Assessment of dopamine D1 receptor affinity and efficacy of three tetracyclic conformationally-restricted analogs of SKF38393

Article abstract of DOI:10.1016/j.bmc.2011.07.057

To assess the effect of conformational mobility on receptor activity, the β-phenyl substituent of dopamine D₁ agonist ligands of the phenylbenzazepine class, (±)-6,6a,7,8,9,13b-hexahydro-5H-benzo[d] naphtho[2,1-b]azepine-11,12-diol (8), and its oxygen and sulfur bioisosteres 9 and 10, respectively, were synthesized as conformationally-restricted analogs of SKF38393, a dopamine D₁-selective partial agonist. Compounds trans-8b, 9, and 10 showed binding affinity comparable to that of SKF38393, but functionally, they displayed only very weak agonist activity. These results suggest that the conformationally-restricted structure of the analogs cannot adopt a binding orientation that is necessary for agonist activity.

Full text of DOI:10.1016/j.bmc.2011.07.057

Bioorganic & Medicinal Chemistry 19 (2011) 5420–5431
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Assessment of dopamine D1 receptor affinity and efficacy of three tetracyclic
conformationally-restricted analogs of SKF38393
⇑
Alia H. Clark, John D. McCorvy, Val J. Watts, David E. Nichols
Department of Medicinal Chemistry and Molecular Pharmacology, College of Pharmacy, Purdue University, West Lafayette, IN 47907, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
To assess the effect of conformational mobility on receptor activity, the b-phenyl substituent of dopamine
D1 agonist ligands of the phenylbenzazepine class, ( )-6,6a,7,8,9,13b-hexahydro-5H-benzo[d]naph-
tho[2,1-b]azepine-11,12-diol (8), and its oxygen and sulfur bioisosteres 9 and 10, respectively, were syn-
thesized as conformationally-restricted analogs of SKF38393, a dopamine D1-selective partial agonist.
Compounds trans-8b, 9, and 10 showed binding affinity comparable to that of SKF38393, but functionally,
they displayed only very weak agonist activity. These results suggest that the conformationally-restricted
structure of the analogs cannot adopt a binding orientation that is necessary for agonist activity.
Ó 2011 Elsevier Ltd. All rights reserved.
Received 12 May 2011
Revised 25 July 2011
Accepted 27 July 2011
Available online 3 August 2011
Keywords:
Dopamine D1
Agonist
Conformation
1. Introduction
The first pharmacological tool available to study D1 receptor
function was SKF38393 (2), which contains a benzazepine ring that
Dopamine (DA, 1) is a catecholamine neurotransmitter involved
in the regulation of a range of physiological functions, including
motor control, cognition, and the ability to experience pleasure.1,2
Abnormalities in the dopaminergic system have been linked to dis-
orders such as Parkinson’s disease, schizophrenia, and addiction.3–6
Dopamine exerts its actions primarily through DA receptors, which
belong to the superfamily of class A (rhodopsin-like) G-protein
coupled receptors (GPCR). The D1-like receptors (D1/D5) activate
Gs proteins and are linked to adenylate cyclase stimulation and
cAMP synthesis, whereas the D2-like receptors (D2/D3/D4) are cou-
pled to Gi/o proteins that inhibit adenylate cyclase activity, among
other effects.6,7 In a continuing effort to develop ligands selective
for specific dopamine receptor-isoforms, we have focused our
research on the design of D1-like receptor agonists.
constrains the ethylamine side chain in a cis-like conformation, and
possesses a b-phenyl substituent (Fig. 1). Discovery of this prototyp-
ical D1 receptor-selective partial agonist led to the development of a
large number of related molecules, many of which demonstrated
selectivity and agonist activity at the dopamine D1 receptor sub-
type.1,7 Structure–activity relationship (SAR) studies of these com-
pounds indicated that a ‘b-phenyldopamine’ pharmacophore, in
which an aryl group is attached to the position b to the amine moi-
ety, may confer selective dopamine D1-like agonist properties (e.g.,
3 and 4).11,12 This b-phenyl moiety is thought to interact with an
accessory binding region in the D1 receptor orthosteric binding site,
and its orientation within this region also may influence receptor
activation.11–13 Despite the cyclic nature of the benzazepines, they
still possess a degree of conformational flexibility. In the case of
Studies of D1-selective dopamine receptor full agonists have
shown they can produce a dramatic reversal of motor deficits in
MPTP-treated monkeys with Parkinson-like symptoms, validating
this receptor as a potential target for treatment of Parkinson’s dis-
ease.8 The D1 dopamine receptor also has been implicated in work-
ing memory processes, suggesting the potential use of D1-selective
agonists for the treatment of cognitive deficits in conditions such as
schizophrenia and age-related cognitive decline.9,10 Another area of
great interest is the role of dopamine in reward phenomena. Studies
of the effect of dopamine D1 agonists in cocaine-seeking animal
models have demonstrated the potential utility of D1 agonists in
the treatment of drug abuse, particularly of psychostimulants.1,4
⇑
Corresponding author. Tel.: +1 765 494 1461; fax: +1 765 494 1414.
E-mail address: drdave@purdue.edu (D.E. Nichols).
Figure 1. Structures of dopamine and D1 selective agonists.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.07.057

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5421
SKF38393 (2), molecular modeling studies indicated that it pos-
sesses two minimum-energy conformations having only a slight
energetic difference (<1 kcal/mol), in which the pendant phenyl
ring is either pseudo-axial or pseudo-equatorial.14 Although several
energetically favorable conformations for the b-phenyl ring are
available to the free ligand, when bound to the receptor, the phenyl
ring is likely to adopt an orientation that optimizes ligand–receptor
interaction. Thus, a number of researchers have put effort into elu-
cidating the bioactive conformation of the b-phenyl substituent.
To study the effect of the orientation of the b-phenyl ring, Ber-
ger et al.14 prepared conformationally-restricted analogs of
SCH23390 (5), a D1 dopamine receptor antagonist belonging to
the phenylbenzazepine series, as a probe for the active conforma-
tion (Fig. 2). Pharmacological evaluation of these analogs indicated
that highest receptor affinity was associated with the trans config-
uration. The 6aS,13bR isomer of 6b, in particular, showed greatest
D1 affinity and receptor selectivity. The lack of activity for the cis
isomer suggested that the axial orientation of the phenyl ring
might be unfavorable for receptor binding. The authors reported
that ring D is locked equatorially in the trans isomer (6b), which
they reported as the preferred low-energy conformation.14 In fact,
however, like 2, the azepine ring can exist in two distinct confor-
mations, one of which places the phenyl into a pseudo-equatorial
position, and the other in which it is in an axial orientation. These
two conformations differ by only 1.7 kcal/mol (H-F, 6-31G⁄), and
in either conformation, the amino group is about 2.5 Å out of
the ring A plane.
The presence of the 7-halogen atom in benzazepine-like ligands
confers antagonist properties, and it has been observed that, in
general, alkylation of the amine nitrogen decreases D1-like agonist
activity.7,15 What caught our attention was the finding that the
dihydroxy analog 7 was reported to have high affinity at D1 recep-
tors, but D1 agonist activity ‘could not be demonstrated’.14 We
hypothesized that the presence of the N-methyl group on the basic
nitrogen might be responsible for the lack of functional activity,
and thus, we had interest in examining the properties of the sec-
ondary amine 8b as a potential dopamine D1 receptor agonist.
Although benzazepine type agonists demonstrate high affinity
and selectivity at the dopamine D1 receptor, they generally do not
possess full intrinsic activity. In endeavors to develop novel dopa-
mine D1 selective agonists, dihydrexidine (DHX) 3 was synthesized
by Nichols et al. as the first potent D1-selective full agonist.16 It is
noteworthy that in 3, the ethylamine fragment is in an extended
trans conformation. Molecular modeling of SKF38393 (2) and DHX
(3) also revealed that although they are both high affinity D1-selec-
tive ligands, the positioning of the b-phenyl relative to the catechol
ring is significantly different between the two molecules.13 For DHX
(3), the b-phenyl ring is nearly coplanar with the catechol ring,
whereas the low energy conformation of SKF38393 (2) displays
an orientation of the b-phenyl moiety that resides in a plane
approximately perpendicular to the catechol ring plane. This obser-
vation led to the hypothesis that the conformation of the b-phenyl
ring relative to the catechol plane is crucial for agonist activity.
In addition, previous work in our laboratory led to the synthesis
of the oxygen and sulfur ‘bioisosteres’ of DHX (3), of which the oxy-
gen analog (doxanthrine, DOX, 4) showed high affinity, selectivity,
and full intrinsic activity at D1 dopamine receptors.17 Encouraged
by these results, and with the use of the facile conjugate chemistry
developed for the synthesis of doxanthrine 4, we have now prepared
8, 9, and 10 as structurally constrained analogs of SKF38393 (2)
(Fig. 2).
2. Results and discussion
2.1. Chemistry
We first envisioned the construction of 8, 9, and 10 through a
key diastereoselective conjugate addition of metallated 4-bromov-
eratrole to the corresponding nitroalkene. Although this method
was efficient for the synthesis of the oxygen and sulfur analogs
(9 and 10), it unfortunately failed for the synthesis of 8, likely
due to the acidity of the benzylic protons that were not present
in the oxygen and sulfur analogs. Consequently, an alternative
method was developed for the preparation of 19 as the key inter-
mediate in the synthesis of 8 (Scheme 1).
Allyl bromide was treated with the organocuprate derivative
of phenyl oxazoline 11 to provide alkene 12, which was easily
purified by vacuum distillation.18 Hydroboration–oxidation of
12, followed by methanolysis of the oxazoline moiety afforded
ester 14, which was brominated to provide intermediate 15. Con-
version of bromide 15 to the corresponding nitro intermediate 16
with sodium nitrite was best carried out in a 2:1 mixture of DMF
and ethylene glycol. Hydrolysis of 16 under acidic conditions
afforded acid 17, which was converted to the corresponding acyl
chloride with SOCl2. Friedel–Crafts acylation of veratrole with this
acyl chloride afforded nitro ketone 18. Henry cyclization with
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as the base afforded
nitroalkene 19. Nitroalcohol 20 was the major side product, but
was readily converted back into 19 with thionyl chloride and
pyridine (Scheme 1).
Although nitroalkene 19 could be prepared by this method, this
multistep synthetic approach was very tedious. Thus, a new meth-
od was developed that greatly reduced the synthetic effort re-
quired to prepare this intermediate. Treatment of
a-tetralone 21
with the Grignard reagent prepared from 4-bromoveratrole, fol-
lowed by acid-catalyzed dehydration of the alcohol readily affor-
ded olefin 22. Nitration of 22 using tetranitromethane then
provided the desired nitroalkene 19 (Scheme 1).
To obtain both the cis and trans isomers of 8, two different
reduction methods for 19 were employed. Treatment with lithium
aluminum hydride (LAH) gave exclusively the cis amine 24a. Acyl-
ation of this amine with chloroacetyl chloride under basic condi-
tions then afforded chloroacetamide 25a, which underwent
photocyclization to give lactam 26a. Reduction of this lactam with
borane, followed by O,O-demethylation of the aryl ethers, afforded
target compound 8a (Scheme 2). To prepare the trans isomer, 19
was first treated with sodium borohydride in a 2:3 solution of
THF/EtOH, which yielded a 2:1 mixture of cis and trans 23, as indi-
cated by NMR. Treatment of this product mixture with ethanolic
NaHCO3 solution resulted in exclusively the trans product 23.
Reduction of 23 with zinc powder and glacial acetic acid afforded
Figure 2. Structures of D1 selective ligands.

