Synthesis and pharmacological evaluation of substituted naphth[1,2,3-de]isoquinolines (dinapsoline analogues) as D1 and D2 dopamine receptor ligands

Article abstract of DOI:10.1016/S0968-0896(02)00561-8

Dinapsoline ((2); (±)-dihydroxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-de]isoquinoline) is a full D₁ dopamine agonist that also has significant D₂ receptor affinity. Based on a similar pharmacophore, dinapsoline has pharmacological sim

Full text of DOI:10.1016/S0968-0896(02)00561-8

Bioorganic & Medicinal Chemistry 11 (2003) 1451–1464
Synthesis and Pharmacological Evaluation of Substituted
Naphth[1,2,3-de]isoquinolines (Dinapsoline Analogues) as D1
and D2 Dopamine Receptor Ligands
Amjad M. Qandil,a,y Mechelle M. Lewis,b Amy Jassen,b Sarah K. Leonard,b
Richard B. Mailmanb and David E. Nicholsa,*
aDepartment of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences,
Purdue University, West Lafayette, IN 47907, USA
bNeuroscience Center, Departments of Psychiatry, Pharmacology, and Medicinal Chemistry,
University of North Carolina School of Medicine, Chapel Hill, NC 27599-7160, USA
Received 23 August 2002; accepted 8 November 2002
Abstract—Dinapsoline ((2); (Æ)-dihydroxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-de]isoquinoline) is a full D1 dopamine agonist
that also has significant D2 receptor affinity. Based on a similar pharmacophore, dinapsoline has pharmacological similarities to
dihydrexidine ((1); (Æ)-trans-10,11-dihydroxy-5,6,6a,7,8,12b-hexahydrobenzo[a]phenanthridine), the first high affinity full D1 ago-
nist. Small alkyl substitutions on the dihydrexidine backbone are known to alter markedly the D1:D2 selectivity of dihydrexidine,
and it was of interest to determine whether similar SAR exists within the dinapsoline series. This report describes the synthesis and
pharmacological evaluation of six analogues of dinapsoline: N-allyl-(3);N-n-propyl- (4); 6-methyl- (5); 4-methyl- (6); 4-methyl-N-
allyl- (7); and 4-methyl-N-n-propyl-dinapsoline (8). As expected from earlier studies with the dihydrexidine backbone, N-allyl (3) or
N-n-propyl (4) analogues had markedly decreased D1 affinity. Unexpectedly, and unlike the dihydrexidine series, these same sub-
stituents did not markedly increase D2 affinity. The addition of a methyl group to position 6 (5) increased D1:D2 selectivity, but less
markedly than did the analogous 2-methyl substituent added to 1. Unlike the analogous 4-methyl substituent of 1, the addition of a
4-methyl-group (6) actually decreased D1 affinity without affecting D2 affinity. These data demonstrate that the dinapsoline (2)
backbone can be modified to produce dopamine agonists with novel properties. Moreover, as rigid ligands in which small sub-
stituents can cause significant changes in selectivity, they are important tools for deriving ‘differential’ SARs of the dopamine
receptor isoforms.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
resemble each other in terms of SAR profiles, and in
being coupled to inhibition of adenylate cyclase or
mechanistically related second messenger systems.9 All
five receptor subtypes belong to the G protein-coupled
receptor (GPCR) superfamily, also known as the seven-
transmembrane (7TM) superfamily. At one time the D2-
receptor subtype was thought to mediate most of the
behavioral effects of dopamine, and to be the subtype
most implicated in the etiology or therapy of dopamine-
related disease states.
The receptors for dopamine are categorized on bio-
chemical and pharmacological properties that divide
them into two families.1,2 The D1-like receptor family in
3
4
mammals includes D1A and D1B (alternatively named
5
D5 ) while the D2-like family includes the D2long
,
D2short, D3, and D4.6À8 The D1-like receptor subtypes
have high sequence homology, very similar SAR pro-
files, and in many cell types lead to stimulation of
adenylate cyclase.9 The D2-like receptor subtypes
Until recently, it had been a general consensus that the
antiparkinsonian effects of levodopa and mixed dopa-
mine agonists were due principally to stimulation of one
or more of the D2–like dopamine receptors.10 After we
had synthesized dihydrexidine (DHX 1), the first high
affinity true full D1 agonist,11,12 we demonstrated that
*Corresponding author. Tel.:+1-765-494-1461; fax:+1-765-494-1414;
e-mail: drdave@pharmacy.purdue.edu
yPresent Address: Department of Medicinal Chemistry and Pharma-
cognosy, Faculty of Pharmacy, Jordan University of Science and
Technology, PO Box 3030, Irbid 22110, Jordan.
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00561-8

1452
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
dihydrexidine could dramatically reduce MPTP-induced
parkinsonism in monkeys (tremor, motor freezing,
abnormal posture, rigidity and bradykinesia)13 contrary
to accepted dogma, but consistent with the notion that
the location of D1 receptors might make them of parti-
cular importance.14 Because DHX has only 10-fold
selectivity for D1 versus D2 receptors in brain, its D2
activity might have been responsible for these profound
antiparkinsonian effects. In acute studies, however, the
antiparkinsonian actions of DHX were completely
blocked by the D1 antagonist SCH23390, but not sig-
nificantly affected by the D2 antagonist remoxepride.2
Subsequent clinical trials demonstrated efficacy in Par-
kinson patients for both dihydrexidine and other D1 full
agonists.15,16 Although these data demonstrated the
importance of D1 activation, the co-activation of both
families of dopamine receptors may also be important.
Table 1. Apparent affinity of dihydrexidine and analogues for dopa-
mine receptors in rat striatal homogenates and cloned receptorsa
Drug
Rat striatum
(nM)
C-6 Glioma cells
(nM)
D1
D2
D2L
D3
Dihydrexidine (DHX; 1)
N-propyl-DHX
2-Methyl-DHX
4-Methyl-DHX
4-Methyl-N-propyl-DHX
6.2
180
8.7
6.6
23.4
58.1
25.7
302
18.0
9.0
1,490
434
—
674
229
170
8.70
—
85
2.08
aFrom references.20,21
ring-substituted isoquinoline-derived synthon. Each
substituted isoquinoline must then be carried through a
parallel linear synthesis. Thus, we were forced to choose
judiciously which ring substituents to examine. The data
reported by Knoerzer et al.,20 showed that a 2-methyl
group in 1 (corresponding to a 6-methyl in 2) gave
enhanced D1 activity, whereas a 4-methyl group in 1
(corresponding to a 4-methyl in 2) led to enhanced D3
potency (Table 1). Thus, we hypothesized that 6-methyl-
2 would have enhanced D1 potency and that 4-methyl-2
would possess enhanced D3 activity. This report, there-
fore, describes the synthesis and dopaminergic proper-
ties of compounds 3–8, and demonstrates that,
surprisingly, analogous substitutions do not always
cause changes in receptor affinities that are predictable
based on the analogous backbone of 1. Compounds
were examined as the racemates because significant
dopaminergic activity resides only in the illustrated
(+)-11bR enantiomer of 2.19
Since the development of DHX (1), we have continued
to study the D1-like receptors and to develop novel
ligands with full D1 agonist properties. One of the pro-
ducts of this research was the second-generation rigid
dopamine D1 agonist dinapsoline (2), first synthesized in
1996.17 More recently, the in vivo pharmacology of 2
has been explored in greater detail, where it was found
that 2 produced a robust response in the unilateral 6-
OH-DA lesioned rat model of Parkinson’s disease.18 In
addition, it now has been established that the predicted
R-(+)-enantiomer (illustrated for 2) possesses the dopa-
minergic effects of the racemate.19
Dinapsoline is of similar affinity and potency to di-
hydrexidine, but is actually less selective for the D1 ver-
sus D2 receptors. With dihydrexidine, it was of
particular interest that relatively subtle substitutions on
its backbone caused marked, often unpredicted, changes
in both the selectivity and affinity of these drugs for the
different dopamine receptor isoforms.20,21 For example,
the addition of an N-n-propyl moiety markedly
increased D2-like affinity, with specific increases in the
affinity for the D3 receptor (Table 1). Further, the
addition of a 2-methyl substituent led to a marked
increase in D1 selectivity. In view of the close structural
similarity between 1 and 2, as well as a high degree of
complementarity of 2 to a D1 dopamine full agonist
model recently developed by Mottola et al.,22 it was of
great interest to examine structural modifications of 2
that were parallel to those reported earlier by Knoerzer
et al.20 In contrast to that work, however, there is no
readily apparent divergent or combinatorial synthetic
approach that allows the synthesis of a variety of
dinapsoline (2) analogues. Rather, each ring-substituted
compound requires the specific synthesis of the requisite
Chemistry and Pharmacology
We have recently reported an improved method for the
synthesis of the key intermediate 9 in the preparation of
dinapsoline.23 Starting from 9 the syntheses of both
N-n-allyl dinapsoline (3) and N-n-propyl dinapsoline (4)
were accomplished (Scheme 1). After stirring 9 with
sulfuric acid at À20 ꢀC, the resulting intermediate 10
was treated with sodium amalgam to remove the
N-(p-toluenesulfonyl) group. This afforded amine 11
that was immediately treated with allyl bromide to give

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
1453
Scheme 1. Reagents and conditions: (a) concd H2SO4, À20 ꢀC (31%); (b) Na(Hg), Na2HPO4, MeOH (74%); (c) allyl bromide, K2CO3, acetonitrile
(94%); (d) BBr3, CH2Cl2, À78 ꢀC (89%); (e) (i) 10% Pd/C, ethanol; (ii) BBr3, CH2Cl2, À78 ꢀC (66%, two steps).
the N-allyl amine 12. Cleavage of the O-methyl groups
then was effected using boron tribromide to yield
N-allyl dinapsoline hydrobromide (3). The same N-allyl
amine 12 was catalytically reduced over hydrogen, and
then treated with boron tribromide to give N-n-propyl
dinapsoline hydrobromide (4).
ted with three equivalents of aluminum chloride at À78–
11 ꢀC, to give the isoquinolones 18 and 19. Treatment
with lithium diisopropyl amide, followed by Comins
reagent29 [1-N,N-bis(trifluoromethylsulfonyl)amino-5-
chloropyridine], furnished the triflates (20 and 21).
Initial attempts to effect a Suzuki cross-coupling
between 20 or 21 and boronic acid 13 using classical
conditions failed, but modified Suzuki conditions
[exclusion of water, stronger base (KBr instead of LiCl),
and toluene in place of 1,2-dimethoxyethane] were used
to afford the coupling products 22 and 23 in excellent
yield (Scheme 4). These protected alcohols were reduced
catalytically with hydrogen over 50% (w/w) of 10%
The retrosynthetic analysis for compounds 5–8 is shown
in Scheme 2. This route represents a convergent synth-
esis that utilizes a Suzuki cross-coupling reaction24,25
between the aryl boronic acid (13)23 and a suitable vinyl
triflate.26 The triflates were synthesized as shown in
Scheme 3. N-p-Toluenesulfonyl glycine methyl ester was
alkylated with the appropriate benzyl chloride to afford
the N-(methylbenzyl)-N-(p-toluenesulfonyl)-glycine methyl
esters 14 and 15.27 The esters then were hydrolyzed with
aqueous sodium hydroxide solution to give the acids 16
and 17 that were converted to acyl chlorides upon treat-
ment with thionyl chloride. Following the precedent of
Schlademan and Partch,28 the acid chlorides were trea-
Scheme 3. Reagents and conditions: (a) K2CO3, NaI, acetone, reflux
(14: 76%, 15: 90%); (b) 10% NaOH (16: 85%, 17: 94%); (c) SOCl2;
AlCl3, À78 !, À11 ꢀC, CH2Cl2 (18: 70%, 19: 70%); (d) (n-BuLi,
HN(iPr)2, THF, À78–0 ꢀC (20: 77%, 21: 77%).
Scheme 2. Retrosynthetic analysis for methyl-substituted analogues of 2.

