Preparation of dopaminergic N-alkyl-benzyltetrahydro-isoquinolines using a 'one-pot' procedure in acid medium

Article abstract of DOI:10.1016/S0968-0896(00)00027-4

The preparation of N-methyl-BTHIQ (4) from N-phenylethyl-phenacetamide (1) by cyclization, reduction and N-alkylation in acid medium has been achieved in good yield in a 'one-pot' procedure. Acylation of imine (2) intermediate afforded the Z and E stereoselectivity in the enamide formation. 6-Hydroxy-BTHIQ (7) shows selectivity for D₂ dopamine receptors, while its N-methylated homologue (8) displays higher affinities for both D₁ and D₂ receptor types, with an unexpected increase in D₁ dopamine receptor affinity. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00027-4

Bioorganic & Medicinal Chemistry 8 (2000) 889±895
Preparation of Dopaminergic N-Alkyl-benzyltetrahydro-
isoquinolines Using a `One-Pot' Procedure in Acid Medium
Inmaculada Andreu, a Diego Cortes, a,* Philippe Protais, b Bruce K. Cassels, c
Abdeslam Chagraoui b and Nuria Cabedo a
a
Â
Departamento de Farmacologõa, Farmacognosia y Farmacodinamia, Facultad de Farmacia, Universidad de Valencia,
46100 Burjassot, Valencia, Spain
Â
Â
Departamento de Quõmica, Facultad de Ciencias, Universidad de Chile, Santiago, Chile
b
Â
Â
Laboratoire de Physiologie, Faculte de Medecine-Pharmacie, Universite de Rouen, 76183 Rouen, France
c
Received 23 April 1999; accepted 15 November 1999
AbstractÐThe preparation of N-methyl-BTHIQ (4) from N-phenylethyl-phenacetamide (1) by cyclization, reduction and N-alkyl-
ation in acid medium has been achieved in good yield in a `one-pot' procedure. Acylation of imine (2) intermediate a€orded the
Z and E stereoselectivity in the enamide formation. 6-Hydroxy-BTHIQ (7) shows selectivity for D2 dopamine receptors, while
its N-methylated homologue (8) displays higher anities for both D1 and D2 receptor types, with an unexpected increase in D1
dopamine receptor anity. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
appropriate chiral modi®ers, in asymmetric syntheses of
chiral 1,2,3,4-tetrahydroisoquinolines.5,6 We now describe
the synthesis of an N-methyl-BTHIQ incorporating a
`one-pot' cyclization±reduction±alkylation sequence.
Brie¯y, after performing the usual POCl3-catalyzed
cyclization the solvent is removed, methanol is added to
form very probably PO(OCH3)3, and the solution is
treated with NaBH4 to a€ord the ®nal product.
Some 1-benzyl-1,2,3,4-tetrahydroisoquinoline (BTHIQ)
alkaloids have been shown to display fairly high anity
for both D1 and D2 dopamine receptors.1 3 The selec-
tive anity for one of these receptor types (D2) appears
to be related to the stereochemistry and the substitution
pattern in the A and C rings, as well as N-substitution.
D1 and D2 dopaminergic anities of two natural
enantiomeric BTHIQs, the N-methylated (S)-reticuline
and the N-unmethylated (R)-coclaurine, have been
reported but without comparision between their corre-
sponding enantiomers.2 We recently accomplished
the stereoselective synthesis of a 6,7,40-trioxygenated
norarmepavine isomer (R)-(+)-nor-roefractine, which
exhibits 6-fold selectivity for D2 versus D1 receptors3
(Fig. 1).
Results and Discussion
N-(3-Benzyloxy-4-methoxyphenylethyl)-phenylacetamide
(1) was ®rst prepared by standard methods,7 starting
from isovanillin, in four steps. The usual POCl3-cata-
lyzed Bischler±Napieralski cyclization of 1, and sub-
sequent reduction of imine 2 by NaBH4 in MeOH, led
to the expected BTHIQ 3. However, when a `one-pot'
procedure was attempted at 40 ꢀC, adding MeOH and
then NaBH4 to the POCl3-containing system, consider-
able N-methylation was found to occur a€ording in
good yield amines 3 and 4 in similar amounts after 5 h
(route A, Fig. 2). We were able to show that N-methyl
derivative 4 can also be obtained from amine 3 (route B)
in POCl3/MeOH in somewhat lower yield. An alter-
native route (C) may be postulated via an intermediate
N-methyliminium species (2a). In all routes the N-
methylation is presumably e€ected by PO(OCH3)3
formed in situ from POCl3 and MeOH.
