Synthesis and biological activities of (R)-5,6-dihydro-N,N-dimethyl-4H- imidazo[4,5,1-ij]quinolin-5-amine and its metabolites

Article abstract of DOI:10.1021/jm960360q

The imidazoquinoline (R)-5,6-dihydro-N,N-dimethyl-4H-imidazo[4,5,1- ij]quinolin-5-amine [(R)3] is a potent dopamine agonist when tested in animals but surprisingly shows very low affinity in in vitro binding assays. When incubated with mouse or monkey liv

Full text of DOI:10.1021/jm960360q

J . Med. Chem. 1997, 40, 639-646
639
Syn th esis a n d Biologica l Activities of (R)-5,6-Dih yd r o-N,N-d im eth yl-
4H-im id a zo[4,5,1-ij]qu in olin -5-a m in e a n d Its Meta bolites
Richard F. Heier,*,† Lester A. Dolak,‡ J . Neil Duncan,§ Deborah K. Hyslop, Michael F. Lipton,| Iain J . Martin,3
Michael A. Mauragis,| Montford F. Piercey, Nanette F. Nichols, Peggy J . K. D. Schreur,
Martin W. Smith,‡ and Malcolm W. Moon†
Structural, Analytical & Medicinal Chemistry, Chemical & Biological Screening, Drug Metabolism Research, CNS Research,
and Chemical Research Preparations, Pharmacia & Upjohn, Inc., Kalamazoo, Michigan 49001-0199
Received May 17, 1996X
The imidazoquinoline (R)-5,6-dihydro-N,N-dimethyl-4H-imidazo[4,5,1-ij]quinolin-5-amine [(R)-
3] is a potent dopamine agonist when tested in animals but surprisingly shows very low affinity
in in vitro binding assays. When incubated with mouse or monkey liver S9 microsomes, (R)-3
is metabolized by N-demethylation and oxidation to (R)-5,6-dihydro-5-(methylamino)-4H-
imidazo[4,5,1-ij]quinolin-2(1H)-one [(R)-6]; intermediate metabolites, where N-demethylation
to the imidazoquinoline (R)-4 and where oxidation to the imidazoquinolinone (R)-5 has taken
place, are also observed in these incubates. A cross-species study on the metabolism of (R)-3
in vitro has shown large variations in the extent of metabolism from species to species.
Imidazoquinolinones (R)-5 and (R)-6 have comparable activity to (R)-3 in animals and also
show good dopaminergic (D2) and serotonergic (5HT1A) activities in binding assays. It is
probable that these metabolites account at least in part for the in vivo activity found for (R)-3.
Efficient syntheses for compounds 3-6 as single enantiomers from quinoline are presented
together with information on the biological activities and metabolic stabilities of these
compounds.
In tr od u ction
ethylamino)imidazoquinoline (R)-3 showed good dopam-
inergic activity when tested in animals but had sur-
prisingly low activity in receptor binding assays.5
Further work showed that this compound was metabo-
lized as outlined in Scheme 1 to imidazoquinolinones
(R)-5 and (R)-6, compounds which show both in vitro
and in vivo biological activities. These compounds were
also more selective for the dopamine D2 receptor
subtype than previously reported compounds. In the
present report, we describe efficient syntheses for the
compounds of Scheme 1 from quinoline and present
their biological activities and metabolic stabilities in rat
hepatocytes. A cross-species study of the metabolism
of (R)-3 by S9 liver microsomes is also presented.
We have previously described a series of 5-aminoimi-
dazoquinolines and related heterocyclic amines which
show dopaminergic and serotonergic activity.1,2 In these
reports, we have described the synthesis and biological
activity of the 5-(dipropylamino)imidazoquinolinone (R)-
1, a potent dopamine D2 agonist which also shows
serotonergic (5HT1A) activity.1 We have also reported
that this compound is metabolized to the 5-(propylami-
no)imidazoquinolinone (R)-2, a compound which has a
similar spectrum of activity but which has much im-
proved oral bioavailability when compared to (R)-1.2 At
Ch em ica l Syn th esis
Compounds 3-6 are readily prepared from quinoline
as outlined in Scheme 26 for the synthesis of the
R-enantiomers. Reduction of quinoline with diisobuty-
laluminum hydride afforded dihydroquinoline 7a as the
major product along with some of the 1,4-dihydroquino-
line isomer.7 In view of the instability of this mixture,
7a was not isolated but immediately converted to the
more stable tert-butoxycarbonyl derivative 7b which was
readily purified by silica gel chromatography.8 This was
reacted with N-bromosuccinimide in aqueous DMSO9
to give the bromohydrin 8 which was treated with
methylamine to give the racemate of the trans-amino
alcohol 10 in excellent yield; while epoxide 9 is an
intermediate in this reaction, its isolation was unneces-
sary. The crude amino alcohol was treated with 0.25
equiv of L-tartaric acid to give (3R)-trans-10 as its hemi-
L-tartrate salt, and this was readily crystallized to
optical purity. The crystalline product, obtained in 37%
yield, was neutralized with sodium hydroxide solution
the time of these reports, dopamine receptors were
divided into two subclasses (D1 and D2). Recently,
molecular biological techniques have established that
there are at least five different dopamine receptor
subtypes, specified as D1-D5.3,4 When (R)-2 was tested
on the new receptors it was found that, like most
available dopamine D2 standards, the compound had
affinity for the dopamine D3 receptor (Table 1). As we
continued work in this area, we found that the 5-(dim-
† Structural, Analytical & Medicinal Chemistry.
‡ Chemical & Biological Screening.
§ Drug Metabolism Research.
CNS Research.
| Chemical Research Preparations.
3 Present address: Astra Pharmaceuticals, U.K.
X Abstract published in Advance ACS Abstracts, February 1, 1997.
S0022-2623(96)00360-3 CCC: $14.00 © 1997 American Chemical Society

640 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5
Heier et al.
Ta ble 1. Dopamine and Serotonin Receptor Binding Data
Sch em e 1. Structures and Metabolic Pathway for
Methylamine Analogues
binding constant, Ki
(nM with 95% confidence intervals)a
compd
(R)-3b
D2 dopamine
D3 dopamine
5HT1A
1281 (860-1908) 2948 (1846-4708) >3521
(S)-3
inactivec
inactive
inactive
(R)-4b
(R)-5b
(S)-5
355 (277-455)d
12 (10-15)
inactivec
2459 (1611-3755) >3257
1929 (1247-2985) 92 (80-105)
inactive
inactive
(R)-6b
(S)-6
9.0 (8.9-9.2)
2333 (1618-3364) 73 (65-82)
inactive
inactivec
inactive
(R)-2b
pergolide
lisuride
2.5 (2.1-3.3)
4.0 (3.4-4.7)
0.5 (0.4-0.7)
36 (31-42)
4.0 (3.7-4.3)
1.7 (1.5-2.0)
87 (71-107)
19 (17-22)
58 (54-63)
4.0 (3.4-4.7)
0.4 (0.3-0.5)
24 (22-27)
1706 (1138-
2557)
bromocriptine 10 (9.5-10.5)
ropinirole
7.2 (6.1-8.4)
a
Receptor binding assays were performed using membranes
prepared from cultured cell lines expressing cloned mammalian
receptors. The following ligands were used D2, [3H]U-86170; D3,
b
[3H]-(+)-7-OH-DPAT; 5HT1A, [3H]-(+)-8-OH-DPAT. These com-
pounds showed low inhibition of radioligand binding (30% or less
at 1 µM) at the following receptors: R1-adrenergic ([3H]prazosin),
R2-adrenergic ([3H]clonidine), â-adrenergic ([3H]dihydroalprenolol),
benzodiazepine ([3H]flunitrazepam), muscarinic cholinergic ([3H]ox-
otremorine-M), D1 dopamine ([3H]SCH 23990), D4 dopamine
([3H]spiperone), opioid ([3H]etorphine), and 5HT2A ([3H]ket-
anserin). c The S-enantiomers showed low inhibition of radioligand
to regenerate the intermediate as the noncrystalline free
base. The mother liquors of the tartrate salt were
converted to the free base and reacted with D-tartaric
acid to give the enantiomeric product (3S)-trans-10 as
its hemi-D-tartrate salt; optical purity of both enanti-
omers was conveniently determined after removal of the
tert-butoxycarbonyl protecting group by HPLC on a
Chiralcel OJ column (see the Experimental Section).
d
binding (50% or less at 1 µM). (R)-4 showed 64% inhibition of
radioligand binding in initial testing at 1 µM.
quinoline (R)-17. Removal of the benzyl protecting
group by hydrogenolysis gave (R)-4 which was methy-
lated with formic acid/formaldehyde to give (R)-3.