1454
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
amine 28 followed by treatment with boron tribromide
afforded N-n-propyl-4-methyl dinapsoline hydro-
bromide (8).
Pharmacology
Each of the new compounds was evaluated for its affi-
nity at D1-like and D2-like receptors in rat striatal
homogenates, as described previously.12 Competition
assays were also carried out in cloned D1A, D2L, and D3
receptors expressed in C-6 glioma cells.21 In addi-
tion, functional assays were carried out both in striatal
homogenate and in C-6 glioma cells to assess the ability
of the new compounds to stimulate cAMP accumulation.
Results
Scheme 4. Reagents and conditions: (a) 13, Pd(PPh3)4, KBr, KOH,
toluene, reflux (22: 91%, 23: 92%); (b) i. Pd/C (10%), H2, ethanol, ii.
3N HCl–THF (1:1) (24: 75%, 25: 79%); (c) concd H2SO4, À20 ꢀC (26:
42%, 27: 62%).
The radioreceptor competition data for the dinapsoline
analogues and reference compounds are presented in
Table 2. As has been reported previously, dinapsoline
has high affinity for either brain, or expressed primate
D1 receptors, with a similar pattern of activity seen in
the rat brain and cloned primate D1 receptors. With the
brain receptors, dinapsoline (2) is relatively non-selec-
tive for D1 versus D2 receptors, having high affinity for
both. The addition of either N-allyl (3) or N-n-propyl
(4) substituents caused a significant decrease in D1-like
affinity, as well as a smaller decrease in D2-like affinity.
The addition of 4-methyl (6) or 6-methyl (5) sub-
stituents did not appreciably affect D2-like affinity, but
in the case of the 4-methyl (6) substituent, it actually
decreased D1-like affinity. The combination of the N-n-
propyl and 4-methyl substitutions in 8 caused a dra-
matic reduction in D1 affinity with no corresponding
decrease in D2 affinity. This result was predictable by
the additive effects of each substituent alone. Studies in
a cell line transiently transfected with the D5 receptor
palladium on carbon, followed by heating to reflux in 3
N HCl/THF (1:1) to yield the deprotected benzyl alco-
hols 24 and 25 in good yields. Intermediates 24 and 25
then were treated with concentrated sulfuric acid at
À20 ꢀC to afford the cyclized products 26 and 27.
Treatment of the cyclized intermediate with sodium
amalgam in buffered methanol effected the removal of
the N-tosyl protecting group, then demethylation with
boron tribromide gave good yields of 5 and 6 as the
hydrobromide salts (Scheme 5). To prepare the N-alkyl
compounds 7 and 8, the amine intermediate resulting
from the N-detosylation of 27 was treated with allyl
bromide in acetone in the presence of potassium car-
bonate to yield 28 that then was demethylated using
boron tribromide to give N-n-allyl-4-methyl dinapsoline
hydrobromide (7). Catalytic reduction of the N-allyl
Scheme 5. Reagents and conditions: (a) (i) Na(Hg), Na2HPO4, methanol; (ii) BBr3, CH2Cl2, À78 ꢀC (5: 61%, 6:73%, 7:62%); (b) (i) Na(Hg),
Na2HPO4, methanol; (ii) allyl bromide, K2CO3, acetonitrile; (c) (i) 10% Pd/C, ethanol; (ii) BBr3, CH2Cl2, À78 ꢀC, (59%, two steps).

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
Table 2. Affinities of dinapsoline analogues for dopamine receptors
1455
Drug
Rat striatum (nM)
D1-like D2-like
0.69
Clonal lines (nM)
D3 C-6
D1A C-6
D2L C-6
D4 CHO
D5 HEK
SCH23390
Chlorpromazine
Dihydrexidine
SKF38393
2 (DNS)
(Æ)-3> (N-allyl-DNS)
(Æ)-4 (N-propyl-DNS)
(Æ)-5 (6-Me-DNS)
(Æ)-6 (4-Me-DNS)
(Æ)-7 (4-Me-N-allyl-DNS)
(Æ)-8 (4-Me-N-propyl-DNS)
—
0.32
—
2.2b
8.6b
6.1
—
0.74
180
—
53
81
39
190
36
—
20
13
—
60
48
17
220
83
63
28
1.0
—
16
80
—
1.2
0.9
18
—
10
14
2.8
28
18
4.3
2.5
5.5
20a
12.1
210
660
11
24
—
25
110
60
57
19
5.0
23
290
1500
8.1
20
690
2400
170
9.8
32
56
200
90
140
1400
49
21
36
23
All receptor data are apparent affinities (K0.5) in nM units. All standard errors were within 15% of the values reported.
aFrom21
bFrom31
Discussion
indicated that none of these compounds had selectivity
for D1 vs. D5 receptors.
The study of these dinapsoline (2) analogues represents
an important continuation of our efforts to develop a
complete and unambiguous picture of the D1 agonist
pharmacophore. As noted earlier, these substitutions
were selected to parallel previously synthesized analo-
gues of dihydrexidine (1) in which the selectivity of the
parent drug for different dopamine receptor isoforms
was altered in unexpected ways.20,21 In the present
study, the parent molecule 2 was, as previously repor-
ted, of affinity similar to 1. The addition of either an
N-allyl (3) or an N-n-propyl (4) substituent caused a
significant decrease in D1–like affinity, consistent with
the effects seen with the dihydrexidine series and as
predicted by the D1 pharmacophore we have developed.
The pattern with the cloned D1A receptor paralleled that
of the rat brain D1 receptor, not surprisingly, because
the rodent striatum contains no detectable D1B/D5
binding sites.30 We further tested the affinity at cloned
D5 receptors and found the pattern of affinities to be
similar to that observed at D1 receptors. As noted, these
Neither the parent compound, nor the N-substituted
compounds 3, 4, and 7 had significant selectivity for D2L
versus D3 receptors. Rather, most of the new com-
pounds possessed selectivity for D3 over D2L receptors,
with compounds 4, 7, and 8 all having about 10-fold
selectivity for D3 over D2L. It is also noteworthy that at
the D4 receptor, there was relative insensitivity of the
dinapsoline structural motif to the types of substituents
that were examined.
The addition of substituents affected the functional
interaction of these drugs with D1 receptors (Figs. 1 and
2). The 6-methyl substitution 5 actually increased
potency relative to the parent, while the 4-methyl sub-
stitution somewhat decreased potency, and surprisingly
rendered the compound a partial agonist. The N-sub-
stituted compounds 3, 4 and 7 had reduced potency and
efficacy, with the 4-methyl-N-allyl DNS 7 compound
being inactive.
Figure 1. Functional effects of DNS and analogues at rat striatal D1
receptors. Some agonists are not depicted because they showed mini-
mal stimulation (see Table 3). Dopamine is included for reference.
Note the significantly greater potency and intrinsic activity of 5 versus
6. Stimulation of D1 receptors by agonists was blocked by the D1
antagonist SCH23390 (data not shown).
Figure 2. Functional effects of DNS and analogues at C-6-mD1
receptors. Dopamine and dihydrexidine are included for reference.
Consistent with data from striatal membranes, notice the unexpected
and significantly greater potency and intrinsic activity of 5 versus 6.
Stimulation of D1 receptors by agonists was blocked by the D1
antagonist SCH23390 (data not shown).

