An efficient synthesis of the potent dopamine D1 agonist dinapsoline by construction and selective reduction of 2′-azadimethoxybenzanthrone

Article abstract of DOI:10.1055/s-2001-10809

8,9-Dihydroxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-de]isoquinoline (dinapsoline, 1) is a potent dopamine D₁ receptor agonist with potential antiparkinsonian activity. A new synthesis was developed with the fully aromatic compound 2 as the key i

Full text of DOI:10.1055/s-2001-10809

262
PAPER
An Efficient Synthesis of the Potent Dopamine D1 Agonist Dinapsoline by
Construction and Selective Reduction of 2’-Azadimethoxybenzanthrone
Tim Sattelkau, Amjad M. Qandil, David E. Nichols*
Department of Medicinal Chemistry and Molecular Pharmacology, School of Pharmacy and Pharmacal Sciences, Purdue University,
West Lafayette, IN 47907, USA
Fax +1(765)4941414; E-mail: drdave@pharmacy.purdue.edu
Received 26 July 2000; revised 25 September 2000
the original synthesis1 and its improved modification2 is
Abstract: 8,9-Dihydroxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-
thereby circumvented, and the convenience of the well-es-
de]isoquinoline (dinapsoline, 1) is a potent dopamine D1 receptor
agonist with potential antiparkinsonian activity. A new synthesis
tablished chemistry of the benzanthrones3 is utilized.
was developed with the fully aromatic compound 2 as the key inter-
The skeleton of dinapsoline (1) resembles that of the
mediate. The synthesis herein described is suitable for a larger scale
benzanthrones. Our newly envisioned approach relied on
preparation of dinapsoline compared to the previously known meth-
the construction of the carbon framework in a highly oxi-
ods. Furthermore, the unproductive protection/deprotection step of
dized state with subsequent selective reduction to the de-
the nitrogen is circumvented by maintaining a high oxidation state
sired functionalities. The aromatic compound 2, a 2’-
azabenzanthrone, was therefore chosen as the primary
synthetic goal. Although only one example for a com-
pound with a ring system like 2 is reported in the litera-
ture,4 the chemistry of benzanthrones and related
compounds was well established in the 1930s, when these
compounds were industrially used as precursors for dyes.3
That knowledge, combined with modern Pd(0)-mediated
methods for the construction of biaryls, was utilized to de-
velop an elegant synthesis of the tetracyclic molecule.
of the isoquinoline moiety throughout the synthesis. The construc-
tion of the framework was accomplished by Friedel-Crafts acyl-
ation and a Suzuki cross-coupling reaction between the commerc-
ially available 4-bromoisoquinoline and aryl boronic acid 5, the lat-
ter demanding the transformation of the lithiation-directing amide
back to a carboxylic acid functionality. The selective reduction was
carried out stepwise with sodium borohydride and sodium cy-
anoborohydride. The new synthesis is high yielding and reduces the
number of transformations in the previously reported methods.
Key words: regioselective reduction, boron, Suzuki cross-cou-
pling, electrophilic aromatic substitution, acylation, hydrogenation,
amide hydrolysis, cyclization
Conversion of 2,3-dimethoxybenzoic acid to its N-meth-
yl-tert-butylamide 4 protected the carbonyl function as
well as allowing functionalization in the ortho position by
the directing properties of this amide in lithiation reac-
tions. Metalation and lithium-boron exchange5 provided
the boronic acid 5 in excellent yields (Scheme). Pd-cata-
lyzed Suzuki cross-coupling with commercially available
4-bromoisoquinoline afforded 6a.2,6,7 This biaryl has con-
siderable steric bulk, as evidenced by its 1H NMR spectra,
in which two distinct atropisomers are evident.
We have previously reported the synthesis and pharmaco-
logical evaluation of 8,9-dihydroxy-2,3,7,11b-tetrahydro-
1H-naphth[1,2,3-de]isoquinoline (dinapsoline, 1),1,2 a po-
tent full dopamine D1 agonist containing a rigid b-phenyl-
dopamine pharmacophore. Meanwhile, very promising,
though so far unpublished data from pharmacological
studies with this substance made it desirable to develop a
more straightforward procedure for a large scale prepara-
tion of dinapsoline (1).