The procedure of Scheme 1 has also been used to
prepare the enantiomeric series of compounds from the
3S-trans-enantiomer of 10. Also, by eliminating the
resolution step in Scheme 1, the racemic series of
products was prepared from bromohydrin 8; physical
properties of the enantiomeric and racemic compounds
are reported in the Experimental Section.
Compound (3R)-trans-10 and its enantiomer have also
been prepared in two steps from the bromohydrin 8. The
bromohydrin was treated with 1.0 equiv of potassium
tert-butoxide to give the racemic epoxide 9; while the
epoxide can be prepared in one step by epoxidation of
7b with acetonitrile/hydrogen peroxide,2 synthesis via
the bromohydrin is a more efficient and rapid procedure.
Epoxide 9 was separated into its enantiomers by
preparative chiral HPLC on a Regis (R,R)-Whelk-O
column. The epoxide enantiomers, higher melting than
the racemic product, were readily crystallized to optical
purity, and this purity could be confirmed by analytical
chiral HPLC. Reaction of the (1aR)-epoxide enantiomer
with methylamine gave the amino alcohol (3R)-trans-
10 as the free base (100% ee, >95% purity) which could
be used without additional purification in further work.
Discu ssion
We have previously reported the D2 dopaminergic and
5HT1A serotonergic activities of compound (R)-2 and the
racemate of 5.1,2 Since these reports, the number of
subclasses of dopamine receptors has increased to
include the dopamine D3 receptor.3,4 The receptor
binding affinities for the R-enantiomers of compounds
2-6 on the dopamine D2 and D3 and serotonin 5HT1A
receptors are shown in Table 1. These data show that
the previously reported compound (R)-2 has affinity for
the dopamine D3 receptor in addition to the activities
already reported. The imidazoquinolines (R)-3 and
(R)-4 showed very low activity in the binding assays,
while imidazoquinolinones (R)-5 and (R)-6 had good
activity. Compound (R)-6, the most active compound,
showed dopaminergic D2 activity and also had affinity
for the 5HT1A serotonin receptor subtype. Its dopam-
inergic activity was more selective for the D2 receptor
subtype (259-fold D2/D3 selectivity) than propylamine
analogue (R)-2 (14-fold selectivity) or other dopaminer-
gic standards (e.g., pergolide, lisuride, bromocriptine,
and ropinirole, 1.0-, 3.4-, 8.7-, and 2.6-fold selectivities,
respectively). As expected from previous work on
related products,1,2 the S-enantiomers of compounds
3-6 were inactive in the binding assays (Table 1).
Despite low in vitro activity (Table 1), imidazoquinolines
(R)-3 and (R)-4 displayed essentially the same activity
in vivo as their imidazoquinolinone counterparts (R)-5
With optical resolution of the products completed,
further conversion of (3R)-trans-10 to the R-enantiomers
of the target compounds 3-6 continued as outlined in
Scheme 2. The compound was reacted with triph-
enylphosphine and diethyl azodicarboxylate10 to gener-
ate the aziridine (1aS)-11. This was reduced with
hydrogen and a palladium catalyst to the amine (R)-12
which was reacted with benzyl chloride to give the
benzylamine (R)-13. This was reacted with sec-butyl-
lithium to generate the anion at the 8-position,1,2,11 and
the anion was reacted with tosyl azide to give azide (R)-
14.12 The azide was reduced with hydrogen and a
palladium catalyst to amine (R)-15. When (R)-15 was
treated with 1.0 equiv of potassium tert-butoxide in THF
at room temperature, imidazoquinolinone (R)-16 was
rapidly formed. Reductive cleavage of the benzyl-
protecting group gave (R)-6 which was formylated with
acetyl formate and then reduced with borane-methyl
sulfide to give (R)-5.
Compound (R)-15 was heated in formic acid to remove
the BOC protecting group and generate the imidazo-

Synthesis and Activities of Imidazoquinolinamines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5 641
Sch em e 2. Synthesis of Methylamine Analogues from Quinoline
and (R)-6 (Table 2, Figure 2). All of the compounds were
potent in producing hypothermia in mice when admin-
istered subcutaneously and also showed good oral
activity (Table 2). The hypothermic effects of (R)-6 could
be completely prevented by pretreatment with sulpiride,
a selective dopamine D2 antagonist, suggesting the
dopaminergic nature of this effect.13
Drug-induced turning in rats with unilateral lesions
of the substantia nigra has been used to characterize
dopamine agonists.14 In this assay, direct-acting ago-
nists produce contralateral turning, indirect agonists
(e.g., amphetamine) produce ipsilateral turning, while
antagonists (e.g., haloperidol) show no turning activity.
The compounds acted as direct agonists in this assay.
All produced intense turning at 3 and 10 mg/kg, with
maximum turning rates (120 turns/10 min) being reached
in 90-120 min and continuing for several hours. Time
courses for compounds (R)-2, (R)-3, and pergolide at 3
mg/kg are shown in Figure 1. Dose-response curves
for all compounds are shown in Figure 2. The four
methylamine compounds were essentially equipotent
and more effective than the propylamine analogue (R)-2
in this assay (Figures 1 and 2). The turning produced
by (R)-6 (3 mg) was completely antagonized by the D2
dopamine antagonist haloperidol (0.1 and 1.0 mg),
indicating the dopaminergic nature of this effect.
Pergolide, in contrast to the R-enantiomers of com-
pounds 2-6, produced intense stereotypic licking
and chewing in the lesioned rats at doses of 0.3 and 3
mg/kg.
In view of the low binding affinities of imidazoquino-
lines (R)-3 and (R)-4, it seemed possible that the
observed dopaminergic effects for these compounds
were a result of metabolism to (R)-5 and (R)-6. N-
Dealkylation reactions are frequently encountered in the
metabolism of biogenic amines,15 and examples of
metabolic oxidation reactions which convert drugs with
an imidazole ring into the corresponding imidazolone

642 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5
Heier et al.
have been reported (e.g., for azathioprine16 and parax-
anthine17). Similar conversion of (R)-3 and (R)-4 to the
corresponding imidazoquinolinones could result in meta-
bolic activation. To investigate this idea, the metabo-
lism of (R)-3 was studied in vitro following incubation
for 2 h in the presence of monkey hepatic S9 mi-
crosomes. The incubate was examined by thermospray
LC/MS/MS to determine levels of parent drug and
metabolites (Figure 3, Table 3). Major amounts of the
primary metabolites (R)-4 and (R)-5 were detected
together with smaller amounts of the secondary me-
tabolite (R)-6, confirming the metabolic pathway out-
lined in Scheme 1; no other metabolites were identified
in this study.
To determine the applicability of these data across
species, (R)-3 was incubated with hepatic microsomal
fractions from rat, mouse, rabbit, dog, mini pig, monkey,
and human; human liver slices were also included in
the study. A large degree of species variation in the
extent of production of the imidazoquinolones (R)-5 and
(R)-6 was found, with microsomes from mouse, monkey,
and rabbit being the most efficient in effecting this
conversion (Table 3). Compound (R)-6 appeared to be
the final metabolite in all cases. With the exception of
human and dog, oxidation of (R)-3 to (R)-5 dominated
over its N-dealkylation to (R)-4. Metabolism of (R)-3
to the imidazoquinolinones (R)-5 and (R)-6 was particu-
larly rapid in mouse. By extrapolation of these results
to the in vivo situation, it is probable that the metabo-
lites account for much of the in vivo hypothermia
observed in this species.
