Total synthesis of the N,C-coupled naphthylisoquinoline alkaloids ancistrocladinium A and B and related analogues

Article abstract of DOI:10.1021/ja9097687

The N,C-coupled naphthyldihydroisoquinoline alkaloids ancistrocladinium A (3) and B (4), which possess an unprecedented iminium-aryl axis and show high in vitro antileishmanial activities, have been synthesized via a short sequence of eight linear steps, without the need of protecting groups. Key steps were a Buchwald-Hartwig amination and a Bischler-Napieralski cyclization, preferentially leading to the naturally predominant M-atropo-diastereomer in the case of 3, while the N,C-axis is configurationally semistable in 4. The highly convergent first access to this type of alkaloids will now facilitate the preparation of structural analogues for structure-activity relationship studies. Its general applicability was shown by the preparation of the sterically even more congested, as yet unnatural N,3′- and N,1′-coupled analogues, ancistrocladinium C (5) and D (6).

Full text of DOI:10.1021/ja9097687

Published on Web 12/31/2009
Total Synthesis of the N,C-Coupled Naphthylisoquinoline
Alkaloids Ancistrocladinium A and B and Related Analogues
Gerhard Bringmann,* Tanja Gulder,† Barbara Hertlein, Yasmin Hemberger, and
Frank Meyer
Institute of Organic Chemistry, UniVersity of Wu¨rzburg, Am Hubland,
D-97074 Wu¨rzburg, Germany
Received November 17, 2009; E-mail: bringman@chemie.uni-wuerzburg.de
Abstract: The N,C-coupled naphthyldihydroisoquinoline alkaloids ancistrocladinium A (3) and B (4), which
possess an unprecedented iminium-aryl axis and show high in vitro antileishmanial activities, have been
synthesized via a short sequence of eight linear steps, without the need of protecting groups. Key steps
were a Buchwald-Hartwig amination and a Bischler-Napieralski cyclization, preferentially leading to the
naturally predominant M-atropo-diastereomer in the case of 3, while the N,C-axis is configurationally
semistable in 4. The highly convergent first access to this type of alkaloids will now facilitate the preparation
of structural analogues for structure-activity relationship studies. Its general applicability was shown by
the preparation of the sterically even more congested, as yet unnatural N,3′- and N,1′-coupled analogues,
ancistrocladinium C (5) and D (6).
in the rainforest in the Democratic Republic of Congo,9,10 we
Introduction
have recently discovered the first N,C-coupled naphthyldihy-
droisoquinolines, ancistrocladinium A (3), which has an un-
precedented rotationally hindered N,C-axis, and B (4), which
is configurationally semistable.9 Because 3 and 4 display very
good antiparasitic properties, especially against the protozoan
pathogen Leishmania major even in the low micromolar range,11
these natural products are promising lead structures12,13 for
urgently needed novel anti-infective drugs. The facile availability
of these compounds in sufficient quantities for further biological
and medicinal studies, for example, by total synthesis, is thus
an important task.
Plants of the families Ancistrocladaceae and Dioncophyl-
laceae are a rich, and the as yet only, source of naphthyliso-
quinoline alkaloids.1,2 More than 140 representatives of these
structurally and biosynthetically unique, bioactive natural
products have so far been isolated from the evergreen vines
indigenous to the rainforests of tropical Africa and South-East
Asia.3 Some of them show promising bioactivities against
pathogens of severe, and widespread, tropical diseases. As an
example, dioncophylline C (1, see Figure 1) exhibits a strong
activity against different Plasmodium species, both in vitro and
in ViVo.4-6 Butler et al.7 described the occurrence of a first,
apparently racemic, N,C-coupled naphthylisoquinoline alkaloid,
ancisheynine (2), which was synthetically accessed and stere-
ochemically characterized by our group.8 From an as yet not
fully identified, possibly new Ancistrocladus species occurring
Most of the numerous “normal”, C,C-coupled naphthyliso-
quinoline alkaloids synthesized so far have been constructed
by using the “lactone method”,14-17 an efficient pathway to
(9) 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.
(10) Bringmann, G.; Spuziak, J.; Faber, J. H.; Gulder, T.; Kajahn, I.; Dreyer,
M.; Heubl, G.; Brun, R.; Mudogo, V. Phytochemistry 2008, 69, 1065.
(11) Ponte-Sucre, A.; Faber, J. H.; Gulder, T.; Kajahn, I.; Pedersen, S. E. H.;
Schultheis, M.; Bringmann, G.; Moll, H. Antimicrob. Agents Chemo-
ther. 2007, 51, 188.