5422
A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
Scheme 1. Synthesis of 4-(3,4-dimethoxyphenyl)-3-nitro-1,2-dihydronaphthalene 19.
Scheme 2. Synthesis of ( )-6,6a,7,8,9,13b-hexahydro-5H-benzo[d]naphtho[2,1-b]azepine-11,12-diol 8.
amine 24b. The trans analog 8b was prepared from amine 24b
using the same methods as for the cis analog 8a.
The synthetic approach for the oxygen and sulfur analogs 9 and
10 was devised based on methodology similar to that employed in

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5423
the synthesis of doxanthrine 4.17 The nitrochromene intermediate
29 was obtained from the reaction of salicylaldehyde 28 with
nitroethene (generated in situ from nitroethanol) in basic solution.
To avoid undesired polymerization, a low concentration of nitro-
ethene was ensured by using a steady and slow addition of nitro-
ethanol through a syringe pump. A 1,4-addition of the Grignard
reagent prepared from 4-bromoveratrole to 29 gave exclusively
trans 30, as indicated by proton NMR. Reduction of the nitro moiety
of 30 with zinc powder and glacial acetic acid afforded amine 31 in
quantitative yield. Chloroacetamide 32, prepared from amine 31
and chloroacetyl chloride, then underwent photocyclization to give
lactam 33. Reduction of the lactam with borane, followed by O,O-
dealkylation of the aryl ethers then afforded the target compound
9 (Scheme 3).
Analogously, the preparation of thio-derivative 10 (Scheme 4)
paralleled the synthetic route developed for the oxygen derivative.
Aldehyde 36 was prepared by ortho-formylation of thiophenol 35
with DMF.19 Due to its instability, this aldehyde was used immedi-
ately in subsequent reactions. Intermediates 37, 38, 39, and 40
were synthesized uneventfully following reaction conditions
established for the oxygen analog. Surprisingly, however, photocy-
lization of 40 proved quite unsatisfactory, producing only a 5–7%
yield of lactam 41, compared to the 31% yield obtained for the oxy-
gen analog. Significant decomposition under the photoirradiation
conditions was observed, and modification of the reaction condi-
tions either with respect to solvent systems or reaction time did
not significantly improve the yield. Sulfur compounds can undergo
various photochemical oxidations, resulting in cleavage and rear-
rangement of the sulfide bond, which could account for the low
yield of the desired product. Nonetheless, despite the unsatisfac-
tory yield for production of lactam 41, an amount of target com-
pound 10 sufficient for pharmacological testing was ultimately
obtained.
intrinsic activity was defined as the maximal activation induced by
10 M of the endogenous neurotransmitter dopamine.
Consistent with the results obtained by Berger et al.14 compari-
l
son of the cis and trans isomers 8a and 8b (Table 1) clearly shows
that higher receptor affinity is associated with the trans isomer. Fur-
thermore, the competition binding experiments also demonstrate
that compounds 8b, 9, and 10 have D1 binding affinities that are
similar to that of SKF38393 (2), with 8b being least potent, showing
a 1.7-fold decrease in affinity compared to SKF38393 (2), and sulfur
analog 10 being the most active, with 1.6-fold improvement in affin-
ity over 2 (Table 1). Within this series, the increase in binding affin-
ity from 8b to its oxygen and sulfur analogs 9 and 10 might be
explained by the slight difference in the conformation of the b-phe-
nyl ring due to the oxygen or sulfur replacement at the three-posi-
tion. Although the size of an oxygen atom is not much different from
a CH2 group, a sulfur atom is closer to the size of an ethylene group.
Thus, substitution of the CH2 group with a sulfur atom would result
in a greater change in the orientation of the b-phenyl ring compared
to the carbocyclic or oxacyclic analogs.
Functionally, 8b, 9, and 10 show a greater than eightfold de-
crease in potency compared to SKF38393 (2), and demonstrate par-
tial intrinsic activity of less than 60% relative to dopamine (Table
2). In our assays the receptors are overexpressed, with the conse-
quence that partial agonists can show full or nearly full agonist
activity. For example, SKF38393 is known to be a weak partial ago-
nist, but in our assays it displays almost full agonist activity (93%
compared to dopamine). Thus, compounds 8b, 9, and 10 in a native
expression system would be expected to have even lower intrinsic
activity than was observed in our experiments.
2.3. Conformational analyses
Although disappointed by the lack of functional activity for 8b, 9,
and 10, we believe these results provide further insight into the nat-
ure of the requirements for dopamine D1 agonist activity. In the full
agonist 3, the amino nitrogen resides essentially in the plane of the
catechol ring (Fig. 3A, right column). Based on what is known from
modern molecular modeling studies about the orthosteric binding
site in the receptor, this placement of the nitrogen is likely crucial
for high agonist activity. The azepine ring of SKF38393 can achieve
a twist conformation that places the nitrogen approximately in the
catechol ring plane, but in doing so the pendant phenyl ring is forced
to twist to avoid a nonbonded interaction between H(9), at the top
of the catechol ring, and the ortho-hydrogen atoms of the appended
phenyl ring. This twist conformation will be of significantly higher
energy than the calculated low energy minima.
2.2. Pharmacology
The conformationally restricted analogs 8a, 8b, 9, and 10 were
compared with SKF38393 (2), DHX (3), and SCH39166 for affinity
at D1- and D2-like receptors in pig striatal homogenate. The D1-like
receptor affinities were determined using [3H]SCH23390
displacement, and D2-like receptor affinities were obtained using
displacement of [3H]spiperone, with added ketanserin to mask
5-HT2A sites.
To establish the functional effects of compounds 8b, 9, and 10,
standard cAMP quantification assays were employed using stably
transfected cell lines expressing cloned human D1 receptors.20 Full
Scheme 3. Synthesis of ( )-trans-6,6a,7,8,9,13b-hexahydrobenzo[d]chromeno[3,4-b]azepine-11,12-diol 9.

5424
A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
Scheme 4. Synthesis of ( )-trans-6,6a,7,8,9,13b-hexahydrobenzo[d]thiochromeno[3,4-b]azepine-11,12-diol 10.
Table 1
Affinity at porcine striatal homogenates (nM)
whereas full agonists such as 3 have their ethylamine fragment
in an extended trans configuration.
Ligand
D1-like Ki
D2-like Ki
These same conformational arguments would apply to the
antagonist molecule SCH39166 (6b). Yet, 6b has very high affinity
for the receptor. Clearly, the structural requirements for antago-
nists must vary significantly from those of agonists. First of all,
replacing one of the catechol hydroxys with a halogen is key to
obtaining antagonist activity. Second, the N-methyl is tolerated
for antagonists, but not for agonists, suggesting a fundamental
realignment of the ligand so as to present its nitrogen atom to
the aspartate residue in helix 3 of the receptor in a complementary
way. Finally, it is probably not necessary for the amine nitrogen to
lie approximately in the plane of the substituted aromatic ring for
antagonists, whereas that may be a crucial structural requirement
for agonist activity.
2
3
5
6b
8a
8b
9
10
151 18
22 1.3
2364 146
320 42
ND
2011 331
88000 8000
8479 891
44840 7900
27980 146
2.67 0.42
0.64 0.06
1.30 0.21
22000 4000
257 42
186 32
92 8.5
Chlorpromazine
ND
Table 2
cAMP assay at HEK human D1 receptor
Ligand
EC50 (nM)
Intrinsic activity (%)
In summary, compounds 8b, 9, and 10 show binding affinity
comparable to SKF38393, with the sulfur analog 10 possessing
the most favorable binding of all. By contrast, the new analogs
are only very weak partial agonists, suggesting that conformational
and other structural features prevent them from adopting shapes
that are complementary to the receptor activation process.
DA
2
3
8b
9
10
37 1.1
56 5.2
7.2 0.6
438 42
570 41
584 63
100
93
103
28
60
21
3
7
2
2
2
1
3. Experimental
3.1. Chemistry
By contrast, for the tetracyclic congeners to adopt an azepine
ring conformation that would place the nitrogen atom approxi-
mately in the catechol ring plane requires the appended phenyl
ring to lie nearly coplanar with the catechol ring. This phenyl ring
is no longer free to twist, and therefore in that conformation a non-
bonded interaction between the analogous hydrogen atom of the
catechol and the hydrogen on the phenyl cannot be avoided
(Fig. 3C and D). Thus, the more rigid tetracyclic analogs cannot
adopt a low energy conformation that places the nitrogen atom
approximately in the plane of the catechol ring. In addition, for
any of the azepine type molecules, even if the nitrogen atom re-
sides in the catechol ring plane, it is displaced from the position
of that in the full agonist 3 by virtue of its cisoid conformation,
3.1.1. Chemistry general
All reagents were commercially available and used without
further purification unless stated otherwise. Flash column chroma-
tography was carried out using silica gel having a particle size of
40–65 lm. Melting points were determined in open capillaries
with a Meltemp apparatus. 1H NMR spectra were obtained using
a 300 MHz Bruker ARX-300 spectrometer or a 500 MHz Bruker
DRX-500 spectrometer. Mass spectra data were obtained by the
Purdue University Campus-wide Mass Spectrometry Center.