Rat S9 microsomes did not appear particularly effec-
tive in metabolizing (R)-3 to imidazoquinolinone prod-
ucts. Studies with rat hepatocytes supported this
observation. The potential of compounds to undergo
hepatic metabolism was investigated in vitro using rat
hepatocytes as outlined in an earlier publication.18 The
metabolic stability (intrinsic clearance) of (R)-3 was
comparable to that of U-93385, a drug standard which
has been reported to have good oral bioavailability in
the rat19 (Table 2). While (R)-3 was less stable than
(R)-6 (stability relative to U-93385 ) 3.1), it had
comparable stability to compounds (R)-4 and (R)-5 in
this assay. It is possible that metabolism of (R)-3 by
the rat in vivo will be slow and that the good dopam-
inergic activity (turning) observed for this compound in
this species arises from an as yet undetermined mech-
anism. However, we feel that metabolism to the active
imidazoquinolones (R)-5 and (R)-6 must remain the
most likely explanation for the biological activity of (R)-3
in vivo.
In summary, we have described a series of four
methylamine analogues, all of which show dopaminergic
activity in animal tests. Of these compounds, only the
imidazoquinolinones (R)-5 and (R)-6 are active in in
vitro receptor binding assays. Evidence is presented to
rationalize these data on the basis of drug metabolism.
These compounds are more selective for the D2 subtype
of the dopamine receptor when compared to previously
described propylamine analogues (R)-1 and (R)-2 and
other published dopamine agonists.
Exp er im en ta l Section
Thin layer chromatographic (TLC) analysis was done on 2.5
× 10 cm glass plates precoated with a 250 µm layer of silica
gel GF (Analtec) with mixtures of ethyl acetate/hexane (E/H)
or methanol/chloroform (M/C) being used to develop the plates.
Developed zones were visualized by UV light or by using
iodine. HPLC purity of products was determined with a setup
consisting of two Waters 6000A pumps, a Waters 660 gradient
programmer, a Waters 486 variable wavelength UV detector
set at 215 nM, a Hewlett-Packard 3390A recorder, and a
Waters C18 reverse-phase column (25 cm × 0.45 cm i.d.).
Aqueous buffer was prepared by adding 5.22 g of sodium
dihydrogen phosphate and 0.76 mL of 85% phosphoric acid to
water (4 L). A linear, 15 min, solvent program from 10%
acetonitrile/aqueous buffer to 85% acetonitrile/aqueous buffer
and a flow rate of 2.0 mL/min was used for all analyses. Chiral
HPLC data were obtained with Beckman Instruments equip-
ment (Beckman 112 solvent delivery module, 165 model UV
detector set at 215 nM, 340 model organizer, and 421 model
controller), a Hewlett-Packard 3390A recorder, and Regis
Whelk-O or Chiral Technologies Chiralcel OD and OJ columns
(25 cm × 0.4 cm i.d.), with 2-propanol/hexane (IPA/H) mixtures
as the mobile phase and a flow rate of 1.0 mL/min. NMR
spectra were run on a Bruker 300 MHz instrument. Elemental
analyses were obtained through the Structural, Analytical and
Medicinal Chemistry Department of Pharmacia and Upjohn,
Inc.
ter t-Bu tyl 1(2H)-Qu in olin eca r boxyla te (7b). A solution
of diisobutylaluminum hydride (3.83 L, 1.0 M in toluene) was
added dropwise over a 1.5 h period to a stirred solution of
quinoline (450 g, 3.48 mol) in THF (3.35 L) while maintaining
the reaction temperature below 45 °C. The crude 1,2-dihyd-
roquinoline (7a ) thus obtained was immediately converted to
its tert-butoxycarbonyl derivative. Triethylamine (535 mL)
was added, and the reaction mixture was stirred at ambient
temperature for 1 h. Di-tert-butyl dicarbonate (950.5 g, 4.36
mol) in toluene (2 L) was added dropwise over a 40 min period
while maintaining the reaction mixture below 45 °C. After
stirring for 1 h, the reaction mixture was cooled to 0-5 °C,
and hydrochloric acid (11.5 L of 1 N) was added. The reaction
mixture was extracted with ethyl acetate (3 × 5 L) and
concentrated to an oil. This was dissolved in THF (4.9 L),
1-methylpiperazine (84.5 g, 0.84 mol) was added, and the
reaction mixture was stirred for 1 h to destroy excess di-tert-
butyl dicarbonate. Evaporation of the solvent gave 1135 g of
material which was chromatographed on 48 kg of silica gel
Despite its apparent activity in vivo, (R)-3 would not
appear to be a suitable candidate for drug development,
either as a drug or as a prodrug, in view of its potential
for extensive and variable cross-species metabolism.
Compounds (R)-4 and (R)-5 would also appear to be less
suitable than (R)-6, a compound which shows better in
vitro activity and equivalent in vivo potency. For these
reasons, we have chosen (R)-6 (U-95666) from this series
of compounds for further development. The compound
shows good metabolic stability in rat hepatocytes (Table
2). Its potent dopamine agonist activity and high
selectivity for the dopamine D2 receptor subtype suggest
that it may be useful as a drug for the treatment of
Parkinson’s disease. Additional results with this com-
pound will be described in forthcoming publications.
with 2% ethyl acetate/hexane as eluant to give 460 g (57.2%)
1
of 7b as a yellow oil: HPLC tR
δ 1.52 (s, 9 H), 4.36 (d of d, 2 H), 5.98 (d of t, 1 H), 6.48 (br d,
) 11.5 min; H NMR (CDCl3)
1 H), 7.03 (m, 3 H), 7.17 (m, 1 H), 7.55 (d of d, 1 H).
ter t-Bu tyl tr a n s-3-Br om o-4-h yd r oxy-3,4-dih yd r o-1(2H)-
qu in olin eca r boxyla te (8). N-Bromosuccinimide (195 g, 1.09
mol) was added over a 0.5 h period to a stirred solution of 7b
(169 g, 0.73 mol) in dimethyl sulfoxide (2.7 L) and water (27
g) while maintaining the reaction temperature at 20-30 °C.
After 10 min, saturated sodium bicarbonate solution (1.25 L)
was added and the product was extracted into 15% ethyl
acetate/hexane (5 × 2.5 L). The organic layers were combined
and evaporated to give 262 g of crude product which was
chromatographed on 30 kg of silica gel with 10% ethyl acetate/
heptane as eluant to give 160 g (65%) of 8 which was
crystallized from ethyl acetate/hexane. A sample was recrys-
tallized from ethyl acetate/hexane for analysis: mp 117-119

Synthesis and Activities of Imidazoquinolinamines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5 643
Ta ble 2. Mouse Hypothermia and Stabilities for Imidazoquinolines and Imidazoquinolinones
sca orala
compd
ED50 (mg/K)
max temp dec (°C)
ED50 (mg/K)
max temp dec (°C)
relative stabilityb
(R)-3
(S)-3
(R)-4
(R)-5
(S)-5
(R)-6
(R)-2
0.073 (0.019-0.028)
17.3 (17.3-17.3)
0.31 (0.16-0.61)
0.41 (0.23-0.74)
17.3 (17.3-17.3)
0.73 (0.27-2.0)
10.5
4.4
12.4
11.0
5.5
1.73 (1.73-1.73)
ntc
0.73 (0.31-1.7)
1.30 (0.76-2.2)
nt
0.41 (0.23-0.74)
0.41 (0.23-0.74)
11.6
nt
11.4
10.2
nt
1.0
nt
0.86
0.71
nt
3.11
2.53
11.4
9.4
9.8
9.6
0.010 (0.0023-0.042)
a
b
ED50 values with 95% confidence limits are shown. Stabilities are relative to U-93385 [(3aR)-cis-2,3,3a,4,5,9b-hexahydro-3-propyl-
1H-benz[e]indole-9-carboxamide], a compound whose biology is described in refs 18 and 19. c nt ) not tested.