† Present address: Department of Chemistry, The Scripps Research
Institute, 10550 North Torrey Pines Road, La Jolla, California 92037.
(1) Bringmann, G.; Pokorny, F. In The Alkaloids; Cordell, G. A., Ed.;
Academic Press: San Diego, CA, 1995; pp 127-171.
(2) Bringmann, G.; Franc¸ois, G.; Ake´ Assi, L.; Schlauer, J. Chimia 1998,
52, 18.
(3) Bringmann, G.; Gu¨nther, 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., Eds.; Springer: Vienna, 2001;
pp 1-293.
(12) Bringmann, G.; Gulder, T.; Hentschel, U.; Meyer, F.; Moll, H.;
Morschha¨user, J.; Ponte-Sucre De Vanegas, A.; Ziebuhr, W.; Stich,
A.; Brun, R.; Mu¨ller, W. E. G.; Mudogo, V. Biofilm-Inhibiting Effect
and Anti-Infective Activity of N,C-Linked Arylisoquinolines and the
Use Thereof; PCT/EP2007/008440, 2007.
(4) Franc¸ois, G.; Timperman, G.; Eling, W.; Ake´ Assi, L.; Holenz, J.;
Bringmann, G. Antimicrob. Agents Chemother. 1997, 41, 2533.
(5) Schwedhelm, K. F.; Horstmann, M.; Faber, J. H.; Reichert, Y.;
Bringmann, G.; Faber, C. ChemMedChem 2007, 2, 541.
(6) Franc¸ois, G.; Timperman, G.; Holenz, J.; Ake´ Assi, L.; Geuder, T.;
Maes, L.; Dubois, J.; Hanocq, M.; Bringmann, G. Ann. Trop. Med.
Parasitol. 1996, 90, 115.
(13) Ponte-Sucre, A.; Gulder, T.; Wegehaupt, A.; Albert, C.; Rikanovic´,
C.; Scha¨flein, L.; Frank, A.; Schultheis, M.; Unger, M.; Holzgrabe,
U.; Bringmann, G.; Moll, H. J. Med. Chem. 2009, 52, 626.
(14) Bringmann, G.; Ochse, M.; Goetz, R. J. Org. Chem. 2000, 65, 2069.
(15) Bringmann, G.; Saeb, W.; Ru¨benacker, M. Tetrahedron 1999, 55, 423.
(16) Bringmann, G.; Holenz, J.; Weirich, R.; Ru¨benacker, M.; Funke, C.;
Boyd, M. R.; Gulakowski, R. J.; Franc¸ois, G. Tetrahedron 1998, 54,
497.
(7) Yang, L.-K.; Glover, R. P.; Yonganathan, K.; Sarnaik, J. P.; Godbole,
A. J.; Soejarto, D. D.; Buss, A. D.; Butler, M. S. Tetrahedron Lett.
2003, 44, 5827.
(8) Bringmann, G.; Gulder, T.; Reichert, M.; Meyer, F. Org. Lett. 2006,
8, 1037.
(17) Bringmann, G.; Jansen, J. R.; Reuscher, H.; Ru¨benacker, M.; Peters,
K.; Von Schering, H. G. Tetrahedron Lett. 1990, 31, 643.
9
10.1021/ja9097687  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 1151–1158 1151

A R T I C L E S
Bringmann et al.
Figure 1. Different coupling types of naphthylisoquinolines: the C,C-linked dioncophylline C (1) and the N,C-coupled analogues ancisheynine (2),
ancistrocladinium A (3), B (4), C (5), and D (6); for 3, 5, and 6, only the respective main atropo-diastereomers 3a, 5b, and 6a are shown.
rotationally stable biaryls established by our group.18-20 Basic
requirements for the applicability of this approach, however,
are the presence of an oxygen function and an (at least cryptic)14
C1 group next to the axis, for the construction of the lactone
bridge. The latter precondition is not present in ancistrocla-
dinium A (3), and because a direct N-arylation, for example,
by Buchwald-Hartwig amination, is restricted to primary and
secondary amines as substrates21-25 and thus cannot be applied
to cyclic imines, it seemed more favorable to prepare the
naphthylisoquinolines 3-6 through an alternative approach. The
only previous total synthesis of an N,C-coupled naphthyliso-
quinoline alkaloid, ancisheynine (2), was achieved by the
condensation of a monocyclic diketone (or the respective
benzopyrylium salt) with an aminonaphthalene, leading to a fully
dehydrogenated isoquinolinium moiety.8
based on the transition-metal-mediated coupling of the known30
S-configured amine 8 with the appropriately brominated naph-
thalene 931 to give the secondary amine 7 (see Figure 2),
followed by Bischler-Napieralski ring closure30,32 to give the
target molecule 3. The late-stage attachment of the isoquinoline
precursor to the naphthalene moiety and thus the separate
preparation of the two molecular portions (and also the C-1-
methyl part, here from acetyl chloride) was expected to permit
a high degree of molecular diversity by individual structural
variations of these three molecular modules. The strategy should
provide the potential to yield a broad variety of structurally
diverse analogues for in-depth SAR investigations. To keep the
synthesis as short and efficient as possible, only directing groups
were used, but no protective groups.