In order to further explore the structural basis of the
anity of the 1,2,3,4-tetrahydroisoquinoline moiety for
dopamine receptors, we have prepared 6,7-dioxy-
genated-BTHIQs unsubstituted in the C ring. Sodium
borohydride reductions of iminium species obtained by
Bischler±Napieralski cyclization are now regularly used
in the syntheses of N-methylated BTHIQs4 and, with
*Corresponding author. Tel.: +34-96-386-49-75; fax: +34-96-386-49-
43; e-mail: dcortes@uv.es
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00027-4

890
I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
Figure 1. 1-Benzyl-1,2,3,4-tetrahydroisoquinoline analogues.
Figure 2. Syntheses of N-methyl-BTHIQ derivative 4. Reagents and conditions: (a) POCl3/CH2Cl2, re¯ux, 3 h, followed by NaBH4/MeOH, 40 ꢀC,
5 h; (b) POCl3/CH2Cl2, re¯ux, 3 h; (c) NaBH4/MeOH, rt, 2 h; (d) POCl3/MeOH, rt, 30 h; (e) POCl3/MeOH, re¯ux, 9 h, followed by NaBH4, rt, 3 h.
Products 3 and 4 are formed in practically identical
quantities (route A, `one-pot'), at 40 ꢀC, at room tem-
perature and at re¯ux (MeOH), suggesting that under
our conditions NaBH4 reacts slowly enough with the
(presumably protonated) imine to allow about half of it
to become N-methylated. The lower yield of product 4
when 3 is methylated with POCl3/MeOH (route B) is an
indication that, under the conditions of route A,
methylation occurs before or concomitantly with, but
not after reduction, and also the speed of product 4
formation is improved with respect to route B (Fig. 2
and Table 1). Similar `one-pot' experiments were carried
out using EtOH as a solvent. Under these conditions,
N-ethyl-BTHIQ derivative (4a) was obtained, pre-
sumably via the corresponding PO(OEt)3, in lower yield
(Fig. 3 and Table 1).
reduction.8,9 The fact that these intermediates must be
isolated ®rst makes the procedure lengthier and less
labor-ecient.
A very important di€erence was observed in the 1H
NMR spectra between secondary (3) and tertiary (4)
amine BTHIQs, presumably due to lack of substitution
on the benzyl moiety in contrast to the usual 40- and
40,50-substituted benzylic ring.2 Compound 3 exists in
solution mainly as an anti rotamer (of one aromatic ring
with regard to the other), with a `protoberberine-like'
conformation, while in the corresponding N-alkyl com-
pounds (4, 4a), the syn `aporphine-like' conformation
is preferred. In this case the anisotropic e€ect of the
unsubstituted benzyl group leads to an unusual
shielding of H-1, H-8 and OCH3-7. A similar but
weaker e€ect was seen comparing the spectra of imine
2, amine 3 and nor-armepavine, in their base forms
(`protoberberine-like'), and as their corresponding salts
(`aporphine-like') (Figs. 1 and 2).
The simultaneous preparation of two homologous (3
and 4) products in a `one-pot' reaction from amide 1 in
very satisfactory overall yield (87%) is a signi®cant
improvement over the usual three-reaction sequence.
