Biosynthesis of naphthylisoquinoline alkaloids: synthesis and?incorporation of an advanced 13C2-labeled isoquinoline?precursor

Article abstract of DOI:10.1016/j.tet.2006.12.022

Biosynthetic studies on naphthylisoquinoline alkaloids involving a specifically [1,1′-¹³C₂]-labeled dihydroisoquinoline 7 are described. The synthesized precursor 7 was fed to callus cultures of Triphyophyllum peltatum and the isolat

Full text of DOI:10.1016/j.tet.2006.12.022

Tetrahedron 63 (2007) 1755–1761
Biosynthesis of naphthylisoquinoline alkaloids: synthesis
and incorporation of an advanced 13C2-labeled
isoquinoline precursor
*
€
Gerhard Bringmann, Joan Mutanyatta-Comar, Marco Greb, Stefan Rudenauer,
Torsten F. Noll and Andreas Irmer
Institute of Organic Chemistry, University of Wu€rzburg, Am Hubland, D-97074 Wu€rzburg, Germany
Received 31 October 2006; revised 1 December 2006; accepted 11 December 2006
Available online 8 January 2007
Abstract—Biosynthetic studies on naphthylisoquinoline alkaloids involving a specifically [1,10-13C2]-labeled dihydroisoquinoline 7 are
described. The synthesized precursor 7 was fed to callus cultures of Triphyophyllum peltatum and the isolated secondary metabolites were
characterized by spectroscopic methods (1H, 13C NMR, and INADEQUATE experiments). The unambiguous incorporation of the precursor
into dioncophylline A and two minor naphthylisoquinolines, together with the formation of the labeled corresponding trans-configured tetra-
hydroisoquinoline, proves the implication of such advanced intermediates in the proposed biosynthetic pathway of naphthylisoquinoline
alkaloids.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Although early feeding experiments involving 14C-labeled
precursors encountered serious setbacks (low incorporation
rates, insufficient quantities of isolated alkaloids), a major
breakthrough succeeded by the establishment of tissue cul-
tures of T. peltatum14 that do produce appreciable amounts
of dioncophylline A (1). With these cell cultures, feeding
experiments with [1,2-13C2]-labeled acetate and INADE-
QUATE analysis of the isolated dioncophylline A (1), assis-
ted by the cryoprobe technique, clearly revealed that the two
molecular portions of 1, the naphthalene and the isoquino-
line part, are each formed from six acetate units, both appar-
ently via the same b-pentaketo precursor 4 (Scheme 1).15,16
Hence, these are the first acetogenic isoquinoline alkaloids
in nature and—together with analogous results for the
Naphthylisoquinoline alkaloids,1,2 like dioncophylline A
(1),3,4 isolated from various species of the tropical plant
families Ancistrocladaceae and Dioncophyllaceae, consti-
tute a rapidly growing class of structurally intriguing natu-
rally occurring biaryl compounds.5 Besides the presence of
stereogenic centers, the compounds possess a rotationally
hindered C,C or C,N biaryl axis, which forms the basis for
axial chirality. Many of these remarkable natural products
exhibit promising bioactivities. As an example, 1 displays
antitrypanosomal6 activities, while other members of this
class of compounds show good antimalarial,7–9 antileishma-
nial,10 or anti-HIV11 activities. The substitution patterns of
these alkaloids do not fit into the—previously general—
biosynthetic pathway to tetrahydroisoquinoline alkaloids
from aromatic amino acids12 but, rather, hint at an as yet
unprecedented origin from acetate units,1,2 which is also
suggested by the co-occurrence of provenly13,14 acetogenic
naphthoquinones like plumbagin (2a) and droserone (2b),
and tetralones like isoshinanolone (3, Fig. 1) in Triphyophyl-
lum peltatum and the related plant species Ancistrocladus
heyneanus.
HO
O
Me
X
R
N
Me
P
R
H
7
1'
Me
OH
Me
MeO
MeO
O
1
2a: X = H
2b: X = OH
HO
O
R
R
Me
OH
Keywords: Naphthylisoquinoline alkaloids; Biosynthesis; Dioncophylline
A; Triphyophyllum peltatum; 13C2 labeling.
3
* Corresponding author. Tel.: +49 931 888 5323; fax: +49 931 888 4755;
e-mail: bringman@chemie.uni-wuerzburg.de
Figure 1. Dioncophylline A (1) and compounds related to its naphthalene
portion, 2a/b and 3, from Triphyophyllum peltatum (Dioncophyllaceae).
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.12.022

1756
G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
2 x 6
CO2Na
Me
OR OH
feeding
Me
5
?
Me
3
Me
6
P
R
Me
reduction
Me
R
N
H
1'
O
7
9
6a: R = H
3'
F mode
1
coupling
9'
O
O
O
6b: R = Me
O-methylation
etc.