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5425
washed with brine, dried over MgSO4, filtered, and the solvents re-
moved under reduced pressure. Silica gel flash column chromatog-
raphy, eluting with 2:3 hexanes/EtOAc, provided 408 mg (54.5%) of
ester 14 as a light yellow oil. 1H NMR (CDCl3, 300 MHz): d 7.88 (dd,
J1 = 7.8 Hz, J2 = 1.5 Hz, 1H, ArH); 7.42 (dt, J1 = 7.2 Hz, J2 = 1.2 Hz, 1H,
ArH); 7.30–7.23 (m, 2H, 2ArH); 3.90 (s, 3H, OCH3); 3.62 (t,
J = 6.0 Hz, 2H, CH2OH); 3.06 (t, J = 7.2 Hz, 2H, ArCH2); 1.96–1.87
(m, 2H, 2-H2). CIMS: m/z (relative intensity): 195 (M+H+, 44), 177
(M+H–H2O, 38), 163 (M+H–CH3OH, 100). Anal. Calcd For
C11H14O3: C, 68.02; H, 7.27. Found: C, 67.84; H, 7.20.
3.1.4. Methyl 2-(3-bromopropyl)benzoate (15)
Triphenylphosphine (8.4 g, 32.13 mmol) was added in small
portions at 0 °C to a flask containing ester 14 (4.16 g, 21.42 mmol)
and carbon tetrabromide (8.5 g, 25.7 mmol) in 100 mL of CH2Cl2.
The reaction was stirred for 1 h, and the solvent was removed un-
der reduced pressure. The residue was washed with ether, and the
white precipitates were removed by filtration. The filtrate was con-
centrated and purified by silica gel flash column chromatography,
eluting with 1:1 hexanes/EtOAc, to give 5.18 g (94%) of bromide 15
as a yellow oil. 1H NMR (DMSO-d6, 300 MHz): d 7.91 (d, J = 7.5 Hz,
1H, ArH); 7.44 (dt, J1 = 8.7 Hz, J2 = 1.5 Hz, 1H, ArH); 7.31–7.26 (m,
2H, 2ArH); 3.90 (s, 3H, OCH3); 3.45 (t, J = 6.6 Hz, 2H, CH2Br); 3.11
(t, J = 7.5 Hz, 2H, ArCH2); 2.18 (q, J = 7.0 Hz, 2H, 2-H2). CIMS: m/z
(relative intensity): 257/259 (M+H+, 47), 177 (M+H–HBr, 100).
3.1.5. Methyl 2-(3-nitropropyl)benzoate (16)
Bromide 15 (3.0 g, 12.34 mmol) was dissolved in 45 mL of a
mixture of 2:1 DMF/ethylene glycol. Sodium nitrite (3.4 g,
49.36 mmol) was then added to the mixture. After stirring at room
temperature for 4 h, the reaction was poured into ice-water and
extracted with CH2Cl2 (3 Â 50 mL). The organic solvent was dried
over MgSO4, filtered, and the solvent removed under reduced pres-
sure to give a yellow oil. Silica gel flash column chromatography,
eluting with 3:2 hexanes/EtOAc, provided 1.62 g of unreacted bro-
mide 15 and 0.65 g (51%) of nitrobenzoate 16: mp 105–106 °C. 1H
NMR (CDCl3, 300 MHz): d 7.94 (d, J = 8.0 Hz, 1H, ArH); 7.47 (dt,
J1 = 7.5 Hz, J2 = 1.2 Hz, 1H, ArH); 7.32 (d, J = 7.5 Hz, 1H, ArH);
7.28–7.24 (m, 1H, 1ArH); 4.44 (t, J = 6.9 Hz, 2H, CH2NO2); 3.89 (s,
3H, OCH3); 3.05 (t, J = 7.2 Hz, 2H, ArCH2); 2.34 (q, J = 7.0 Hz, 2H,
2-H2). CIMS: m/z (relative intensity): 224 (M+H+, 100).
Figure 3. Space filling energy-minimized representations of: (A) the full agonist
DHX (3, top row), (B) the partial agonist SKF38393 2, (C) 8b with the pendant
phenyl ring in a pseudo-equatorial position, and (D) 8b with the pendant phenyl
ring in an axial position. Structures on the left are viewed from the top, whereas the
right column is viewed edge-on, with the amine towards the viewer.
3.1.2. 3-(2-(4,4-Dimethyl-4,5-dihydrooxazol-2-yl)phenyl)pro
pan-1-ol (13)
Alkene 12 (2.0 g, 9.3 mmol) was dissolved in 20 mL of dry THF
and 28 mL (14 mmol) of a 0.5 M solution of 9-BBN in THF was
added drop-wise at 0 °C. After stirring at room temperature for
4 h, 5 mL of water were added, followed by 15 mL of H2O2
(30 wt % in H2O) in 2 N NaOH (20 mL).The reaction was heated at
60 °C for 2.5 h. The reaction was cooled to room temperature, ex-
tracted with EtOAc (3 Â 60 mL). The organic solvent was washed
with brine, dried over MgSO4, filtered, and the solvents removed
under reduced pressure. Silica gel flash column chromatography,
eluting with 1:4 hexanes/EtOAc, provided 1.9 g (87.7%) of alcohol
13 as a colorless oil. 1H NMR (CDCl3, 300 MHz): d 7.74 (dd,
J1 = 9.0 Hz, J2 = 1.5 Hz, 1H, ArH); 7.51 (dt, J1 = 7.5 Hz, J2 = 1.2 Hz,
1H, ArH); 7.29–7.22 (m, 2H, 2ArH); 5.90 (br s, 1H, OH); 4.10 (s,
2H, Oxazolinyl-CH2); 3.43 (t, J = 4.8 Hz, 2H, CH2OH); 3.16 (t,
J = 6.3 Hz, 2H, ArCH2); 1.98–1.92 (m, 2H, 2-H2); 1.41 (s, 6H,
2CH3). CIMS: m/z (relative intensity): 234 (M+H+, 100). Anal. Calcd
For C14H19NO2Á0.11C4H8O2: C, 71.37; H, 8.25; N, 5.76. Found: C,
71.38; H, 8.06; N, 5.62.
3.1.6. 2-(3-Nitropropyl)benzoic acid (17)
Nitrobenzoate 16 (6.5 g, 29.12 mmol) was dissolved in a mix-
ture of 100 mL of AcOH and 40 mL of 6 M HCl. The reaction was
heated at reflux for 14 h, then cooled to room temperature and ex-
tracted with CH2Cl2 (3 Â 80 mL). The extract was dried over
MgSO4, filtered, and the solvent removed under reduced pressure
to give a yellow oil. The crude product was dissolved in EtOAc
and extracted with saturated NaHCO3 solution (3 Â 100 mL). The
combined aqueous solution was acidified with 1 M HCl and ex-
tracted with CH2Cl2 (3 Â 80 mL). The organic extract was dried
over MgSO4, filtered, and the solvent removed under reduced pres-
sure to yield 4.5 g (73.9%) of acid 17. 1H NMR (CDCl3, 500 MHz): d
8.11 (dd, J1 = 8.0 Hz, J2 = 1.0 Hz, 1H, ArH); 7.53 (dt, J1 = 7.5 Hz,
J2 = 1.0 Hz, 1H, ArH); 7.38–7.30 (m, 2H, 2ArH); 4.51 (t, J = 7.0 Hz,
2H, CH2NO2); 3.13 (t, J = 7.5 Hz, 2H, ArCH2); 2.34 (q, J = 7.3 Hz,
2H, 2-H2). CIMS: m/z (relative intensity): 208 (M–H, 100).
3.1.7. (3,4-Dimethoxyphenyl)(2-(3-nitropropyl)phenyl)
methanone (18)
3.1.3. Methyl 2-(3-hydroxypropyl)benzoate (14)
Alcohol 13 (0.9 g, 3.86 mmol) was dissolved in 50 mL of 6 M
methanolic HCl and heated at reflux for 30 h. The solvent was re-
moved under reduced pressure, and the residue was partitioned
between water and CH2Cl2. The aqueous solution was extracted
twice more with CH2Cl2. The combined organic solvent was
Thionyl chloride (1.87 mL, 25.6 mmol) was added to a flask con-
taining acid 17 (1.34 g, 6.41 mmol) in 30 mL of benzene. The
reaction was heated at reflux for 2.5 h. Excess thionyl chloride
was then azeotropically removed from the reaction with toluene.
The resulting acyl chloride was dried under vacuum and used

5426
A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
without further purification. To the flask containing the acyl chlo-
ride in CH2Cl2, AlCl3 (1.28 g, 9.61 mmol) was added, followed by
drop-wise addition of veratrole (1.2 mL, 9.61 mmol) at 0 °C. The
reaction was stirred overnight at room temperature, and then
was quenched with water and extracted with CH2Cl2 (3 Â 60 mL).
The organic solution was washed with brine, dried over MgSO4, fil-
tered, and the solvent removed under reduced pressure. Silica gel
flash column chromatography, eluting with 1:1 hexanes/EtOAc,
provided 1.68 g of ketone 18: mp 70–72 °C. 1H NMR (CDCl3,
300 MHz): d 7.55 (d, J = 2.7 Hz, 1H, ArH); 7.49–7.42 (m, 1H, ArH);
7.35–7.29 (m, 2H, 2ArH); 7.21 (dd, J1 = 8.4 Hz, J2 = 2.1 Hz, 1H,
ArH); 6.91 (d, J = 3.0 Hz, 1H, ArH); 6.84 (d, J = 8.4 Hz, 1H, ArH);
4.36 (t, J = 6.9 Hz, 2H, CH2NO2); 3.95 (s, 3H, OCH3); 3.89 (s, 3H,
OCH3); 2.75 (t, J = 6.9 Hz, 2H, ArCH2); 2.32 (p, J = 7.2 Hz, 2H, 2-
H2). CIMS: m/z (relative intensity): 330 (M+H+, 100). Anal. Calcd
for C18H19NO5: C, 65.64; H, 5.81; N, 4.25. Found: C, 65.55; H,
5.69; N, 4.14.
2.5 h. After cooling to room temperature, the reaction was
quenched with 40 mL of ice cold 6 M HCl, and stirred for another
1 h. The mixture was extracted with EtOAc (3 Â 50 mL), the extract
was dried over MgSO4, filtered, and the solvents were removed un-
der reduced pressure. Recrystallization from EtOH/hexane yielded
5.3 g (58%) of alkene 22: mp 66–68 °C. 1H NMR (CDCl3, 300 MHz): d
7.19–7.03 (m, 4H, 4ArH); 6.90–6.87 (m, 3H, 3ArH); 6.07 (t,
J = 9.3 Hz, 1H, 3-H); 3.92 (s, 3H, OCH3); 3.86 (s, 3H, OCH3); 2.85
(t, J = = 7.7 Hz, 2H, 1-H2); 2.43–2.38 (m, 2H, 2-H2). CIMS: m/z (rel-
ative intensity): 267 (M+H+, 100). Anal. Calcd For C18H18O2: C,
81.17; H, 6.8. Found: C, 80.86; H, 6.73.
3.1.10. ( )-Trans-1-(3,4-dimethoxyphenyl)-2-nitro-1,2,3,4-tetra
hydronaphthalene (23)
Nitronaph-thalene 19 (1.5 g, 4.82 mmol) was dissolved in
250 mL of
a mixture of 2:3 THF/EtOH. Sodium borohydride
(729 mg, 19.27 mmol) was then added to the solution. The reaction
was stirred at room temperature for 16 h, at which time two more
molar equivalents of sodium borohydride were added. After stir-
ring for another 7 h, the reaction was quenched with diluted aque-
ous urea/AcOH solution. The solvent was removed under reduced
pressure, and the residue was partitioned between water and
CH2Cl2. The aqueous solution was extracted twice more with
CH2Cl2 (2 Â 50 mL). The combined organic extract was dried over
MgSO4, filtered, and the solvents removed under reduced pressure
to afford a mixture of a 2:1 ratio of cis and trans 23, as determined
from the 1H NMR spectrum of the product. Epimerization of cis 23
was achieved by dissolving the cis/trans mixture of 23 in 100 mL of
NaHCO3-saturated 95% EtOH. The mixture was heated at reflux for
6 h. The solvent was then removed under reduced pressure, and
the residue partitioned between water and CH2Cl2. The aqueous
solution was extracted twice more with CH2Cl2 (2 Â 50 mL). The
combined organic extract was dried over MgSO4, filtered, and the
solvents removed under reduced pressure. Recrystallization from
EtOH yielded 1.28 g (84.4%) of trans 23: mp 129–131 °C. 1H NMR
3.1.8. 4-(3,4-Dimethoxyphenyl)-3-nitro-1,2-dihydronaphtha
lene (19)
3.1.8.1. Method 1. Ketone 18 (720 mg, 2.19 mmol) was dissolved
in 60 mL of a mixture of 1:1 MeOH/THF. DBU (1 mL, 6.56 mmol)
was then added to this solution, and the reaction was heated at re-
flux for 24 h. The solvent was then removed under reduced pres-
sured and the residue was partitioned between 1 M HCl and
CH2Cl2. The aqueous layer was extracted twice more with CH2Cl2.
The combined organic extract was dried over MgSO4, filtered,
and the solvent removed under reduced pressure. Silica gel flash
column chromatography, eluting with 3:2 hexanes/EtOAc, pro-
vided 230 mg (33.8%) of nitronaphthalene 19 and 216 mg (30%)
of alcohol 20 as the major side product, the latter being easily
dehydrated to give desired nitronaphthalene 19.
3.1.8.2. Method 2. Alkene 22 (8.5 g, 31.9 mmol) and pyridine
(2.5 g, 31.9 mmol) were dissolved in 300 mL of dry MeCN. Into this
stirring solution, tetranitromethane (3.8 mL, 31.9 mmol) was added
drop-wise at 0 °C. The reaction was stirred for 3.5 h at room tem-
perature, and quenched by pouring into ice-cold water. The mixture
was concentrated under reduced pressure. The residue was parti-
tioned between water and CH2Cl2. The aqueous solution was ex-
tracted twice more with CH2Cl2 (100 mL). The organic extract was
dried over MgSO4, filtered, and the solvent removed under reduced
pressure. Recrystallization from MeOH yielded 7.65 g (83.5%) of
nitronaphthalene 19 as a yellow powder: mp 150–158 °C. 1H
NMR (CDCl3, 300 MHz): d 7.33–7.23 (m, 2H, 2ArH); 7.15 (dt,
J1 = 8.2 Hz, J2 = 2.5 Hz, 1H, ArH); 6.92 (d, J = 8.4 Hz, 1H, ArH); 6.87
(d, J = 7.5 Hz, 1H, ArH); 6.75 (dd, J1 = 8.1 Hz, J2 = 2.1 Hz, 1H, ArH);
6.69 (d, J = 1.8 Hz, 1H, ArH); 3.93 (s, 3H, OCH3); 3.84 (s, 3H,
OCH3); 3.10–3.00 (m, 4H, ArCH2, CH2CNO2). EIMS: m/z (relative
intensity): 311 (M+, 100). Anal. Calcd for C18H17NO4 0.18C2H6O: C,
68.99; H, 5.70; N, 4.38. Found: C, 68.87; H, 5.50; N, 4.38.
(CDCl3, 500 MHz):
d
7.19–7.15 (m, 2H, 2ArH); 7.09 (dt,
J1 = 7.7 Hz, J2 = 1.9 Hz, 1H, ArH); 6.84 (d, J = 7.7 Hz, 1H, ArH); 6.80
(d, J = 7.9 Hz, 1H, ArH); 6.67 (dd, J1 = 8.7 Hz, J2 = 1.9 Hz, 1H, ArH);
6.61 (d, J = 1.9 Hz, 1H, ArH); 4.93 (ddd, J1 = 14.7 Hz, J2 = 8.8 Hz,,
J3 = 1.9 Hz, 1H, CHNO2); 4.86 (d, J = 8.8 Hz, 1H, 1-H); 3.86 (s, 3H,
OCH3); 3.80 (s, 3H, OCH3); 3.10–3.03 (m, 2H, 4-H2); 2.49–2.47
(m, 2H, 3-H2). EIMS: m/z (relative intensity): 313 (M+, 100). Anal.
Calcd for C18H19NO4: C, 68.99; H, 6.11; N, 4.47. Found: C, 69.01;
H, 6.03; N, 4.48.
3.1.11. ( )-Cis-1-(3,4-dimethoxyphenyl)-1,2,3,4-tetrahydronaph
thalen-2-amine (24a)
Nitronaphthalene 19 (1.0 g, 3.21 mmol) in 30 mL dry THF was
added drop-wise to a flask containing LAH (488 mg, 12.85 mmol)
in 40 mL of dry THF. The reaction was heated at reflux for 3 h,
cooled to room temperature, and quenched with 2 M HCl. After
stirring for another 10 min, 2 M NaOH was added to the reaction.
The mixture was extracted with CH2Cl2 (3 Â 50 mL), the extract
was dried over MgSO4, filtered, and the solvent removed under re-
duced pressure to afford a dark oil. The crude product was dis-
solved in EtOAc and extracted with aqueous 2 N HCl solution
(3 Â 60 mL). The combined aqueous layer was basified with 2 N
NaOH solution and extracted with CH2Cl2 (3 Â 40 mL). The organic
solution was then dried over MgSO4, filtered, and the solvent re-
moved under reduced pressure to yield 720 mg (79%) of amine
24a: mp (HCl salt) 272–275 °C. 1H NMR (CDCl3, 300 MHz): d
7.06–6.96 (m, 3H, 3ArH); 6.89 (d, J = 7.5 Hz, 1H, ArH); 6.70 (d,
J = 8.1 Hz, 1H, ArH); 6.57 (d, J = 8.0 Hz, 1H, ArH); 6.48 (d,
J = 7.9 Hz, 1H, ArH); 4.12 (d, J = 4.5 Hz, 1H, 1-H); 3.80–3.75 (m,
1H, CHNH2); 3.77 (s, 3H, OCH3); 3.72 (s, 3H, OCH3); 3.05–2.80
(m, 2H, 4-H2); 1.75–1.72 (m, 2H, 3-H2). ESIMS: m/z (relative inten-
Alcohol 20: mp 150–158 °C. 1H NMR (CDCl3, 500 MHz): d 7.33–
7.19 (m, 4H, 4ArH); 6.77 (d, J = 2.0 Hz, 1H, ArH); 6.72 (d, J = 8.5 Hz,
1H, ArH); 6.49 (dd, J1 = 8.0 Hz, J2 = 2.0 Hz, 1H, ArH); 5.04 (dd,
J1 = 11.0 Hz, J2 = 4.0 Hz, 1H, NO2CH); 3.83 (s, 3H, OCH3); 3.79 (s,
3H, OCH3); 3.16 (td, J1 = 17.5 Hz, J2 = 3.5 Hz, 1H, ArCHa); 3.10–
3.02 (m, 1H, ArCHb); 2.38–2.34 (m, 2H, CH2CNO2). EIMS: m/z (rel-
ative intensity): 311 (M+, 100). Anal. Calcd for C18H19NO5: C, 65.64;
H, 5.81; N, 4.25. Found: C, 65.59; H, 5.80; N, 4.32.
3.1.9. 1-(3,4-Dimethoxyphenyl)-3,4-dihydronaphthalene (22)
To a stirred solution of 4-bromoveratrole (11.1 g, 0.051 mol) in
100 mL of dry THF was added magnesium powder-50 mesh (1.7 g,
0.068 mol), and the mixture was heated at reflux for 2 h to form
the Grignard reagent.
a-Tetralone (5 g, 0.034 mol) was added
drop-wise to the refluxing mixture, which was heated for another