F igu r e 3. Mass chromatogram of 120 min monkey S9 liver
microsome incubation of compound (R)-3. Units on right-hand
axis are ion current in exponential notation, while those on
the left are arbitrary. RIC refers to reconstituted ion current
over the whole scan range.
F igu r e 1. Turning activity for several compounds at 3 mg/
kg sc. Time course from 0 to 360 min is shown together with
means (with standard errors) for each 10 min interval for the
treatment groups (n ) 6): (0) vehicle, (1) pergolide, (b) (R)-
2, and ([) (R)-3.
Ta ble 3. Extent of Metabolism of (R)-3 in Various Speciesa
specimen
(R)-3
(R)-4
(R)-5
(R)-6
rat S9
rabbit S9
dog S9
mini pig S9
monkey S9
mouse S9
human microsomes
human liver slices (male)
++++a
+
++++
++++
+++
++++
++++
++++
+
+
trace
+
+
++++
trace
+
++
+
++
+
++++
++++
trace
++
+
+
+
+++
+
a
++++, largest observed peak; +++, 50-80% of largest
observed peak; ++, 20-50% of largest observed peak; +, <20% of
largest observed peak.
tr a n s-1,2,3,4-t et r a h ydr o-3-h ydr oxy-N-m et h yl-4-qu in oli-
namine whose chiral purity was determined on a Chiralcel OJ
column (30% IPA/H, tR ) 12.1 min).
A suspension of the above salt (70.7 g, 0.2 mol) in ether (500
mL) and 4 N sodium hydroxide solution (50 mL, 0.2 mol) was
stirred until conversion of the salt to the ether-soluble free
base was complete (40 min). The ether was separated, and
the aqueous phase was extracted with ether (100 mL). The
oil obtained after evaporation of the ether was dissolved in
ethyl acetate/toluene and re-evaporated to remove residual
water; the 66.0 g of (3R)-trans-10 thus obtained was used
without further purification in subsequent reactions: HPLC
F igu r e 2. Turning activity for several compounds; area under
the curve (AUC) from 0 to 210 min after dosing.
°C; HPLC tR ) 10.2 min; 1H NMR (CDCl3) δ 1.54 (s, 9 H), 4.00
(m, 1 H), 4.34 (m, 2 H), 4.84 (m, 1 H), 7.14 (d of d, 1 H), 7.30
(d of d of d, 1 H), 7.44 (d of d, 1 H), 7.73 (d of d, 1 H). Anal.
(C14H18BrNO3) C, H, N, Br.
1
tR ) 6.7 min; H NMR (CDCl3) δ 1.52 (s, 9 H), 2.1 (br s, 1 H),
2.5 (s, 3 H), 3.49 (d, 1 H), 3.71-3.91 (m, 3 H), 7.05 (m, 1 H),
7.23 (m, 2 H), 7.73 (d, 1 H).
ter t-Bu tyl (3R)-tr a n s-3,4-Dih yd r o-3-h yd r oxy-4-(m eth y-
la m in o)-1(2H)-qu in olin eca r boxyla te [(3R)-tr a n s-10].
A
solution of 8 (328 g, 1 mol) in ethanol (300 mL) and methy-
lamine (400 mL of 40% in water, 4 mol) was refluxed for 3 h.
The bulk of the solvents was removed under reduced pressure,
and the residue was partitioned between ether (500 mL) and
4 N sodium hydroxide solution (50 mL). The ether phase was
washed with water and evaporated. The crude product was
dissolved in DMF (2 L) at 100 °C, and L-tartaric acid (35 g,
0.22 mol) in DMF (20 mL) was added. The solution was
stirred, allowed to cool to 50 °C, and filtered to give 140 g of
the hemitartrate salt of (3R)-trans-10. This was recrystallized
from DMF to give 104.8 g of product, mp 216 °C dec. Anal.
(C15H22N2O3‚0.5C4H6O6) C, H, N. A sample of the product was
dissolved in trifluoroacetic acid for 15 min to generate (3R)-
ter t-Bu tyl (3S)-tr a n s-3,4-Dih yd r o-3-h yd r oxy-4-(m eth y-
la m in o)-1(2H)-qu in olin eca r boxyla te [(3S)-tr a n s-10]. The
original mother liquors from the previous reaction were
converted to the free base and dissolved in DMF (2 L), and
D-tartaric acid (35 g, 0.22 mol) in DMF (20 mL) was added.
The solution was stirred, allowed to cool to 50 °C, and filtered
to give 134.6 g of product which was recrystallized from DMF,
mp 216 °C dec. A sample of the product was dissolved in TFA
for 15 min to generate (3S)-trans-1,2,3,4-tetrahydro-3-hydroxy-
N-methyl-4-quinolinamine whose chiral purity was determined
on a Chiralcel OJ column (30% IPA/H, tR ) 9.1 min).
ter t-Bu t yl 1a ,7b -Dih yd r ooxir en o[c]q u in olin e-3(2H )-
ca r boxyla te (9). Potassium tert-butoxide (508.3 mL of a 1.69

644 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5
Heier et al.
M solution, 0.86 mol) was added dropwise over a 30 min period
to a stirred solution of 8 (281.9 g, 0.86 mol) in THF (4.43 L)
while maintaining the reaction temperature below 30 °C.
After stirring for 30 min, the reaction mixture was washed
with saturated sodium chloride solution (2 × 750 mL). The
organic layer was evaporated, and the product was slurried
of (R)-13 as a clear oil: HPLC tR ) 10.5 min; chiral HPLC
(Chiralcel OJ column, 5% 2-propanol/hexane) tR ) 9.8 min; 1H
NMR (CDCl3) δ 1.52 (s, 9 H), 2.31 (s, 3 H), 2.79-2.88 (dd, J )
10.5, 15.7 Hz, 1 H), 2.9-3.04 (m, 2 H), 3.29-3.37 (dd, J ) 12.6
Hz, 1 H), 3.68 (d, J ) 14 Hz, 1 H), 3.69 (d, J ) 14 Hz, 1 H),
4.08-4.24 (ddd, J ) 1.5, 4.5, 9.0 Hz, 1 H), 7.0 (m, 1 H), 7.13
(m, 2 H), 7.27 (m, 5 H), 7.57 (d, J ) 8.1 Hz, 1 H). (S)-13: chiral
HPLC (Chiralcel OJ column, 5% 2-propanol/hexane) tR ) 7.5
min. p-Toluenesulfonate salt of the racemate of 13: mp 86-
89 °C. Anal. (C15H20N2O2‚C7H8O3S‚H2O) C, H, N, S.
in hexane (1 L) and filtered to give 180.2 g of product.
A
sample was recrystallized from ethyl acetate/hexane for analy-
sis: mp 83-86 °C; HPLC tR ) 7.8 min; 1H NMR (CDCl3) δ
1.48 (s, 9 H), 3.03 (d, J ) 14.3 Hz, 1 H), 3.85-3.91 (m, 2 H),
4.83 (dd, J ) 14.3 Hz, 1 H), 7.1 (m, 1 H), 7.27-7.39 (m, 3 H).
Anal. (C14H17NO3) C, H, N.
ter t-Bu tyl (R)-3-[Meth yl(ph en ylm eth yl)am in o]-8-azido-
3,4-d ih yd r o-1(2H)-qu in olin eca r boxyla te [(R)-14]. sec-Bu-
tyllithium (14.4 mL of a 1.3 M solution in cyclohexane, 18.7
mmol) was added to a stirred solution of (R)-13 (3.3 g, 9.36
mmol) in dry THF (65 mL) at -78 °C under nitrogen. After
stirring for 15 min at -78 °C, tosyl azide (4.6 g, 23.3 mmol)
was added and the solution was allowed to stir at room
temperature for 1 h. The solvent was removed under reduced
pressure, and the mixture was partitioned between ethyl
acetate and water. The organic layer was evaporated to give
6.37 g of an orange oil. This was chromatographed on silica
gel with ethyl acetate/hexane (1:19) as the initial eluant to
give 3.06 g (83%) of (R)-14 as a yellow oil which was used
P r ep a r a tion of th e En a n tiom er s of 9. The racemic
epoxide 9 (30 g) was chromatographed by loading the com-
pound in 1 g aliquots as a 100 mg/mL solution in methylene
chloride onto a Regis (R,R)-Whelk-O preparative-scale column
(5.1 cm × 25 cm) with 10% 2-propanol/hexane as the eluant.