Earlier attempts to hydrolyze the analogous diethylamide
(6, R1 = R2 = Et) prepared by essentially the same method,
were not successful. Indeed, the recalcitrant nature of
N,N-dialkylbenzamides is well recognized, and especially
2,6-disubstituted secondary benzamides are essentially
inert to hydrolysis.8-10 We therefore employed the meth-
odology of Reitz,9 using the N-methyl-tert-butylamide as
a directing group, which can be cleaved stepwise under
relatively mild conditions. Deprotection of amide 6a
was achieved in high overall yield by first cleaving off the
t-butyl with trifluoroacetic acid, giving methylamide 6b,
followed by nitrosation and hydrolysis of the N-nitros-
amide to give the free carboxylic acid 7. This acid is best
isolated and stored as its hydrochloride or trifluoroacetate,
due to the hygroscopic character of the amphoteric com-
pound.
The original synthesis was rather long and, due to the cen-
tral ring-closure step that proved to be unsuitable for
scaleup, only small quantities could be obtained. We re-
port herein an alternative formal synthesis of dinapsoline
(1) using a different strategy for the construction of the
carbon framework, employing the aromatic intermediate
2. The unproductive protection/deprotection sequence of
H3CO
O
OH
H3CO
HO
H
Formation of 7 was always accompanied by small
amounts of its ethyl ester. This byproduct, however, could
easily be hydrolyzed to the acid. Although this deprotec-
N
N
H
1
2
Synthesis 2001, No. 2, 262–266 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Dinapsoline
263
Br
OCH3
OCH3
OCH3
OCH3
O
OCH3
OCH3
1. HONO2
2. KOH, EtOH,
-CH2N2
1. BuLi, TMEDA
2. B(OCH3)3
N
OCH3
OCH3
COOH
NR1R2
O
C
O
C
Pd(PPh3)4
2M Na2CO3
(89% from 6a)
(99%)
N
N
N
(OH)2B
N
(80%)
6a: R1 = Me, R2 = t-Bu
6b: R1 = Me, R2 = H
TFA
4
5
7
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
O
NaBH4, TFA
ClSO3H (neat)
(80%)
NaCNBH3
75%
9a,b
+
N
N
HN
Silica Oxidation
10, 75%
2
11
Scheme
tion protocol is high yielding and suitable for a large scale, accomplish the transformation of both moieties in one
it involves the generation of diazomethane in the final step by means of catalytic hydrogenation. Under a variety
step, which demands the use of special glassware and pre- of conditions however, ring B was usually the target of at-
cautions. In attempts to circumvent this problem, we also tack and partially saturated. For example, hydrogenation
prepared an oxazoline protected analogue of boronic acid with PtO2 in acetic acid yielded the azaphenanthrene 8,
5 from 2,3-dimethoxybenzoic acid (3) by brominating while Pd/C in ethanol led to a mixture of the isomers 9a,b.
with dibromodimethylhydantoin (DBDMH),11 formation
of the oxazoline by fusing with the appropriate amino al-
cohol,12 halogen-metal exchange, and quenching with
trimethyl borate. This compound however, in our hands
failed to give satisfactory results in the cross-coupling re-
action with 4-bromoisoquinoline, probably due to the
even greater steric bulk of the projected biaryl, compared
to 6a.
OCH3
OCH3
OCH3
OCH3
OH
OCH3
OH
OCH3
OH
N
N
N
In another approach to prepare carboxylic acid 7, we at-
tempted to switch the reactivities of the two partners in the
cross-coupling reaction. There are two literature reports of
4-boronated isoquinolines that had been employed in a
Suzuki reaction. Nevertheless, our attempts to couple ei-
8
9a
9b
Although it is usually possible to control the regioselectiv-
ity of the hydrogenation of isoquinolines,17 no catalytic
reduction of the heterocyclic ring could be achieved by
this means. We therefore examined selective methods for
biarylketone reduction. The action of triethylsilane in
TFA18 on 2 afforded one isomer of 9 (the regiochemistry
of the double bond was not elucidated), while sodium
borohydride/AlCl3 reduction in THF19 afforded no isol-
able products. Finally, the method of Gribble,20 using
NaBH4 in TFA yielded the desired benzanthrene 10, but
this method also formed small amounts of 9. Fortuitously,
during chromatographic workup 9 was found to reoxidize
back to the starting material 2, which could be recycled.