1456
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
Table 3. Functional effects (EC50s) of dinapsoline analogues at D1
and D2 dopamine receptors
Functional characterization of these compounds pro-
vided other unexpected results. Previous studies had
determined dinapsoline (2) to be a full agonist at rat
striatal23 and cloned31 D1 receptors. Characterization of
the analogues of 2 in both these preparations revealed
that a 6-methyl addition to dinapsoline (5) increased the
potency of the drug 5-fold, while a 4-methyl addition (6)
decreased the potency as expected from the affinity, but
changed the compound to a partial agonist (Figs. 1 and
2). Interestingly, while the addition of the N-n-allyl 3
decreased the potency of the molecule, the combination
of the 4-methyl and the N-n-allyl-7 rendered the mole-
cule inactive. As expected, the inclusion of the N-n-pro-
pyl-4 greatly diminished the activity at D1 receptors.
The functional activity of 8 was not determined due to
its low affinity for these receptors.
Drug
Rat striatum
D1-like (nM)
D1A C-6
glioma cells (nM)
Dopamine
Dihydrexidine
SKF38393
2 (DNS)
1000
100
100a
350
220
34
520b
55
(Æ)-3 (N-allyl-DNS)
(Æ)-4 (N-propyl-DNS)
(Æ)-5 (6-Me-DNS)
(Æ)-6 (4-Me-DNS)
(Æ)-7 (4-Me-N-allyl-DNS)
(Æ)-8 (4-Me-N-propyl-DNS)
ꢁ500
NR
170
900
NR
ND
500
1200
10
48
NR
ND
All receptor data are EC50s in nM units. All standard errors were
within 15% of the values reported.
NR=Non-responsive.
ND=Not determined.
aFrom .12
Together, these data indicate that a valid pharmaco-
phore is available to describe adequately the recognition
characteristics of both the rat and human D1 receptor.
Yet while this pharmacophore has been highly pre-
dictive, it is clear that it has certain aspects that can be
probed more accurately with rigid analogues of these
types. By contrast, the differences in substituent effects
for these two rigid backbones clearly indicate that a
different pharmacophore will need to be developed to
accommodate the interaction of these ligand families
with the D2-like receptors, including the D2 and D3
receptors. Thus, dinapsoline and dihydrexidine ana-
logues should be extremely useful for testing hypotheses
about the docking of ligands to these D2-like receptors
for several reasons: (1) they are both based on a rigid
backbone; (2) one or more members of each family have
high affinity for the receptors in question; and (3) small
substituents can cause substantial changes in affinity.
bFrom .31
results with the D1 receptors provide additional vali-
dation for the well-defined D1 agonist pharmacophore
that has been developed in our laboratories.12,20,22
Conversely, although we had expected that substituent
effects on D2-like affinity would be very similar to those
seen with the dihydrexidine backbone, there were many
unexpected findings. Knoerzer et al.20 demonstrated
that addition of a 2-methyl to 1 caused a marked
decrease in D2 affinity, whereas a methyl substituent at
the 4-position in 2 increased D2 affinity (Table 1). The
compound analogous to 2-Me-DHX, 5, thus was pre-
dicted to have lower D2 affinity and higher D1 selectivity
than the parent drug 2. In fact, whereas there was some
increase in D1:D2 selectivity, it was of smaller magni-
tude than seen with the dihydrexidine series. Similarly,
6, analogous to 4-methyl-DHX, was predicted to have
higher D2 affinity and thus lower D1:D2 selectivity than
the parent. While lower D1:D2 selectivity was found,
this was a result solely of an unexpected loss of D1 affi-
nity, a change that was paralleled by a loss of full ago-
nist character.
It also has been reported that dihydrexidine and several
of its D2-selective analogues were atypical agonists at
D2-like receptors, having a property we have termed
‘functional selectivity’ (e.g., they could bind the D2L
receptor in an expression system and cause agonist
effects at one function, and antagonist effects at
another).
Differences in how these two series of rigid dopamine
agonists interact with D2-like receptors also were seen
In preliminary studies, dinapsoline has been shown to
differ from dihydrexidine in notable ways, but like
dihydrexidine, appears to have this unique property of
functional selectivity. The preliminary data suggest that
detailed and careful functional characterization is
necessary for the D2-like properties of these drugs. It is
clear, however, that whereas useful predictions can be
made using our D1 pharmacophore, at the D2-like
receptors analogous substituents clearly lead to unex-
pected effects.
when comparing their interactions with the cloned D2L
,
D3, and D4 receptors. Dinapsoline (2) had much higher
D2L affinity than dihydrexidine (1), and slightly higher
D3 affinity (compare Table 2 with Table 1). The addi-
tion of an N-n-propyl substituent increased affinity
with both series, but somewhat more with the di-
hydrexidine than dinapsoline backbones (Tables 1 and
4 vs. 2). Similarly, compounds with the highest D3
affinity in both series were the 4-methyl-N-n-propyl
derivatives, yet both the absolute D3 affinity, and the
D3 versus D2 selectivity were higher within the dihy-
drexidine series. The pattern of affinities of the dinap-
soline series at cloned D4 receptors was similar to that
observed at D2L receptors, as seen in Table 2. These
results suggest that these rigid analogues may be useful
for developing agonist pharmacophores for these two
receptor subtypes.
Conclusion
In summary, although dihydrexidine and dinapsoline
arose from the same D1 agonist pharmacophore par-
entage, they have significant differences in both their
pharmacology, and in the pharmacology of their sub-
stituted analogues. Such differences provide avenues for

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
1457
extremely interesting uses as research tools and in clin-
ical applications. Not only can one design drugs with
different degrees of relative receptor activation, but
simple substituents may also affect pharmacokinetic
properties if they can intrinsically inhibit specific oxi-
dative or conjugative enzymes.
N-Allyl-8,9-dimethoxy-2,3,7,11b-tetrahydro-1H-naphth
[1,2,3-de]isoquinoline (12). To the amine 11 (480 mg,
1.71 mmol) in 50 mL of acetone, 236 mg (1.71 mmol) of
potassium carbonate and 0.15 mL (210 mg, 1.73 mmol)
of allyl bromide were added. The mixture was allowed
to stir under argon for 5 h and then was diluted with
100 mL of diethyl ether, filtered and concentrated. The
residue was purified with rotary chromatography (silica
gel, 10% ethylacetate–hexane under ammonia) to afford
516 mg (94%) of a clear oil that was used immediately
in the next reaction: 1H NMR (500 MHz, CDCl3): d
2.56 (t, 1H, Ar2–CH–CH2–N, J 32;=32; 11.5 Hz); 3.31
(dt, 1H, N–CH2–CH¼CH2, J 32;=6.5 Hz, 1 Hz); 3.41
(d, 1H, Ar–CH2–N, J=15 Hz); 3.49 (dd, 1H, Ar–CH2–
Ar, J 32;= 17.5 Hz, 3 Hz); 3.80 (m, 1H, Ar–CH–CH2–
Ar); 3.83 (s, 6H, OCH3); 3.86 (m, 1H, Ar–CH–CH2–
Ar); 3.98 (d, 1H, Ar–CH2–N, J=15 Hz); 4.32 (d, 1H,
Ar–CH2–Ar, J=17.5 Hz); 5.24 (dd, 1H, , N–CH2–
CH¼CH2, J=10.0 Hz, 1.5 Hz); 5.32 (dd, 1H, , N–CH2–
CH¼CH2, J=17.0 Hz, 1.5 Hz); 6.01 (m, 1H, , N-CH2–
CH¼CH2); 6.76 (d, 1H, ArH, J=8.0 Hz); 6.90 (d, 1H,
ArH, J=8.5 Hz); 6.92 (d, 1H, ArH, J=8.5 Hz); 7.09 (t,
1H, ArH, J=7.5 Hz); 7.17 (d, 1H, ArH, J=8 Hz).
CIMS m/z: 322 (M+H+, 100%). HRMS: calcd:
321.1729. Found: 321.1741.
Experimental
Chemistry
General procedures. Melting points were determined with
a Thomas-Hoover apparatus and are uncorrected. 1H
NMR spectra were obtained with a Varian VXR-5000s
(500 MHz) or a Bruker-AXR (300 MHz) NMR instrument
in CDCl3, DMSO-d6 or CD3OD and chemical shifts are
reported in d values (ppm) relative to an internal reference
of CHCl3 (d 7.24), DMSO-d5 (d 2.49) or CD3OH (d 3.3)
respectively. Chemical ionization (CI) and electron ioni-
zation (EI) mass spectra were obtained with a Finnegan
4000 quadrupole mass spectrometer. High resolution CI
and EI mass spectra were obtained on a Kratos MS 50
spectrometer and were within 0.0015 m/z, unless otherwise
noted. Ionization gas for CIMS and high resolution
CIMS was isobutane, unless otherwise noted. Elemental
analyses were performed by the microanalysis laboratory
in the Chemistry Department at Purdue University.
N-Allyl-8,9-dihydroxy-2,3,7,11b-tetrahydro-1H-naphth
[1,2,3-de]isoquinoline hydrobromide (3). The N-allyl
amine 12 (280 mg, 0.87 mmol) was dissolved in 100 mL
of dichloromethane and stirred at À78 ꢀC under argon.
To this solution 4.5 mL (4.5 mmol) of boron tribromide
(1 M solution in dichloromethane) was added and the
mixture was allowed to stir at À78 ꢀC for 2 h, then the
cooling bath was removed and the reaction mixture was
stirred at room temperature overnight. The reaction
mixture was re-cooled to À78 ꢀC, and 15 mL of anhy-
drous methanol were added. The mixture was stirred for
15 min at room temperature and then evaporated to
dryness. This was repeated three times to quench all the
boron species in the reaction mixture. The residue was
crystallized from ethanol–ethyl acetate to afford 290 mg
8,9-Dimethoxy-(N-p-toluenesulfonyl)-2,3,7,11b-tetrahydro-
1H-naphth[1,2,3-de]isoquinoline (10). To sulfuric acid
(50 mL) cooled to À20 ꢀC and stirred with a mechanical
stirrer under argon was added the alcohol 923 (500 mg,
1.1 mmol) as a solid in portions. After 30 min at À20 ꢀC
the reaction mixture was poured into 150 g of ice and
the aqueous solution was extracted with dichloro-
methane (3Â150 mL). The combined extracts were
washed with aqueous 1 N NaOH solution (2Â50 mL),
dried (MgSO4), filtered and the solvent was evaporated.
The residue was subjected to column chromatography
(silica gel, 30% ethylacetate–hexane) to afford 150 mg
(31%) of a white solid. The reaction was repeated sev-
eral times to accumulate more of the product: mp 210–
ꢀ
1
(89%) of a tan solid: mp 228 C dec; H NMR: d 3.48
(dd, 1H, Ar–CH2–Ar, J=17.5 Hz, 3 Hz); 3.546 (t, 1H,
Ar2–CH–CH2–N, J=11.5 Hz); 4.08 (m, 3H, Ar2–CH–
CH2–N and N–CH2–CH¼CH2); 4.38–4.59 (m, Ar–
CH2–Ar and Ar–CH2–N); 4.46 (m, 1H, Ar2–CH–N);
4.59 (d, 1H, Ar–CH2–N, J=15.5); 5.72 (d, 1H, N–CH2–
CH¼CH2, J=10.5 Hz); 5.75 (d, 1H, , N–CH2–
CH¼CH2, J=16.5 Hz); 6.15 (m, 1H, N–CH2–
CH¼CH2); 6.59 (d, 1H, ArH, J=8.0 Hz); 6.71 (d, 1H,
ArH, J=8.0 Hz);7.11 (d, 1H, ArH, J=8.0 Hz);7.27 (t,
1H, ArH, J=7.5 Hz);7.35 (d, 1H, ArH, J=7.5 Hz).
PDMS m/z: 294. Anal. (C19H20BrNO5) C, H, N.
212 ꢀC (lit.17 204–206 ꢀC); H NMR and chemical ioni-
1
zation MS were identical to material prepared by the
method of Ghosh et al.17
8,9-Dimethoxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-de]
isoquinoline (11). The cyclized compound 10 (1 g, 2.29
mmol) was dissolved in 100 mL of anhydrous methanol
and 11.2 g of 6% sodium-amalgam and 1.3 g (6.87
mmol) of Na2HPO4 was added. The reaction mixture
was maintained at reflux overnight, then it was decanted
into an Erlenmeyer flask and diluted with 200 mL of
water. The mixture was extracted with dichloromethane
(3Â200 mL) and the combined organic layers were dried
(Na2SO4). The solvent was filtered and evaporated, and
the residue was purified by rotary chromatography
(50% ethylacetate–hexane, under ammonia) to yield 480
mg (75%) of a yellow oil that was immediately carried
into the next reaction. The 1H NMR of the product was
identical to material prepared by the method of Ghosh
et al.17
N - n - Propyl - 8,9 - dihydroxy - 2,3,7,11b - tetrahydro - 1H-
naphth[1,2,3-de]iso-quinoline hydrobromide (4). The
N-allyl amine 12 (180 mg, 0.56 mmol) was dissolved in
100 mL of ethanol and then shaken with 18 mg of Pd/C
(10%) in a Parr hydrogenator at 50 psi for 3 h. The
catalyst was filtered and the solvent was evaporated.
The residue was placed under high vacuum for 8 h, then
it was dissolved in 100 mL CH2Cl2 and stirred under
argon at À78 ꢀC. To the solution, 2.8 mL (2.8 mmol) of