Classical N-alkylation methods (to give N-alkyl-BTHIQ
derivatives in good yield) start from the corresponding
secondary amine or from imine and subsequent
One of the intermediates in the N-alkyl-BTHIQ synth-
esis, imine 2, was N-acylated in order to prepare the two
stereoisomeric (Z)- and (E)-enamides, and then study

I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
Table 1. N-Alkyl-BTHIQ derivatives obtained by acid reductive alkylation
891
Compounds
Reagents
Time
Reaction compounds
Yield (%)
1
(1) POCl3 (re¯ux);
(2) NaBH4/MeOH (rt)
POCl3 (re¯ux); NaBH4/MeOH ( 40 ꢀC)
POCl3 (re¯ux); NaBH4/EtOH (re¯ux)
POCl3/MeOH (re¯ux); NaBH4 (rt)
POCl3/MeOH (rt)
3 h
2 h
8 h
20 h
12 h
30 h
30 h
2
83
63
87
82
76
31
0
3
3 (45%)+4 (42%)
3 (69%)+4a (13%)
3 (39%)+4 (37%)
1
1
2
3
3
4
4
HCl/MeOH (re¯ux)
1
the e€ect of the phenyl group on the H NMR spectra
in an enamide system. Certain b-alkoxycarbonylated
enamides have been found to be good starting materials
for the synthesis of optically active b-amino acids.10
The geometric isomers (Z)- and (E)-N-acetyl-1-benzyl-
idene-6-benzyloxy-7-methoxy-3,4-dihydroisoquinolines
(5 and 6), formed in similar yields, have very di€erent
high resolution mass spectrometry (HREIMS). All four
BTHIQ amines (3, 4, 7 and 8) were able to displace both
[3H]-SCH 23390 (a D1 dopamine receptor-selective
ligand) and [3H]-raclopride (a D2 dopamine receptor-
selective ligand) from their speci®c binding sites in rat
striatal membranes;1 3 enamides 5 and 6 would not be
expected to show anity for dopamine receptors because
of their delocalized nitrogen electron pair (non-basic
character).1 6-Benzyloxy-BTHIQ (3), displays rather
low anities for both D1 and D2 binding sites, but its
N-methylated homologue (4) shows increased anities
for both receptor types. The 6-hydroxy-BTHIQ (7), as
compared to its O-benzylated precursor (3), only has
enhanced anity for [3H]-raclopride binding sites, and
therefore shows selectivity for D2 dopamine receptors
(ratio D1/D2=13.5). Its N-methyl homologue (8), how-
ever, has increased anities for both receptor types and
is non-selective (Table 2 and Fig. 4). The selectivity of
bis-BTHIQ and tetrahydroprotoberberine alkaloids for
D2 dopamine receptors has been reported previously.12,13
Moreover, these binding data suggest that a `proto-
berberine-like' conformation of BTHIQs with an
unsubstituted benzyl moiety such as 7 increases their
selectivity for D2 dopamine receptors, while an `apor-
phine-like' conformation such as 8 displays high a-
nities for both D1 and D2 dopamine receptors, with an
unexpected increase in D1 dopamine receptor anity
(12.5-fold 8 versus 7, see Table 2 and Fig. 4). The `one-pot'
reaction sequence we have demonstrated is a convenient
means for obtaining pairs of N-alkylated and unalkylated
BTHIQs for future pharmacological studies.
1
UV and H NMR spectroscopic properties. Signi®cant
bathochromic and hyperchromic e€ects are seen in the
UV of the trans-stylbene chromophore of 5 as compared
to its cis-isomer 6.11 The (Z)-enamide (5) shows a
1
shielding of the N-acetyl methyl group in the H NMR,
while the (E)-enamide (6) shows a shielding of the
OCH3-7 group, both shieldings attributable to the ani-
sotropic e€ect of the benzylidene aromatic ring. Thus
the (Z)-isomer has the N-acyl group on the same side as
the benzylidene aromatic ring and the (E)-isomer has
the N-acyl group and the benzylidene aromatic ring
on opposite sides (Fig. 3). Moreover, the enamide
carbonyl group in the (Z)-isomer (5) a€ects both 3a
and 3b protons (shielding one: m, d 3.18, and deshield-
ing the other: m, d 5.01), indicating that this group lies
predominantly syn to C-3, probably due to electronic
repulsion by the benzylidene aromatic ring (t, d 3.99,
J3 4=6.4 Hz, CH2-3 in (E)-isomer, 6).
Compounds 3 and 4 were debenzylated by re¯uxing
with equal volumes of EtOH and concentrated HCl,
a€ording the corresponding phenolic compounds, 7
and 8 (Fig. 3), in good yield (ꢁ80%). The molecular
formulae of compounds 7 and 8 were determined by
Figure 3.

892
I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
Table 2. Comparative IC50 of BTHIQs on speci®c 3[H]-SCH 23390
(D1 receptors) and 3[H]-raclopride (D2 receptors) binding to rat stria-
tal membranes
Experimental
General experimental procedures
UV spectra were recorded on a Shimadzu UV-2101PC.