The effective yield of 10 was therefore 75% (Scheme).
ther
diethyl(4-isoquinolyl)borane13
or
(4-iso-
quinolyl)boronic acid14 with methyl 2-bromo-5,6-
dimethoxybenzoate15 under the reported conditions failed.
The intramolecular Friedel-Crafts acylation as a ring-
closing step is well established in the chemistry of an-
thraquinones and benzanthrones,3 and we found the reac-
tion conditions for a similar carbocyclic framework16
readily applicable to heterocyclic acid 7. Although the re-
action conditions are rather harsh, ketone 2 could never-
theless be obtained in high yields and purity. Trials to
utilize oleum as the cyclization reagent or the use of an or-
ganic solvent gave inferior results.
The conversion of 2 to 1 demanded the selective reduction
of both the biarylketone and the heterocyclic ring of the
isoquinoline. In our initial trials to reduce 2, we hoped to
Dinapsoline dimethyl ether (11) was obtained by further
reduction of 10 with sodium cyanoborohydride. This
Synthesis 2001, No. 2, 262–266 ISSN 0039-7881 © Thieme Stuttgart · New York

264
T. Sattelkau et al.
PAPER
(2.50 g), toluene (250 mL), EtOH (120 mL), and 2 M Na2CO3 solu-
tion (100 mL, 200 mmol) were heated at reflux under Ar for 6 h. Af-
ter cooling, the mixture was separated, the organic phase washed
once with brine, and the combined aqueous phases were extracted
with toluene. The combined organic phases were washed with H2O
and filtered through a pad of Celite. After evaporation 33.8 g of
crude product was obtained that was recrystallized from Et2O/
CH2Cl2/hexanes to give 17.70 g of product in the first crop and an
additional 6.21 g upon evaporation of the mother liquors and a sec-
ond crystallization. The combined yield was 23.91 g (80%); mp 136
∞C. An analytical sample was recrystallized from Et2O; mp 138 ∞C.
1H NMR (300 MHz, CDCl3): d = 9.20 (br s, 1 H, Ar-H), 8.61 and
8.27 (2 s, 1 H, Ar-H), 7.96 and 7.93 (m and d, 1 H, J = 7.2 Hz, Ar-
H), 7.75-7.55 (m, 3 H, Ar-H), 7.02 (m, 2 H, Ar-H), 3.93 (s, 3 H,
OCH3), 3.91 and 3.90 (2 s, 3 H, OCH3), 2.75 and 2.37 (2 s, 3 H,
NCH3), 1.00 and 0.98 (2 s, 9 H, t-C4H9).
compound was identical to the product prepared by the
original method. Demethylation of 11 with BBr3 gave di-
napsoline (1) as described previously.1
Melting points were determined on a Thomas-Hoover melting point
apparatus and are uncorrected. 1H NMR spectra were recorded on a
Bruker ARX 300 MHz spectrometer. Chemical shifts are reported
in d values (ppm) relative to an internal standard of TMS in CDCl3.
High resolution CI and EI mass spectra were obtained within 0.0015
amu. The ionization gas for CIMS and High Resolution CIMS was
isobutane. Microanalyses were obtained from the Purdue Microan-
alytical Laboratory. Solvents and reagents were used as purchased
unless noted otherwise.
2,3-Dimethoxy-N-tert-butyl-N-methylbenzamide (4)
MS (CI): m/z (%) = 379 (M+ + 1, 100%).