1458
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
boron tribromide (1 M solution in dichloromethane)
was added and the mixture was stirred for 2 h, then the
cooling bath was removed and the reaction was stirred
at room temperature overnight. The reaction was re-
cooled to À78 ꢀC, 15 mL of anhydrous methanol was
added, and the mixture was stirred for 15 min and then
evaporated to dryness. This was repeated three times to
quench the boron species in the reaction. The residue
was crystallized from ethanol–ethyl acetate to afford
150 mg (66%) of a tan solid: mp 228 C dec; H NMR
(300 MHz, CD3OD): d1.11 (t, 1H, N-CH2–CH2–CH3,
J=8 Hz); d1.97 (m, 1H, N–CH2–CH2–CH3, J=8.0 Hz);
3.3–3.39 (m, 4H, Ar–CH2–Ar, Ar2–CH–CH2–N and N–
CH2–CH2–CH2); 4.08 (m, 1H, Ar2–CH–CH2–N); 4.36–
4.56 (m, 3H, Ar2–CH–CH2–N, Ar–CH2–Ar and Ar–
CH2–N); 4.65 (d, 1H, Ar–CH2–N, J=17.5 Hz); 6.61 (d,
1H, ArH, J =8.5 Hz); 6.72 (d, 1H, ArH, J=8.5 Hz);
7.12 (d, 1H, ArH, J=8.0 Hz); 7.27 (t, 1H, A–H, J=8.0
Hz); 7.35 (d, 1H, ArH, J=8.0 Hz). CIMS m/z: 296
(M+H+, 100%).
N-(p-Methylbenzyl)-N-(p-toluenesulfonyl) glycine (16).
To the ester 14 (70 g; 0.2 mol), 400 mL of 10% sodium
hydroxide solution was added and the mixture was stir-
red vigorously at room temperature overnight with a
mechanical stirrer. The mixture was then acidified and
extracted with dichloromethane (3Â150 mL). The
organic layers were dried (MgSO4), filtered, and the
solvent was evaporated. The residue was crystallized
from ethyl acetate–hexane to yield 57.2 g (85%): mp
142–144 C; H NMR (500, CDCl3): d 2.26 (s, 3H, Ar–
CH3); 2.439 (s, 3H, Ar–CH3); 3.76 (s, 2H, CO–CH2–N);
4.33 (s, 2H, Ar–CH2–N); 7.07 (d, 2H, ArH, J=8 Hz);
7.12 (d, 2H, ArH, J=8.0 Hz); 7.39 (d, 2H, ArH, J=8.0
Hz); 7.73 (d, 2H, ArH, J=8.0 Hz). CIMS m/z: 334
(M+H, 40%); 178 (M+H-SO2C7H7, 19%); 157
(SO2C7H7, 18%); (105 (C8H9, 100%). Anal.
(C17H19NO4S) C, H, N.
ꢀ
ꢀ
1
1
N-(o-Methylbenzyl)-N-(p-toluenesulfonyl) glycine (17).
To the ester 15 (70 g; 0.2 mol), 400 mL of 10%
sodium hydroxide solution was added and the mixture
was stirred vigorously overnight at room temperature
with a mechanical stirrer. The mixture was then acid-
ified and extracted with dichloromethane (3Â150 mL).
The organic layers were dried (MgSO4), filtered, and
the solvent was evaporated. The residue was crystal-
lized from ethyl acetate–hexane to yield 59.9 g (94%):
N-(p-Methylbenzyl)-N-(p-toluenesulfonyl) glycine methyl
ester (14). N-(p-Toluene-sulfonyl) glycine methyl ester
(80 g; 0.329 mol) was dissolved in 1000 mL of acetone.
The solution was placed under argon and 2.46 g (5
mol%) of sodium iodide, and 45.47 g (0.329 mol) of
potassium carbonate were added. After 30 min at reflux,
52.3 mL (55.49 g, 0.395 mol) of 4-methylbenzyl chloride
was added. Reflux was maintained for 6 h, and then the
mixture was cooled, filtered, and concentrated. The resi-
due was dissolved in 300 mL of dichloromethane and
washed extensively with water. The organic layer was
then dried (MgSO4), filtered and evaporated. The residue
was crystallized from ethyl acetate–hexane to yield 87.3 g
mp 116–118 ꢀC; H NMR (500, CDCl3): d 2.28 (s, 3H,
1
Ar–CH3); d 2.41 (s, 3H, Ar–CH3); d 3.82 (s, 2H, CO–
CH2–N); 4.46 (s, 2H, Ar–CH2–N); 7.06 (d, 1H, ArH,
J=8 Hz); 7.12 (t, 1H, ArH, J=7.5 Hz); 7.136 (d, 2H,
ArH, J=7.5 Hz); 7.19 (d, 2H, Ar-H, J=8 Hz); 7.296
(d, 2H, ArH, J=8.5 Hz); 7.73(d, 2H, ArH, J=8.5
Hz); 9.1 (bs, 1H, COOH). CIMS m/z: 334 (M+H,
63%); 178 (M+H–SO2C7H7, 19%); 157 (SO2C7H7,
100%); (105 (C8H9, 87%). Anal. (C17H19NO4S) C, H,
N.
(76%): mp 62–64 ꢀC; H NMR (500, CDCl3): d 2.30 (s,
1
3H, Ar–CH3); d 2.42 (s, 3H, Ar–CH3); d 3.52 (s, 3H, O–
CH3); d 3.88 (s, 2H, CO–CH2–N); d 4.42 (s, 2H, Ar–
CH2–N); d 7.09 (s, 4H, ArH); d 7.31 (d, 2H, ArH, J=8
Hz); d7.75 (d, 2H, ArH, J=8 Hz). CIMS m/z: 348,
(M+H, 50%); 192 (M+H–SO2C7H7, 22%), 105,
(C8H9, 100%). Anal. (C18H21NO4S) C, H, N.
6-Methyl-N-(p-toluenesulfonyl)-1,2,3,4-tetrahydroiso-
quinol-4-one (18). The acid 16 (62 g; 0.186 mol) was
dissolved in 35 mL of thionyl chloride under an atmo-
sphere of argon. After dissolution was complete the
excess thionyl chloride was distilled off under vacuum.
The residue was triturated with 500 mL of hexane. The
solid residue was then dissolved in 200 mL of dry di-
chloromethane and the solution was added slowly to a
stirred suspension of 74.4 g (0.558 mol) of aluminum
chloride in 200 mL of dry dichloromethane cooled to
À78 ꢀC under argon. The mixture was allowed to warm
to À11 ꢀC over 4 h and it was then carefully added to
2.0 L of an HCl–ice mixture and stirred for 30 min. The
layers was separated, and the aqueous layer was extrac-
ted with dichloromethane (3Â300 mL). The organic
layers then were washed with water, and dried (MgSO4).
The solvent was filtered, evaporated, and the residue
was crystallized from dichloromethane–diethyl ether to
yield 41 g (70%): mp 150–152 ꢀC; 1H NMR (500,
CDCl3): d 2.32 (s, 3H, Ar–CH3); 2.37 (s, 3H, Ar–CH3);
3.96 (s, 2H, CO–CH2–N); 4.44 (s, 2H, Ar–CH2–N); 7.11
(d, 1H, ArH, J=8 Hz); 7.23 (d, 1H, ArH, J=8 Hz);
7.23 (d, 1H, ArH, J=8 Hz); 7.61 (d, 2H, ArH, J=8
Hz); 7.66 (d, 2H, ArH, J=8 Hz). Anal. (C17H17NO3S)
C, H, N.
N-(o-Methylbenzyl)-N-(p-toluenesulfonyl) glycine methyl
ester (15). N-(p-Toluenesulfonyl) glycine methyl ester
(70 g; 0.288 mol) was dissolved in 1000 mL of acetone.
The solution was placed under argon and then 2.15 g
(5 mol%) of sodium iodide and 39.70 g (0.288 mol)
potassium carbonate were added. After 30 min at
reflux, 45.6 mL (48.47 g, 0.344 mol) of 2-methylbenzyl
chloride was added. Reflux was maintained for 6 h, and
then the mixture was cooled, filtered, and concentrated.
The residue was dissolved in 300 mL of dichloro-
methane and washed extensively with water. The
organic layer was then dried (MgSO4), filtered and con-
centrated. The residue was crystallized from ethyl ace-
tate–hexane to yield 90.2 g (90%): mp 66–68 ꢀC; 1H
NMR (500, CDCl3): d 2.30 (s, 3H, Ar–CH3); 2.42 (s,
3H, Ar–CH3); 3.52 (s, 3H, O–CH3); 3.88 (s, 2H, CO–
CH2–N); 4.42 (s, 2H, Ar–CH2–N); 7.09 (s, 4H, ArH);
7.31 (d, 2H, ArH, J=8 Hz); 7.75 (d, 2H, ArH, J=8.0
Hz). CIMS m/z: 348, (M+H, 100%); 192 (M+H–
SO2C7H7, 22%), 105, (C8H9, 62%). Anal.
(C18H21NO4S) C, H, N.