The flow rate was maintained at 90 mL/min until the first
peak eluted [(1aR)-9, 24 min] and then increased to 120 mL/
min until the second peak of (1aS)-9 eluted. Pooling the
appropriate fractions gave 7.9 g of (1aR)-9 and 7.5 g of (1aS)-
9. The enantiomers were rechromatographed on silica gel with
20% EtOAc/hexane as the eluant to remove minor impurities
and crystallized from EtOAc (10 mL)/hexane (15 mL). (1aR)-
9: mp 105-108 °C; [R]D ) -119.1° (MeOH, c ) 1.0); chiral
HPLC [(R,R)-Whelk-O, 20% IPA/H] tR ) 9.6 min (100%). Anal.
(C14H17NO3) C, H, N. (1aS)-9: mp 106-108 °C; [R]D ) +116.6°
(MeOH, c ) 1.0); chiral HPLC [(R,R)-Whelk-O, 20% IPA/H] tR
) 12.5 min (100%). Anal. (C14H17NO3) C, H, N.
ter t-Bu tyl (1aS)-1,1a,2,7b-Tetr ah ydr o-1-m eth yl-3H-azir i-
n o[2,3-c]qu in olin e-3-car boxylate [(1aS)-11]. Diethyl azodi-
carboxylate (5.55 g, 0.032 mol) was added to a stirred solution
of (3R)-trans-10 (5.91 g, 0.021 mol) and triphenylphosphine
(8.34 g, 0.032 mol) in anhydrous THF (120 mL). After 40 min,
the solvent was removed under reduced pressure and the oil
was dissolved in EtOAc (10 mL)/hexane (50 mL). The pre-
cipitate of triphenylphosphine oxide/diethyl hydrazinodicar-
boxylate (7.68 g) was filtered off, and the filtrate was chro-
matographed on silica gel using 30% EtOAc/hexane as eluant
to give 4.62 g of product as a clear oil. A sample was
crystallized from 10% EtOAc/hexane for analysis: mp 71-73
°C; [R]D ) +48.5° (MeOH, c ) 1.0); HPLC tR ) 7.6 min; GC tR
) 4.3 min; chiral HPLC [(R,R)-Whelk-O column, 10% 2-pro-
panol/hexane] tR ) 17.0 min (100% ee); 1H NMR (CDCl3) δ
1.48 (s, 9 H), 2.32 (s, 2 H), 2.47 (s, 3 H), 2.73 (d, J ) 13.4 Hz,
1 H), 4.70 (d, J ) 13.4 Hz, 1 H), 7.05 (m, 1 H), 7.18 (m, 1 H),
7.3 (m, 2 H). Anal. (C15H20N2O2) C, H, N. (1aR)-11: obtained
as an oil; chiral HPLC [(R,R)-Whelk-O column, 10% 2-pro-
panol/hexane] tR ) 13.5 min. Racemic 11: mp 101-102 °C.
Anal. (C15H20N2O2) C, H, N.
ter t-Bu tyl (R)-3,4-Dih ydr o-3-(m eth ylam in o)-1(2H)-qu in -
olin eca r boxyla te [(R)-12]. A mixture of (1aS)-11 (4.3 g, 16.5
mmol) and 10% palladium/charcoal (0.5 g) in 100 mL of
absolute ethanol was hydrogenated until uptake of hydrogen
ceased (1 h). The solution was filtered to remove catalyst,
evaporated, and chromatographed on silica gel using chloro-
form as the initial eluant to give 3.45 g of (R)-12 as a yellow
oil: HPLC tR ) 7.7 min; 1H NMR (CDCl3) δ 1.52 (s, 9 H), 2.52
(s, 3 H), 2.61 (d of d, J ) 6.9, 15.6 Hz, 1 H), 2.99 (m, 1 H), 3.05
(d of d, J ) 5.5, 15.6 Hz, 1 H), 3.40 (d of d, J ) 7.4, 12.8 Hz,
1 H), 4.00 (d of d, J ) 3.7, 12.7 Hz, 1 H), 6.99 (m, 1 H), 7.08
(m, 2 H), 7.64 (m, 1 H). A portion of this product was converted
to the hydrochloride salt: mp 213 °C dec; [R]D ) -23.0°
(MeOH, c ) 1.0). Anal. (C15H22N2O2‚HCl) C, H, N, Cl.
Hydrochloride salt of (S)-12: mp 213 °C dec; [R]D ) +23.9°
(MeOH, c ) 1.0). Hydrochloride salt of the racemate of 12:
mp 198 °C dec. Anal. (C15H22N2O2‚HCl) C, H, N, Cl.
ter t-Bu tyl (R)-3,4-Dih yd r o-3-[m eth yl(p h en ylm eth yl)-
a m in o]-1(2H)-qu in olin eca r boxyla te [(R)-13]. Benzyl bro-
mide (2.4 g, 14 mmol) was added to a stirred mixture of (R)-
12 (3.3 g, 12.5 mmol) and sodium carbonate (3 g, 28 mmol) in
DMF (10 mL). After 1 h, the solvent was evaporated and the
material was partitioned between EtOAc and 4 N NaOH. The
layers were separated, and the organic phase was evaporated
to give 4.97 g of a yellow oil. This was chromatographed on
silica gel with 10% EtOAc/hexane as the eluant to give 3.68 g
1
without further purification: HPLC tR ) 11.1 min; H NMR
(CDCl3) δ 1.47 (s, 9 H), 2.28 (s, 3 H), 2.70-2.88 (br m, 2 H),
3.10-3.3 (br m, 1 H), 3.64 (d, 2 H), 6.94-6.98 (m, 2 H), 7.09-
7.14 (m, 1 H), 7.14-7.32 (m, 5 H).
ter t-Bu tyl (R)-8-Am in o-3-[m eth yl(ph en ylm eth yl)am in o]-
3,4-d ih yd r o-1(2H)-qu in olin eca r boxyla te [(R)-15]. A mix-
ture of (R)-14 (3.1 g, 7.9 mmol) and 10% palladium/charcoal
(0.2 g) in ethanol (100 mL) was hydrogenated for 45 min at
50 lb of initial hydrogen pressure. The solution was filtered
to remove catalyst, evaporated, and chromatographed on silica
gel using ethyl acetate/hexane (1:9) as the initial eluant to give
2.2 g (78%) of (R)-15 as a yellow oil. The product was
crystallized from hexane (7 mL) to give 1.97 g of white
crystals: mp 69-72 °C; [R]D ) +1.5° (MeOH, c ) 1); HPLC tR
) 8.5 min. Anal. (C22H29N3O2) C, H, N. (S)-15 and the
racemate of 15 were obtained as oils.
(R)-5,6-Dih yd r o-5-[m et h yl(p h en ylm et h yl)a m in o]-4H -
im id a zo[4,5,1-ij]q u in olin -2(1H)-on e [(R)-16]. Potassium
tert-butoxide (7.5 mL of a 1 M solution in THF, 7.5 mmol) was
added to a stirred solution of (R)-15 (1.84 g, 5 mmol) in THF
at room temperature. The solvent was removed under reduced
pressure, and the mixture was partitioned between ethyl
acetate and water. The organic layer was evaporated to give
a yellow oil. This was crystallized from ethanol (15 mL) to
give 1.26 g of (R)-16 as white crystals: mp 162-164 °C; [R]D
) +2.0° (MeOH, c ) 1.0); HPLC tR ) 6.1 min; 1H NMR (CDCl3)
δ 2.39 (s, 3 H), 3.02-3.04 (m, 2 H), 3.29 (br m, 1 H), 3.61-
3.68 (dd, 1 H), 3.76 (s, 2 H), 4.24-4.30 (dd, 1 H), 6.86-6.99
(m, 3 H), 7.24-7.34 (m, 5 H), 9.94 (s, 1 H); chiral HPLC [(S,S)-
Whelk-O column, 10% 2-propanol/hexane] tR ) 12.8 min. Anal.