A solution of 2,3-dimethoxybenzoic acid (3; 24.75 g, 136 mmol) in
CH2Cl2 (500 mL) was cooled by means of an external ice-bath and
oxalyl chloride (30 mL) was added dropwise over 10 min, followed
by DMF (2.0 mL). The mixture was stirred for 70 min at 0 ∞C and
then 2 h at r.t. The solution was then evaporated to dryness (40 ∞C
bath temperature) and dried in vacuo to give 27.5 g (quant.) of a sol-
id. A part of this crude acid chloride (8.32 g, 41.5 mmol) was dis-
solved in anhyd THF (15 mL) and added to a cooled (0 ∞C) solution
of methyl-tert-butylamine (5.0 mL, 40.9 mmol) and Et3N (5.82 mL,
41.8 mmol) in anhyd THF (100 mL). The mixture was stirred over-
night, filtered, the precipitate washed with THF and the filtrate
evaporated to yield 10.13 g of crude product. This solid was recrys-
tallized from Et2O/hexanes to give pure amide 4, yield: 9.48 g
(91%); mp 101.5 ∞C.
Anal. calcd for C23H26N2O3: C, 72.99; H, 6.92; N 7.40. Found: C,
73.07; H, 6.86; N 7.48.
4-[3,4-Dimethoxy-2-(methylcarbamoyl)phenyl]isoquinoline
(6b)
TFA (100 mL) was slowly added to the amide 6a (23.88 g,
63.23 mmol). After the solid had dissolved, the mixture was heated
at reflux for 2.5 h, until no starting material remained according to
TLC. After cooling, the solution was evaporated to yield the TFA
salt of 6b as an amber gum. This material was used in the next step
without further purification. An analytical sample was recrystal-
lized from MeOH/CHCl3/Et2O to give white crystals; mp 196 ∞C.
1H NMR (300 MHz, CDCl3, TFA salt): d = 9.60 (s, 1 H, Ar-H), 8.50
(s, 1 H, Ar-H), 8.40 (d, 1 H, J = 8.0 Hz, Ar-H), 8.08 (dt, 1 H,
J = 7.9, 1.1 Hz, Ar-H), 7.96 (m, 2 H, Ar-H), 7.39 (br s, 1 H, NH),
7.19 (d, 1 H, J = 8.5 Hz, Ar-H), 7.11 (d, 1 H, J = 8.5 Hz, Ar-H),
3.99 (s, 3 H, OCH3), 3.96 (s, 3 H, OCH3), 2.71 (d, 3 H, J = 4.9 Hz,
NCH3).
1H NMR (300 MHz, CDCl3): d = 7.01 (t, 1 H, J = 7.8 Hz, Ar-H),
6.83 (dd, 1 H, J = 7.7, 1.0 Hz, Ar-H), 6.75 (dd, 1 H, J = 7.8, 1.0 Hz,
Ar-H), 3.81 (s, 3 H, OCH3), 3.80 (s, 3 H, OCH3), 2.73 (s, 3 H,
NCH3), 1.48 (s, 9 H, t-C4H9).
MS (CI): m/z (%) = 252 (M+ + 1, 100), 196 (28).
Anal. calcd for C14H21NO3: C, 66.91; H, 8.42; N 5.57. Found: C,
67.11; H, 8.36; N 5.45.
MS (CI): m/z = 323 (M+ + 1, 100).
Anal. calcd for C21H19F3N2O5: C, 57.80; H, 4.39; N 6.42. Found: C,
57.62; H, 4.31; N 6.44.
3,4-Dimethoxy-2-(tert-butylmethylcarbamoyl)phenylboronic
Acid (5)
4-(2-Carboxy-3,4-dimethoxyphenyl)isoquinoline (7)
Compound 4 (20.88 g, 83.19 mmol) in anhyd THF (100 mL) was
added to a solution of TMEDA (13.8 mL, 91.5 mmol), 1.3 M BuLi
in cyclohexane (70 mL, 91.5 mmol) and anhyd THF (450 mL) at
-78 ∞C. The resulting creamy suspension was maintained at -78 ∞C
for 45 min, and trimethyl borate (28.4 mL, 250 mmol) was added at
once. This mixture was stirred for 6 h while gradually warming to
r.t. The solution was acidified with 1 N HCl and brine was added.