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
1459
8-Methyl-N-(p-toluenesulfonyl)-1,2,3,4-tetrahydroiso-
quinol-4-one (19). The acid 17 (58 g; 0.174 mol) was
dissolved in 35 mL of thionyl chloride under an atmo-
sphere of argon. After dissolution was complete the
excess thionyl chloride was distilled off under vacuum.
The residue was triturated with 500 mL of hexane. The
solid residue then was dissolved in 200 mL of dry di-
chloromethane and the solution was added slowly to a
stirred suspension of 69.6 g (0.522 mol) of aluminum
chloride in 200 mL of dry dichloromethane cooled to
À78 ꢀC under argon. The mixture was allowed to warm
to À11 ꢀC over 4 h and was then carefully added to 2.0
L of an HCl–ice mixture and stirred for 30 min. The
layers was separated and the aqueous layer was extrac-
ted with dichloromethane (3Â300 mL). The organic
layers then were washed with water and dried (MgSO4).
The solvent was filtered, evaporated, and the residue
was crystallized from dichloromethane–diethyl ether to
yield 38.3 g (70%): mp 148–150 ꢀC; 1H NMR (500,
CDCl3): d 2.31 (s, 3H, Ar–CH3); 2.35 (s, 3H, Ar–CH3);
3.97 (s, 2H, CO–CH2–N); 4.43 (s, 2H, Ar–CH2–N); 7.21
(m, 4H, ArH,); 7.34 (d, 1H, ArH, J=7.5 Hz); 7.605 (d,
2H, ArH, J=8 Hz); 7.72 (d, 1H, ArH, J=8 Hz). CIMS
m/z: 316 (M+H, 100%); 160 (M+H– SO2C7H7, 36%).
Anal. (C17H17NO3S) C, H, N.
h and then a solution of 1-[N,N-bis(trifluoro-methyl-
sulfonyl)]amino-5-chloropyridine (20 g, 50.94 mmol) in
150 mL of dry THF was added slowly. The mixture was
stirred and left to warm to room temperature overnight.
The reaction was quenched with 50 mL of water and
concentrated, and then diluted with 500 mL of diethyl
ether. The organic layers were washed with cold 2 N
sodium hydroxide solution and then dried (MgSO4).
The residue after filtration and solvent evaporation was
subjected to column chromatography over neutral alu-
mina (5% ethyl acetate–hexane) to afford 16.35 g (77%)
of the triflate: mp 76–78 ꢀC; 1H NMR (500, DMSO–d6):
d 2.32 (s, 3H, Ar–CH3); 2.37 (s, 3H, Ar–CH3); 4.59 (s,
2H, Ar–CH2–N); 6.91 (s, 1H, CO(Tf)=CH–N); 7.14 (d,
1H, ArH, J=8 Hz); 7.21 (d, 1H, ArH, J=8 Hz); 7.33
(d, 1H, ArH, J=8 Hz); 7.43 (d, 2H, ArH, J=8 Hz);
7.78 (d, 2H, ArH, J=8 Hz). CIMS m/z: 448 (M+1);
292 (M–SO2C7H7); 276 (M–OSO2CF3). Anal.
(C16H16F3NO5S2) C, H, N.
4-(3,4-Dimethoxy-2-methoxymethoxymethylphenyl)-6-
methyl-(N-p-toluene-sulfonyl)-1,2-dihydroisoquinoline (22).
To a solution of triflate 20 (4 g, 8.94 mmol) in 50 mL
of
degassed
toluene,
tetrakis(triphenylpho-
sphine)palladium(0) (0.5 g, 5 mol%) was added. After 5
min, the boronic acid 13 (2.98 g, 11.62 mmol), potas-
sium bromide (3.19 g, 26.8 mmol), and potassium
hydroxide (1.5 g, 26.8 mmol) were added and the mix-
ture was heated to 90 ꢀC for 1 h. The mixture was then
allowed to cool to room temperature, it was diluted with
500 mL of toluene, and the resulting suspension was
washed with 2 N sodium hydroxide solution (3Â70 mL).
The organic layer was dried (MgSO4) and then filtered.
The residue after evaporation of the solvent was sub-
jected to column chromatography (neutral alumina,
30% ethyl acetate-hexane) to afford 4.1 g (90%) as a
white solid. An analytical sample was crystallized from
ethyl acetate–hexane: mp 116–118 ꢀC; 1H NMR
(500 MHz, DMSO-d6): d 2.08 (s, 3H, Ar–CH3); 2.36 (s,
3H, Ar–CH3); 2.95 (s, 3H, OCH3); 3.77 (s, 3H, OCH3);
3.85 (s, 3H, OCH3); 3.97 (d, 1H, Ar–CH2–O, J=9.5
Hz); 4.3 (d, 1H, O–CH2–O, J=4.0 Hz); 4.6 (m, 2H, Ar–
CH2–N and O–CH2–O); 4.64 (d, 1H, Ar–CH2–N,
J=14.0 Hz); 6.26 (s, 1H, Ar–C¼CH–N); 6.65 (s, 1H,
ArH); 6.91 (d, 1H, ArH, J=8.5 Hz); 6.98 (d, 1H, ArH,
J=7.0 Hz); 7.08 (d, 1H, ArH, J=8.5 Hz); 7.155 (2, 1H,
ArH, J=7.5 Hz); 7.41 (d, 2H, Ar–H, J=7.5 Hz); 7.75
(d, 2H, Ar–H, J=7.5 Hz). CIMS m/z: 509 (M+, 32%);
478 (M+-OCH3, 32%); 448 (M+–OCH2OCH3, 77%);
292, (M+–OCH2OCH3–SO2C7H7, 100%); 276. Anal.
(C28H31NO6S) C, H, N.
N-(p-Toluenesulfonyl)-6-methyl-4-(trifluoromethylsulfonyl-
oxy)-1,2-dihydro-isoquinoline (20). A round bottomed
flask was charged with 200 mL of freshly distilled THF
under argon. Next, 4 mL (2.89 g, 28.56 mmol) of diiso-
propylamine was added, followed by 17.8 mL of n-BuLi
(1.6 M solution in hexane). After 5 min, the mixture was
cooled to À78 ꢀC, and a solution of isoquinolone 18 (7.5
g; 23.8 mmol) in 100 mL of dry THF was added drop-
wise. The mixture was stirred for 1 h and then a solution
of 1-[N,N-bis(trifluoromethylsulfonyl)]amino-5-chlor-
opyridine (10 g, 25.47 mmol) in 100 mL of dry THF was
slowly added. The mixture was stirred, and left to warm
to room temperature overnight. The reaction was
quenched with 15 mL of water and concentrated, and
then diluted with 500 mL of diethyl ether. The organic
layers were washed with 2 N sodium hydroxide solu-
tion, dried (MgSO4), then filtered. The residue after
solvent evaporation was subjected to column chroma-
tography (neutral alumina, 5% ethyl acetate–hexane) to
afford 8.2 g (77%) of the triflate: mp 70–72 ꢀC; 1H
NMR (500, DMSO–d6): d2.32 (s, 3H, Ar–CH3); 2.37 (s,
3H, Ar–CH3); 4.59 (s, 2H, Ar–CH2–N); 6.91 (s, 1H,
CO(Tf)¼CH-N); 7.14 (d, 1H, ArH, J = 8 Hz); 7.21 (d,
1H, ArH, J=8 Hz); 7.33 (d, 1H, ArH, J=8 Hz); 7.43
(d, 2H, ArH, J=8 Hz); 7.78 (d, 2H, Ar–H, J=8 Hz).
CIMS m/z: 448 (M+1); 292 (M–SO2C7H7); 276 (M–
OSO2CF3). Anal. (C16H16F3NO5S3) C, H, N.
4-(3,4-Dimethoxy-2-methoxymethoxymethylphenyl)-8-
methyl-(N-p-toluenesulfonyl)-1,2-dihydroisoquinoline (23).
To a solution of triflate 21 (14.6 g, 32.63 mmol) in 150
mL of degassed toluene, tetrakis (triphenylphosphine)
palladium(0) (1.82 g, 5 mol%) was added. After 5 min,
the boronic acid 13 (10.86 g, 42.42 mmol), potassium
bromide (11.5 g, 96.6 mmol), and potassium hydroxide
(4.5 g, 80.2 mmol) were added and the mixture was
heated to 90 ꢀC for 1 h. The reaction mixture was then
allowed to cool to room temperature, it was diluted with
1000 mL of toluene, and the resulting suspension was
N-(p-Toluenesulfonyl)-8-methyl-4-(trifluoromethylsulfonyl-
oxy)-1,2-dihydro-isoquinoline (21). A round bottomed
flask was charged with 300 mL of freshly distilled THF
and was flushed with argon. Next, 4 mL (5.78 g, 57.12
mmol) of diisopropylamine was added, followed by 35.6
mL of n-BuLi (1.6 M solution in hexane). After 5 min,
the mixture was cooled to À78 ꢀC and a solution of
isoquinolone 19 (15 g; 47.6 mmol) in 200 mL of dry
THF was added dropwise. The mixture was stirred for 1