(C18H19N3O) C, H, N. (S)-16: mp 162-165 °C; [R]D ) -1.6°
(MeOH, c ) 1.0); chiral HPLC [(R,R)-Whelk-O column, 10%
2-propanol/hexane] tR ) 7.5 min. Anal. (C18H19N3O) C, H,
N. Racemic 16: mp 208-210 °C. Anal. (C18H19N3O) C, H,
N.
(R)-5,6-Dih yd r o-N-m et h yl-N-(p h en ylm et h yl)-4H -im i-
d a zo[4,5,1-ij]qu in olin -5-a m in e [(R)-17]. A mixture of (R)-
15 (2.0 g, 5.4 mmol) and formic acid (20 mL) was heated at
110 °C for 1 h. The solvents were evaporated, and the residual
oil was partitioned between ethyl acetate and 4 N sodium
hydroxide solution. The ethyl acetate was evaporated, and the
residual oil was chromatographed on silica gel with 10% ethyl
acetate/hexane as the eluant to give 1.43 g of (R)-17 as an oil:
HPLC tR ) 4.8 min. A portion of the product was converted
to the dihydrochloride salt: mp 235-237 °C dec; [R]D ) -6.7°
(MeOH, c ) 1.0). Anal. (C18H19N3‚2HCl) C, H, N, Cl. The
free base of (S)-17 was obtained as an oil. Free base of the
racemate of 17: mp 93-96 °C. Anal. (C18H19N3) C, H, N.
(R)-5,6-Dih ydr o-N-m eth yl-4H-im idazo[4,5,1-ij]qu in olin -
5-a m in e [(R)-4]. A mixture of (R)-17 (3.4 g, 12.4 mmol) and
10% palladium/charcoal (2.0 g) in ethanol (100 mL) was
hydrogenated at 55 °C (50 lb of initial hydrogen pressure) for
18 h. The catalyst was filtered off, the solvent was removed

Synthesis and Activities of Imidazoquinolinamines
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5 645
under reduced pressure, the residual oil was dissolved in
methanol, and excess ethereal HCl was added. The precipitate
of the dihydrochloride salt of (R)-4 (2.68 g) was filtered off:
binding sites were membranes prepared from CHO-K1 cells
permanently expressing cloned mammalian receptors.21 In-
cubation of the 0.9 mL binding reaction mixtures was for 1 h
at room temperature. Reactions were stopped by vacuum
filtration using ice cold 50 mM Tris buffer containing 5 mM
MgCl2 at pH 7.4. Compounds were initially screened at 1.0
µM. Compounds showing >50% inhibition of radioligand
binding were then further evaluated to determine Ki values;
although compounds (R)-3 and (R)-4 showed low (<50%)
activity in the initial tests, Ki values were determined for these
compounds. These experiments employed 11 half-log dilutions
of drugs run in duplicate tubes. IC50 values were estimated
by fitting the data to a one-site model by nonlinear least-
squares minimization using GraphPad Prism. Ki values were
calculated according to Cheng-Prusoff.22 These are expressed
in Table 1 with 95% confidence intervals associated with three
to five determinations.
Hyp oth er m ia Assa y. Upjohn or Charles River mice (18-
22 g) were individually caged in clear plastic cages with
sawdust bedding for at least 20 min prior to testing. Rectal
temperatures (°F) were measured by using a YSI Telether-
mometer. The test compound was then administered either
by sc injection in 0.1 mL of solution or orally in 0.2 mL of
solution via an 18 gauge needle. Twenty minutes after drug
mp >280 °C dec; chiral HPLC (Chiralcel OJ , 10% IPA/H) tR
)
15.7 min. Anal. (C11H13N3‚2HCl) C, H, N, Cl. Dihydrochlo-
ride salt of (S)-4: mp >280 °C dec; chiral HPLC (Chiralcel
OJ , 10% IPA/H) tR ) 20.7 min.
(R)-5,6-Dih ydr o-N,N-dim eth yl-4H-im idazo[4,5,1-ij]qu in -
olin -5-a m in e [(R)-3]. A mixture of (R)-4 (780 mg, 4.2 mmol),
formic acid (7 mL), and 40% aqueous formaldehyde solution
(0.7 mL) was heated at 100 °C for 1 h. The solvents were
evaporated, and the residual oil was partitioned between ethyl
acetate and 4 N sodium hydroxide solution. The ethyl acetate
was evaporated, the residual oil was dissolved in methanol,
and excess ethereal HCl was added. The precipitate of the
dihydrochloride salt of (R)-3 was filtered: mp 264 °C dec; [R]D
) -6.2° (MeOH, c ) 1.0); chiral HPLC (Chiralcel OD, 30%
IPA/H) tR ) 9.8 min. Anal. (C12H15N3‚2HCl) C, H, N, Cl.
Dihydrochloride salt of (S)-3: mp 267 °C dec; [R]D ) +5.6°
(MeOH, c ) 1.0); chiral HPLC (Chiralcel OD, 30% IPA/H) tR
) 12.8 min. Anal. (C12H15N3‚2HCl) C, H, N, Cl.
(R)-5,6-Dih yd r o-5-(m eth yla m in o)-4H-im id a zo[4,5,1-ij]-
qu in olin -2(1H)-on e [(R)-6]. A mixture of (R)-16 (0.89 g, 3.0
mmol) and 10% palladium/charcoal (0.9 g) in ethanol (100 mL)
was hydrogenated (50 lb of initial hydrogen pressure) for 18
h. The catalyst was filtered off, the solvent was removed under
reduced pressure, and the residual oil was chromatographed
on silica gel with 1% methanol/chloroform as the initial eluant
to give 0.56 g of product. Crystallization from ethyl acetate
(7 mL) gave 0.48 g of (R)-6: mp 156-158 °C; [R]D ) -20.9°
(MeOH, c ) 1.0); HPLC tR ) 3.2 min. Anal. (C11H13N3O) C,
H, N. Hydrochloride salt of (R)-6: mp >310 °C; [R]D ) -30.3°
(MeOH, c ) 1.0). Anal. (C11H13N3O‚HCl) C, H, N, Cl. Maleate
salt of (R)-6: mp 216 °C dec; [R]D ) -27.3° (water, c ) 1.0).
Anal. (C11H13N3O‚C4H4O4) C, H, N. (S)-6: mp 154-157 °C;
[R]D ) +20.7° (MeOH, c ) 1.0). Anal. (C11H13N3O) C, H, N.
Racemic 6: mp 142-144 °C. Anal. (C11H13N3O) C, H, N.
(R)-5,6-Dih yd r o-5-(d im eth yla m in o)-4H-im id a zo[4,5,1-
ij]qu in olin -2(1H)-on e [(R)-5]. The hydrochloride salt of
(R)-6 (12.0 g, 0.05 mol) was dissolved in sodium hydroxide
solution (13.0 mL of 4.0 N), and the solution was stirred during
the addition of acetonitrile (50 mL). Acetyl formate (prepared
by mixing 20 g of formic acid with 36 g of acetic anhydride)
was added, and the solution was stirred for 1 h. The solution
was filtered to remove sodium chloride (2.7 g) and evaporated,
and methanol was added to destroy excess anhydride. After
1 h, the solution was again evaporated. HPLC analysis
showed complete conversion of (R)-6 into its formyl derivative
[(R)-N-(5,6-dihydro-2-oxo-4H-imidazo[4,5,1-ij]quinolin-5-yl)-N-
methylformamide (HPLC tR ) 5.5 min)]. This was dissolved
in methanol/chloroform (30 mL of 1:1) and chromatographed
on silica gel with 1% methanol/chloroform as the eluant to give
9.5 g of product as an oil. The formyl compound was dissolved
in THF (75 mL), borane-methyl sulfide (9.0 mL) was added,
and the solution was refluxed for 1 h until HPLC showed
conversion to (R)-5 (tR ) 3.5 min, 27%) and its borane adduct
(tR) 7.9 min, 73%) was complete. The solvent was removed,
methanolic hydrogen chloride (150 mL of 5 N) was added, and
the solution was refluxed for 1 h to destroy excess borane and
the borane adduct of (R)-5. The solution was evaporated: the
residual solid was triturated with 1:1 methanol/ether and
filtered to give 7.3 g of solid. This was recrystallized from
methanol/ether to give 6.4 g of (R)-5 as the hydrochloride
salt: mp 225-240 °C; [R]D ) -23.4° (MeOH, c ) 1.0); HPLC
tR ) 3.5 min. Anal. (C12H15N3O‚HCl) C, H, N, Cl. Free base
of (R)-5: mp 96-99 °C; [R]D ) -10.2° (MeOH, c ) 1.0); chiral
HPLC (Chiralcel OJ , 10% IPA/H) tR ) 8.9 min. Anal.