IR spectra (®lm) were run on a Perkin±Elmer 1750
FTIR spectrometer. EIMS, LSIMS and HREIMS were
determined on a VG Auto Spec Fisons instrument.
NMR spectra were recorded on Bruker AC-250 or
Varian Unity-400 spectrometer at 250 or 400 MHz for
1H, and 100 MHz for 13C. Multiplicities of 13C NMR
signals were assigned by DEPT experiments. NOEDIFF
irradiations were recorded at 250 MHz and COSY 45
and HMQC correlations at 400 MHz. All reactions
were monitored by analytical TLC with silica gel 60 F254
Compounds
IC50 (mM)a on speci®c binding of
Ratio
3[H]-SCH 23390
3[H]-raclopride
D1/D2
3
4
7
8
57.1 (8.15±399.4)
16.0 (2.30±112.0)
40.3 (5.80±282.1)
3.2 (0.72±14.3)
25.7 (3.67±179.6)
8.6 (1.20±60.0)
3.0 (0.52±17.0)
3.5 (0.78±15.6)
2.2
1.9
13.5
0.9
aIC50s and their 95% con®dence limits were calculated by the method
of Litch®eld and Wilcoxon14 from concentration±e€ect curves with
four determinations for each concentration.
Figure 4. Comparative displacement curves of speci®c [3H]-SCH 23390 and [3H]-raclopride binding by BTHIQ compounds 3 and 7, and N-methyl-
BTHIQ compounds 4 and 8. Displacement curves correspond to four determinations at each concentration.

I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
893
(Merck 5554). The residues were puri®ed through 60 H
silica gel column (5±40 mm, Merck 7736), and by ¯ash
chromatography (230±400 mm, Merck 9385).
H-20, 60), 7.27±7.32 (5H, m, Ph), 7.36 (1H, t, J=6.8 Hz,
H-40), 7.42 (2H, d, J=6.8 Hz, H-30,50); 13C NMR
(CDCl3, 100 MHz) d 170.8 (CO), 148.3 and 147.9 (C-3
and C-4), 137.0, 134.8 and 131.0 (C-1, C-10 and C-100),
129.4-127.3 (Ph), 121.3 (C-6), 114.6 (C-2), 111.9 (C-5),
71.0 (OCH2Ph-3), 56.1 (OCH3-4), 43.8 (CH2-CO), 40.6
(CH2-a), 34.9 (CH2-b); EIMS m/z (%) 375 [M]+ (11),
284 (2), 240 (80), 137 (4.5), 91 (100).
Bioassays
Binding experiments were performed on striatal mem-
branes. Each striatum was homogenized in 2 mL ice-
cold Tris±HCl bu€er (50 mM, pH=7.4 at 22 ꢀC) with a
Polytron (4 s, maximal scale) and immediately diluted
with Tris bu€er. The homogenate was centrifuged either
twice ([3H]-SCH 23390 binding experiments) or four
times ([3H]-raclopride binding experiments) at 20 000Âg
for 10 min at 4 ꢀC with resuspension in the same volume
of Tris bu€er between centrifugations. For [3H]-SCH
23390 binding experiments, the ®nal pellet was resus-
pended in Tris bu€er containing 5 mM MgSO4, 0.5 mM
EDTA and 0.02% ascorbic acid (Tris±Mg bu€er) and
the suspension was brie¯y sonicated and diluted to a
protein concentration of 1 mg/mL. A 100 mL aliquot of
freshly prepared membrane suspension (100 mg of stria-
tal protein) was incubated for 1 h at 25 ꢀC with 100 mL
Tris bu€er containing [3H]-SCH 23390 (0.25 nM ®nal
concentration) and 800 mL of Tris±Mg bu€er containing
the required drugs. Non-speci®c binding was deter-
mined in the presence of 30 mM SK&F 38393 and
represented around 2 to 3% of total binding. For [3H]-
raclopride binding experiments, the ®nal pellet was
resuspended in Tris bu€er containing 120 mM NaCl,
5 mM KCl, 1 mM CaCl2, 1 mM MgCl2 and 0.1%
ascorbic acid (Tris±ions bu€er), and the suspension was
treated as described above. A 200 mL aliquot of freshly
prepared membrane suspension (200 mg of striatal pro-
tein) was incubated for 1 h at 25 ꢀC with 200 mL of Tris
bu€er containing [3H]-raclopride (0.5 nM ®nal con-
centration) and 400 mL of Tris±ions bu€er containing
the drug being investigated. Non-speci®c binding was
determined in the presence of 50 mM apomorphine and
represented around 5 to 7% of total binding. In both
cases, incubations were stopped by addition of 3 mL of
ice-cold bu€er (Tris±Mg bu€er or Tris±ions bu€er, as
appropriate) followed by rapid ®ltration through
Whatman GF/B ®lters. Tubes were rinsed with 3 mL
ice-cold bu€er, and ®lters were washed with 3Â3 mL
ice-cold bu€er. After the ®lters had been dried, radio-
activity was counted in 4 mL BCS scintillation liquid at
an eciency of 45%. Filter blanks corresponded to
approximately 0.5% of total binding and were not
modi®ed by drugs.
