific binding
was estimated by adding chlorpromazine (1 µM). Triplicate
determinations were made for each drug concentration. Assay
tubes were incubated at 37 °C for 15 min, and binding was
terminated by rapid filtration as described above for assays
with rat striatum.
1
(52); H NMR (500 MHz) δ 2.42-2.58 (m, 6H), 3.11-3.15 (m,
2H), 3.46 (dd, J ) 15.5, 5.0 Hz, 1H), 3.63 (d, J ) 14.5 Hz, 1H),
3.91 (d, J ) 15.5 Hz, 1H), 7.07 (d, J ) 7.5 Hz, 1H), 7.18 (t, J
) 7.5 Hz, 1H), 7.23 (t, J ) 7.5 Hz, 1H), 7.30 (d, J ) 8.5 Hz,
1H), 7.32 (s, 1H). Anal. (C15H18N3Br) C, H, N, Br.
tr a n s-5a,11b-6-P r opyl-1,5,5a,6,7,11b-h exah ydr opyr azolo-
[3,4-b]p h en a n th r id in e (6d ). Employing 6c, the reaction
procedure was identical to the one described for 5c. The
product was chromatographed on silica gel using CHCl3-
CH3CN-MeOH-NH4OH (360:40:7:2) resulting in 6d (68%) as
a colorless foam: mp 215-220 °C (monohydrochloride salt,
becomes glasslike); CIMS m/z 268 ([M + H]+, 100), 173 (8),
1
172 (14); H NMR δ 0.92 (t, J ) 7.4 Hz, 3H), 1.55-1.65 (m,
2H), 2.47-2.62 (m, 3H), 2.68-2.82 (m, 2H), 3.06-3.12 (m, 2H),
3.50 (dd, J ) 15.8, 5.0 Hz, 1H), 3.37 (d, J ) 14.9 Hz, 1H), 3.96
(d, J ) 15.2 Hz, 1H), 7.06 (d, J ) 7.2 Hz, 1H), 7.13-7.21 (m,
2H), 7.31 (d, J ) 7.6 Hz, 1H), 7.36 (s, 1H). Anal. (C17H21N3) C,
H, N.
P h a r m a cology P r oced u r es. Str ia ta l tissu e p r ep a r a -
tion : Adult male Sprague-Dawley rats weighing 250-400 g
were obtained from Charles River Breeding Laboratories
(Raleigh, NC). Rats were killed by decapitation and the brains
removed rapidly and placed on ice. Striatum was dissected,
and samples were placed in microcentrifuge tubes and stored
frozen at -80 °C until the day of the receptor binding assays.
Clon ed r ecep tor p r ep a r a tion : C-6 glioma cells expressing
rat D2L or D3 receptors were obtained from Dr. Kim Neve
(Oregon Health Sciences University) and grown in 75-cm2
flasks with DMEM-H medium containing 4500 mg/L glucose,
L-glutamine, 5% fetal bovine serum, and 700 ng/mL G418.
Cells were grown to confluence, then rinsed and lysed with
10 mL of ice-cold lysis buffer (1 mM HEPES, 2 mM EDTA;
pH 7.4) for 10 min at 4 °C. Cells were then scraped from the
flasks using a sterile cell scraper obtained from Baxter
(McGraw Park, IL). The combined cell suspension was spun
at 24000g (Sorvall RC-26 Plus/SS-34, DuPont, Wilmington,
DE) at 4 °C for 20 min. The pellet was resuspended in ice-
cold storage buffer (50 mM HEPES, pH 8.0), homogenized with
a Wheaton glass-Teflon homogenizer (5 strokes), and then
stored in 1-mL aliquots to yield an estimated final concentra-
tion of ca. 1.0 mg of protein/mL. Aliquots of the final homo-
genate were stored in microcentrifuge tubes at -80 °C. Prior
to their use for radioligand binding assays, protein levels for
each membrane preparation were quantified using the bicin-
choninic acid (BCA) protein assay reagent (Pierce, Rockford,
IL) adapted for use with a microplate reader (Molecular
Devices, Menlo Park, CA).