OH Me
O
O
MeO
MeO
SR
O
OH
5'
Me
1
R
R
O
5
4
N
R = COSCoA,
COOH or H
13C2] unit
OH Me
7
[
Scheme 1. Postulated biogenetic pathway to naphthylisoquinoline alkaloids; isotope labeling of 13C2 units (in red) proven for 115 and analogously assumed for
the putative intermediates 4–7.
related compounds 2a/b and 313,14—the first experimentally
proven examples of the so-called F folding mode17 in higher
plants (i.e., with two intact acetate derived C2 units in the
first ring).
dihydroisoquinoline 7. These two molecular halves, 6 and
7, hence divergently formed from identical precursors
should then, convergently, be joined together according to
the principle of phenol-oxidative biaryl coupling5 thus, after
some follow-up reactions, eventually yielding the complete
alkaloid, here, e.g., dioncophylline A (1). In this paper, we
report on the synthesis of a two-fold 13C-labeled enantio-
merically pure dihydroisoquinoline 7 (Scheme 2) and its
successful incorporation into dioncophylline A (1) and other
naphthylisoquinoline alkaloids, using our alkaloid produc-
ing tissue cultures of T. peltatum.
According to our biosynthetic hypothesis (Scheme 1), the
open-chain hexaketide molecule 4 should undergo aldol
condensation and aromatization, to provide a monocyclic
diketone of type 5. Further aldol cyclization should give
the corresponding naphthalene portion 6 or, by reductive
amination of the (more reactive) acetonyl keto function of
5 (e.g., by transamination), deliver the respective primary
amine (not shown), which would condense to form the
2. Results and discussion
Starting from the primary amine 9, which is readily available
from the arylacetone 8,18 the synthesis of the [1,10-13C2]-
labeled dihydroisoquinoline 7 was achieved based, in
part, on a previously established19 related synthetic proce-
dure, which was further optimized in order to avoid loss
of labeled material. The 13C2 labeling was introduced by
conversion of 9 into the amide 10 by using commercially
available [1,2-13C2]-acetyl chloride, followed by Bischler–
Napieralski cyclization to give the dihydroisoquinoline 11.
Deoxygenation at C-6 was performed by regioselective
O-demethylation of 11,20 O-triflation of the resulting phenol
12,21 and subsequent Pd-catalyzed hydrogenation leading
to the monooxygenated dihydroisoquinoline 14 in excellent
yields. The hydrogenation step (with NH+4HCO2ꢀ, Pd/C)
had to be monitored carefully by TLC to avoid over reaction
leading to the cis-configured 8-O-methyltetrahydroisoqui-
noline. The synthesis of the target molecule 7 was completed
by O-demethylation of 14 with sodium thiomethoxide in
almost quantitative yield (Scheme 2).
Me
NH2
Me
MeO
MeO
MeO
HO
R
O
MeO
MeO
MeO
5
NEt3
+
O
8
9
Cl
Me
Me
Me
Me
MeO
POCl3
82%
3
N
R
R
N
O
7
H
1
Me
MeO
Me
10
HBr
11
69%
Me
TfO
NaH
Tf2O
92%
R
R
N
N
MeO
MeO
MeO
Me
Me
97%
12
13
With the labeled precursor 7 in hand, feeding experiments
were performed using the alkaloid producing cell cultures
of T. peltatum previously established by our group.14 After
the incubation, the callus cultures were harvested and the
CH2Cl2/MeOH (1:1) extracts thereof analyzed by HPLC,
which showed two significant peaks with the UV spectrum
of naphthylisoquinoline alkaloids and a third peak with
a UV trace of an isoquinoline molecular half.
-
NH4+ HCO2
Pd/C
Me
Me
NaSMe
97%
R
R
N
N
1'
Me
HO
Me
7
14
[
13C2] unit
Upon preparative HPLC isolation, the largest of these peaks
provided dioncophylline A (1) as the main metabolite (Fig. 2),
Scheme 2. Preparation of the [1,10-13C2]-labeled assumed precursor 7.

G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
1757
Me
R
N
HO
Me
7
Me
Me
R
R
HO
Me
P
M
R N
R
N
H
H
OH
OH
16
Me
Me
MeO
RO
MeO
MeO
1: R=Me
15: R=H
Me
Me
R
R
N
R
N
H
T. peltatum
Me
OH
HO
Me
17
[
13C2] unit
Figure 2. 13C2-labeled metabolites isolated from callus cultures of T. peltatum after feeding [1,10-13C2]-7.
7
initially contaminated by small quantities (ca. 11%) of its
minor 50-O-demethyl analog 15,22 but subsequently purified
by HPLC (see Section 4.3). The second peak gave the side-
chain oxygenated analog habropetaline A (16),23 and the
most polar one provided residual dihydroisoquinoline 7,
along with the trans-configured tetrahydroisoquinoline 17—
the latter two initially as a mixture, which was further
resolved by HPLC (see Section 4.3).
hint at a close organizational cooperation of the polyketide
synthase (PKS) involved, with the respective reductase or
the coupling enzyme, which may thus act as tailoring en-
zymes.25 This in turn, may be the reason why the isoquino-
line portion of naphthylisoquinoline alkaloids has (nearly)
never been found free in nature,26 never coupled to each
other, but always to a naphthalene portion. From the fact
that 17 is diluted by natural material, while 7 is fully labeled,
one can further conclude that 7 and 17 are not in a (rapid) re-
dox equilibrium with each other, because apparently 7 is re-
duced to 17, but 17 is not oxidized back to 7 to a noticeable
degree.