1-Benzyl-6-benzyloxy-7-methoxy-3,4-dihydroisoquinoline
(2). A solution of 1 (500 mg, 1.33 mmol) in CH2Cl2 (7
mL) was treated with POCl3 (0.5 mL, 5.31 mmol) and
the mixture re¯uxed for 3 h. The reaction mixture was
diluted with H2O, made basic (pHꢁ9) and extracted
with CH2Cl2. The organic solution was washed with
H2O, dried and concentrated to give a brown oil. This
residue was puri®ed through 60 H silica gel column
(CH2Cl2:MeOH 96:4). Compound 2 (415 mg, 83%) was
obtained and its HCl salt was crystallized from MeOH
to give yellow crystals, mp 178±181ꢀ; C24H23NO2; IR
1
(®lm) nmax 2935, 1657 (CˆN), 1566, 1541 cm
;
1H
NMR (CDCl3, 400 MHz) d 2.79 (2H, t, J=7.6 Hz,
CH2-4), 3.78 (3H, s, OCH3-7), 3.83 (2H, t, J=7.6 Hz,
CH2-3), 4.11 (2H, s, CH2-a), 5.15 (2H, s, OCH2Ph-6),
6.68 (1H, s, H-8), 7.00 (1H, s, H-5), 7.19±7.41 (10H, m,
2Ph); 13C NMR (CDCl3, 100 MHz) d 166.4 (C-1), 150.5
(C-7), and 147.8 (C-6), 137.6, 136.4, 131.8 and 121.3 (C-
10, C-100, C-8a and C-4a), 128.7±126.6 (Ph), 112.3 (C-5),
110.4 (C-8), 70.7 (OCH2Ph-6), 56.1 (OCH3-7), 46.4 (C-
3), 42.9 (CH2-a), 25.6 (C-4); EIMS m/z (%) 357 [M]+
(72), 356 [M 1]+ (93), 326 (43), 266 (89), 236 (16), 91
(100).
1-Benzyl-6-benzyloxy-7-methoxy-1,2,3,4-tetrahydroiso-
quinoline (3). NaBH4 (20 mg, 0.529 mmol) was added
portionwise to a solution of 2 (400 mg, 1.120 mmol) in
MeOH (4 mL) over 30 min, and then stirred for 1.5 h at
room temperature. The reaction solution was diluted
with H2O and extracted with CH2Cl2 after making
alkaline (pHꢁ9) with NH3 (aq). The organic solution
was washed with H2O, dried, concentrated, and the
residue puri®ed by ¯ash chromatography on silicagel
(CH2Cl2:MeOH:DEA 96:4:0.1). 3 (305 mg, 76%) was
obtained: C24H25NO2; IR (®lm) nmax 3304, 2926, 1606,
1
1511, 1454 cm 1; H NMR (CDCl3, 400 MHz) d 2.62
(2H, m, CH2-4), 2.86 and 3.14 (4H, 2m, CH2-3 and a),
3.73 (3H, s, OCH3-7), 4.10 (1H, m, H-1), 5.05 (2H, s,
OCH2Ph-6), 6.55 and 6.56 (2H, 2s, H-8 and 5), 7.17±
7.40 (10H, m, 2Ph); 13C NMR (CDCl3, 100 MHz) d
147.6 and 146.8 (C-6 and C-7), 139.0, 137.2, 130.8 and
127.7 (C-10, C-100, C-8a and C-4a), 129.4±126.5 (Ph),
114.6 (C-5), 110.2 (C-8), 71.0 (OCH2Ph-6), 56.8 (C-1),
56.1 (OCH3-7), 42.6 (C-3), 40.5 (CH2-a), 29.2 (C-4);
LSIMS m/z 360 [MH]+; EIMS m/z (%) 359 [M]+ (1),
268 (100), 177 (67), 91 (77).