Ra t r a d ior ecep tor a ssa ys: Frozen tissue samples from rat
striatum were homogenized gently in 8.0 mL of ice-cold assay
buffer (50 mM HEPES, 4 mM MgCl2; pH 7.4) by seven strokes
in a Wheaton glass-Teflon homogenizer. Tissue was pelleted
by centrifugation (10 min at 15000g), after which the super-
natant was removed. The pellet then was rehomogenized by
five strokes in 8 mL of buffer, followed by repelleting. The
supernatant was decanted, and the pellet was resuspended
by homogenization to a final concentration of 2.0 mg/mL
(estimated protein concentration of ca. 0.14-0.20 mg of
protein/assay tube).
An a lysis of r a d ioliga n d com p etition cu r ves: The re-
sulting concentration curves were analyzed by nonlinear
regression using the algorithms in Prism (GraphPad, Inc.). The
data were initially fit using the sigmoid model to provide a
slope coefficient. All of these data are expressed as K0.5, in
which the experimental IC50 is corrected for the radioligand
concentration using the following formula: K0.5 ) IC50/(1 +
L*/KD*), where L* is the radioligand concentration used in the
assay and KD* is the KD of the radioligand. In cases where
the Hill slope (nH) ) 1, K0.5 ) KI(KD). Using the K0.5 obviates
the need to pick a more complex model (e.g., two vs three site)
when nH < 1, while allowing comparison of IC50’s from
experiments in which different [L*]’s were used.
Ack n ow led gm en t. This work was supported by
PHS Grants MH42705 and MH40537, Center Grants
HD03310 and MH33127, and Training Grant GM07040.
The authors wish to extend their appreciation to the
Swiss National Science Foundation for providing a
fellowship for M.K-H.D. The authors thank J ennifer
Sloan for her technical assistance and interest in this
project.
The amount of tissue added to each assay tube was ca. 0.8
mg, in a final assay volume of 1.0 mL for experiments using
the Skatron harvester (Skatron, Inc., Sterling, VA) and 0.5
mL for experiments using the Packard harvester (Meridan,
Connecticut). Tubes were incubated with either 0.3 nM [3H]-
SCH23390 (D1 receptor assays) or 0.07 nM [3H]spiperone (D2
receptor assays) and varying concentrations of test compounds
at 37 °C for 15 min. Assays with 0.07 nM [3H]spiperone also
included 50 nM ketanserin tartrate in all tubes to mask
Refer en ces
(1) Kebabian, J . W.; Calne, D. B. Multiple receptors for dopamine.
Nature 1979, 277, 93-96.
(2) Van Tol, H. H. M.; Buznow, J . R.; Guan, H. C.; et al. Cloning of
the gene for the human dopamine D4 receptor with high affinity
for the antipsychotic clozapine. Nature 1991, 350, 610-614.

940 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 5
Brief Articles
(3) Sokoloff, P.; Giros, B.; Martres, M.-P.; Bouthenet, M.-L.; Schwartz,
J .-C. Molecular cloning and characterization of a novel dopamine
receptor (D3) as a target for neuroleptics. Nature 1990, 347, 146-
151.
(4) Sunahara, R.-K.; Guan, H. C.; O’Dowd, B. F.; Seeman, P.;
Laurier, L. G.; Ng, G.; George, S. R.; Torchia, J .; Van Tol, H. H.;
Niznik, H. B. Cloning of the gene for a human dopamine D5
receptor with higher affinity for dopamine than D1. Nature 1991,
350, 614-619.
(5) Brewster, W. K.; Nichols, D. E.; Riggs, R. M.; Lovenberg, T. W.;
Mottola, D. M.; Lee, R. C.; Lewis, M. H.; Mailman, R. B. Trans-
10,11-Dihydroxy-5,6,6a,7,8,12b-hexahydrobenzo[a]phenanthiri-
dine: A highly potent selective dopamine D1 full agonist. J . Med.
Chem. 1990, 33, 1756-1764.