(C12H15N3O) C, H, N. (S)-5: mp 121-124 °C; [R]D ) +10.1°
treatment, rectal temperatures were again measured.
A
decrease of 2 °F or more was regarded as a positive hypoth-
ermic response. Drug doses started at 30 mg/kg and were
decreased by half-log increments until zero out of four mice
showed a hypothermic response. The Spearman-Karber
method23 was used to determine ED50’s and confidence inter-
vals. Mean maximum temperature drop was taken as an
index of drug efficacy and an indirect estimate of intrinsic
activity.
6-OHDA Lesion s of th e Su bsta n tia Nigr a . Male Spra-
gue-Dawley rats from Harlan (225-250 g) were pretreated
with desmethylimipramine (25 mg/kg ip) 1 h before surgery.
They were anesthetized with chloropent (2.5 mL/kg ip) and
then placed in a stereotaxic apparatus. A small hole was
drilled through the skull, and 30 gauge stainless steel tubing
was lowered to the right substantia nigra. 6-OH-dopamine‚-
HBr solution was injected into the substantia nigra at 12 µg/2
µL (8 µg of free base) in 0.9% saline/0.1% ascorbic acid at 1
µL/min. Following surgery, the scalp incision was closed with
clips, and ointment containing a local anesthetic and antibiot-
ics was applied to the incision area. Rats were returned to
home cages, and clips were removed after 1 week.
Tu r n in g Measu r em en ts. After 2 weeks, rats were screened
to assess the degree of lesion. This was done by measuring
their turning response to 0.5 mg/kg sc apomorphine‚HCl in
0.9% saline. Total turns were recorded in discrete 10 min
intervals by an automated Roto-Scan system developed in
conjunction with Omnitech Electronics, Columbus, OH. Each
rat was connected by a lightweight harness and tether to a
rotometer at the top of a clear plastic cylindrical cage. Rats
were used for experiments if they had at least 30 contralateral
turns/10 min in this screen (contralateral to the side of the
lesion). Groups (6 animals/dose) were compared statistically
by Student’s t-test and one-way analysis of variance, for
individual data points and for area under the curve (AUC).
In cu ba tion w ith Ra t Hep a tocyte. Hepatocytes from
male Sprague-Dawley rats were prepared by a modification
of the cell perfusion technique. Cell viability was determined
by trypan blue dye exclusion. Cells (5 mL of 2 × 106 cells/
mL) were suspended at pH 7.4 in phosphate-buffered saline
containing glucose (0.1%, w/v) and fetal calf serum (10%, v/v).
The test compound was dissolved in water (1 mg/mL) and
incubated with the hepatocytes at a concentration of 20 µg/
mL (free base, ca. 0.09 mM). Incubations were carried out at
37 °C for 2 h. After incubation, the flask contents were stored
at -20 °C to await analysis. The metabolic stability relative
to U-93385 was determined as described in ref 18.
(MeOH, c ) 1.0); chiral HPLC (Chiralcel OJ , 10% IPA/H) tR
)
10.6 min. Anal. (C12H15N3O) C, H, N. Racemic 5: mp 129-
131 °C. Anal. (C12H15N3O) C, H, N. The dihydrochloride salt
of the racemate is reported in ref 1.
In for m a t ion on in Vitr o Syst em s. Buffers for the
preparation of S9 microsomes were used ice cold. The centri-
fuge rotor and homogenizing vessels were stored overnight at
4 °C prior to use. Livers (mouse, rat, and rabbit) were removed
into 0.01 M phosphate-buffered saline (PBS); dog and monkey
Recep tor Bin d in g Assa ys. The agonist radioligands (all
tritiated) used were U-8617020 (D2 sites, 62 Ci/mmol, 1-2 nM),
(+)-7-OH-DPAT (D3 sites, 146 Ci/mmol, 0.9 nM), and 8-OH-
DPAT (5HT1A sites, 85 Ci/mmol, 1.2 nM). The sources of

646 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 5
Heier et al.
1960, 3249-3257. Birch, A. J .; Lehman, P. G. 1,4-Dihydro-
quinoline. J . Chem. Soc., Perkin Trans. I 1973, 2754-2759.
Natsume, M.; Kumadaki, S.; Kanda, Y.; Kiuchi, K. Reduction of
Quinoline with Sodium Hydride. Tetrahedron Lett. 1973, 2335-
2338. Forrest, T. P.; Dauphinee, G. A.; Deraniyagala, S. A. On
the Mechanism of Disproportionation Reactions of 1,2-Dihyd-
roquinoline. Can. J . Chem. 1985, 63, 412-417.
livers were thawed in the same buffer. The livers were roughly
chopped (razor blade) and homogenized in 3 vol of 0.025 M
PBS using an Ultra Turrax homogenizer (J anke & Kunkel
GmbH & Co. KG). The homogenizing vessel was encased in
ice to dissipate local heat generated by the homogenizer. The
homogenate from each liver was split between two centrifuge
tubes (25 × 89 mm polyallomer, Beckman Instruments) and
0.025 M PBS added to bring the volume of the homogenate to
about 20 mm below the top of each tube. The tubes were
capped and centrifuged in a Beckman L8-60M ultracentrifuge
at 10 500 rpm (9000g) for 20 min at 4 °C. The supernatants,
representing the S9 fraction, were frozen at -20 °C and then
stored at -80 °C prior to use. Compound (R)-3 (20 µg/mL)
was added, and after 2 h at 37 °C, the incubates were frozen
at -20 °C to await analysis.
Rat, mouse, and rabbit S9 microsomes were prepared from
fresh liver; dog and monkey liver was stored at -80 °C prior
to use. Human liver slices were supplied by IIAM (Exton, PA)
and stored at 4 °C in Belzers cold storage medium during
shipment. Human microsomes were also supplied by IIAM
and stored at -80 °C prior to use.
An a lysis of In cu ba tes. Samples were analyzed with an
HPLC setup including Millipore manual U6K injector, Waters
600MS pump, Brownlea RP8 guard column (15 × 3.2 mm),
and Zorbax RX C8 column (250 × 4.6 mm). An aliquot of the
incubation solution (100 µL, equivalent to 100 µL of the
original incubate) was injected with 10% (v/v) acetonitrile/0.1%
(v/v) heptafluorobutyric acid in Milli-Q water (Millipore). For
thermospray (TSP)/LC/MS a Finnigan MAT TSQ 70 mass
spectrometer coupled via a Finnigan TSP II interface was used.
The spectrometer was operated in the Q3 MS mode to collect
the TSP/LC/MS data and in the TSP/LC/MS/MS mode to
obtain the collisionally induced dissociation data.
(8) Under the conditions described in the Experimental Section, the
byproduct 1,4-dihydroquinoline does not react with di-tert-butyl
dicarbonate.
(9) Dalton, D. R.; Dutta, V. P.; J ones, D. C. Bromohydrin Formation
in Dimethyl Sulfoxide. J . Am. Chem. Soc. 1968, 90, 5498.