1460
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
washed with 2 N sodium hydroxide solution (3Â100
mL). The organic layer was dried (MgSO4) and then
filtered. The residue after evaporation of the solvent was
subjected to column chromatography (neutral alumina,
30% ethyl acetate–hexane) to afford 15.35 g (92%) of a
white solid. An analytical sample was crystallized from
ethyl acetate–hexane: mp 112–114 ꢀC; 1H NMR
(500 MHz, DMSO–d6): d 2.26 (s, 3H, Ar–CH3); 2.35 (s,
3H, Ar–CH3); 2.97 (s, 3H, OCH3); 3.76 (s, 3H, OCH3);
3.84 (s, 3H, OCH3); 3.97 (d, 1H, Ar–CH2–O, J=10.5
Hz); 4.32 (s, 1H, O–CH2–O); 4.36 (d, 1H, O–CH2–O,
J=10 Hz); d4.41 (d, 1H, Ar–CH2–N, J=14.0 Hz); 4.71
(d, 1H, Ar–CH2–N, J=14.0 Hz); 6.24 (d, 1H, ArH,
J=7.5 Hz); 6.63 (s, 1H, Ar–C¼CH–N); 6.89 (dd, 1H,
ArH, J=8.5 Hz, 1 Hz); 6.96 (t, 1H, ArH, J=7.5 Hz);
7.02 (d, 1H, ArH, J=7.5 Hz); 7.09 (dd, 1H, ArH,
J=8.5 Hz, 2 Hz); 7.4 (2, 1H, ArH, J=7.5 Hz); 7.41 (d,
2H, ArH, J=7.5 Hz); 7.75 (d, 2H, ArH, J=7.5 Hz).
CIMS m/z: 509 (M+, 18%); 478 (M+–OCH3, 32%);
448 (M+–OCH2OCH3, 95%); 292, (M+–OCH2OCH3–
SO2C7H7, 100%); 276. Anal. (C28H31NO6S) C, H, N.
acetate–hexane to afford 2.89 g (79%) as a white solid:
mp 162–164 ꢀC; H NMR (500 MHz, CDCl3): d 2.23 (s,
1
3H, Ar–CH3); 2.39 (s, 3H, Ar–CH3); d 3.23 (dd, 1H,
Ar2CH–CH2–N, J=12.0 Hz, 7.0 Hz); 3.6 (dd, 1H,
Ar2CH–CH2–N, J=12.0, 5.0 Hz); 3.81 (s, 3H, OCH3);
3.89 (s, 3H, OCH3); 4.18 (d, 1H, Ar–CH2–N, J=15.0
Hz); 4.28 (d, 1H, Ar–CH2–N, J=15.0 Hz); 4.56 (t, 1H,
Ar–CH–Ar, J=6.5 Hz); 4.71 (bs, 2H, Ar–CH2–OH);
6.56 (d, 1H, ArH, J=8.5 Hz); 6.69 (t, 1H, ArH, J=8.0
Hz); 6.72 (d, 1H, ArH, J=8.5 Hz); 6.99 (m, 2H, ArH);
7.27 (d, 2H, ArH, J=8.0 Hz); 7.65 (d, 2H, ArH, J=8.0
Hz). CIMS m/z: 458 (M+H+, 35%); 450 (M+H+–
H2O, 100%). Anal. (C26H29NO5S). C, H, N.
8,9-Dimethoxy-6-methyl-(N-p-toluenesulfonyl)-2,3,7,11b-
tetrahydro-1H-naphth-[1,2,3-de]isoquinoline (26). To 50
mL of sulfuric acid at À20 ꢀC stirred with a mechanical
stirrer was added the solid alcohol 24 (500 mg, 1.07
mmol) in portions. After 30 min at À20 ꢀC the mixture
was poured into 150 mL of ice, extracted with di-
chloromethane (3Â100 mL), and the combined extracts
were washed with aqueous 1 N NaOH solution (2Â50
mL). The organic layers were dried (MgSO4), filtered,
and the solvent was evaporated. The residue was sub-
jected to column chromatography (silica gel, 30% ethyl
acetate–hexane) to afford 202 mg (42%) as a white
solid. This reaction was repeated two times to accumu-
late more of the product: mp 192–194 ꢀC; 1H NMR
(500 MHz, CDCl3): d2.39 (s, 3H, Ar–CH3); 2.41 (s, 3H,
Ar–CH3); 2.8 (t, 1H, Ar2–CH–CH2–N,J =11.0 Hz);
3.23 (dd, 1H, Ar–CH2–Ar, J=18.0, 3.5 Hz); 3.834 (s,
3H, OCH3); 3.837 (s, 3H, OCH3); 3.88 (m, 2H, Ar2CH–
CH2–N and Ar–CH2–N); 4.48 (d, 1H, Ar–CH2–Ar,
J=18.0 Hz); 4.70 (m, 2H, Ar–CH2–N and Ar–CH–Ar);
6.78 (d, 1H, ArH, J=9.0 Hz); 6.84 (d, 1H, ArH, J=8.0
Hz); 6.88 (d, 1H, ArH, J=8.5 Hz); 7.00 (d, 1H, ArH,
J=8.0 Hz); 7.33 (d, 2H, ArH, J=8.0 Hz); 7.8 (d, 2H,
ArH, J=8.0 Hz). CIMS m/z: 450 (M+H+, 29%); 198
(C9H12O2S, 100%). Anal. (C26H29NO5S) C, H, N.
4-(3,4-Dimethoxy-2-hydroxymethylphenyl)-6-methyl-(N-
p-toluenesulfonyl)-1,2,3,4-tetrahydroisoquinoline
(24).
The enamide 22 (4 g, 7.85 mmol) was dissolved in 200
mL of ethanol containing 2 g of Pd/C (10%). The mix-
ture was shaken at 50 psi in a Parr hydrogenator for 72
h. The catalyst was removed by filtration and the sol-
vent was evaporated. The residue was taken up into 30
mL of THF, 30 mL of 3 N HCl was added, and the
reaction was maintained at 50 ꢀC for 7 h. The reaction
was then allowed to cool and was diluted with 50 mL of
water and extracted with dichloromethane (3Â100 mL).
The combined organic extracts were then dried
(MgSO4), filtered and the solvent was evaporated. The
residue was crystallized from ethyl acetate–hexane to
afford 2.75 g (75%) as a white solid: mp 138–149 ꢀC; 1H
NMR (500 MHz, CDCl3): d 2.15 (s, 3H, Ar-CH3); 2.38
(s, 3H, Ar-CH3); 3.19 (dd, 1H, Ar2CH-CH2-N, J=12.0
Hz, 7.0 Hz); 3.59 (dd, 1H, Ar2CH-CH2-N, J=12.0 Hz,
5.0 Hz); 3.82 (s, 3H, OCH3); 3.89 (s, 3H, OCH3); 4.26
(d, 1H, Ar-CH2-N, J=15.0 Hz); 4.32 (d, 1H, Ar-CH2-
N, J=15 Hz); 4.53 (t, 1H, Ar-CH-Ar, J=6.5 Hz); 4.72
(bs, 2H, Ar-CH2-OH); 6.57 (d, 1H, ArH, J=8.5 Hz);
6.64 (s, 1H, ArH); 6.73 (d, 1H, ArH, J=8.5 Hz); 6.96 (s,
2H, ArH); 7.26 (d, 2H, ArH, J=8.0 Hz); 7.63 (d, 2H,
ArH, J=8.0 Hz). CIMS m/z: 450 (M+H+ÀH2O,
100%). Anal. (C26H29NO5S) C, H, N.
8,9-Dimethoxy-4-methyl-(N-p-toluenesulfonyl)-2,3,7,11b-
tetrahydro-1H-naphth[1,2,3-de]isoquinoline (27). To 50
mL of sulfuric acid at À20 ꢀC stirred with a mechanical
stirrer was added the alcohol 25 (500 mg, 1.07 mmol) as
a solid in portions. After 30 min at À20 ꢀC the mixture
was poured into 150 mL of ice, extracted with di-
chloromethane (3Â100 mL) and the combined extracts
were washed with aqueous 1 N NaOH solution (2Â50
mL). The organic layers were dried (MgSO4), filtered,
and the solvent was evaporated. The residue was sub-
jected to column chromatography (silica gel, 30% ethyl
acetate-hexane) to afford 300 mg (62%) as a white solid.
This reaction was repeated several times to accumulate
more of the product: mp 160–162 ꢀC; 1H NMR
(500 MHz, CDCl3): d 2.19 (s, 3H, Ar–CH3); 2.44 (s, 3H,
Ar–CH3); 2.79 (t, 1H, Ar2–CH–CH2–N, J=11.5 Hz);
3.496 (dd, 1H, Ar–CH2–Ar, J=18.0 Hz, 3.5 Hz); 3.76
(d, 1H, Ar–CH2–Ar, J=16.0 Hz); 3.84 (s, 3H, OCH3);
3.85 (s, 3H, OCH3); 3.93 (m, 1H, Ar2CH–CH2–N); 4.29
(d, 1H, Ar–CH2–Ar, J=18.0 Hz); 4.73 (m, 2H, Ar2CH–
CH2–N and Ar–CH–Ar); 6.78 (d, 1H, ArH, J=8.5 Hz);
6.89 (d, 1H, ArH, J=8.0 Hz); 6.99 (d, 1H, ArH, J=7.0
Hz); 7.13 (d, 1H, ArH, J=7.5 Hz); 7.36 (d, 2H, ArH,
4-(3,4-Dimethoxy-2-hydroxymethylphenyl)-8-methyl-(N-
p-toluenesulfonyl)-1,2,3,4-tetrahydroisoquinoline
(25).
The enamide 23 (4 g, 7.85 mmol) was dissolved in 200
mL of ethanol and to the solution was added 2 g Pd/C
(10%). The mixture was shaken at 50 psi in a Parr
hydrogenator for 72 h. The catalyst was removed by
filtration and the solvent was then evaporated. The
residue was taken up into 30 mL of THF, 30 mL of 3 N
HCl was added, and the reaction was maintained at
50 ꢀC for 7 h. The reaction was then allowed to cool and
was diluted with 50 mL of water and extracted with di-
chloromethane (3Â100 mL). The combined organic
extracts were dried (MgSO4), filtered, and the solvent
was evaporated. The residue was crystallized from ethyl