(6) Ghosh, D.; Snyder, S. E.; Watts, V. J .; Mailman, R. B.; Nichols,
D. E. 8,9-Dihydroxy-2,3,7,11b-tetrahydro-1H-naph[1,2,3-de]iso-
quinoline: a potent full dopamine D1 agonist containing a rigid
â-phenyl dopamine pharmacophore. J . Med. Chem. 1996, 39,
549-555.
(13) Ninomiya, I.; Naito, T. Application of enamide cyclizations in
alkaloid synthesis. In The Alkaloids; Brossi, A., Ed.; Academic
Press Inc.: New York, 1983; pp 189-279.
(14) Ninomiya, I.; Naito, T. Enamide photocyclization and its ap-
plication to the synthesis of heterocycles. Heterocycles 1981, 15,
1433-1462.
(15) Ninomiya, I.; Naito, T.; Kiguchi, T. Photocyclization of enamides.
Part III. Photocyclisation of N-benzoylenamines of cyclohexanone
and 2-substituted cyclohexanones. J . Chem. Soc., Perkin Trans.
1 1973, 2257-2261.
(16) Ninomiya, I.; Naito, T.; Kiguchi, T. Stereoselective syntheses of
trans-1,2,3,4,4a,5,6,10b-octahydrophenanthridines: Photocy-
clization of cyclohexanone N-benzoylenamines. Tetrahedron Lett.
1970, 4451-4453.
(17) Ninomiya, I.; Naito, T.; Mori, T. Photocyclisation of enamides.
Part I. Photochemical Reactions of N-acyl enamides. J . Chem.
Soc., Perkin Trans. 1 1973, 505-509.
(18) Meyers, A. I.; Berney, D. Asymmetric synthesis of the (4aS,
8aR,8S)-hydrolilolidone system. A formal total synthesis of the
unnatural (+)-aspidospermine. J . Org. Chem. 1989, 54, 4673-
4676.
(7) Smith, H. P.; Nichols, D. E.; Mailman, R. B.; Lawler, C. P.
Locomotor inhibition, yawning, and vacuous chewing induced
by a novel dopamine D2 postsynaptic receptor agonist. Eur. J .
Pharmacol. 1997, 323, 27-36.
(8) Watts, V. J .; Lawler, C. P.; Knoerzer, T.; Mayleben, M. A.; Neve,
K. A.; Nichols, D. E.; Mailman, R. B. Hexahydrobenzo[a]-
phenanthridines: Novel dopamine D3 receptor ligands. Eur. J .
Pharmacol. 1993, 239, 271-273.
(19) Schaus, J . M.; Huser, D. L.; Titus, R. D. Synthesis of the
dopamine agonist (-)-quinpirole. Synth. Commun. 1990, 20,
3553-3562.
(20) Gigg, R.; Conant, R. N-Allyl and N-benzyl derivatives of 2-amino-
2-deoxy-D-glucose. Carbohydr. Res. 1982, 100, c5-c9.
(21) Laguzza, B. C.; Ganem, B. A new protecting group for amines.
Synthesis of anticapsin from L-tyrosine. Tetrahedron Lett. 1981,
22, 1483-1486.
(22) Lemaire-Audoire, S.; Savignac, M.; Dupuis, C.; Genet, J . P.
π-Allyl palladium methodology for selective deprotection of
allylamines. Practical synthesis of secondary amines. Bull. Soc.
Chim. Fr. 1995, 1157-1166.
(9) Gackenheimer, S. L.; Schaus, J . M.; Gehlert, D. R. [3H]-
Quinelorane binds to D2 and D3 dopamine receptors in the rat
brain. J . Pharmacol. Exp. Ther. 1995, 274, 1558-1565.
(10) Malmberg, A.; Mohell, N. Characterization of [3H]-quinpirole
binding to human dopamine D2A and D3 receptors: Effects of
ions and guanidine nucleotides. J . Pharmacol. Exp. Ther. 1995,
274, 790-797.
(11) Ninomiya, I.; Naito, T.; Mori, T. Photochemical reactions of
N-Acylenamines. Tetrahedron Lett. 1969, 2259-2262.
(12) Lenz, G. R. The photochemistry of enamides. Synthesis 1978,
489-518.
J M9804533