The compound investigated first by 13C NMR was the
apparently newly formed trans-configured tetrahydroisoqui-
noline 17, which had not been visible in the chromatogram
of the untreated cell cultures and whose formation had
thus been triggered by the application of the labeled
compound 7. As expected, it showed a very high labeling
degree of ca. 94%. That this substrate was not an entirely
unnatural metabolite of a likewise unnatural dihydroisoqui-
noline 7, was made plausible by the exclusive formation
of the respective trans-isomer; the corresponding cis-diaste-
reomer was not detected. This hints at a specific, and thus
an enzymatic process, since the chemical reduction of such
dihydroisoquinolines, by catalytic hydrogenation or using
common hydride transfer reagents like NaBH4, normally
produces cis,19 while the directed chemical preparation
of trans is synthetically more demanding and requires par-
ticular reaction conditions.24 Moreover, the fact that the
labeling degree is not complete (i.e., 100%), but ‘only’
With all this information at hand—that 7 is at least accepted
by a specific reductase, and is thus apparently accessible to
the enzymes of the cells—the thrilling question was whether
an incorporation had taken place into the isolated main alka-
loid, the complete naphthylisoquinoline dioncophylline A
(1). Thus the most important result of the present study
was the unambiguous incorporation of 7 into 1 to the signif-
icant extent of ca. 5%, as clearly seen from the two distinct
doublets for C-1 and Me-1 at 50.0 and 18.1 ppm, respec-
tively (Fig. 3). These new peaks even exceeded the size of
those resulting from ‘non-labeled’ dioncophylline A (1),
i.e., with the natural 13C abundance of ca. 1%. The results
evidence the first successful incorporation of a precursor
more advanced and differentiated than acetate, into a naph-
thylisoquinoline alkaloid.
1
94% (determined by H and 13C NMR), reveals that the
[1,10-13C2]-labeled compound 17 is diluted by non-
labeled—hence natural—material, so that 17 must be a
normal, constitutional metabolite in the cell cultures of
T. peltatum.
The quantities of the likewise present minor alkaloid, 50-O-
demethyldioncophylline A (15), by contrast, were not suffi-
cient to provide a good 13C NMR spectrum after complete
purification; still, the initial 11:89 mixture of 15 with the
main alkaloid dioncophylline A (1) clearly showed an incor-
poration into 15, too, exactly as that for 1 (Fig. S3, Supple-
mentary data).
The recovered dihydroisoquinoline 7, by contrast, was found
to be fully labeled with no signs of the respective singlets for
the non-labeled analog, which should just possess the natural
13C abundance of ca. 1% (Fig. S2, Supplementary data).
Nonetheless this as yet merely synthetic compound, which
has never been found as such in nature, still might occur
even in the non-treated cell cultures—in steady-state con-
centrations that can be very low, because it is usually imme-
diately reduced to 17 and then coupled to dioncophylline A
(1) or, vice versa, first coupled and then reduced. This may
In the case of the other minor alkaloid, habropetaline A (16),
by contrast, sufficient material was obtained after complete
purification, which again, as already in the case of 1, permit-
ted to prove a significant incorporation of 7 to an extent of ca.
4% (for the respective spectra, see Fig. S19, Supplementary
data).

1758
G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
20
C-1
Me-1
40
60
80
100
120
140
160
180
200
220
C-1
Me-1
Me
R
Me
P
R
N
H
OH
40
Me
MeO
MeO
1
50.2
49.8
18.4
18.0
180
160
140
120
100
80
60
20
0
13C (ppm)
Figure 3. INADEQUATE spectrum (MeOH-d4) of dioncophylline A (1) after feeding [1,10-13C2]-dihydroisoquinoline 7 to callus cultures of T. peltatum. The
expanded region (see included window left) indicates the two doublets of the 13C2 unit of C-1 (50.0 ppm) and Me-1 (18.1 ppm) into which the 13C2 label was
incorporated.
3. Conclusions
silica gel precoated glass plates (60 F254, Merck). Visuali-
zation was accomplished by irradiation with UV light at
254 nm and 365 nm. Flash chromatography was performed
using silica gel (0.063 mm, Merck). Melting points were
obtained on a Stuart Scientific SMP10 instrument and are
This work constitutes the first proven incorporation of a more
advanced biosynthetic precursor, viz. the dihydroisoquino-
line 7, into naphthylisoquinoline alkaloids, proving, for the
first time, that the two molecular portions, the naphthalene
and the isoquinoline, are formed separately, and are coupled
to each other in a quite advanced form (i.e., not as an
open-chain or monocyclic precursor). The smooth diastereo-
selective in vivo conversion of 7 to the respective trans-
tetrahydroisoquinoline, 17 (and not to the cis-configured
one, not shown), clearly indicates that 7 has access to the
enzymes of the cell and is a substrate to its redox system
and, in connection with its incorporation into the naphthyl-
isoquinolines 1, 15, and 16, possibly also to the respective
coupling enzymes. The observed slight dilution of the label-
ing degree in 17, but not in the reisolated 7, does not neces-
sarily mean that 17 is the immediate coupling precursor; it
might also be that 7 is coupled and that normally very small
portions are reduced to the tetrahydroisoquinoline 17, while
now, under the feeding conditions, larger quantities get
reduced and thus remain uncoupled. Thus, it remains to be
established, which of these two compounds, 7 or 17, is the
authentic coupling substrate, and also what exactly the naph-
thalene precursor is—whether it is a dihydroxy compound
6a or its methyl ether 6b. This work, as well as the search
for the involved enzymes, is presently in progress.