N-(3-Benzyloxy-4-methoxyphenylethyl)-phenylacetamide
(1). Prepared from b-(3-benzyloxy-4-methoxy-phenyl)-
ethylamine and phenylacetyl chloride by standard pro-
cedure,7 as white crystals obtained from EtOH. Com-
pound 1: mp 111±113ꢀ; C24H25NO3; IR (®lm) nmax 3296
1
(NH), 2927, 1640 (amide), 1585, 1543 cm 1; H NMR
(CDCl3, 400 MHz) d 2.62 (2H, t, J=6.8 Hz, CH2-b),
3.38 (2H, td, J=6.8 Hz, J0=6.0 Hz, CH2-a; COSY 45
and double resonance, with CH2-b and NH amide), 3.49
(2H, s, CH2-CO), 3.86 (3H, s, OCH3-4), 5.08 (2H, s,
OCH2Ph-3), 5.27 (1H, br signal, NHCO), 6.56 (1H dd,
J=8.4 Hz, J0=1.6 Hz, H-6), 6.64 (1H, d, J0=1.6 Hz, H-
2), 6.75 (1H, d, J=8.4 Hz, H-5), 7.14 (2H, d, J=6.8 Hz,
Preparation of N-alkyl-BTHIQ derivatives
N-Methyl-1-benzyl-6-benzyloxy-7-methoxy-1,2,3,4-tetra-
hydroisoquinoline (4). Compound 1 (196 mg, 0.522
mmol) was treated as above in CH2Cl2 (5 mL) with
POCl3 (0.33 mL, 3.5 mmol), re¯uxed for 3 h and con-
centrated. The acid residue was dissolved with MeOH

894
I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
(7 mL) and NaBH4 (17 mg, 0.449 mmol) was added in
small portions over 1 h and the solution was stirred for
4 h at 40 ꢀC. The acid methanolic solution, after
removing the solvent under reduced pressure, was dis-
solved in H2O, basi®ed (pHꢁ9) with NH3 (aq) and
extracted with CH2Cl2. The organic layer was washed
with H2O and brine, dried over anhydrous Na2SO4 and
concentrated. The residue was puri®ed through 60 H
silica gel column (toluene:AcOEt:DEA 96:2:2) to a€ord
compounds 3 (85 mg, 45%) and 4 (81 mg, 42%). Com-
pound 4: C25H27NO2; IR (®lm) nmax 3383, 2925, 1608,
4), 3.18 and 5.01 (2H, m, CH2-3), 3.93 (3H, s, OCH3-7),
5.15 (2H, s, OCH2Ph-6), 6.65, 6.79 and 7.18 (3H, 3s, H-5,
H-8 and H-a), 7.32±7.47 (10H, m, 2Ph); EIMS m/z (%)
399 [M]+ (27), 308 (100), 266 (17), 91 (25).
(E)-Enamide (6): C26H25NO3; UV, lmax, EtOH, nm
1
(log E) 228 (4.19), 290 (3.94); H NMR (CDCl3, 400
MHz) d 2.30 (3H, s, NCOCH3), 2.85 (2H, t, J=6.4 Hz,
CH2-4), 3.35 (3H, s, OCH3-7), 3.99 (2H, t, J=6.4 Hz,
CH2-3), 5.12 (2H, s, OCH2Ph-6), 6.46, 6.64 and 6.65
(3H, 1brs and 2s, H-5, H-8 and H-a), 7.32±7.43 (10H,
m, 2Ph); EIMS m/z (%) 399 [M]+ (22), 308 (100), 266
(10), 91 (49).