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5427
sity): 284 (M+H+, 100). Anal. Calcd for C18H21NO2ÁHCl: C, 67.60; H,
CIMS: m/z (relative intensity): 360 (M+H+, 100). Anal. Calcd for
20H22ClNO3: C, 66.75; H, 6.16; N, 3.89. Found: C, 66.45; H, 6.07;
6.93; N, 4.38. Found: C, 67.31; H, 6.78; N, 4.39.
C
N, 3.89.
3.1.12. ( )-Trans-1-(3,4-dimethoxyphenyl)-1,2,3,4-tetrahydrona
phthalen-2-amine (24b)
3.1.15. ( )-Cis-11,12-dimethoxy-6a,7,9,13b-tetrahydro-5H-
benzo[d]naphtho[2,1-b]azepin-8(6H)-one (26a)
Naph-thalene 23 (500 mg, 1.60 mmol) was dissolved in 40 mL
of AcOH and 1.05 g (16.0 mmol) of powdered zinc were added.
The mixture was stirred vigorously at room temperature for 16 h,
at which time the reaction was filtered through Celite, and the fil-
tered solids were rinsed on the filter with AcOH. The filtrate was
concentrated to about 1/3 volume and water (80 mL) was added.
The aqueous solution was then basified with conc ammonium
hydroxide and extracted with CH2Cl2 (3 Â 40 mL). The organic sol-
vent was then washed twice with brine, dried over MgSO4, filtered,
and the solvents removed under reduced pressure to give a yellow
oil. Acid/base purification of crude product (same method as in
24a) yielded 440 mg (97.4%) of amine 24b: mp (HCl salt) 248–
250 °C. 1H NMR (CD3OD, 300 MHz): d 7.20–7.13 (m, 2H, 2ArH);
7.05 (dt, J1 = 8.1 Hz, J2 = 1.9 Hz, 1H, ArH); 6.98 (d, J = 9.0 Hz, 1H,
ArH); 6.78 (d, J = 1.9 Hz, 1H, ArH); 6.75–6.71 (m, 2H, 2ArH); 4.06
(d, J = 9.1 Hz, 1H, 1-H); 3.84 (s, 3H, OCH3); 3.77 (s, 3H, OCH3);
4.93 (dt, J1 = 10.5 Hz, J2 = 3.3 Hz,, 1H, CHNH2); 3.16–3.11 (m, 1H,
4-Ha); 3.06–2.99 (m, 1H, 4-Hb); 2.32–2.28 (m, 1H, 3-Ha); 2.03–
1.95 (m, 1H, 3-Hb). CIMS: m/z (relative intensity): 284 (M+H+,
100). Anal. Calcd for C18H21NO2ÁHCl: C, 67.60; H, 6.93; N, 4.38.
Found: C, 65.14; H, 6.90; N, 4.25.
A solution of chloroacetamide 25a (800 mg, 2.22 mmol) and
NaHCO3 (560 mg, 6.67 mmol) in 500 mL of a mixture of 3:1
MeOH/H2O was irradiated for 55 min at room temperature with
a medium pressure mercury lamp, with cooling. The resulting solu-
tion was reduced in volume under pressure, and partitioned be-
tween water and CH2Cl2. The aqueous solution was extracted
twice more with CH2Cl2. The pooled organic solution was dried
over MgSO4, filtered, and the solvents removed under reduced
pressure. Silica gel flash column chromatography, eluting with
EtOAc, provided 180 mg (25.1%) of lactam 26a: mp 220–222 °C.
1H NMR (CDCl3, 300 MHz): d 7.19–7.16 (m, 1H, ArH); 7.15–7.11
(m, 1H, ArH); 7.00 (d, J = 7.2 Hz, 1H, ArH); 6.83 (d, J = 8.4 Hz, 1H,
ArH); 6.61 (d, J = 2.1 Hz, 1H, ArH); 6.51 (dd, J1 = 8.4 Hz,
J2 = 2.1 Hz, 1H, ArH); 5.43 (br s, 1H, NH); 4.60–4.53 (m, 1H, 6a-
H); 4.37 (d, J = 5.7 Hz, 1H, 13b-H); 4.08 (s, 2H, 9-H2); 3.90 (s, 3H,
OCH3); 3.86 (s, 3H, OCH3); 3.13–3.10 (m, 2H, 5-H2); 1.97–1.88
(m, 2H, 6-H2). CIMS: m/z (relative intensity): 324 (M+H+, 100).
Anal. Calcd for C20H21NO3: C, 74.28; H, 6.55; N, 4.33. Found: C,
74.04; H, 6.56; N, 4.32.
3.1.16. ( )-Trans-11,12-dimethoxy-6a,7,9,13b-tetrahydro-5H-
benzo[d]naphtho[2,1-b]azepin-8(6H)-one (26b)
3.1.13. ( )-Cis-2-chloro-N-(1-(3,4-dimethoxyphenyl)-1,2,3,4-tet
rahydronaphthalen-2-yl)acetamide (25a)
A solution of chloroacetamide 25b (700 mg, 1.95 mmol) and
NaHCO3 (1.31 g, 15.56 mmol) in 500 mL of a mixture of 3:1
MeOH/H2O was irradiated for 55 min at room temperature with
a medium pressure mercury lamp, with cooling. The resulting solu-
tion was reduced in volume under reduced pressure, and parti-
tioned between water and CH2Cl2. The aqueous solution was
extracted twice more with CH2Cl2. The pooled organic solvent
was dried over MgSO4, filtered, and the solvent removed under re-
duced pressure. Silica gel flash column chromatography, eluting
with EtOAc, provided 154 mg (24.5%) of lactam 26b: mp 232–
234 °C. 1H NMR (CDCl3, 300 MHz): d 7.23–7.15 (m, 3H, 3ArH);
7.08 (d, J = 7.3 Hz, 1H, ArH); 6.76 (s, 1H, ArH); 6.42 (s, 1H, ArH);
5.59 (s, 1H, NH); 4.62 (d, J = 10.8 Hz, 1H, 13b-H); 4.13, 3.65 (ABq,
JAB = 15.9 Hz, 2H, 9-H2); 3.89 (s, 3H, OCH3); 3.65 (s, 3H, OCH3);
3.54 (dt, J1 = 11.0 Hz, J2 = 3.9 Hz,, 1H, 6a-H); 2.96–2.91 (m, 2H, 5-
H2); 2.05–1.99 (m, 1H, 6-Ha); 1.97–1.89 (m, 1H, 6-Hb). ESIMS: m/
Amine 24a (750 mg, 2.65 mmol) was dissolved in 100 mL of
Et2O, and Et3N (402 mg, 3.97 mmol) was added. The reaction was
cooled to 0 °C, and chloroacetyl chloride (330 mg, 2.92 mmol)
was added drop-wise to the reaction mixture. After stirring for
16 h at room temperature, water was added to quench the reac-
tion. The mixture was extracted with CH2Cl2 (3 Â 30 mL), the or-
ganic extract was dried over MgSO4, filtered, and concentrated
under reduced pressure to give the crude product, which was
recrystallization from EtOH to afford 670 mg (71.1%) acetamide
25a: mp 127–130 °C. 1H NMR (CDCl3, 500 MHz): d 7.02–7.01 (m,
2H, 2ArH); 7.00–6.99 (m, 1H, ArH); 6.88 (d, J = 7.8 Hz, 1H, ArH);
6.72 (d, J = 8.4 Hz, 1H, ArH); 6.50 (d, J = 1.8 Hz, 1H, ArH); 6.40
(dd, J1 = 8.4 Hz, J2 = 2.1 Hz, 1H, ArH); 4.45–4.37 (m, 1H, CHN);
4.25 (d, J = 5.7 Hz, 1H, 1-H); 3.96 (s, 2H, CH2Cl); 3.77 (s, 3H,
OCH3); 3.73 (s, 3H, OCH3); 3.00–2.95 (m, 2H, 4-H2); 1.84–1.75
(m, 2H, 3-H2). CIMS: m/z (relative intensity): 360/362 (M+H+,
100). Anal. Calcd for C20H22ClNO3: C, 66.75; H, 6.16; N, 3.89. Found:
C, 66.49; H, 6.01; N, 3.91.
z
C
(relative intensity): 324 (M+H+, 100). Anal. Calcd for
20H21NO3: C, 74.28; H, 6.55; N, 4.33. Found: C, 74.10; H, 6.45;
N, 4.34.
3.1.14. ( )-Trans-2-chloro-N-(1-(3,4-dimethoxyphenyl)-1,2,3,4-
tetrahydronaphthalen-2-yl)acetamide (25b)
3.1.17. ( )-Cis-11,12-dimethoxy-6,6a,7,8,9,13b-hexahydro-5H-
benzo[d]naphtho[2,1-b]azepine hydrochloride (27a)
Amine 24b (5.6 g, 19.76 mmol) was dissolved in 250 mL of
CH2Cl2, and Na2CO3 (8.4 g, 15.32 mmol) was added. The reaction
was cooled to 0 °C, and chloroacetyl chloride (6.7 g, 59.29 mmol)
was added drop-wise to the reaction mixture. After stirring for
16 h at room temperature, water was added to quench the reac-
tion. The mixture was extracted with CH2Cl2 (3 Â 30 mL), the or-
ganic solvent was dried over MgSO4, filtered, and concentrated
under reduced pressure to give the crude product, which was
recrystallized from EtOH to yield 6.82 g (95.9%) of acetamide
25b: mp 171–174 °C. 1H NMR (CDCl3, 500 MHz): d 7.18 (d,
J = 4.0 Hz, 1H, ArH); 7.10–7.07 (m, 1H, ArH); 6.86 (d, J = 8.0 Hz,
1H, ArH); 6.78 (d, J = 8.5 Hz, 1H, ArH); 6.67–6.57 (m, 3H, 3ArH);
4.32 (qd, J1 = 8.0 Hz, J2 = 2.5 Hz, 1H, CHN); 4.00 (d, J = 7.5 Hz, 1H,
1-H); 3.99, 3.90 (ABq, JAB = 20.0 Hz, 2H, CH2Cl); 3.85 (s, 3H,
OCH3); 3.81 (s, 3H, OCH3); 3.10–3.07 (m, 1H, 4-Ha); 2.96–2.91
(m, 1H, 4-Hb); 2.23–2.21 (m, 1H, 3-Ha); 1.84–1.57 (m, 1H, 3-Hb).
Lactam 26a (100 mg, 0.31 mmol) was suspended in 60 mL of
dry THF, and 1.86 mL (1.86 mmol) of a 1 M solution of BH3 in
THF was added. This solution was stirred at reflux for 24 h, then
cooled and quenched with excess MeOH. The solvent was removed
under reduced pressure and the residue was stirred for 4 h with
30 mL of 6 M ethanolic HCl solution at reflux. The solvent was re-
moved, and MeOH was added and removed under vacuum three
times. The residue was partitioned between 2 M aqueous NaOH
and CH2Cl2. The aqueous solution was extracted twice more with
CH2Cl2 (2 Â 40 mL). The combined organic extract was dried over
MgSO4, filtered, and the solvent removed under reduced pressure.
The crude amine was dissolved in a small amount CH2Cl2, and 2 M
HCl in ether was added to the solution. The solvent was removed
under reduced pressure, and the hydrochloride salt was recrystal-
lized from EtOH to give 50 mg (52.6%) of amine hydrochloride 27a:
mp >175 °C. 1H NMR (CDCl3, 300 MHz): 7.21–7.18 (m, 2H, 2ArH);