1
uncorrected. H NMR spectra were recorded on a Bruker
Avance 400 spectrometer (400 MHz). Optical rotations were
measured on a JASCO P-1020 polarimeter. IR spectra were
taken on a JASCO FT-IR-410 spectrometer. HRMS were ob-
tained on a micrOTOF-focus mass spectrometer (Bruker
Daltonik GmbH). The following chemicals were purchased
and used as received: NaH (Fluka), Tf2O (Sigma Aldrich),
€
ammonium formate (Riedel-de Haen), Pd/C (Fluka), and
NaSMe (ACROS). CH3CN, CH2Cl2, DMF, and toluene
were dried over CaH2; MeOH was dried over sodium,
and all solvents were freshly distilled prior to use.
[1,2-13C2]-Labeled acetyl chloride (99% labeling degree)
was purchased from Aldrich and used without any further
purification. Deactivated silica gel was prepared by mixing
silica gel with 7.5 vol % NH3. All labeled synthetic com-
1
pounds showed a 13C2 degree of ꢁ98.5 according to H
and 13C NMR. The IR data of compounds 10, 11, and 12
are in agreement with those of the non-labeled analogs19
and are given in the Supplementary data.
4.2. Synthesis of [1,10-13C2]-8-hydroxy-3,4-dihydro-
1,3-dimethylisoquinoline
4. Experimental
4.1. General
4.2.1. N-[(R)-1-(3,5-Dimethoxyphenyl)propan-2-yl]-
[1,2-13C2]-acetamide (10). Amine 9$HCl (2.74 g,
11.8 mmol, 1.0 equiv) was placed under nitrogen in a
flame-dried flask. Dichloromethane (80 mL) and freshly
distilled NEt3 (2.63 g, 26.0 mmol, 2.2 equiv) were added
and the mixture was cooled to 0 ꢂC. After addition of
All reactions were carried out under a nitrogen atmosphere
with magnetic stirring. Analytical TLC was performed on

G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
1759
[1,2-13C2]-acetyl chloride (1.00 g, 12.4 mmol, 1.05 equiv),
1H, Hax-4), 2.51 (ddd, 1JHC¼131.0 Hz, 2JHC¼6.1 Hz, JHH
¼
the reaction mixture was stirred for 30 min at room temper-
ature and then quenched with 2 N HCl. The organic layer
was washed twice with 2 N HCl, then with water, dried
over MgSO4, and the solvent was removed under reduced
pressure to give 10 in quantitative yield as white crystals;
mp 83–84 ꢂC (CH2Cl2); lit.19,27 mp 85 ꢂC (CH2Cl2/cyclo-
hexane); [a]2D0 +23.5 (c 0.08, CH2Cl2); lit.19,27 [a]D20 +10.1
(c 0.5, MeOH); 1H NMR (400 MHz, CDCl3) d 6.33 (s, 3H,
H-20, H-40, and H-60), 5.30 (br s, 1H, NH), 4.22–4.28
(m, 1H, H-2), 3.78 (s, 6H, OCH3), 2.79 (dd, 2JHH¼13.4 Hz,
0.9 Hz, 3H, CH3-1), 1.33 (d, 3JHH¼6.4 Hz, 3H, CH3-3); 13
C
NMR (100 MHz, CD3OD) d 169.2 (d, JCC¼41.2 Hz, C-1),
167.5 (C-6 and C-8), 142.7 (C-10), 115.6 (C-5), 101.8
(C-7), 55.8 (OCH3), 49.1 (C-3), 36.7 (C-4), 23.9 (d, JCC
¼
41.2 Hz, CH3-1), 18.9 (CH3-3); EIMS (70 eV) m/z (rel
int.) 207.2 (100) [M]+, 192.2 (37) [M+ꢀCH3]; HRESIMS
208.1242 ([M+H]+, 208.1242 calcd for [13C2]-C10H16NO2).
4.2.4. [1,10-13C2]-(3R)-3,4-Dihydro-8-methoxy-1,3-di-
methylisoquinolin-6-yl trifluoromethanesulfonate (13).