The phases were separated, the aqueous layer washed with Et2O,
and the combined organic phases washed with aq sat Na2CO3 solu-
tion and brine. After drying (Na2SO4) and evaporation 24.8 g (99%)
of product was obtained as a white powder with a melting range of
153-160 ∞C. This crude material was used directly in the next step.
1H NMR (300 MHz, CDCl3): d = 7.66 (d, 1 H, J = 8.3 Hz, Ar-H),
6.91 (d, 1 H, J = 8.3 Hz, Ar-H), 6.24 [br s, 2 H, B(OH)2], 3.87 (s, 3
H, OCH3), 3.81 (s, 3 H, OCH3), 2.76 (s, 3 H, NCH3), 1.55 (s, 9 H, t-
C4H9).
MS (CI): m/z (%) = 296 (M+ + 1, 100), 278 (23), 252 (28), 222 (20),
196 (11).
The crude TFA salt of 6b obtained above was dissolved in a mixture
of AcOH (70 mL) and Ac2O (400 mL). The solution was cooled to
0 ∞C and stirred with a mechanical stirrer while NaNO2 (100.0 g)
was added in portions over 3 h. The mixture was stirred overnight
while cooling was maintained with an ice bath. After filtration, the
solids were washed with Ac2O and Et2O, and the combined organic
phases were evaporated, with the bath temperature not exceeding
45 ∞C. The residue was dissolved in CH2Cl2 (400 mL) and washed
with 2 N NaOH until the washings remained basic. The combined
aqueous layers were washed with CH2Cl2, and the organic solution
was dried (MgSO4). After evaporation and drying in vacuo 22.5 g
of the N-nitrosamide was obtained as an orange solid. This material
was dissolved in a mixture of Et2O (200 mL) and EtOH (20 mL) and
added dropwise to a solution of KOH (15.0 g) in 80% EtOH
(150 mL), maintained at ~90 ∞C in a DIAZALD™ apparatus with
glacial AcOH in the receiving flask. After that a mixture of Et2O (50
mL) and EtOH (50 mL) was added dropwise to the reaction mixture
to ensure that all of the generated diazomethane had been removed
by azeotropic distillation. H2O (50 mL) was added and the mixture
was heated to reflux for 1 h. After evaporation of EtOH, H2O
(200 mL) was added and the mixture was extracted with CH2Cl2.
Evaporation of the solvent yielded 1.40 g (6%) of the ethyl ester
of acid 7. The remaining aqueous layer was made slightly acidic
by addition of 2 N HCl and extracted thoroughly with CHCl3.
Evaporation of the solvent yielded 20.8 g of material that was
HRMS: m/z Calcd for C14H22BNO5 295.9686. Found: 295.9678.
4-[3,4-Dimethoxy-2-(tert-butylmethylcarbamoyl)phenyl]iso-
quinoline (6a)
The crude boronic acid 5 (24.8 g, 82.37 mmol) from the previous
step, 4-bromoisoquinoline (16.48 g, 79.23 mmol), (Ph3P)4Pd
Synthesis 2001, No. 2, 262–266 ISSN 0039-7881 © Thieme Stuttgart · New York

PAPER
Synthesis of Dinapsoline
265
dissolved in 2 N HCl (250 mL) and extracted with CHCl3. The
CHCl3 phase was reextracted with 2 N HCl and the combined aque-
ous phases were evaporated to dryness. Residual H2O was removed
by azeotropic distillation with toluene on the rotary evaporator. The
remaining solid was dissolved in absolute EtOH and precipitated
with Et2O to afford 18.55 g (85%) of the HCl salt of acid 7 as a
slightly amber powder. The ethyl ester of acid 7 obtained above
could be converted to the free acid by standard procedures, giving a
combined yield of 89% starting from N-methyl-tert-butylamide 6a.
For analytical purposes, the TFA-salt of acid 7 was prepared; mp
165 ∞C.