Some of the substituted naphthalenes, such as the 8-bromo
compound 9, can be obtained by literature procedures.31 The
enantiomerically pure amine 8 should be accessible by a
regioselective ring-opening of the N-Boc-functionalized aziridine
11 using a C-nucleophile generated in situ.
We now report on the first total synthesis of the highly anti-
infective N,C-coupled naphthyldihydroisoquinoline alkaloids
ancistrocladinium A (3) and B (4) by a Buchwald-Hartwig
amination f Bischler-Napieralski cyclization sequence. The
flexible synthetic strategy also paves the way for the construction
of conceivable other members of this class, natural or unnatural,
and thus permits easy access to a broad variety of structurally
diverse compounds for structure-activity relationship studies
(SAR). The general use of the developed method, including
sterically even more demanding representatives, was demon-
strated by the preparation of the related regioisomeric N,3′- and
N,1′-coupled analogues 5 and 6 (see Figure 1), likewise potential
natural products, which have not yet been found in nature so
far.
Preparation of the Primary Amine 8. A key intermediate in
the total synthesis of both alkaloids, ancistrocladinium A (3)
(18) Bringmann, G.; Gulder, T.; Gulder, T. A. M. In Asymmetric Syn-
thesis - The Essentials; Bra¨se, S., Christmann, M., Eds.; Wiley-VCH:
Weinheim, 2006; pp 246-250.
(19) Bringmann, G.; Mortimer, A. J. P.; Keller, P. A.; Gresser, M. J.;
Garner, J.; Breuning, M. Angew. Chem., Int. Ed. 2005, 44, 5384.
(20) Bringmann, G.; Menche, D. Acc. Chem. Res. 2001, 34, 615.
(21) Hartwig, J. F. Nature 2008, 455, 314.
(22) Hartwig, J. F. Synlett 2006, 1283.
(23) Hartwig, J. F. In Handbook of Organopalladium Chemistry for Organic
Synthesis; Negishi, E.-i., Ed.; John Wiley & Sons Inc.: New York,
2002; pp 1051-1096.
(24) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046.
(25) Schlummer, B.; Scholz, U. AdV. Synth. Catal. 2004, 356, 1599.
(26) Bringmann, G.; Ru¨benacker, M.; Weirich, R.; Ake´ Assi, L. Phy-
tochemistry 1992, 31, 4019.
Results and Discussion
(27) Bringmann, G.; Holenz, J.; Weirich, R.; Ru¨benacker, M.; Funke, C.;
Boyd, M. R.; Gulakowski, R. J.; Franc¸ois, G. Tetrahedron 1998, 54,
497.
Retrosynthetic Analysis. In contrast to the total synthesis of
the “normal”, C,C-linked naphthylisoquinoline alkaloids, such
as dioncophylline C (1),26,27 in which the key step is the stepwise
construction of the central biaryl axis by the “lactone
method”,3,18-20 a different approach had to be envisaged for
the new alkaloids 3 and 4. This strategy was also intended to
avoid possible problems of regioselectivity and overoxidation
as expected from a more biomimetically oriented28 oxidative
N,C-cross coupling of the corresponding dihydroisoquinoline
and naphthalene portions.29 As exemplarily illustrated for
ancistrocladinium A (3), a more promising strategy should be
(28) Like for C,C-coupled analogues,70 a separate formation of the
isoquinoline and naphthalene portion is assumed for the biosynthetic
formation of 3 (as also for 2 and 4), followed by a phenol-oxidative
cross coupling.
(29) For a successful biomimetic cross coupling to give N,C-coupled dimers
in the field of carbazole alkaloids, see: Bringmann, G.; Tasler, S.;
Endress, H.; Kraus, J.; Messer, K.; Wohlfahrt, M.; Lobin, W. J. Am.
Chem. Soc. 2001, 123, 2703.