6-Hydroxydihydroisoquinoline 12 (103 mg, 0.49 mmol,
1.0 equiv) was placed under nitrogen in a flame-dried
Schlenk flask and CH2Cl2 (8 mL) was added. After cooling
the flask to 0 ꢂC, NaH (44.0 mg of a 60% dispersion in oil,
1.09 mmol, 2.2 equiv) was added and the solution was
stirred for 30 min at 0 ꢂC. A solution of Tf2O (115 mg,
0.54 mmol, 1.1 equiv) in CH2Cl2 (15 mL) was added drop-
wise at 0 ꢂC over a period of 15 min and after stirring for
5 min, the reaction mixture was directly filtered through
a short pad of silica gel and then washed with CH2Cl2 giving
13 as a yellow oil (155 mg, 0.45 mmol, 92%); [a]2D0 ꢀ65.4
(c 0.1, CH2Cl2); IR (KBr) nmax 2967, 1606, 1466, 1606,
2
3
3JHH¼5.8 Hz, 1H, H-1), 2.64 (dd, JHH¼13.4 Hz, JHH
¼
1
2
7.2 Hz, 1H, H-1), 1.94 (dd, JHC¼128.0 Hz, JHC¼6.0 Hz,
3H, COCH3), 1.12 (d, 3JHH¼6.7 Hz, 3H, CH3-3); 13C NMR
(100 MHz, CDCl3) d 172.2 (d, JCC¼51.0 Hz, COCH3),
163.6 (C-30, C-50), 143.0 (C-10), 110.3 (C-20 and C-40),
101.3 (C-60), 58.1 (OCH3), 48.8 (C-2), 45.5 (C-1), 26.3 (d,
JCC¼51.0 Hz, COCH3), 22.8 (CH3-3); EIMS (70 eV) m/z
(rel int.) 239.0 (17) [M]+, 178.0 (100) [M+ꢀNHCOCH3];
HRESIMS 262.1327 ([M+Na]+, 262.1330 calcd for [13C2]-
C13H20NO3Na).
4.2.2. [1,10-13C2]-(3R)-3,4-Dihydro-6,8-dimethoxy-1,3-di-
methylisoquinoline (11). Under nitrogen, a solution of 10
(2.83 g, 11.8 mmol, 1.0 equiv) and freshly distilled POCl3
(2.13 g, 13.8 mmol, 1.15 equiv) in CH3CN (40 mL) was
stirred at 70 ꢂC for 45 min. After removal of the solvent un-
der reduced pressure, the residue was carefully quenched by
addition of an aqueous ammonia solution at 0 ꢂC and the
aqueous layer was extracted with CH2Cl2. Drying of the
organic layer over MgSO4 and removal of the solvent under
reduced pressure yielded the dihydroisoquinoline 11 as
a brown oil (2.30 g, 10.4 mmol, 82%, 2 steps); [a]2D0+69
(c 0.10, MeOH); lit.19,27 [a]2D0+139 (c 0.6, MeOH); 1H
NMR (250 MHz, CDCl3) d 6.35 (s, 1H, H-7), 6.31 (s, 1H,
H-5), 3.82 (s, 6H, OCH3), 3.24–3.36 (m, 1H, H-3), 2.59
1
1305, 1244, 1217, 1141, 1120, 979, 834 cmꢀ1; H NMR
(250 MHz, CDCl3) d 6.71 (s, 2H, H-5, H-7), 3.88 (s, 3H,
OCH3), 3.25–3.35 (m, 1H, H-3), 2.65 (dd, JHH¼15.8 Hz,
2
1
3JHH¼4.6 Hz, 1H, Heq-4), 2.43 (ddd, JHC¼128.0 Hz,
2JHC¼7.0 Hz, JHH¼0.9 Hz, 3H, CH3-1), 2.36 (dd,
3
2JHH¼15.8 Hz, JHH¼13.7 Hz, 1H, Hax-4), 1.38 (d,
3JHH¼6.7 Hz, 3H, CH3-3); 13C NMR (100 MHz, CDCl3)
d 162.5 (d, JCC¼50.2 Hz, C-1), 158.9 (C-8), 150.9 (C-6),
3
143.4 (C-10), 118.7 (q, JCF¼320.9 Hz), 112.6 (C-5 and
C-9), 104.2 (C-7), 56.3 (OCH3), 51.5 (C-3), 34.8 (C-4),
27.7 (d, JCC¼50.2 Hz, CH3-1), 22.0 (CH3-3); EIMS
(70 eV) m/z (rel int.) 339.0 (13) [M]+, 206.1 (100)
[M+ꢀTf]; HRESIMS 340.0730 ([M+H]+, 340.0736 calcd
for [13C2]-C11H15F3NO4S).
2
3
(dd, JHH¼15.6 Hz, JHH¼4.2 Hz, 1H, Heq-4), 2.42 (ddd,
2
1JHC¼128.0 Hz, JHC¼6.7 Hz, JHH¼1.8 Hz, 3H, CH3-1),
2
3
2.34 (dd, JHH¼15.5 Hz, JHH¼13.1 Hz, 1H, Hax-4), 1.35
4.2.5. [1,10-13C2]-(3R)-3,4-Dihydro-8-methoxy-1,3-di-
methylisoquinoline (14). A suspension of 13 (101 mg,
0.29 mmol, 1.0 equiv), ammonium formate (78.0 mg,
1.23 mmol, 4.0 equiv), and 5 mg Pd/C in MeOH (7 mL)
was stirred under nitrogen at 80 ꢂC (using a preheated oil
bath) for 5 min. The reaction mixture was filtered through a
short pad of Celite and washed with MeOH. After removal of
the eluent under reduced pressure, the obtained residue was
purified by column chromatography on deactivated silica gel
using an increasing gradient of CH2Cl2/MeOH (100:3 up to
100:7) giving 14 (55.0 mg, 0.28 mmol, 97%) as a yellow oil;
[a]2D0 ꢀ96.4 (c 0.1, CH2Cl2); IR (KBr) nmax 2963, 2940,
1592, 1473, 1326, 1275, 1159, 1091, 1033 cmꢀ1; 1H NMR
3
(d, JHH¼6.7 Hz, 3H, CH3-3); 13C NMR (100 MHz,
CDCl3) d 166.0 (d, JCC¼48.8 Hz, C-1), 164.7 (C-6), 161.8
(C-8), 145.1 (C-10), 107.0 (C-5), 99.9 (C-7 and C-9), 58.1
(OCH3), 54.0 (C-3), 38.0 (C-4), 30.2 (d, JCC¼48.8 Hz,
CH3-1), 24.5 (CH3-3); EIMS (70 eV) m/z (rel int.) 221.2
(100) [M]+, 206.1 (44) [M+ꢀCH3], 191.1 (17)
[M+ꢀ2CH3]; HRESIMS 222.1397 ([M+H]+, 222.1399
calcd for [13C2]-C13H18NO2).