5428
A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
7.13–7.10 (m, 1H, ArH); 7.01 (s, 1H, ArH); 6.96–6.93 (m, 1H, ArH);
6.81 (s, 1H, ArH); 4.45 (m, 1H, 13b-H); 3.70 (m, 1H, 6a-H); 3.88 (s,
3H, OCH3); 3.85 (s, 3H, OCH3); 3.24–3.19 (m, 3H, 8-H2, 9-Ha); 3.05–
2.91 (m, 2 H, 5-H2); 2.63 (m, 1H, 9-CHb); 1.86–1.65 (m, 2H, 6-H2).
CIMS: m/z (relative intensity): 310 (M+H+, 100). Anal. Calcd for
300 MHz): d 7.22–7.19 (m, 3H, 3ArH); 7.08–7.06 (m, 1H, ArH);
6.70 (s, 1H, ArH); 6.97 (s, 1H, ArH); 4.67 (d, J = 8.2 Hz, 1H, 13b-
H); 3.65 (ddd, J1 = 13.2 Hz, J2 = 5.1 Hz, J3 = 1.5 Hz, 1H, 6a-H); 3.44
(t, J = 114.7 Hz, 1H, 8-CHa); 3.31–3.24 (m, 1 H, 5-Ha); 3.02 (t,
J = 12.9 Hz, 1H, 8-Hb); 2.92–2.85 (m, 3H, 9-H2, 5-Hb); 2.27–2.21
(m, 1H, 6-Ha); 1.80–1.74 (m, 1H, 6-Hb). ESIMS: m/z 282 (M+H+,
100). Anal. Calcd for C18H20BrNO2: C, 59.68; H, 5.56; N, 3.87.
Found: C, 59.44; H, 5.65; N, 3.94.
C20H24ClNO2: C, 69.45; H, 6.99; N, 4.05. Found: C, 69.22; H, 6.85;
N, 4.02.
3.1.18. ( )-Trans-11,12-dimethoxy-6,6a,7,8,9,13b-hexahydro-
5H-benzo[d]naphtho[2,1-b]azepine hydrochloride (27b)
Lactam 26b (300 mg, 0.93 mmol) was suspended in 60 mL of
dry THF, and 7.42 mL (7.42 mmol) of a 1 M solution of BH3 in
THF was added. This solution was stirred at reflux for 24 h, then
cooled and quenched with excess MeOH. The solvent was removed
under reduced pressure. The residue was then stirred for 4 h at re-
flux with 50 mL of 6 M ethanolic HCl solution. The solvent was re-
moved and MeOH was added and removed under vacuum three
times. The residue was partitioned between 2 M aqueous NaOH
and CH2Cl2. The aqueous solution was extracted twice more with
CH2Cl2 (2 Â 40 mL). The combined organic extract was dried over
MgSO4, filtered, and the solvents removed under reduced pressure.
The crude amine was dissolved in a small amount of CH2Cl2, and
2 M HCl in ether was added to the solution. The solvent was re-
moved under reduced pressure, and the hydrochloride salt was
recrystallized from EtOH to provide 250 mg (77.9%) of amine
hydrochloride 27b: mp 279–282 °C. 1H NMR (CD3OD, 300 MHz):
d 7.23–7.22 (m, 3H, 3ArH); 7.06 (d, J = 5.1 Hz, 1H, ArH); 6.92 (s,
1H, ArH); 6.07 (s, 1H, ArH); 4.75 (d, J = 8.4 Hz, 1H, 13b-H); 3.82
(s, 3H, OCH3); 3.68 (dd, J1 = 14.2 Hz, J2 = 5.1 Hz, 1H, 6a-H); 3.56–
3.45 (m, 2H, 8-H2); 3.45 (s, 3H, OCH3); 3.09–3.01 (m, 2H, 5-H2);
2.91–2.88 (m, 2H, 9-CH2); 2.27–2.21 (m, 1H, 6-Ha); 1.83–1.77 (m,
1H, 6-Hb). ESIMS: m/z (relative intensity): 310 (M+H+, 100). Anal.
Calcd for C20H24ClNO2: C, 69.45; H, 6.99; N, 4.05. Found: C,
69.19; H, 6.82; N, 4.10.
3.1.21. 3-Nitro-2H-chromene (29)
In a three-necked flask equipped with a Dean Start trap, con-
denser and mechanical stirring, salicylaldehyde (17.4 g,
0.142 mol) was dissolved in 400 mL of toluene containing di-n-bu-
tyl amine (9.2 g, 0.071 mol) and phthalic anhydride (42.14 g,
0.285 mol). The mixture was heated at reflux with vigorous stirring
as 2-nitroethanol (27.1 mL, 0.384 mol) was added very slowly into
the reaction flask using a syringe pump set at a rate of approxi-
mately 0.5 mL/h. After the addition of 2-nitroethanol was com-
pleted (ca. 55 h), the reaction was stirred at reflux for another
12 h. The mixture was cooled to room temperature and was fil-
tered through Celite. The filtrate was washed with 2 N NaOH
(3 Â 250 mL), brine (100 mL), and dried over MgSO4. The organic
solvent was then filtered, and removed by rotary evaporation to
yield a dark oil. Silica gel flash column chromatography, eluting
with 4:1 hexanes/EtOAc, provided 11.08 g (43.9%) of 29 as an oil
that crystallized upon standing, and which could be easily recrys-
tallized from EtOH: mp 65–69 °C. 1H NMR (CDCl3, 300 MHz): d 7.78
(s, 1H, CHNO2); 7.33–7.23 (m, 2H, 2ArH); 6.99–6.85 (m, 2H, 2ArH);
5.24 (s, 2H, OCH2). EIMS: m/z (relative intensity): 130 (MÀOH, –
NO, 100), 160 (MÀOH, 94), 177 (M+, 81). Anal. Calcd For
C9H7NO3: C, 61.02; H, 3.98; N, 7.91. Found: C, 60.88; H, 4.11; N,
7.90.
3.1.22. ( )-Trans-4-(3,4-dimethoxyphenyl)-3-nitrochroman (30)
To a solution of 4-bromoveratrole (25.73 g, 0.117 mol) in
150 mL of dry THF was added magnesium powder-50 mesh
(3.09 g, 0.125 mol), and the mixture was heated at reflux for 2 h
to form the Grignard reagent. The reaction mixture was then
cooled to À78 °C, and a solution of nitrochromene 29 (7.4 g,
0.042 mol) in 50 mL of dry THF was added drop-wise through an
addition funnel. After stirring for another 30 min, 100 mL of aque-
ous NH4Cl solution was added to quench the reaction. The mixture
was extracted with CH2Cl2 (3 Â 60 mL), the extract was washed
with brine, dried over MgSO4, filtered, and concentrated under
vacuum to provide a dark oil. Silica gel flash column chromatogra-
phy, eluting with 4:1 hexanes/EtOAc, afforded a yellow oil, which
was crystallized from MeOH gave 8.01 g (60.9%) of 30: mp 123–
126 °C. 1H NMR (CDCl3, 300 MHz): d 7.21–7.15 (m, 1H, ArH);
6.93–6.90 (m, 2H, 2ArH); 7.38 (d, J = 9 Hz, 1H, ArH); 6.64–6.60
(m, 2H, 2ArH); 4.91–4.83 (m, 2H, OCH2); 4.59 (ddd, J1 = 10.8 Hz,
J2 = 6.0 Hz, J3 = 1.5 Hz, 1H, CHNO2); 4.42 (dd, J1 = 10.8 Hz,
J2 = 3.0 Hz, 1H, ArCH); 3.84 (s, 3H, OCH3); 3.80 (s, 3H, OCH3). CIMS:
3.1.19. ( )-Cis-6,6a,7,8,9,13b-hexahydro-5H-benzo[d]naphtho[2,
1-b]azepine-11,12-diol hydrobromide (8a)
Amine 27a (40 mg, 0.129 mmol) was dissolved in 15 mL of
CH2Cl2 and cooled to À78 °C. Into this flask, 0.39 mL (0.39 mmol)
of a 1 M BBr3 solution in CH2Cl2 was added drop-wise. The reaction
was stirred at À78 °C for 2 h and another 4 h at room temperature.
The reaction mixture was then cooled to 0 °C and 5 mL of MeOH
were added drop-wise. The solvents were then removed under re-
duced pressure, and 5 mL of MeOH were once again added and re-
moved. The process was repeated one more time. The residue was
then recrystallized from EtOH to afford 20 mg (42.8%) of the hydro-
bromide salt of 8a: mp >300 °C. 1H NMR (D6MSO, 500 MHz): d 8.87
(s, 1H, OH); 8.76 (s, 1H, OH); 7.21–7.18 (m, 3H, 3ArH); 6.83 (br s,
1H, ArH); 6.62 (s, 1H, ArH); 5.91 (br s, 1H, ArH); 4.52 (s, 1H, 13b-
H); 3.64 (s, 1H, 6a-H); 3.16–3.15 (m, 3H, 8-H2, 9-CHa); 2.92–2.89
(m, 2 H, 5-H2); 2.67 (br s, 1H, 9-CHb); 1.86–1.85 (m, 1H, 6-Ha);
1.66–1.62 (m, 1H, 6-Hb). High res. ESIMS for C18C19NO2 (M+): calcd
282.1494, found 282.1493.
m
/z (relative intensity): 269 (M+H–NO2, 100), 316 (M+H+, 32).
Anal. Calcd for C17H17NO5: C, 64.75; H, 5.43; N, 4.44. Found: C,
64.62; H, 5.54; N, 4.47.
3.1.20. ( )-Trans-6,6a,7,8,9,13b-hexahydro-5H-benzo[d]naph
tho[2,1-b]azepine-11,12-diol hydrobromide (8b)
Amine 27b (300 mg, 0.97 mmol) was dissolved in 50 mL of
CH2Cl2 and cooled to À78 °C. Into this flask, 3.4 mL (3.4 mmol) of
a 1 M BBr3 solution in CH2Cl2 was added drop-wise. The reaction
was stirred at À78 °C for 2 h and another 4 h at room temperature.
The reaction mixture was then cooled to 0 °C and 5 mL of MeOH
were added drop-wise. The solvents were then removed under re-
duced pressure, and 5 mL of MeOH were once again added and re-
moved. The process was repeated one more time. The residue was
then recrystallized from EtOH to afford 300 mg (85.5%) of the
hydrobromide salt of 8b: mp >310 °C dec 1H NMR (CD3OD,
3.1.23. ( )-Trans-4-(3,4-dimethoxyphenyl)chroman-3-amine
(31)
Nitrochroman 30 (2.9 g, 9.2 mmol) was dissolved in 100 mL of
AcOH and 3.6 g (55.2 mmol) of powdered zinc were added. The
mixture was stirred vigorously at room temperature for 16 h, at
which time the reaction was filtered through Celite, and the
filtered solids were rinsed on the filter with AcOH. The filtrate
was concentrated to 1/3 of its volume and water (300 mL) was
added. The aqueous solution was then basified with conc ammo-