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
1461
J=8.0 Hz); 7.82 (d, 2H, ArH, J=8.0 Hz). CIMS m/z:
450 (M+H+, 35%); 198 (C9H12O2S, 100%). Anal.
(C26H29NO5S) C, H, N.
cooling bath was removed, and the reaction mixture was
allowed to stir at room temperature overnight. The sus-
pension was then re-cooled to À78 ꢀC, 20 mL of an-
hydrous methanol was added, and the solution was
stirred for 15 min at room temperature. The solvents
were evaporated, and three more portions of anhydrous
methanol were added and evaporated to destroy all the
remaining BBr3. The residue was then crystallized from
ethanol–ethyl acetate to afford 260 mg (73%): mp
266 C dec; H NMR (500, CD3OD): d 2.26 (s, 3H, Ar–
CH3); 3.4 (t, 1H, Ar2–CH–CH2–N, J=11.5 Hz); 4.34
(d, 1H, Ar–CH2–Ar, J=19.0 Hz); 4.03 (m, 1H, Ar–CH–
Ar); 4.35 (m, 3H, Ar2–CH–CH2–N, Ar–CH2–Ar and
Ar–CH2–Ar); 4.43 (d, 1H, Ar–CH2–N, J=15.5 Hz);
6.58 (d, 1H, ArH, J=8.5 Hz); 6.7 (d, 1H, ArH, J=8.0
Hz); 7.1 (d, 1H, ArH, J=8.0 Hz); 7.22 (d, 1H, ArH,
J=8.0 Hz); 7.17 (d, 1H, ArH, J=8.0 Hz). PDMS m/z:
268. Anal. (C17H18BrNO2) C, H, N.
8,9-Dihydroxy-6-methyl-2,3,7,11b-tetrahydro-1H-naphth
[1,2,3-de]isoquinoline Hydrobromide (5). The cyclized
compound 26 (400 mg, 0.89 mmol) was dissolved in 100
mL of anhydrous methanol and 5.5 g of 6% sodium-
amalgam and 505 mg (3.56 mmol) of Na2HPO4 were
added. The reaction mixture was maintained at reflux
overnight and the mixture was then decanted into an
Erlenmeyer flask, diluted with 200 mL of water, and
extracted with dichloromethane (3Â200). The combined
extracts were dried (Na2SO4), and the solvent was fil-
tered and evaporated. The residue was purified with
radial chromatography (Chromatotron; silica gel, 50%
ethyl acetate–hexane, under ammonia) to yield 211 mg
(73%) of a yellow oil that was immediately used for the
next reaction. The amine (150 mg, 0.51 mmol) was dis-
solved in 50 mL of dichloromethane under argon and
the solution was cooled to À78 ꢀC. To this was added
2.6 mL (2.6 mmol) of boron tribromide (1 M solution in
dichloromethane) and the mixture was stirred for 2 h.
The cooling bath was removed and stirring was con-
tinued overnight at room temperature. The suspension
was re-cooled to À78 ꢀC, 20 mL of anhydrous methanol
was added, the solution was stirred for 30 min, then the
solvents were evaporated. Three more portions of
anhydrous methanol were added, and evaporated to
destroy all the remaining BBr3. The residue then was
crystallized from ethanol:ethyl acetate to afford 109 mg
ꢀ
1
N-Allyl-8,9-dimethoxy-4-methyl-2,3,7,11b-tetrahydro-
1H-naphth[1,2,3-de]-isoquinoline (28). To the amine
intermediate resulting from the detosylation of 27 (300
mg, 1.02 mmol) in 50 mL of acetone, 141 mg (1.02
mmol) of potassium carbonate and 0.1 mL (139 mg,
1.15 mmol) of allyl bromide were added. The reaction
mixture was allowed to stir under argon for 5 h, then it
was diluted with 100 mL of diethyl ether, filtered, and
concentrated. The residue was purified with radial
chromatography (silica gel, 10% ethyl acetate–hexane
under ammonia) to afford 320 mg (94%) of a clear oil
that was used immediately in the next reaction: 1H
NMR (300 MHz, CDCl3): d 2.19 (s, 3H, CH3); 2.56 (t,
1H, Ar2–CH–CH2–N, J=11.5 Hz); 3.39 (dt, 1H, N–
CH2–CH¼CH2, J=7.0, 1.0 Hz); 3.26 (d, 1H, Ar–CH2–
N, J=17.0 Hz); 3.49 (dd, 1H, Ar–CH2–Ar, J=18.0 Hz,
3.0 Hz); 3.82 (m, 1H, Ar2–CH–CH2–N); 3.86 (s, 6H,
OCH3); 3.91 (dd, 1H, Ar2CH–CH2–N, J=11.5 Hz, 6.0
Hz); 4.01 (d, 1H, Ar–CH2–N, J=17.0 Hz); 4.32 (d, 1H,
Ar–CH2–Ar, J=18.0 Hz); 5.28 (dd, 1H, N–CH2–
CH¼CH2, J=11 Hz, 1.5 Hz); 5.38 (dd, 1H, N–CH2–
CH¼CH2, J=17.5 Hz, 1.5 Hz); 6.06 (m, 1H, N–CH2–
CH¼CH2); 6.77 (d, 1H, ArH, J=8.0 Hz); 6.94 (d, 1H,
ArH, J=8.0 Hz); 6.98 (d, 1H, ArH, J=8.0 Hz); 7.12 (d,
1H, ArH, J=8.0 Hz). CIMS m/z: 336, (M+H+,
100%), HRMS: calcd: 335.1885. Found: 335.1876.
1
(61%): mp H NMR (500, CD3OD): d 2.42 (s, 3H, Ar–
CH3); 3.27 (dd, 1H, Ar–CH2–Ar, J=18.0 Hz, 3.0 Hz);
3.39 (t, 1H, Ar2–CH–CH2–N, J=13.0 Hz); 4.04 (m, 1H,
Ar–CH–Ar); 4.33 (dd, 1H, Ar2–CH–CH2–N, J=13.0
Hz, 12.0 Hz); 4.38 (s, 2H, Ar–CH2–N); 4.53 (d, 1H, Ar–
CH2–Ar, J=18.0 Hz); 4.81 (d, 1H, Ar–CH2–Ar,
J=17.5 Hz; 1.5 Hz); 6.6 (d, 1H, ArH, J=8.5 Hz); 6.73
(d, 1H, ArH, J=8.5 Hz); 7.03 (d, 1H, Ar-H, J=8.0 Hz);
7.16 (d, 1H, ArH, J=8.0 Hz); 7.17 (d, 1H, ArH, J=8
Hz). CIMS m/z: 268 (M+H+, 100%). Anal.
(C17H18BrNO2) C, H, N.
8,9-Dihydroxy-4-methyl-2,3,7,11b-tetrahydro-1H-naphth
[1,2,3-de]isoquinoline Hydrobromide (6). The cyclized
compound 27 (600 mg, 1.33 mmol) was dissolved in 100
mL of anhydrous methanol and to the solution was
added 8.2 g of 6% sodium-amalgam and 755 mg (5.32
mmol) of Na2HPO4 was added. The reaction was
maintained at reflux overnight and the mixture was then
decanted into an Erlenmeyer flask, diluted with 200 mL
of water, and extracted with dichloromethane (3Â200).
The combined extracts were dried (Na2SO4), the solvent
was filtered, and evaporated. The residue was purified
with radial chromatography (silica gel, 50% ethyl ace-
tate–hexane, over ammonia) to yield 298 mg (76%) of a
yellow oil that was immediately used for the next reac-
tion. This reaction was repeated two times to accumu-
late more of the amine. The amine (300 mg, 1.02 mmol)
was dissolved in 100 mL of dichloromethane, and
cooled to À78 ꢀC under argon. To this was added 5.1
mL (5.1 mmol) of boron tribromide (1 M solution in
dichloromethane). After stirring for 2 h at À78 ꢀC the
N-Allyl-8,9-dihydroxy-4-methyl-2,3,7,11b-tetrahydro-
1H-naphth[1,2,3-de]-isoquinoline Hydrobromide (7). The
N-allyl amine 30 (162 mg, 0.48 mmol) was dissolved in
40 mL of dichloromethane under argon. The solution
was cooled to À78 ꢀC, and 2.5 mL (2.5 mmol) of boron
tribromide (1 M solution in CH2Cl2) was added. The
mixture was allowed to stir for 2 h, then the cooling
bath was removed and the reaction was stirred at room
temperature overnight. The reaction mixture was then
re-cooled to À78 ꢀC, 10 mL of anhydrous methanol was
added, the mixture was stirred for 15 min, then evapo-
rated to dryness. Methanol was added and evaporated
three times to quench the boron species. The residue
was crystallized from ethanol–ethyl acetate to afford
115 mg (62%) of a tan solid: mp 258 ꢀC dec; H NMR
1
(300 MHz, CD3OD): d 1.97 (s, 3H, CH3); 3.12 (m, 2H,
Ar2–CH–CH2–N and Ar–CH2–Ar); 3.84 (m, 2H, N–

1462
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
CH2–CH¼CH2 and Ar2–CH–CH2–N); 3.97 (d, 1H, Ar–
CH2–N, J=17.0 Hz); 4.06 (d, 1H, Ar–CH2–Ar, J=18.0
Hz); 4.19 (dd, 1H, Ar2–CH–CH2–N, J=12.5 Hz, 6.0
Hz); 4.32 (d, 1H, Ar–CH2–N, J=17.0 Hz); 5.45 (dd,
1H, N–CH2–CH¼CH2, J=11 Hz); 5.51 (d, 1H, , N–
CH2–CH¼CH2, J=19.0 Hz); 5.9 (m, 1H, N–CH2–
CH¼CH2); 6.29 (d, 1H, ArH, J=9.0 Hz); 6.41 (d, 1H,
ArH, J=9.0 Hz); 6.82 (d, 1H, ArH, J=7.5 Hz); 6.92 (d,
1H, ArH, J=7.5 Hz). CIMS m/z: 308, (M+H+,
100%). Anal. (C20H22BrNO2) C, H, N.
H media containing 4500 mg/L glucose, l-glutamine,
5% fetal bovine serum supplemented with 1.9 mg/mL
puromycin (D1A) or 600 ng/mL G418 (D2L and D3).
CHO-D4 cells were a gift from NIH and grown in
Ham’s/F12 media containing 10% bovine serum, 1%
MEM and supplemented with 50,000 U penicillin, 50
mg/L streptomycin, 10 mg/L gentamycin and 0.5 mg/
mL G418. HEK cells were grown in DMEM containing
10% bovine serum. Cells were transiently transfected
using Lipofectamine according to the manufacturer’s
protocol. All cells were maintained in a humidified
incubator at 37 ꢀC with 95% O2 and 5% CO2.
N-n-Propyl-8,9-dihydroxy-4-methyl-2,3,7,11b-tetrahydro-
1H-naphth[1,2,3-de]isoquinoline Hydrobromide (8). The
N-allyl amine 30 (148 mg, 0.44 mmol) was dissolved
in 100 mL of ethanol, then shaken in a Parr hydro-
genator at 50 psi for 3 h over 15 mg of Pd/C (10%).
The catalyst was then filtered and the solvent was
evaporated. After placing the residue under high
vacuum for 8 h, it was dissolved in 100 mL of di-
chloromethane, placed under argon, then cooled to
À78 ꢀC. To the solution 2.2 mL (2.2 mmol) of boron
tribromide (1M solution in dichloromethane) was
added, and the mixture was stirred for 2 h. Then the
cooling bath was removed, and the reaction was stirred
at room temperature overnight. The reaction mixture
was re-cooled to À78 ꢀC, 15 mL of anhydrous methanol
was added, the mixture was stirred for 15 min and then
evaporated to dryness. Methanol was added, and eva-
porated three times to quench all the boron species in
the reaction. The residue was crystallized from ethanol–
ethyl acetate to afford 101 mg (59%) of a tan solid: mp
240 ꢀC dec; 1H NMR (300 MHz, CD3OD): d 1.10 (t, 1H,
N–CH2–CH2–CH3, J=8.0 Hz); 1.97 (m, 1H, N–CH2–
CH2–CH3, J=8.0 Hz); 2.23 (s, 3H, Ar–CH3); 3.38–3.52
(m, 4H, Ar–CH2–Ar, Ar2–CH–CH2–N and N–CH2–
CH2–CH2); 4.06 (m, 1H, Ar2–CH–CH2–N); d4.25 (d,
1H, Ar–CH2–N, J=17.5 Hz); 4.32 (d, 1H, Ar–CH2–Ar,
J=19.0 Hz); 4.47 (dd, 1H, Ar2–CH–CH2–N, J=13.0
Hz, 6.0 Hz); 4.59 (d, 1H, Ar–CH2–N, J=17.5 Hz); 6.57
(d, 1H, ArH, J=9.0 Hz); 6.65 (d, 1H, ArH, J=9.0 Hz);
7.07 (d, 1H, ArH, J=8.0 Hz); 7.2 (d, 1H, ArH, J=8.0
Hz). CIMS m/z: 310 (M+H+, 100%), HRMS: calcd:
309.1729. Found: 309.1721.
Membrane preparation for receptor binding studies
Cells were rinsed with hypoosmotic buffer (HOB; 1 mM
Hepes, and 2 mM EDTA, pH 7.4), and then incubated
with 7 mL HOB for 5–10 min at 4 ꢀC. Cells were then
scraped off the bottom of the flask using a rubber
policeman, collected into 50 mL tubes, and centrifuged
at 28,000 g at 4 ꢀC for 20 min. The resulting pellet was
resuspended in binding buffer (50 mM Hepes, pH
7.4), homogenized with a Brinkman Polytron on a
setting of 5, and either used immediately or stored in
1 mL aliquots at À80 ꢀC until use. Protein content
was assessed using the BCA protein assay reagent
(Pierce, Rockford, IL).
Radioreceptor binding studies
Competition binding studies were done to evaluate the
affinity of the different agonists cell lines. Membranes
were incubated with 0.3 nM [3H]-SCH23390 or 0.07 nM
[3H]spiperone in 50 mM Hepes, 4 mM MgCl2, 0.01%
ascorbic acid, pH 7.4 and increasing concentrations of
competing drug. Tubes were run in triplicate in a final
volume of 500 mL. After incubation at 37 ꢀC for 15 min,
96-tube plates were filtered rapidly through Packard 96
GF/C filters, and rinsed with 5 mL of ice-cold wash
buffer using a Packard Micro Cell Harvester (Packard
Instruments, Downers Grove, IL). Filters were allowed
to dry, and then 30 mL 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 Micro-
plate scintillation counter (Packard, Downers Grove, IL).
Pharmacology Methods
Materials
Functional assays; measurement of cAMP production
[3H]-SCH23390 was synthesized according to the
method of Wyrick et al.32 Dihydrexidine, dinapsoline
and their analogues were synthesized according to pub-
lished methods.17,20 R(+)-SCH23390 was purchased
from Research Biochemicals, Inc. (Natick, MA),
whereas chlorpromazine HCl was a gift of SmithKline
Beecham. [3H]-Spiperone was purchased from Amer-
sham (Piscataway, NJ).
Adenylate cyclase assays in rat striatal or cell
homogenates were performed as described by Watts
et al.21 A 20 mL aliquot of a 2.5 mg/mL membrane
suspension was added to each reaction tube. Base-
line values of cAMP were subtracted from the total
amount of cAMP produced for each drug condition.
Data for each drug were expressed as fmol cAMP per
sample.
Cell cultures
Data analysis
33
D1A
,
D
,
34 or D335 receptors expressed in C-6 glioma
Dose–response curves were analyzed by nonlinear
regression using an algorithm for sigmoid curves in the
curve-fitting program Prism (Graphpad, Inc; San
2L
cells (courtesy of Dr. Kim Neve, Portland VA Medical
Center) were used. The C-6 cells were grown in DMEM-