4.2.3. [1,10-13C2]-(3R)-3,4-Dihydro-6-hydroxy-8-meth-
oxy-1,3-dimethylisoquinoline (12). Dimethoxydihydro-
isoquinoline 11 (575 mg, 2.59 mmol, 1.0 equiv) was stirred
in 12 mL of 62% hydrobromic acid for 45 min at 110 ꢂC.
Excess HBr was removed under reduced pressure, the resi-
due was treated with water and all volatiles were removed
in vacuo. After column chromatography on deactivated sil-
ica gel, using an increasing gradient of CH2Cl2/MeOH
(100:3 up to 100:7) (v/v) as an eluent, the monohydroxy
compound 13 (376 mg, 1.78 mmol, 69%) was obtained as a
yellow-brownish solid; mp 213 ꢂC (CH2Cl2/MeOH);19,28
[a]2D0+284 (c 0.1, MeOH);19,28 1H NMR (250 MHz, CD3OD)
d 6.00 (s, 1H, H-7), 5.98 (s, 1H, H-5), 3.82 (s, 3H, OCH3),
3
3
(400 MHz, CDCl3) d 7.28 (dd, J¼7.3 Hz, J¼8.2 Hz, 1H,
H-6), 6.84 (d, J¼8.2 Hz, 1H, H-5 or H-7), 6.77 (d,
J¼7.3 Hz, 1H, H-5 or H-7), 3.86 (s, 3H, OCH3), 3.25–3.35
2
3
(m, 1H, H-3), 2.63 (dd, JHH¼15.6 Hz, JHH¼4.6 Hz,
1
2
1H, Heq-4), 2.46 (ddd, JHC¼128.0 Hz, JHC¼7.0 Hz,
2
JHH¼2.1 Hz, 3H, CH3-1), 2.35 (dd, JHH¼15.8 Hz,
3JHH¼14.7 Hz, 1H, Hax-4), 1.38 (d, JHH¼6.7 Hz, 3H,
3
CH3-3); 13C NMR (100 MHz, CDCl3) d 164.9 (d,
JCC¼48.7 Hz, C-1), 158.2 (C-8), 140.4 (C-10), 132.0 (C-
6), 119.9 (C-5 and C-9), 110.3 (C-7), 55.4 (OCH3), 50.9
(C-3), 34.3 (C-4), 27.0 (d, JCC¼48.7 Hz, CH3-1), 21.1
(CH3-3); EIMS (70 eV) m/z (rel int.) 191.1 (100) [M]+,
2
3
3.55–3.75 (m, 1H, H-3), 2.82 (dd, JHH¼15.6 Hz, JHH
¼
4.9 Hz, 1H, Heq-4), 2.59 (dd, 2JHH¼15.6 Hz, 3JHH¼11.8 Hz,

1760
G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
176.1 (33) [M+ꢀCH3]; HRESIMS 192.1294 ([M+H]+,
available authentic 1 and synthetically prepared 7 and 17.
The isolated natural products 1, 15, and 16 were spectro-
scopically identical with the materials isolated earlier.3,22,23
In order to establish the C–C connectivities for labeled
dioncophylline A (1) and its 50-O-demethylated analog, 15, a
2D-INADEQUATE spectrum was measured on a Bruker
DMX600 spectrometer. The labeling degree of all isolated
compounds was determined by 13C NMR.
192.1294 calcd for [13C2]-C10H16NO).
4.2.6. [1,10-13C2]-(3R)-3,4-Dihydro-8-hydroxy-1,3-di-
methylisoquinoline (7). To a solution of 14 (90.0 mg,
0.47 mmol, 1.0 equiv) in DMF (4 mL), NaSMe (148 mg,
2.11 mmol, 4.5 equiv) was added under nitrogen and the
mixture was stirred at 160 ꢂC for 45 min. Repeatedly, water
was added and all volatiles were removed in vacuo. The re-
sulting residue was purified by column chromatography on
deactivated silica gel, using CH2Cl2/MeOH (100:8) as an
eluent, to afford 7 (81.0 mg, 0.47 mmol, 97%) as yellow
crystals; mp 195–197 ꢂC (dec., CH2Cl2/MeOH); [a]2D0 +67.0
(c 0.1, MeOH); IR (KBr) nmax 2924, 1685, 1606, 1465, 1205,
1135, 800 cmꢀ1; 1H NMR (400 MHz, CD3OD) d 7.53 (dd,
Acknowledgements
The authors acknowledge the DFG (SPP 1152 ‘Evolution
of Metabolic Diversity’) and the Fonds der Chemischen
Industrie for financial support of this work. J.M.-C. is
grateful to the Alexander-von-Humboldt Foundation for her
Georg-Forster research fellowship. Further thanks are due
3
3J¼8.9 Hz, J¼7.0 Hz, 1H, H-6), 6.93 (d, J¼8.9 Hz, 1H,
H-5 or H-7), 6.86 (d, J¼7.0 Hz, 1H, H-5 or H-7), 3.93–
€
to Dr. M. Grune and T. A. M. Gulder for their assistance.