(10) (a) Carlock, J . T.; Mack, M. P. A Mild Quantitative Method for
the Synthesis of a Variety of Heterocyclic Systems. Tetrahedron
Lett. 1978, 52, 5153-5156. (b) Pfister, J . R. A One-Pot Synthesis
of Aziridines from 2-Aminoethanols. Synthesis 1984, 969.
(11) Beak, P.; Lee, W.-K. R-Lithioamine Synthetic Equivalents from
Dipole Stabilized Carbanions. The t-BOC Group as an Activator
for R-Lithiation of Carbamates. Tetrahedron Lett. 1989, 30,
1197-1200.
(12) This alkylation reaction also works well for the related 3-dipro-
pylamino compound (see ref 1) but fails when applied to the close
analogue tert-butyl 3-(dimethylamino)-1(2H)-quinolinecarboxy-
late. The dimethylamine analogue undergoes deprotonation at
the 2- and 4-positions followed by elimination of the dimethy-
lamine substituent to form olefin 7a and its 1,4-dihydro isomer
instead of undergoing deprotonation at C-8 (M. W. Moon,
unpublished results).
(13) Hypothermia produced by 10 mg of (R)-6 is completely blocked
by sulpiride (100 mg); we thank Dr. A. H. Tang for this
unpublished information.
(14) Ungerstedt, U.; Arbuthnott, G. W. Quantitative Recording of
Rotational Behavior in Rats after 6-Hydroxydopamine Lesions
of the Nigrostriatal Dopamine System. Brain Res. 1970, 24,
485-493.
(15) Rose, J .; Castagnoli, N., J r. The Metabolism of Tertiary Amines.
Med. Res. Rev. 1983, 3, 73-88.
(16) Chalmers, A. H.; Knight, P. R.; Atkinson, M. R. Conversion of
Azathioprine into Mercaptopurine and Mercaptoimidazole De-
rivatives in Vitro and During Immunosuppressive Therapy.
Aust. J . Exp. Biol. Med. Sci. 1967, 45, 681-691.
(17) Lelo, A.; Kjellen, G.; Birkett, D. J .; Miners, J . O. Paraxanthine
Metabolism in Humans: Determination of Metabolic Partial
Clearances and Effects of Allopurinol and Cimetidine. J .
Pharmacol. Exp. Ther. 1989, 248, 315-319.
(18) Haadsma-Svensson, S. R.; Svensson, K.; Duncan, J . N.; Smith,
M. W.; Lin, C.-H. C-9 and N-Substituted Analogues of cis-(3aR)-
(-)-2,3,3a,4,5,9b-Hexahydro-3-propyl-1H-benz[e]indole-9-car-
boxamide: 5HT1A Receptor Agonists with Various Degrees of
Metabolic Stability. J . Med. Chem. 1995, 38, 725-734.
(19) Lin, C.-H.; Haadsma-Svensson, S. R.; Phillips, G.; McCall, R.
B.; Piercey, M. F.; Smith, M. W.; Svensson, K.; Carlsson, A.;
Chidester, C. G.; VonVoigtlander, P. F. Synthesis and Biological
Activity of cis-(3aR)-(-)-2,3,3a,4,5,9b-Hexahydro-3-propyl-1H-
benz[e]indole-9-carboxamide: A Potent and Selective 5HT1A
Receptor Agonist with Good Oral Activity. J . Med. Chem. 1993,
36, 2208-2218.
Refer en ces
(1) Moon, M. W.; Morris, J . K.; Heier, R. F.; Chidester, C. G.;
Hoffmann, W. E.; Piercey, M. F.; Althaus, J . S.; VonVoigtlander,
P. F.; Evans, D. L.; Figur, L. M.; Lahti, R. A. Dopaminergic and
Serotonergic Activities of Imidazoquinolines and Related Com-
pounds. J . Med. Chem. 1992, 35, 1076-1092.
(2) Moon, M. W.; Morris, J . K.; Heier, R. F.; Hsi, R. S. P.; Manis,
M. O.; Royer, M. E.; Walters, R. R.; Lawson, C. F.; Smith, M.
W.; Lahti, R. A.; Piercey, M. F.; Sethy, V. H. Medicinal Chemistry
of Imidazoquinolinone Dopamine Receptor Agonists. Drug Des.
Discovery 1992, 9, 313-322.
(3) Civelli, O.; Bunzow, J . F.; Grandy, D. K. Molecular Diversity of
the Dopamine Receptors. Annu. Rev. Pharmacol. Toxicol. 1993,
32, 281-307.
(4) Sokoloff, P.; Martres, M.-P.; Giros, B.; Bouthenet, M.-L.; Schwartz,
J .-C. The Third Dopamine Receptor (D3) as a Novel Target for
Antipsychotics. Biochem. Pharmacol. 1992, 43, 659-666.
(5) A preliminary communication of this work has been made; see:
Moon, M. W.; Heier, R. F.; Duncan, J . N.; Martin, I. J .; Harris,
D. W.; Piercey, M. F.; Smith, M. W.; Nichols, N. F.; Schreur, P.
J . K. D. Division of Medicinal Chemistry, ACS National Meeting,
Chicago, IL, Aug. 20-25, 1995; Abstract 127.
(6) A brief description of the synthesis of (R)-1 and (R)-2 from
quinoline has already been presented without experimental
details (ref 2). The synthesis shown in Scheme 2 has several
improvements over that published. Preparation of the bromo-
hydrin makes isolation of the epoxide 9 unnecessary and also
provides an improved procedure for preparation of the epoxide
if isolation of this product is desired. The one-step, base-induced
cyclization of (R)-15 to (R)-16 is also a significant improvement
(20) Moon, M. W.; Hsi, R. S. P. Synthesis of (R)-5-(Di[2,3-3H2]-
propylamino)-5,6-dihydro-4H-imidazo[4,5,1-ij]-quinolin-2(1H)-
one ([3H]U-86170) and (R)-5-([2,3-3H2]Propylamino)-5,6-dihydro-
4H-imidazo[4,5,1-ij]quinolin-2(1H)-one ([3H]U-91356). J . Labelled
Compds. Radiopharm. 1992, 31, 933-943.
(21) Chio, C. L.; Hess, G. F.; Graham, R. S.; Huff, R. M. A Second
Molecular Form of D2 Dopamine Receptor in Rat and Bovine
Caudate Nucleus. Nature 1990, 343, 266. Chio, C. L.; Lajiness,
M. E.; Huff, R. M. Activation of Heterologously Expressed D3
Dopamine Receptors: Comparison With D2 Dopamine Receptors.
Mol. Pharmacol. 1994, 45, 51. Fargin, A.; Raymond, J . R.;
Lohse, M. J .; Kobilka, B. K.; Caron, M. G.; Lefkowitz, R. J . The
Genomic Clone G-21 Which Resembles a â-Adrenergic Receptor
Sequence Encodes the 5-HT1A Receptor. Nature 1988, 335, 358.
(22) Cheng, Y. C.; Prusoff, W. H. Relationship Between the Inhibition
Constant (Ki) and the Concentration of Inhibitor Which Causes
50 Percent Inhibition (IC50) of an Enzymatic Reaction. Biochem.
Pharmacol. 1973, 22, 3099.
over the three-step procedure described to effect
conversion in ref 2.
a similar
(7) Minter, D. E.; Stotter, P. L. Efficient Conversions of Quinolines
to N-(Carboalkoxy)-1,2-dihydroquinolines. J . Org. Chem. 1981,
46, 3965-3970. Katayama, H.; Ohkoshi, M.; Yasue, M. A
Convenient Preparation of N-Acyl-1,2-dihydroquinoline. Chem.
Pharm. Bull. 1980, 28, 2226-2228. Braude, E. A.; Hannah, J .;
Linstead, R. Hydrogen transfer. Part XVI. Dihydrides of
Nitrogenous Heterocycles as Hydrogen Donors. J . Chem. Soc.
(23) Finney, D. J . Spearman-Karber Method for ED50 and Confidence
Interval Determination. Statistic Methods in Biological Assay;
Hafner Publishing Co.: New York, 1952; p 524.
J M960360Q