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5429
nium hydroxide and extracted with CH2Cl2 (3 Â 40 mL). The organ-
ic solution was then washed twice with brine, dried over MgSO4,
filtered, and the solvent removed under reduced pressure to pro-
vide 2.6 g (quant yield) of amine 31: mp 84–88 °C. 1H NMR (CDCl3,
300 MHz): d 7.17–7.11 (m, 1H, ArH); 6.91–6.81 (m, 4H, 4ArH);
6.72–6.65 (m, 2H, 2ArH); 4.23 (dd, J1 = 13.6 Hz, J2 4.1 Hz, 1H,
ArCH); 3.93–3.78 (m, 2H, OCH2); 3.88 (s, 3H, OCH3); 3.81 (s, 3H,
OCH3); 3.37–3.31 (m, 1H, CHNH2). CIMS: m/z (relative intensity):
286 (M+H+, 100), 369 (MÀNH3, 60). Anal. Calcd for C17H19NO3: C,
71.56; H, 6.71; N, 4.91. Found: C, 71.61; H, 6.82; N, 4.94.
tered, and the solvent removed under reduced pressure to yield
336 mg (87.7%) of amine 34: mp 106–109 °C. 1H NMR (CDCl3,
300 MHz): d 7.19 (dt, J1 = 7.5 Hz, J2 = 1.9 Hz, 1H, ArH); 7.13–7.08
(m, 1H, ArH); 6.99–6.93 (m, 2H, 2ArH); 6.70 (s, 1H, ArH); 6.22 (s,
1H, ArH); 4.44 (d, J = 7.8 Hz, 1H, 13b-H); 4.15 (dd, J1 = 10.3 Hz,
J2 = 4.0 Hz, 1H,6-Ha); 3.88 (s, 3H, OCH3); 3.61 (t, J = 10.3 Hz, 1H,6-
Hb); 3.56 (s, 3H, OCH3); 3.40–3.31 (m, 1H, 6a-H); 3.30–3.24 (m,
1H, 8-Ha); 3.05–2.97 (m, 1H, 8-Hb); 2.86–2.73 (m, 2H, 9-H2). CIMS:
m
/z 312 (M+H+, 100). Anal. Calcd for C19H21NO3: C, 73.29; H, 6.80;
N, 4.50. Found: C, 72.94; H, 6.79; N, 4.42.
3.1.24. ( )-Trans-2-chloro-N-(4-(3,4-dimethoxyphenyl)chro
man-3-yl)acetamide (32)
3.1.27. ( )-Trans-6,6a,7,8,9,13b-hexahydrobenzo[d]chromeno
[3,4-b]azepine-11,12-diol hydrobromide (9)
Amine 31 (500 mg, 1.75 mmol) was dissolved in 80 mL of
CH2Cl2, and powd Na2CO3 (0.74 g, 7.0 mmol) was added. The reac-
tion was cooled to 0 °C, and chloroacetyl chloride (0.6 g,
5.26 mmol) was added drop-wise to the reaction mixture. After
stirring for 16 h at room temperature, water was added to quench
the reaction. The mixture was extracted with CH2Cl2 (3 Â 30 mL),
the organic solvent was dried over MgSO4 and removed under
pressure to give the crude product, which upon recrystallization
with EtOH yielded 550 mg (87%) acetamide 32: mp 105–109 °C.
1H NMR (CDCl3, 300 MHz): d 7.25–7.20 (m, 1H, ArH); 7.00–6.91
(m, 4H, 4ArH); 6.83 (d, J = 1.8 Hz, 1H, ArH); 6.78 (d, J = 8.4 Hz,
1H, ArH); 6.56 (dd, J1 = 8.4 Hz, J2 = 2.1 Hz, 1H, ArH); 4.33–4.30
(m, 1H, ArCH); 4.16 (dd, J1 = 11.4 Hz, J2 = 1.5 Hz, 2H, OCH2); 4.08–
4.06 (m, 1H, CHNH); 4.04 (d, J = 1.2 Hz, 2H, CH2Cl); 3.85 (s, 6H,
2OCH3). CIMS: m/z (relative intensity): 362/364 (M+H+, 100). Anal.
Calcd for C19H20ClNO4: C, 63.07; H, 5.57; N, 3.87. Found: C, 63.04;
H, 5.62; N, 3.88.
Amine 34 (336 mg, 1.08 mmol) was dissolved in 40 mL of
CH2Cl2 and cooled to À78 °C. A 1 M BBr3 solution in CH2Cl2
(4.32 mL, 4.32 mmol) was then added drop-wise. This solution
was then warmed to room temperature and stirred overnight.
The reaction mixture was then cooled to 0 °C and 5 mL of MeOH
were added drop-wise. The solvents were removed under reduced
pressure, and 5 mL of MeOH were again added and removed. The
process was repeated one more time. The residue was then recrys-
tallized from MeOH/EtOAc and dried under vacuum to afford
147 mg (37.4%) of the hydrobromide salt of 9: mp >300 °C dec 1H
NMR (DMSO-d6, 300 MHz): d 9.11 (br s, 2 H, NH2+); 8.82 (s, 1H,
OH); 8.77 (s, 1H, OH); 7.26 (dt, J1 = 8.4 Hz, J2 = 2.1 Hz, 1H, ArH);
7.12–6.98 (m, 3H, 3ArH); 6.66 (s, 1H, ArH); 5.93 (s, 1H, ArH);
4.72 (d, J = 8.7 Hz, 1H, 13b-H); 4.34 (dd, J1 = 10.2 Hz, J2 = 4.5 Hz,
1H, 6-Ha); 3.72 (t, J1 = 10.8 Hz, 1H, 6-Hb); 3.63–3.55 (m, 1H, 6a-
H); 3.44–3.27 (m, 2H, 8-H2); 2.98–2.84 (m, 1H, 9-Ha); 2.48 (dd,
J1 = 15.6 Hz, J2 = 5.1 Hz, 1H, 9-Hb). ESIMS: m/z 284 (M+H+, 100).
Anal. Calcd for C17H18BrNO3: C, 56.06; H, 4.98; N, 3.85. Found: C,
55.70; H, 4.95; N, 3.68.
3.1.25. ( )-Trans-11,12-dimethoxy-6a,7,9,13b-tetrahydrobenzo
[d]chromeno[3,4-b]azepin-8(6H)-one (33)
A solution of chloroacetamide 32 (800 mg, 2.21 mmol) and
NaHCO3 (1.5 g, 17.86 mmol) in 500 mL of a mixture of 1:1
MeOH/H2O was irradiated for 55 min at room temperature with
a medium pressure mercury lamp, with cooling. The resulting solu-
tion was reduced in volume under vacuum, and partitioned be-
tween water and CH2Cl2. The aqueous solution was extracted
twice more with CH2Cl2. The pooled organic solvent was dried over
MgSO4, filtered, and the solvent was removed under reduced pres-
sure. The resulting crude product was recrystallized from EtOH to
give 220 mg (30.6%) of lactam 33: mp 220–224 °C. 1H NMR (CDCl3,
300 MHz): d 7.21–7.18 (m, 2H, 2ArH); 6.99 (dt, J1 = 9.0 Hz,
J2 = 1.2 Hz, 2H, 2ArH); 6.78 (s, 1H, ArH); 6.52 (d, 1H, ArH); 6.18
(br s, 1H, NH); 4.78 (d, J = 10.8 Hz, 1H, 13b-H); 4.3–4.30, 3.58
(ABq, JAB = 16.2 Hz, 2H, 9-H2); 4.17 (dd, J1 = 10.2 Hz, J2 = 3.9 Hz,
1H, 6-Ha); 3.91 (t, J = 10.8 Hz, 1H, 6a-H); 3.89 (s, 3H, OCH3); 3.70
(dd, J1 = 11.1 Hz, J2 = 4.3 Hz, 1H, 6-Hb); 3.67 (s, 3H, OCH3). EIMS:
m/z 325 (M+, 100). Anal. Calcd for C19H19NO4 Á0.11H2O: C, 69.71;
H, 5.92; N, 4.28. Found: C, 69.69; H, 6.22; N, 4.18.
3.1.28. 3-Nitro-2H-thiochromene (37)
A solution of thiosalicylaldehyde 36 (24.3 g, 0.176 mol) in
300 mL of toluene containing di-n-butyl amine (8.1 mL,
0.044 mol) and phthalic anhydride (14.3 g, 0.097 mol) was placed
into a three-necked flask equipped with a Dean-Stark trap, con-
denser, and mechanical stirring. The mixture was heated at reflux
with vigorous stirring and then 2-nitroethanol (6.2 mL, 0.088 mol)
was added very slowly into the reaction flask using a syringe pump
set at a rate of approximately 0.5 mL/h (about 12 h). After addition
of 2-nitroethanol was completed, the reaction was stirred at reflux
for another 12 h. The reaction mixture was cooled to room temper-
ature and filtered through Celite. The filtrate was washed with 2 N
NaOH (3 Â 200 mL), brine (60 mL), and dried over MgSO4. The or-
ganic solvent was then filtered and concentrated under reduced
pressure to yield a dark oil. Silica gel flash column chromatogra-
phy, eluting with 4:1 hexanes/EtOAc, provided a red oil, which
was crystallized from EtOAc/Hexane to give 10.2 g (60%) of 37:
mp 37–41 °C. 1H NMR (CDCl3, 300 MHz): d 7.91 (s, 1H, CHNO2);
7.46–7.18 (m, 4H, 4ArH); 4.10 (s, 2H, SCH2). CIMS: m/z (relative
intensity): 194 (M+H+, 100), 147 (MÀNO2, 26). Anal. Calcd For
C9H7NO2S: C, 55.94; H, 3.65; N, 7.25. Found: C, 56.02; H, 3.70; N,
7.20.
3.1.26. ( )-Trans-11,12-dimethoxy-6,6a,7,8,9,13b-hexahydro
benzo[d]chromeno[3,4-b]azepine (34)
Lactam 33 (400 mg, 1.23 mmol) was suspended in 100 mL of
dry THF and 7.83 mL (7.38 mmol) of a 1 M solution of BH3 in THF
was added. This solution was stirred at reflux for 24 h, cooled
and quenched with excess MeOH, and the solvent was removed
under reduced pressure. The residue was then stirred for 4 h at re-
flux with 50 mL of 6 N ethanolic HCl solution. The solvent was re-
moved and MeOH was added and removed under vacuum 3 times.
The crude product was dissolved and EtOAc and extracted with
aqueous 2 N HCl solution (3 Â 60 mL). The combined aqueous layer
was basified with 2 N NaOH solution and extracted with CH2Cl2
(3 Â 40 mL). The organic solution was then dried over MgSO4, fil-
3.1.29. ( )-Trans-4-(3,4-dimethoxyphenyl)-3-nitrothiochroman
(38)
To a solution of 4-bromoveratrole (13.5 g, 0.062 mol) in 80 mL
of dry THF was added magnesium powder (mesh?) (2.0 g,
0.082 mol), and the mixture was heated at reflux for 2 h to form
the Grignard reagent. The reaction mixture was then cooled to
À78 °C, and a solution of thionitrochromene 37 (4.0 g, 0.021 mol)
in 30 mL of dry THF was added drop-wise through an addition fun-
nel. After stirring for another 30 min, 100 mL of aqueous NH4Cl