1
1520, 1455 cm 1; H NMR (CDCl3, 400 MHz) d 2.54
(3H, s, NCH3), 2.55 (2H, m, CH2-4), 2.78 and 3.18 (4H,
2m, CH2-3 and a), 3.50 (3H, s, OCH3-7), 3.73 (1H, m,
H-1), 5.08 (2H, s, OCH2Ph-6), 5.97 (1H, s, H-8), 6.59
(1H, s, H-5), 7.09±7.44 (10H, m, 2Ph); NOEDIFF
(CDCl3, 400 MHz) H-1 with NMe, CH2-a and H-8; H-8
with OCH3-7; H-5 with OCH2Ph-6; 13C NMR (CDCl3,
100 MHz) d 146.9 and 146.5 (C-6 and C-7), 140.0, 137.3,
129.8 and 125.7 (C-10, C-100, C-8a and C-4a), 129.8±
125.9 (Ph), 114.0 (C-5), 111.6 (C-8), 71.0 (OCH2Ph-6),
64.8 (C-1), 55.5 (OCH3-7), 46.7 (C-3), 42.6 (NCH3),
41.0 (CH2-a), 25.4 (C-4); LSIMS m/z 374 [MH]+; EIMS
m/z (%) 282 (100), 191 (92), 178 (31), 161 (42), 91 (56).
Selective hydrolysis of the benzyloxy protective group
1-Benzyl-6-hydroxy-7-methoxy-1,2,3,4-tetrahydroisoquino-
line (7). BTHIQ 3 (45 mg, 0.125 mmol) was re¯uxed for
3 h with a mixture of equal volumes of ethanol and
concentrated HCl (14 mL). The reaction solution was
made basic (pHꢁ9) and extracted with CH2Cl2. The
organic layer was washed with H2O, dried, concentrated
and puri®ed over a 60 H silica gel column (CH2Cl2:
AcOEt:MeOH:NH4OH 40:50:10:0.1). 7 (27 mg, 80%)
was obtained: C17H19NO2; IR (®lm) nmax 3330, 2926,
2848, 1601, 1510, 1452, 1328, 1272, 1111, 1029, 963,
863, 801, 754, 700 cm 1; 1H NMR (CDCl3, 400 MHz) d
2.70 (2H, m, CH2-4), 2.94 and 3.19 (4H, 2m, CH2-3 and
a), 3.48 (1H, s, NH; COSY 45 with CH2-3), 3.79 (3H, s,
OCH3-7), 4.16 (1H, m, H-1), 6.56 (1H, s, H-8), 6.64
(1H, s, H-5), 7.23±7.35 (5H, m, Ph); EIMS m/z (%) 269
[M]+ (6), 268 [M 1]+ (35), 192 (74), 178 (47), 163 (47),
147 (16), 134 (67), 91 (100); HREIMS m/z 268.13379
[M 1]+ (268.13375 calcd for C17H18NO2), 178.08229
(178.08680 calcd for C10H12NO2), 163.09301 (163.09971
calcd for C10H13NO), 91.05377 (91.05477 calcd for
C7H7).
Preparation of 4 from 2. Compound 2 (46.2 mg, 0.13
mmol) dissolved in MeOH (5 mL) was treated with
POCl3 (0.1 mL, 1.05 mmol). After re¯uxing for 9 h,
NaBH4 (4 mg, 0.10 mmol) was added in small portions
to the reaction mixture over 1 h, which was then stirred
for 2 h. Following the usual work up, the residue was
puri®ed through 60 H silica gel column (toluene:
AcOEt:DEA 96:2:2), to a€ord 3 (18 mg, 39%) and 4 (18
mg, 37%).
Preparation of 4 from 3. Compound 3 (46 mg, 0.128
mmol) dissolved in MeOH (10 mL) was treated with
POCl3 (0.24 mL, 2.5 mmol) at rt for 30 h. 4 (15 mg, 31%)
was obtained following the same procedure as above.
N-Methyl-1-benzyl-6-hydroxy-7-methoxy-1,2,3,4-tetra-
hydroisoquinoline (8). Using similar conditions as for 7,
4 (26 mg, 0.069 mmol) was converted into 8 (16 mg,
82%): C18H21NO2; IR (®lm) nmax 3320, 2933, 2846,
1586, 1470, 1449, 1346, 1323, 1272, 1246, 1224, 1209,
N-Ethyl-1-benzyl-6-benzyloxy-7-methoxy-1,2,3,4-tetra-
hydroisoquinoline (4a). Following the same procedure
as for 4, but with EtOH instead of MeOH and re¯uxing
for 20 h, 1 (10 mg, 0.027 mmol) was converted into 3
(6.7 mg, 69%) and 4a (1.3 mg, 13%). 4a: C26H29NO2;
LSIMS m/z 387 [M]+, 205 [M 2 Ph]+.