1H NMR (300 MHz, CDCl3, TFA salt): d = 9.64 (d, 1 H, J = 2.1 Hz,
Ar-H), 8.47 (s, 1 H, Ar-H), 8.39 (d, 1 H, J = 8.2 Hz, Ar-H), 7.97 (t,
1 H, J = 7.3 Hz, Ar-H), 7.87 (t, 1 H, J = 7.3 Hz, Ar-H), 7.72 (d, 1
H, J = 8.2 Hz, Ar-H), 7.33 (d, 1 H, J = 8.5 Hz, Ar-H), 7.19 (d, 1 H,
J = 8.5 Hz, Ar-H), 3.94 (s, 3 H, OCH3), 3.85 (s, 3 H, OCH3).
the solution was evaporated to dryness and filtered through a short
column of silica gel with CHCl3/MeOH/NH4OH (100:4:1) as the
solvent. This procedure gave 0.38 g (75%) of 9 as bright orange
crystals. On heating in a melting point apparatus the crystals started
to darken at 162 ∞C, and finally decomposed at 204 ∞C. On heating
in an open container in an oven at 140 ∞C overnight the compound
quantitatively reoxidized back to ketone 2.
1H NMR (300 MHz, CDCl3): d = 10.41 (br s, 1 H, OH), 9.49 (s, 1
H, Ar-H), 8.35 (dd, 1 H, J = 9.0, 2.7 Hz, Ar-H), 8.06 (s, 1 H, Ar-H),
7.53 (dd, 1 H, J = 9.3, 2.8 Hz, Ar-H), 6.56 (dd, 1 H, J = 9.8, 2.0 Hz,
Ar-H), 6.29 (m, 1 H, Ar-H), 4.07 (s, 3 H, OCH3), 4.00 (s, 3 H,
OCH3), 3.46 (br s, 2 H).
MS (CI): m/z (%) = 294 (M+ + 1, 100), 278 (61).
HRMS: m/z Calcd for C18H15NO3: 294.1130. Found: 294.1131.
8,9-Dimethoxy-7H-naphth[1,2,3-de]isoquinoline (10)
MS (CI): m/z (%) = 310 (M+ + 1, 100), 292 (38).
Ketone 2 (4.50 g, 15.46 mmol) in TFA (90 mL) was cooled to 0 ∞C.
NaBH4 pellets (3.01 g, 82.0 mmol) were added one by one in a con-
stant stream of argon. The mixture was stirred at 0 ∞C for 30 min and
then 4.5 h at r.t., when TLC indicated the complete conversion of
starting material. The solution was hydrolyzed by the addition of
H2O (40 mL) and most of the liquid was evaporated. It was then
cooled to 0 ∞C and basified by addition of 25% NaOH. This mixture
was extracted with CHCl3 and evaporated to yield 3.54 g of crude
product, which was purified by flash chromatography (CHCl3/Et2O/
NH4OH, silica gel) to give 2.64 g (62%) of the product; mp 183 °C.
Some starting material (0.81 g, 18%) was also recovered.
1H NMR (300 MHz, CDCl3): d = 9.04 (s, 1 H, Ar-H), 8.95 (s, 1 H,
Ar-H), 7.84 (d, 1 H, J = 8.8 Hz, Ar-H), 7.78 (dd, 1 H, J = 7.9, 1.4
Hz, Ar-H), 7.62 (dd, 1 H, J = 7.2, 1.6 Hz, Ar-H), 7.56 (t, 1 H,
J = 7.5 Hz, Ar-H), 6.95 (d, 1 H, J = 8.8 Hz, Ar-H), 5.53 (br s, 2 H,
CH2), 3.94 (s, 3 H, OCH3), 3.93 (s, 3 H, OCH3).
Anal. calcd for C20H20F3NO5: C, 56.74; H, 3.81; N 3.31. Found: C,
56.52; H, 3.90; N 3.21.