4.05 (m, 1H, H-3), 2.85 (dd, 1JHC¼131.0 Hz, 2JHC¼6.2 Hz,
2
3
3H, CH3-1), 3.11 (dd, JHH¼16.4 Hz, JHH¼5.2 Hz, 1H,
2
3
Supplementary data
Heq-4), 2.87 (dd, JHH¼16.4 Hz, JHH¼11.5 Hz, 1H,
3
Hax-4), 1.45 (d, JHH¼6.7 Hz, 3H, CH3-3); 13C NMR
Spectroscopic details for compounds 17 and 18. Copy of the
INADEQUATE spectrum of the mixture of 1 and 15. Copies
of the 1H and 13C NMR spectra for 1, 7, 10–14 and 16–18.
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.tet.2006.12.022.
(100 MHz, CD3OD) d 178.3 (d, JCC¼40.4 Hz, C-1), 163.0
(C-8), 140.3 (C-10), 139.8 (C-6), 120.6 (C-5), 117.6 (C-7
and C-9), 50.2 (C-3), 34.6 (C-4), 24.8 (d, JCC¼40.1 Hz,
CH3-1), 18.0 (CH3-3); EIMS (70 eV) m/z (rel int.) 177.2
(100) [M]+, 162.2 (44) [M+ꢀCH3], 148.2 (43); HRESIMS
178.1138 ([M+H]+, 178.1143 calcd for [13C2]-C11H14NO).
References and notes
4.3. Feeding experiments
1. Bringmann, G.; Pokorny, F. In The Alkaloids; Cordell, G. A.,
Ed.; Academic: New York, NY, 1995; Vol. 46, pp 127–271.
ꢀ
2. Bringmann, G.; Franc¸ois, G.; Ake Assi, L.; Schlauer, J. Chimia
1998, 52, 18–28.
3. Bringmann, G.; R€ubenacker, M.; Jansen, J. R.; Scheutzow, D.;
Callus cultures of T. peltatum were cultivated as previously
described.14 The [1,10-13C2]-labeled precursor 7 (77.6 mg)
was administered aseptically, in triplicate, to 5.84 g (dry
weight) callus for a period of nine weeks. The harvested,
dried callus material was extracted with CH2Cl2/MeOH
(1:1) and after evaporation of the solvent in vacuo, the crude
organic extract was purified by preparative HPLC on a Sym-
metry RP18 column (Waters, 19ꢃ300 mm, 7 mm) employing
the following gradient: H2O (A)/CH3CN (B) and 0.05% TFA
(v/v); flow rate 12 mL minꢀ1; 0 min 5% B, 30 min 70% B,
35 min 100% B, 40 min 100% B, 41 min 5% B, 46 min
5% B. This yielded a mixture of dioncophylline A (1) along
with its 50-O-demethylated analog (15, tR¼20.6 min,
14 mg), habropetaline A (16, tR¼16.4 min, 4 mg, 4% 13C2
labeling degree), and a mixture of trans-configured tetrahy-
droisoquinoline 17 together with reisolated 7 (tR¼11.3 min,
14 mg). Further purification on a Chromolith Semi Prepara-
tive RP18 column (100ꢃ10 mm) using H2O (A)/CH3CN (B)
and 0.05% TFA (v/v); flow rate 10 mL minꢀ1; 0 min 0% B,
2 min 0% B, 8 min 15% B, 23 min 35% B, 25 min 50% B,
27 min 50% B, 27.5 min 0% B, 30 min 0% B, gave dionco-
phylline A (1, 4.9 mg, 5% 13C2 labeling degree), while the
tetrahydroisoquinoline 17 (7 mg, 94% 13C2 labeling degree)
and its dihydro analog 7 (7 mg, 98.5% 13C2 labeling degree)
were separated by the following gradient: flow rate
12 mL minꢀ1; 0 min 0% B, 4 min 0% B, 4.5 min 2% B,
10 min 2% B, 10.5 min 5% B, 16 min 5% B, 16.5 min
25% B, 20 min 0% B, 30 min 0% B.
ꢀ
Ake Assi, L. Tetrahedron Lett. 1990, 31, 639–642.
4. Bringmann, G.; Jansen, J. R.; Reuscher, H.; R€ubenacker, M.;
Peters, K.; von Schnering, H. G. Tetrahedron Lett. 1990, 31,
643–646.
5. Bringmann, G.; G€unther, C.; Ochse, M.; Schupp, O.; Tasler, S.
In Progress in the Chemistry of Organic Natural Products;
Herz, W., Falk, H., Kirby, G. W., Moore, R. E., Tamm, C.,
Eds.; Springer: Wien, New York, NY, 2001; Vol. 82, pp 111–
123.