(30) Bringmann, G.; Weirich, R.; Reuscher, H.; Jansen, J. R.; Kinzinger,
L.; Ortmann, T. Liebigs Ann. Chem. 1993, 877.
(31) Bringmann, G.; Hamm, A.; Schraut, M. Org. Lett. 2003, 5, 2805.
(32) Rizzacasa, M. A.; Sargent, M. V.; Skelton, B. W.; White, A. H. Aust.
J. Chem. 1990, 43, 79.
9
1152 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

Naphthylisoquinoline Alkaloids Ancistrocladinium A and B
A R T I C L E S
Scheme 1. Improved Synthesis of the S-Configured Primary
Amine 8
Figure 2. Retrosynthetic analysis of N,C-linked naphthyldihydroisoquino-
line alkaloids, exemplified for ancistrocladinium A (3).
Scheme 2. Analogous Preparation of the R-Configured Amine
ent-8
and B (4), is the enantiopure 1-aryl-2-propylamine 8. Different
from an enantioselective seven-step synthesis based on the
reductive amination of 1-aryl-2-propanones,30 we have now
further developed an approach via chiral aziridines initially used
by Hoye et al. for the directed synthesis of, albeit 3R-configured,
naphthylisoquinoline alkaloids.33,34 N-Tosyl activated aziridines
(not shown) have proven to be valuable intermediates in organic
synthesis because of their ability to undergo highly regio- and
stereoselective ring-opening reactions with a wide variety of
nucleophiles.35,36 Their strain-induced reactivity permits rapid
conversion into a wide range of derivatives,35-40 making the
use of an aziridine for the preparation of the enantiopure
2-propylamine 8 a rewarding option. Because of the harsh
reaction conditions often required for the subsequent removal
of the N-substituents,41-43 like the previously used N-tosyl
groups,33,34 we searched for a more suitable directing group
that should activate the aziridine toward a nucleophilic ring-
opening at the less substituted carbon atom and, simultaneously,
would permit its mild cleavage. For this purpose, the N-Boc
group turned out to be ideally suited.
reacted with 11 in the presence of catalytic CuBr ·SMe2 (10
mol %), furnishing the N-Boc amine 13 in an excellent yield
of 91%, without any loss of optical purity. Final removal of
the N-Boc directing group of 13 yielded the primary amine 8
in quantitative yield (Scheme 1). Enantiomerically pure 8 was
thus available in five steps and 81% overall yield, which
constitutes a significant improvement over previously established
pathways to 8 (seven and five steps in 33%30 and 36% yield,33,34
respectively).
Hence, a short and practicable procedure for the preparation
of 1-aryl-2-propyl amines of type 8 as the central precursor for
the planned first total synthesis of N,C-coupled naphthyldihy-
droisoquinolines on a multigram scale had become available,
utilizing cheap optically pure starting material from the chiral
pool. The strategy should also permit variation of the absolute
configuration and of the identity of the eventual C-3 substituent
(here Me) by using configurationally or constitutionally different
amino acids. This was demonstrated by the analogous prepara-
tion of the enantiomeric primary amine ent-8 from D-alanine
(ent-12) (Scheme 2) in a similar yield (78%).
Synthesis of Ancistrocladinium A (3). For the synthesis of
the N,8′-coupled alkaloid ancistrocladinium A (3), the bro-
monaphthalene 931 had to be transformed into the secondary
amine 7 by Buchwald-Hartwig amination with 8 (Scheme 3).
The use of chelating ligands such as BINAP48,49 or DPPF50 for
the nitrogen-carbon bond formation is known to provide high
yields for the reaction of primary alkylamines and arylbro-
mides,49 as these ligands prevent rapid ꢀ-hydrogen elimination24
and diarylation48 of the amine. On the other hand, intermolecular
palladium-catalyzed N-arylation can proceed with loss of optical
purity for substrates bearing a stereogenic center in the R-posi-
tion to the nitrogen atom, such as the amine 8, if monodentate
44
After reduction of L-alanine (12) with LiAlH4 and N-
functionalization with (Boc)2O,45 cyclization was achieved by
intramolecular SN2 displacement of the likewise, in situ,
generated O-tosyl group,46 giving the aziridine 11 in 90% overall
yield. The Grignard reagent prepared in situ from 1047 was
(33) Hoye, T. R.; Chen, M. Tetrahedron Lett. 1996, 37, 3099.
(34) Hoye, T. R.; Chen, M.; Hoang, B.; Mi, L.; Priest, O. P. J. Org. Chem.
1999, 64, 7184.