5430
A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
solution was added to quench the reaction. The mixture was ex-
tracted with CH2Cl2 (3 Â 60 mL), the extract was washed with
brine, dried over MgSO4, filtered, and the solvent removed to give
a dark oil. Silica gel flash column chromatography, eluting with 4:1
hexanes/EtOAc, provided an yellow oil, which crystallized from
MeOH gave 3.0 g (43.7%) of 38: mp 137–140 °C. 1H NMR (CDCl3,
between water and CH2Cl2. The aqueous solution was extracted
twice more with CH2Cl2. The pooled organic extract was dried over
MgSO4, filtered, and concentrated under reduced pressure. Silica
gel flash column chromatography, eluting with 1:9 hexanes:EtOAc,
provided a yellow oil that crystallized from EtOH to afford 43 mg
(6.8%) of lactam 41: mp >240 °C. 1H NMR (DMSO-d6, 300 MHz): d
7.70 (d, J = 3.0 Hz, 1H, ArH); 7.30 (dd, J1 = 6.6 Hz, J2 = 0.9 Hz, 1H,
ArH); 7.16 (dt, J1 = 7.5 Hz, J2 = 1.2 Hz, 1H, ArH); 7.16 (dt,
J1 = 7.5 Hz, J2 = 1.2 Hz, 1H, ArH); 6.92 (s, 1H, ArH); 6.63 (s, 1H,
ArH); 4.28–4.19 (m, 1H, 13b-H); 3.87–3.82 (m, 2H, 6a-H, 9-Ha);
3.80 (s, 3H, OCH3); 3.63 (s, 3H, OCH3); 3.23 (dd, J1 = 11.1 Hz,
J2 = 5.7 Hz, 2H, 6-Ha, 9-Hb); 3.12 (t, J = 11.1 Hz, 1H, 6-Hb). CIMS:
m/z (relative intensity): 342 (M+H+, 100). Anal. Calcd For
C19H19NO3SÁ0.45H2O: C, 65.29; H, 5.74; N, 4.01. Found: C, 65.29;
H, 5.68; N, 3.97.
300 MHz):
d
7.20–7.11 (m, 2H, 2ArH); 7.05 (dt, J1 = 8.0 Hz,
J2 = 2.0 Hz, 1H, ArH); 6.90 (d, J = 8.0 Hz, 1H, ArH); 6.82 (d,
J1 = 8.0 Hz, 1H, ArH); 6.67 (dd, J1 = 8.0 Hz, J2 = 2.0 Hz, 1H, ArH);
6.62 (d, J = 2.0 Hz, 1H, ArH); 5.17 (dt, J1 = 9.0 Hz, J2 = 3.0 Hz, 1H,
CHNO2); 4.79 (d, J = 9.0 Hz, 1H, ArCH); 3.87 (s, 3H, OCH3); 3.81
(s, 3H, OCH3); 3.61 (dd, J1 = 12.9 Hz, J2 = 8.6 Hz, 1H, SCHa); 3.61
(dd, J1 = 12.9 Hz, J2 = 3.4 Hz, 1H, SCHb). EIMS: m/z (relative inten-
sity): 313 (M+, 100). Anal. Calcd For C17H17NO4S: C, 61.61; H,
5.17; N, 4.23. Found: C, 61.36; H, 5.35; N, 4.19.
3.1.30. ( )-Trans-4-(3,4-dimethoxyphenyl)thiochroman-3-
amine (39)
3.1.33. ( )-Trans-11,12-dimethoxy-6,6a,7,8,9,13b-hexahydro
benzo[d]thiochromeno[3,4-b]azepine (42)
Thionitrochroman 38 (3.69 g, 11.13 mmol) was dissolved in
150 mL of AcOH and 4.4 g (66.8 mmol) of powdered zinc were
added. The mixture was stirred vigorously at room temperature
for 16 h, at which time the reaction was filtered through Celite,
and the filtered solids were rinsed on the filter with AcOH. The fil-
trate was concentrated to 1/3 of its volume and water (300 mL)
was added. The aqueous solution was then basified with conc
ammonium hydroxide and extracted with CH2Cl2 (3 Â 40 mL).
The organic extract was then washed twice with brine, dried over
MgSO4, filtered, and the solvent removed under reduced pressure
to give 3.35 g (quant yield) of amine 39: mp (HCl salt) 228–
230 °C. 1H NMR (CDCl3, 300 MHz): d 7.21–7.10 (m, 2H, 2ArH);
7.00–7.90 (m, 2H, 2ArH); 6.79 (d, J = 8.1 Hz, 1H, ArH); 6.63 (d,
J = 1.5 Hz, 1H, ArH); 6.56 (dd, J1 = 8.4 Hz, J2 = 1.4 Hz, 1H, ArH);
3.97–3.88 (m, 1H, ArCH); 3.85 (s, 3H, OCH3); 3.81 (s, 3H, OCH3);
3.60–3.51 (m, 1H, CHNH2); 3.11 (dd, J1 = 12.6 Hz, J2 = 2.4 Hz, 1H,
SCHa); 2.76 (dd, J1 = 12.6 Hz, J2 = 6.0 Hz, 1H, SCHb). CIMS: m/z (rel-
ative intensity): 302 (M+H+, 100). Anal. Calcd For C17H20NO2SÁHCl:
C, 60.43; H, 5.97; N, 4.15. Found: C, 60.21; H, 5.95; N, 4.22.
Lactam 41 (120 mg, 0.35 mmol) was suspended in 50 mL of dry
THF and 3.51 mL (3.51 mmol) of a 1 M solution of BH3 in THF was
added. This solution was stirred at reflux for 24 h, then cooled and
quenched with excess MeOH. The solvent was removed under re-
duced pressure and the residue was then stirred for 4 h at reflux
with 30 mL of 6 N ethanolic HCl solution. The solvent was removed
and MeOH was added and removed under vacuum 3 times. The
crude product was dissolved and EtOAc and extracted with aque-
ous 2 N HCl solution (3 Â 30 mL). The combined aqueous layer
was basified with 2 N NaOH solution and extracted with CH2Cl2
(3 Â 20 mL). The organic solution was then dried over MgSO4, fil-
tered, and the solvent removed under reduced pressure to yield
80 mg (70%) of amine 42: mp (HCl salt) 170–190 °C. 1H NMR
(CDCl3, 300 MHz): d 7.31–7.03 (m, 4H, 4ArH); 6.76 (s, 1H, ArH);
6.25 (s, 1H, ArH); 5.06 (d, J = 8.4 Hz, 1H, 13b-H); 4.90 (dd,
J1 = 8.6 Hz, J2 = 4.3 Hz, 1H, 6a-H); 4.22 (t, J = 10.8 Hz, 1H, 6-Ha);
3.93 (s, 3H, OCH3); 3.87–3.83 (m, 2H, 6-Hb, 8-Ha); 3.60 (s, 3H,
OCH3); 3.55–3.45 (m, 1H, 8-Hb); 3.05–2.98 (m, 2H, 9-H2). ESIMS:
m/z (relative intensity): 328 (M+H+, 100). Anal. Calcd For
C
19H22ClNO2SÁ1.26H2O: C, 59.03; H, 6.39; N, 3.62. Found: C,
3.1.31. ( )-Trans-2-chloro-N-(4-(3,4-dimethoxyphenyl)thiochro
man-3-yl)acetamide (40)
59.01; H, 6.61; N, 3.62.
Amine 39 (6.78 g, 22.5 mmol) was dissolved in 300 mL of
CH2Cl2, and powd Na2CO3 (9.5 g, 90.0 mmol) was added. The reac-
tion was cooled to 0 °C with stirring, and chloroacetyl chloride
(7.6 g, 67.5 mmol) was added drop-wise to the reaction mixture.
After stirring for 16 h at room temperature, water was added to
quench the reaction. The mixture was extracted with CH2Cl2
(3 Â 30 mL), the organic solution was dried over MgSO4, filtered,
and concentrated under reduced pressure to give a dark oil, which
upon crystallization from MeOH yielded 8.06 mg (94.8%) of acet-
amide 40: mp 94–97 °C. 1H NMR (CDCl3, 300 MHz): d 7.23–7.16
(m, 2H, 2ArH); 7.08–7.98 (m, 2H, 2ArH); 6.76 (d, J = 8.1 Hz, 1H,
ArH); 6.72 (d, J = 2.1 Hz, 1H, ArH); 6.48 (dd, J1 = 8.7 Hz,
J2 = 2.4 Hz, 1H, ArH); 4.65–4.58 (m, 1H, ArCH); 4.31–4.29 (m, 1H,
CHNH); 4.03 (d, J = 1.8 Hz, 1H, CH2Cl); 3.84 (s, 3H, OCH3); 3.83 (s,
3H, OCH3); 3.11 (dd, J1 = 6.7 Hz, J2 = 2.4 Hz, 1H, SCHa); 2.76 (dd,
J1 = 6.7 Hz, J2 = 1.1 Hz, 1H, SCHb). CIMS: m/z (relative intensity):
378 (M+H+, 35) 285 (M+-NH2COCH2Cl, 25), 240 (M+-2(OCH3)Ph,
100). Anal. Calcd For C19H20ClNO3SÁ0.38H2O: C, 59.32; H, 5.44; N,
3.64. Found: C, 59.32; H, 5.40; N, 3.63.
3.1.34. ( )-Trans-6,6a,7,8,9,13b-hexahydrobenzo[d]thiochro
meno[3,4-b]azepine-11,12-diol hydrobromide (10)
Amine 42 (50 mg, 0.153 mmol) was dissolved in 20 mL of
CH2Cl2 and cooled to À78 °C. Into this flask, 0.54 mL (0.54 mmol)
of a 1 M BBr3 solution in CH2Cl2 was added drop-wise. This solu-
tion was then warmed to room temperature and stirred overnight.
The reaction mixture was then cooled to 0 °C and 5 mL of MeOH
were added drop-wise. The solvents were removed under reduced
pressure, and 5 mL of MeOH were once again added and removed.
The process was repeated one more time. The residue was then
recrystallized from MeOH/EtOAc and dried under vacuum to afford
30 mg (51.7%) of the hydrobromide salt of 10: mp >300 °C dec 1H
NMR (DMSO-d6, 500 MHz): d 8.96 (br s, 2H, NH2+); 8.75 (s, 1H,
OH); 8.67 (s, 1H, OH); 7.31 (d, J = 7.3 Hz, 1H, ArH); 7.20 (t,
J = 4.5 Hz, 1H, ArH); 7.15 (d, J = 4.5 Hz, 1H, ArH); 7.05 (d,
J = 4.5 Hz, 1H, ArH); 6.61 (s, 1H, ArH); 5.56 (s, 1H, ArH); 4.71–
4.69 (m, 1H, 13b-H); 3.51 (dd, J1 = 11.9 Hz, J2 = 1.2 Hz, 1H, 6a-H);
3.45–3.35 (m, 2H, 6-H2); 3.05 (d, J = 11.5 Hz, 1H, 9-Ha); 2.91 (t,
J = 12.5 Hz, 1H, 8-Ha); 2.81 (t, J = 12.5 Hz, 1H, 8-Ha); 2.73 (dd,,
J1 = 15.5 Hz, J2 = 4.5 Hz, 1H, 9-Hb). ESIMS: m/z 300 (M+H+, 100).
3.1.32. ( )-Trans-11,12-dimethoxy-6a,7,9,13b-tetrahydrobenzo
[d]thiochromeno[3,4-b]azepin-8(6H)-one (41)
A solution of chloroacetamide 40 (700 mg, 1.85 mmol) and
NaHCO3 (1.5 g, 17.9 mmol) in 500 mL of a mixture of 1:1 MeOH:
H2O was irradiated for 55 min at room temperature with a medium
pressure mercury lamp, with cooling. The resulting solution was
concentrated under reduced pressure, and the residue partitioned
3.2. Pharmacology
3.2.1. Materials
Radioligands used for this study were [3H]SCH23390 (73.1 Ci/
mmol), [3H]-N-methylspiperone (75 Ci/mmol), and [3H] cyclic