1
1138, 1111, 1026, 948, 867, 783, 735, 715, 695 cm 1; H
NMR (CDCl3, 250 MHz) d 2.55 (3H, s, NCH3), 2.60
(2H, m, CH2-4), 2.78 and 3.21 (4H, 2m, CH2-3 and a),
3.49 (3H, s, OCH3-7), 3.74 (1H, m, H-1), 5.86 (1H, s, H-
8), 6.63 (1H, s, H-5), 7.09±7.30 (5H, m, Ph); EIMS m/z
(%) 283 [M]+ (2), 282 [M 1]+ (13), 192 (42), 178 (100),
163 (14), 161 (13), 148 (72), 132 (25), 91 (55); HREIMS
m/z 283.15249 [M]+(283.15722 calcd for C18H21NO2),
192.09990 (192.10245 calcd for C11H14 NO2), 178.08428
(178.08680 calcd for C10H12NO2), 91.05496 (91.05477
calcd for C7H7).
Preparation of N-acyl-enamides
(Z)- and (E)-N-Acetyl-1-benzylidene-6-benzyloxy-7-meth-
oxy-3,4-dihydroisoquinolines (5, 6). Treatment of 2
(8 mg, 0.022 mmol) with dry pyridine (0.2 mL) and
acetic anhydride (0.5 mL) a€orded N-acetyl-enamides 5
and 6, after stirring for 4 h at rt. After work up the oily
residue was fractionated by 60 H silica gel column (hex-
ane: AcOEt 60:40), to give 5 (3 mg, 34%) and 6 (3.5 mg,
40%).
Acknowledgements
(Z)-Enamide (5): C26H25NO3; UV, lmax, EtOH, nm
(log E) 231 (4.21), 322 (4.17); 1H NMR (CDCl3, 400 MHz)
d 1.75 (3H, s, NCOCH3), 2.64 and 3.16 (2H, 2m, CH2-
This research was supported by the Spanish DGICYT
under grant SAF 97-0013 and, in part, by the Presi-
dential Chair in Sciences (Chile, BKC).

I. Andreu et al. / Bioorg. Med. Chem. 8 (2000) 889±895
895
References
8. Wert, K. L.; Chackalamannil, S.; Miller, E.; Dalton, D. R.
J. Org. Chem. 1982, 47, 5141.
1. Protais, P.; Arbaoui, J.; Bakkali, E. H.; Bermejo, A.;
Cortes, D. J. Nat. Prod. 1995, 58, 1475.
2. Bermejo, A.; Protais, P.; Blazquez, M. A.; Rao, K. S.;
Zafra-Polo, M. C.; Cortes, D. Nat. Prod. Lett. 1995, 6, 57.
3. Cabedo, N.; Protais, P.; Cassels, B. K.; Cortes, D. J. Nat.
Prod. 1998, 61, 709.
4. He, X. S.; Brossi, A. J. Nat. Prod. 1993, 56, 973.
5. Polniaszek, R. P.; McKee, J. A. Tetrahedron Lett. 1987, 28,
4511.
9. Orito, K.; Miyazawa, M.; Kanbayashi, R.; Tokuda, M.;
Suginome, H. J. Org. Chem. 1999, 64, 6583.
10. Lubell, W. D.; Kitamura, M.; Noyori, R. Tetrahedron:
Asymmetry 1991, 2, 543.
11. Kitamura, M.; Hsiao, Y.; Ohta, M.; Tsukamoto, M.;
Otha, T.; Takaya, H.; Noyori, R. J. Org. Chem. 1994, 59,
297.
12. Cortes, D.; Figadere, B.; Saez, J.; Protais, P. J. Nat. Prod.
1992, 55, 1281.
6. Polniaszek, R. P.; Kaufman, C. R. J. Am. Chem. Soc. 1989,
111, 4859.
13. Cortes, D.; Arbaoui, J.; Protais, P. Nat. Prod. Lett. 1993,
2, 233.
7. Chen, C. M.; Fu, Y. F.; Yang, T. H. J. Nat. Prod. 1995, 58,
1767.
14. Litch®eld, Jr., L. T.; Wilcoxon, F. J. Pharmacol. Exp.
Ther. 1947, 96, 99.