6. Bringmann, G.; Hoerr, V.; Holzgrabe, U.; Stich, A. Pharmazie
2003, 58, 343–346.
ꢀ
7. Franc¸ois, G.; Timperman, G.; Eling, W.; Ake Assi, L.; Holenz,
J.; Bringmann, G. Antimicrob. Agents Chemother. 1997, 41,
2533–2539.
8. Franc¸ois, G.; Chimanuka, B.; Timperman, G.; Holenz, J.;
ꢀ
Plaizier-Vercammen, J.; Ake Assi, L.; Bringmann, G.
Parasitol. Res. 1999, 85, 935–941.
´
9. Bringmann, G.; Feineis, D. Act. Chim. Therapeut. 2000, 26,
151–171.
10. Bringmann, G.; Kajahn, I.; Reichert, M.; Pedersen, S. E. H.;
Faber, J. H.; Gulder, T.; Brun, R.; Christensen, S. B.; Ponte-
Sucre, A.; Moll, H.; Heubl, G.; Mudogo, V. J. Org. Chem.
2006, 71, 9348–9356.
11. Boyd, M. R.; Hallock, Y. F.; Cardellina, J. H., II; Manfredi,
K. P.; Blunt, J. W.; McMahon, J. B.; Buckheit, R. W., Jr.;
€
Structural elucidation of the isolated compounds was
achieved by 1D- (for 1, 7, 15, 16, and 17) and 2D-NMR
(HMQC, HMBC, COSY, and NOESY, for 1, 15, and 16) ex-
periments and, in the case of 1, 7, and 17, by co-elution with
Bringmann, G.; Schaffer, M.; Cragg, G. M.; Thomas, D. W.;
Jato, J. G. J. Med. Chem. 1994, 37, 1740–1745.
12. Dewick, P. M. Medicinal Natural Products; Wiley-VCH:
Weinheim, 2002; pp 315–326.

G. Bringmann et al. / Tetrahedron 63 (2007) 1755–1761
1761
€
ꢀ
23. Bringmann, G.; Messer, K.; Schwobel, B.; Brun, R.; Ake Assi,
13. Bringmann, G.; Wohlfarth, M.; Rischer, H.; R€uckert, M.;
Schlauer, J. Tetrahedron Lett. 1998, 39, 8445–8448.
ꢀ
14. Bringmann, G.; Rischer, H.; Wohlfarth, M.; Schlauer, J.; Ake
Assi, L. Phytochemistry 2000, 53, 339–343.
L. Phytochemistry 2003, 62, 345–349.
24. Bringmann, G.; Jansen, J. R.; Rink, H.-P. Angew. Chem., Int.
Ed. Engl. 1986, 25, 913–915.
15. Bringmann, G.; Wohlfarth, M.; Rischer, H.; Gr€une, M.;
Schlauer, J. Angew. Chem., Int. Ed. 2000, 39, 1464–1466.
16. Bringmann, G.; Feineis, D. J. Exp. Bot. 2001, 52, 2015–2022.
17. Thomas, R. ChemBioChem. 2001, 2, 612–627.
25. (a) Shen, B. In Topics in Current Chemistry; Leeper, F. J.,
Vederas, J. C., Eds.; Springer: Berlin, Heidelberg, 2000; Vol.
200, pp 1–51; (b) Rix, U.; Fischer, C.; Remsing, L. L.; Rohr,
J. Nat. Prod. Rep. 2002, 19, 542–580.
18. Knupp, G.; Frahm, A. W. Arch. Pharm. 1985, 318, 535–542.
19. Bringmann, G.; Weirich, R.; Reuscher, H.; Jansen, J. R.;
Kinzinger, L.; Ortmann, T. Liebigs Ann. Chem. 1993, 877–888.
20. Likewise obtained were small quantities (1.6% isolated) of the
8-OH regioisomer (18, Fig. S22, Supplementary data).
21. The initially observed formation of the respective O,N-bistri-
flated compound as an undesired byproduct was avoided by
first diluting the triflic anhydride, see Section 4.
26. For few exceptions, see: (a) Hallock, Y. F.; Manfredi, K. P.;
€
Blunt, J. W.; Cardellina, J. H., II; Schaffer, M.; Gulden, K.-P.;
Bringmann, G.; Lee, A. Y.; Clardy, J.; Franc¸ois, G.; Boyd,
M. R. J. Org. Chem. 1994, 59, 6349–6355; (b) Hallock, Y. F.;
Cardellina, J. H., II; Kornek, T.; Gulden, K.-P.; Bringmann,
G.; Boyd, M. R. Tetrahedron Lett. 1995, 36, 4753–4756; (c)
Bringmann, G.; Messer, K.; Wohlfarth, M.; Kraus, J.;
€
Dumbuya, K.; Ruckert, M. Anal. Chem. 1999, 71, 2678–
€
22. Bringmann, G.; Saeb, W.; God, R.; Schaffer, M.; Franc¸ois, G.;
Peters, K.; Peters, E.-M.; Proksch, P.; Hostettmann, K.; Ake
2686.
27. Lit.19 value is for the non-labeled analog.
28. Lit.19 value is only given for the hydrobromide.
ꢀ
Assi, L. Phytochemistry 1998, 49, 1667–1673.