8,9-Dimethoxy-7-oxo-7H-naphth[1,2,3-de]isoquinoline (5,6-
Dimethoxy-bz-2-azabenzanthrone, 2)
Chlorosulfonic acid (18.0 mL) was slowly added to the HCl salt of
7 (4.00 g, 11.58 mmol) at 0 ∞C. This mixture was stirred at r.t. for
44 h, then cooled back to 0 ∞C and carefully hydrolyzed by the drop-
wise addition of H2O. The mixture was made basic with 25% NaOH
and extracted thoroughly with CHCl3. Drying (MgSO4) and evapo-
ration of the solvent yielded 2.69 g (80%) 2 as dark orange crystals
in high purity, melting at 208 ∞C. A sample was recrystallized from
DMF and dried over P2O5, yielding 2 as tiny yellow needles; mp
222-223 ∞C.
1H NMR (300 MHz, DMSO): d = 9.60 (s, 1 H, Ar-H), 9.46 (s, 1 H,
Ar-H), 8.66 (dd, 1 H, J = 7.5, 1.1 Hz, Ar-H), 8.56 (dd, 1 H, J = 7.8,
1.1 Hz, Ar-H), 8.48 (d, 1 H, J = 9.0 Hz, Ar-H), 8.00 (t, 1 H, J = 7.8
Hz, Ar-H), 7.57 (d, 1 H, J = 8.9 Hz, Ar-H), 3.94 (s, 3 H, OCH3),
3.88 (s, 3 H, OCH3).
MS (CI): m/z = 278 (M+ + 1, 100%).
Anal. calcd for C18H15NO2: C, 77.96; H, 5.45; N 5.05. Found: C,
77.58; H, 5.33; N 4.96.
MS (CI): m/z = 292 (M+ + 1, 100%).
8,9-Dimethoxy-2,3,7,11b-tetrahydro-1H-naphth[1,2,3-de]iso-
quinoline (11)
Anal. calcd for C18H13NO3: C, 74.22; H, 4.50; N 4.81. Found: C,
74.10; H, 4.33; N 4.78.
The benzanthrene 10 (3.05 g, 11.01 mmol) was suspended in
anhyd MeOH (100 mL) and dissolved upon addition of 3 M HCl in
MeOH (15 mL). To this solution powdered NaBH3CN (4.15 g,
66.06 mmol) was added in portions under argon. A few drops of
methanolic bromocresol green solution were added which turned
the color of the reaction to brown-blue. The reaction mixture was
acidified (the color of the reaction turned yellow) and maintained
acidic by adding 3 M methanolic HCl solution through a dropping
funnel. When the yellow color persisted, and further addition of
HCl did not result in gas evolution, the solution was evaporated to
dryness. The residue was partitioned between Na2CO3 solution and
CH2Cl2, the phases were separated and the aqueous layer was ex-
tracted with CH2Cl2. The organic phase was acidified with a solu-
tion of 48% HBr (2.50 g) in EtOH (30 mL) and evaporated to
dryness. The crude salt was recrystallized once from EtOH/toluene
and a second time from EtOH/EtOAc to yield 2.99 g (75%) of 11
HBr-salt as very fluffy yellowish needles; mp 247 ∞C. This material
showed identical spectroscopic properties to the material obtained
by the original method.1
8,9-Dimethoxy-7-hydroxy-5,6-dihydro-4H-naphth[1,2,3-de]iso-
quinoline (8)
Ketone 2 (1.00 g, 3.44 mmol) was dissolved in AcOH (40 mL) and
shaken for 9 h with PtO2 (10 mg) in a Parr apparatus at 60 psi hy-
drogen pressure. The solution was evaporated to dryness and fil-
tered through a short column of silica with CHCl3/MeOH/NH4OH
(100:4:1) as the solvent. After drying (MgSO4) and evaporation of
the solvent the remaining solids were recrystallized from CH2Cl2 /
Et2O yielding 0.63 g (62%) of 8 as deep red crystals; mp 133 ∞C.
1H NMR (300 MHz, DMSO): d = 9.70 (s, 1 H, Ar-H), 8.74 (d, 1 H,
J = 8.9 Hz, Ar-H), 8.30 (s, 1 H, Ar-H), 7.58 (d, 1 H, J = 9.0 Hz, Ar-
H), 4.04 (s, 3 H, OCH3), 4.00 (s, 3 H, OCH3), 3.33 (m, 4 H), 2.95
(m, 2 H).