A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
1463
Diego, CA). For all of the assays incorporated into
these studies, analysis of the residuals indicated an
excellent fit with r values being greater than 0.90. Data
are expressed as apparent inhibition constants (K0.5),
and where the Hill slopes were significantly less than 1,
this was so noted.
grich, J. A.; Godinot, N.; Bertrand, L.; Yang Feng, T. L.;
Fremeau, R. T., Jr.; Caron, M. G. Proc. Natl. Acad. Sci.
U.S.A. 1991, 88, 7491.
5. 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. Nature 1991, 350, 614.
6. Monsma, F. J., Jr.; McVittie, L. D.; Gerfen, C. R.; Mahan,
L. C.; Sibley, D. R. Nature 1989, 342, 926.
7. Sokoloff, P.; Giros, B.; Martres, M. P.; Bouthenet, M. L.;
Schwartz, J. C. Nature 1990, 347, 146.
Elemental analysis data
8. Van Tol, H. H.; Bunzow, J. R.; Guan, H. C.; Sunahara, R. K.;
Seeman, P.; Niznik, H. B.; Civelli, O. Nature 1991, 350, 610.
9. Clark, D.; White, F. J. Synapse 1987, 1, 347.
10. Starr, M. S. Synapse 1995, 19, 264.
11. Lovenberg, T. W.; Brewster, W. K.; Mottola, D. M.; Lee,
R. C.; Riggs, R. M.; Nichols, D. E.; Lewis, M. H.; Mailman,
R. B. Eur. J. Pharmacol. 1989, 166, 111.
12. Brewster, W. K.; Nichols, D. E.; Riggs, R. M.; Mottola,
D. M.; Lovenberg, T. W.; Lewis, M. H.; Mailman, R. B.
J. Med. Chem. 1990, 33, 1756.
13. Taylor, J. R.; Lawrence, M. S.; Redmond, D. E., Jr.; Els-
worth, J. D.; Roth, R. H.; Nichols, D. E.; Mailman, R. B. Eur.
J. Pharmacol. 1991, 199, 389.
14. Mailman, R. B.; Schulz, D. W.; Lewis, M. H.; Staples, L.;
Rollema, H.; DeHaven, D. L. Eur. J. Pharmacol. 1984, 101,
159.
15. Blanchet, P. J.; Fang, J.; Gillespie, M.; Sabounjian, L.;
Locke, K. W.; Gammans, R.; Mouradian, M. M.; Chase, T. N.
Clin. Neuropharmacol. 1998, 21, 339.
16. Rascol, O.; Blin, O.; Thalamas, C.; Descombes, S.; Sou-
brouillard, C.; Azulay, P.; Fabre, N.; Viallet, F.; Lafnitzegger,
K.; Wright, S.; Carter, J. H.; Nutt, J. G. Ann. Neurol. 1999, 45,
736.
17. Ghosh, D.; Snyder, S. E.; Watts, V. J.; Mailman, R. B.;
Nichols, D. E. J. Med. Chem. 1996, 39, 549.
Compd C (%)
H (%)
N (%)
Calcd Found Calcd Found Calcd Found
(Æ)-3
(Æ)-4
(Æ)-5
(Æ)-6
(Æ)-7
(Æ)-8
14
15
16
17
18
19
20
21
22
23
24
25
26
60.97 60.82 5.39 5.01
60.65 60.28 5.89 6.06
58.63 58.33 5.21 5.17
58.63 58.28 5.21 5.04
61.86 61.50 5.71 5.68
61.54 59.11 6.20 5.87
62.23 62.22 6.09 6.05
62.23 62.31 6.09 6.05
61.24 61.00 5.47 5.66
61.24 61.38 5.74 5.76
64.74 64.74 5.43 5.33
64.74 64.48 5.43 5.29
48.32 48.41 3.60 3.25
48.32 48.48 3.60 3.55
65.99 65.84 6.13 6.03
65.99 65.60 6.13 6.09
66.79 66.60 6.25 6.45
66.79 66.67 6.25 6.27
69.46 69.49 6.05 5.91
69.46 69.48 6.05 6.01
3.74 3.73
3.72 3.59
4.02 3.84
4.02 3.99
3.61 3.51
3.59 3.39
4.03 4.08
4.03 4.07
4.20 4.25
4.20 4.12
4.44 4.44
4.44 4.57
3.13 3.47
3.13 3.21
2.75 2.60
2.75 2.69
3.00 2.97
3.00 2.93
3.12 3.06
3.12 3.08
18. Gulwadi, A. G.; Korpinen, C. D.; Mailman, R. B.;
Nichols, D. E.; Sit, S. Y.; Taber, M. T. J. Pharmacol. Exp.
Ther. 2001, 296, 338.
27
19. Sit, S. Y.; Xie, K.; Jacutin-Porte, S.; Taber, M. T.; Gul-
wadi, A. G.; Korpinen, C. D.; Burris, K. D.; Molski, T. F.;
Ryan, E.; Xu, C.; Wong, H.; Zhu, J.; Krishnananthan, S.;
Gao, Q.; Verdoorn, T.; Johnson, G. J. Med. Chem. 2002, 45,
3660.
20. Knoerzer, T. A.; Watts, V. J.; Nichols, D. E.; Mailman,
R. B. J. Med. Chem. 1995, 38, 3062.
21. Watts, V. J.; Lawler, C. P.; Knoerzer, T.; Mayleben,
M. A.; Neve, K. A.; Nichols, D. E.; Mailman, R. B. Eur. J.
Pharmacol. 1993, 239, 271.
22. Mottola, D. M.; Laiter, S.; Watts, V. J.; Tropsha, A.;
Wyrick, S. D.; Nichols, D. E.; Mailman, R. B. J. Med. Chem.
1996, 39, 285.
23. Qandil, A. M.; Ghosh, D.; Nichols, D. E. J. Org. Chem.
1999, 64, 1407.
Disclosure: Drs. Mailman and Nichols have a significant
financial interest in DarPharma Inc, Chapel Hill, NC,
the company that currently holds license rights to
dinapsoline and its analogues. The interpretation and dis-
cussion in this article are those of the authors alone and do
not reflect the views of DarPharma Inc, the University of
North Carolina at Chapel HIll, or Purdue University.
Acknowledgements
The authors would like to acknowledge the technical
assistance of Mr. Stan Southerland. This work was
supported in part by PHS grants MH42705 and
MH40537, training grants DA07244 and AA07573, and
by Center Grants HD01130 and MH03327.
24. Miyaura, N.; Yanagi, T.; Suzuki, A. Syn. Comm. 1981, 11,
513.
25. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
26. Wustrow, D.; Wise, L.; Turdybekov, K.; Lindeman, S.;
Koleniskov, A.; Yu, S. Synthesis 1999, 1001, 993.
27. Rozhkov, V.; Sladkov, V.; Turdybekov, K.; Lindeman, S.;
Koleniskov, A.; Yu, S. J. Org. Chem. -USSR 1989, 1162.
28. Schlademan, J.; Partch, R. J. Chem. Soc. 1972, 213.
29. Comins, D.; Dehghani, A. Tetrahedron Lett. 1992, 33,
6299.
References and Notes
1. Garau, L.; Govoni, S.; Stefanini, E.; Trabucchi, M.; Spano,
P. F. Life Sci. 1978, 23, 1745.
2. Kebabian, J. W.; Calne, D. B. Nature 1979, 277, 93.
3. Dearry, A.; Gingrich, J. A.; Falardeau, P.; Fremeau, R. T.,
Jr.; Bates, M. D.; Caron, M. G. Nature 1990, 347, 72.
4. Tiberi, M.; Jarvie, K. R.; Silvia, C.; Falardeau, P.; Gin-
30. Montague, D. M., Striplin, C. D., Overcash, J. S., Drago,
J., Mailman, R. B., and Lawler, C. P. Quantification of D1B
(D5) receptors in dopamine D1A receptor-deficient mice.
Synapse, (in press).
31. Lewis, M. M.; Watts, V. J.; Lawler, C. P.; Nichols, D. E.;
Mailman, R. B. J. Pharmacol. Exp. Ther. 1998, 286, 345.

1464
A. M. Qandil et al. / Bioorg. Med. Chem. 11 (2003) 1451–1464
32. Wyrick, S.; McDougald, D. L.; Mailman, R. B. J. Label.
Comp. Radiopharm. 1986, 23, 685.
34. Neve, K. A.; Cox, B. A.; Henningsen, R. A.; Spanoyannis,
A.; Neve, R. L. Mol. Pharmacol. 1991, 39, 733.
33. Machida, C. A.; Searles, R. P.; Nipper, V.; Brown, J. A.;
Kozell, L. B.; Neve, K. A. Mol. Pharmacol. 1992, 41, 652.
35. Cox, B. A.; Rosser, M. P.; Kozlowski, M. R.; Duwe,
K. M.; Neve, R. L.; Neve, K. A. Synapse 1995, 21, 1.