A. H. Clark et al. / Bioorg. Med. Chem. 19 (2011) 5420–5431
5431
AMP (30 Ci/mmol) and were purchased from PerkinElmer Life
Sciences (Massachusetts, United States). Test ligands used were
chlorpromazine, SKF38393, SCH23390, (+)-butaclamol, dopamine
hydrochloride and all were purchased from Sigma–Aldrich Chemical
Company (St. Louis, MO, United States). SCH 39166 and ketanserin
were purchased from Tocris Bioscience (Ellisville, MO, United
States). DHX was previously synthesized in our laboratory.16 Striatal
tissue used for competition binding experiments was dissected from
porcine brain tissue obtained from Purdue Butcher Block and
prepared as described previously.17
scint-O was added to each well, and radioactivity was counted using
a Packard Topcount scintillation counter. Standard curves ranging
from 0.01 to 300 pmol of cAMP were used to estimate the concentra-
tion of cAMP in each cell lysate sample.
3.2.6. Data analysis
Nonlinear regression for radioligand displacement and dose–re-
sponse curves was performed using GraphPad Prism, version 4.00
for Windows (GraphPad Software, San Diego California USA,
www.graphpad.com.) For radioligand displacement, Hill slopes
were fixed and the bottoms of curves were set to nonspecific bind-
ing values to generate IC50 values for test compounds. Ki values
were calculated by the Cheng-Prusoff equation using the radioli-
gand concentration and previously established porcine striatal Kd
values of 0.44 and 0.075 nM for D1- and D2-like binding, respec-
tively. For cAMP dose–response curves, nonlinear regression sig-
moidal dose–response analysis (variable slope) was used to
generate EC50 values. Intrinsic activity (IA) was calculated as per-
cent maximum response of cAMP accumulation compared to
3.2.2. Competition binding experiments
Competition binding experiments were done with porcine stria-
tal tissue and utilized 1.5 nM [3H]SCH23390 for D1-like binding and
0.2 nM [3H]-N-methylspiperone for D2-like binding. For D2-like
binding, 50 nM ketanserin was included to block native 5-HT2A
receptors. All experiments were performed with receptor binding
buffer (50 mM HEPES, 4 mM MgCl2, pH 7.4) and used 96 well assay
tubes with drug dilutions, radioligand, and 75
membrane protein per tube. Nonspecific binding was defined in
the presence of 5 M (+)-butaclamol. All experiments were incu-
lg porcine striatal
10 lM dopamine.
l
bated at 37 °C for 30 min and terminated by rapid filtration with a
3.3. Molecular modeling
96-well Packard Filtermate cell harvester with ice cold wash buffer
(10 mM Tris, 0.9% NaCl). Filter plates were dried and 40
l
L of Pack-
Molecules were minimized as the protonated species in a vac-
uum using Spartan ’10 for Windows, version 1.0.1 (Wavefunction
Inc., Irvine, CA). All possible conformers of each molecule were
manually constructed and geometry optimized using the semi-
empirical AM-1 potentials in Spartan. The lowest energy structure
obtained for each molecule was then selected and geometry opti-
mized using Hatree-Fock, with 6-31G⁄ force fields.
ard Microscint-O was added to each filter well. Radioactivity was
counted as counts per minute (CPM) using a Packard Topcount scin-
tillation counter.
3.2.3. Cell line and culture methods
Human embryonic kidney cells (HEK 293) stably transfected
with pcDNA V5 His TOPO subcloned with the human dopamine
D1 receptor (hD1) were used to assess cAMP accumulation. The
hD1 receptor expressed at approximately 1800 fmol/mg protein,
as determined by saturation binding with [3H]SCH23390. Cells
were grown in DMEM supplemented with 5% fetal clone serum,
Acknowledgments
This work was funded by NIH Grants MH42705 (D.E.N.) and
MH60397 (V.J.W).
5% bovine calf serum, 100 U/mL Penicillin, 100
lg/mL streptomy-
cin, 0.25 g/mL amphotericin B, and maintained in 100
l
l
g/mL
References and notes
G418 for selection of hD1. Cells were grown at 37 °C in a humidi-
fied incubator with 5% CO2.
1. Huang, X.; Lawler, C. P.; Lewis, M. M.; Nichols, D. E.; Mailman, R. B. Int. Rev.
Neurobiol. 2001, 48, 65.
2. Missale, C.; Nash, S. R.; Robinson, S. W.; Jaber, M.; Caron, M. G. Physiol. Rev.
1998, 78, 189.
3.2.4. Cyclic AMP accumulation assay
3. Hurley, M. J.; Jenner, P. Pharmacol. Ther. 2006, 111, 715.
4. Nutt, D.; Lingford-Hughes, A. Brit. J. Pharmacol. 2008, 154, 397.
5. Meltzer, H. Y.; Stahl, S. M. Schizophr. Bull. 1976, 2, 19.
6. Zhang, J.; Xiong, B.; Zhen, X.; Zhang, A. Med. Res. Rev. 2008, 29, 272.
7. Zhang, A.; Neumeyer, J. L.; Baldessarini, R. J. Chem. Rev. 2007, 107, 274.
8. Taylor, J. R.; Lawrence, M. S.; Redmond, D. E., Jr.; Elsworth, J. D.; Roth, R. H.;
Nichols, D. E.; Mailman, R. B. Eur. J. Pharmacol. 1991, 199, 389.
9. Hersi, A. I.; Rowe, W.; Gaudreau, P.; Quirion, R. Neuroscience 1995, 69,
1067.
10. Arnsten, A. F.; Cai, J. X.; Murphy, B. L.; Goldman-Rakic, P. S. Psychopharmacology
(Berl) 1994, 116, 143.
11. Riggs, R. M.; McKenzie, A. T.; Byrn, S. R.; Nichols, D. E.; Foreman, M. M.; Truex,
L. L. J. Med. Chem. 1987, 30, 1914.
Cells were seeded into 96 wells and allowed to grow to 90–95%
confluency until the day of the assay. All drug dilutions were made
in EBSS assay buffer (Earle’s balanced salt solution containing 2%
bovine calf serum, 0.025% ascorbic acid, and 15 nM HEPES pH
7.4). First, all media was decanted and 500 lM isobutylmethylxan-
thine (IBMX) in EBSS buffer was added to all wells. Next, drug dilu-
tions were added in duplicate, and allowed to incubate for 15 min
in a 37 °C water bath. Cyclic AMP accumulation assays were termi-
nated by decanting the media followed by addition of 3% trichloro-
acetic acid. Plates were stored at 4 °C for at least 24 h until
quantification of cyclic AMP using the cyclic AMP binding assay.
12. 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.
13. Snyder, S. E.; viles-Garay, F. A.; Chakraborti, R.; Nichols, D. E.; Watts, V. J.;
Mailman, R. B. J. Med. Chem. 1995, 38, 2395.
14. Berger, J. G.; Chang, W. K.; Clader, J. W.; Hou, D.; Chipkin, R. E.; McPhail, A. T. J.
Med. Chem. 1989, 32, 1913.
15. Iorio, L. C.; Barnett, A.; Billard, W.; Gold, E. H. Adv. Exp. Med. Biol. 1986, 204, 1.
16. 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.
17. Cueva, J. P.; Giorgioni, G.; Grubbs, R. A.; Chemel, B. R.; Watts, V. J.; Nichols, D. E.
J. Med. Chem. 2006, 49, 6848.
18. Ozaki, S.; Adachi, M.; Sekiya, S.; Kamikawa, R. J. Org. Chem. 2003, 68, 4586.
19. Toste, F. D.; Lough, A. J.; Still, I. W. J. Tetrahedron Lett. 1995, 36, 6619.
20. Watts, V. J.; Neve, K. A. Mol. Pharmacol. 1996, 50, 966.
3.2.5. Cyclic AMP binding assay
Cyclic AMP accumulation assays were quantified using a previ-
ously described protocol.20 Cellular lysate was added to 96 well
assay tubes containing [3H]cAMP (1 nM final concentration) and bo-
vine adrenal gland cAMP binding protein (100–150 lg in 500 lL of
buffer) diluted in cAMP binding buffer (100 mM Tris–HCl, pH 7.4,
100 mM NaCl, 5 mM EDTA). Assay tubes were allowed to incubate
on ice at 4 °C for 2 h, and terminated by harvesting with ice-cold
wash buffer (10 mM Tris, 0.9% NaCl) using a 96-well Packard Filter-
mate cell harvester. Filter plates were dried, 40 lL of Packard Micro-