MS (CI): m/z (%) = 296 (M+ + 1, 85), 278 (100).
Anal. calcd for C18H17NO3: C, 73.20; H, 5.80; N 4.74. Found: C,
73.05; H, 5.63; N 4.71.
8,9-Dimethoxy-7-hydroxy-H-naphth[1,2,3-de]isoquinoline (9)
Ketone 2 (0.50 g, 1.72 mmol) was dissolved in TFA (5.5 mL). Tri-
ethylsilane (1.89 mL, 11.86 mmol) was added at once, and the mix-
ture was stirred at r.t. for 15 h. After addition of a few drops of H2O,
Acknowledgement
The authors gratefully acknowledge support by Bristol-Myers-
Squibb Corporation and PHS grant MH42705.
Synthesis 2001, No. 2, 262–266 ISSN 0039-7881 © Thieme Stuttgart · New York

266
T. Sattelkau et al.
PAPER
(13) (a) Ishikura, M.; Mano, T.; Oda, I.; Terashima, M.
References
Heterocycles 1984, 22, 2471.
(b) Ishikura, M.; Oda, I.; Terashima, M. Heterocycles 1987,
26, 1603.
(1) Ghosh, D.; Snyder, S.E.; Watts, V.J.; Mailman, R.B.; Nichols,
D.E. J. Med. Chem. 1996, 39, 549.
(2) Qandil, A.M.; Ghosh, D.; Nichols, D.E. J. Org. Chem. 1999,
64, 1407.
(3) Bayer, O. Houben-Weyl; 4. Aufl., Band VII/3c, Thieme:
Stuttgart, 1979.
(4) Bayer, O.; Ebel, F. DRP 621455 (1934), also cited in Ref. 3, p
347.
(14) Chu, L.; Fisher, M.H.; Goulet, M.T.; Wyvratt, M.J.
Tetrahedron Lett. 1997, 38, 3871.
(15) Kametani, T; Honda, T.; Inoue, H.; Fukumoto, K.
Heterocycles 1975, 3, 1091.
(16) Kalischer, G; MÜller, R., Frister, F. DRP 426347 (1923), also
cited in Ref. 3, p 314.
(5) Alo, B.I.; Kandil, A.; Patil, P.A.; Sharp, M.J.; Siddiqui, M.A.,
Snieckus, V.; Josephy, P.D. J. Org. Chem. 1991, 56, 3763.
(6) Miller, R.B.; Svoboda, J.J. Synth. Commun. 1994, 24, 1187.
(7) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
(8) Snieckus, V. Chem. Rev. 1990, 90, 879.
(9) Reitz, D.B.; Massey, S.M. J. Org. Chem. 1990, 55, 1375.
(10) Collins, S.; Hong, Y. Tetrahedron Letters 1987, 28, 4391.
(11) Auerbach, J.; Weissmann, S.A.; Blacklock, T.J.; Angeles,
M.R.; Hoogsteen, K. Tetrahedron Lett. 1993, 34, 931.
(12) Meyer, A.I.; Gabel, R.; Mihelich, E.D. J. Org. Chem. 1978,
43, 1372.
(17) Augustine, R.L. Heterogenous Catalysis for the Synthetic
Chemist; Marcel Dekker: New York, 1996.
(18) West, C.T.; Donnelly, S.J.; Kooistra, D.A.; Doyle, M.P. J.
Org. Chem. 1973, 38, 2675.
(19) Ono, A.; Maruyama, T.; Suzuki, N. Synth. Commun. 1987, 17,
1001.
(20) Gribble, G.W.; Kelly, W.J.; Emery, S.E. Synthesis 1978, 763.
Article Identifier:
1437-210X,E;2001,0,02,0262,0266,ftx,en;M01800SS.pdf
Synthesis 2001, No. 2, 262–266 ISSN 0039-7881 © Thieme Stuttgart · New York