(35) Sweeney, J. In Aziridines and Epoxides in Organic Synthesis; Yudin,
A. K., Ed.; Wiley-VCH: Weinheim, 2006; pp 117-144.
(36) Sweeney, J. B. Chem. Soc. ReV. 2002, 31, 247.
(37) Hu, X. E. Tetrahedron 2004, 60, 2701.
(38) Tanner, D. Angew. Chem., Int. Ed. Engl. 1994, 33, 599.
(39) Pineschi, M. Eur. J. Org. Chem. 2006, 4979.
(40) McCoull, W.; Davis, F. A. Synthesis 2000, 1347.
(41) ProtectiVe Groups in Organic Synthesis; Greene, T. W., Wuts,
P. G. M., Eds.; Wiley & Sons: New York, 1999.
(42) Fleming, I.; Frackenpohl, J.; Hiriyakkanavar, I. J. Chem. Soc., Perkin
Trans. 1 1998, 1229.
(43) Alonso, D. A.; Andersson, P. G. J. Org. Chem. 1998, 63, 9455.
(44) Hsiao, Y.; Hegedus, L. S. J. Org. Chem. 1997, 62, 3586.
(45) Posakony, J. J.; Grierson, J. R.; Tewson, T. J. J. Org. Chem. 2002,
67, 5164.
(46) Wessig, P.; Schwarz, J. Synlett 1997, 893.
(47) By using dimethoxybromobenzene 10 instead of the corresponding
chlorobenzene derivative, as applied by Hoye et al.,33,34 the reaction
time for generating the Grignard reagent was reduced from 12 to
2.5 h.
(48) Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc. 1996,
118, 7215.
(49) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1144.
(50) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 7217.
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 1153

A R T I C L E S
Bringmann et al.
Scheme 3. Synthesis of Ancistrocladinium A (3) by a
Buchwald-Hartwig Amination f Bischler-Napieralski Ring
Closure Sequence
Figure 3. Comparison of the CD spectrum of synthetic ancistrocladinium
A (3) and its enantiomer, ent-3, with that of the natural-derived alkaloid 3
to confirm the absolute configuration of the major atropo-diastereomers.
Scheme 4. Preparation of the Unnatural (3R)-Enantiomers of
Ancistrocladinium A
ligands are utilized.48 The use of a chelating ligand like rac-
BINAP was thus necessary for the N-arylation of the chiral
amine 8 with the bromonaphthalene 9. Despite the electron-
rich character of this arylhalide, the secondary amine 7 was
obtained in as much as 63% yield when using Pd2(dba)3 as the
transition-metal catalyst and KOtBu as the base.51 The catalyst
loading was decreased to 1 mol % without diminishing the
chemical yield if the reaction was run in refluxing toluene.
N-Acetylation of amine 7 gave the amide 14, whose Bischler-
Napieralski cyclization afforded ancistrocladinium A (3) as a
2.5:1 mixture of atropo-diastereomers in an excellent yield
(83%). Despite extensive variations of the reaction parameters
(temperature, Lewis acid, and solvent), the diastereomeric excess
could not be changed substantially during the cyclization
reaction. As already reported for the isolated natural product
3,9 all attempts to resolve the atropo-diastereomers, 3a and 3b,
failed, even on chiral HPLC phases.
The chromatographic, physical, and spectroscopic properties
of synthetic 3 were identical to those of the authentic natural
product in all respects, including the prevalence of the naturally
predominant M-atropo-diastereomer 3a in the synthetic product,
so that the only difference in the NMR spectra was the relative
intensity of the signals of 3a versus those of 3b due to a different
diastereomeric ratio of synthetic and natural 3, which was 10:1
(M:P) for ancistrocladinium A (3) isolated from Ancistrocladus
ikela.9 The circular dichroism (CD) spectrum of ancistrocla-
dinium A (3) obtained by total synthesis was in good agreement
with the CD curve of isolated 3 (Figure 3), thus confirming the
absolute configuration at the iminium-carbon axis of the major
synthetic diastereomer, 3a, as M.
enantiomer ent-3 of ancistrocladinium A (3), now with R-
configuration at C-3, succeeded (Scheme 4). Pd-catalyzed
Buchwald-Hartwig N,C-coupling of the two molecular portions,
ent-8 and 9, gave the secondary amine ent-7 in 61% yield.
Formation of the acetamide and Bischler-Napieralski cycliza-
tion delivered ent-3 in 83% yield, again in a 2.5:1-ratio of the
respective atropo-diastereomers, ent-3a and ent-3b.
As expected, the CD spectrum of synthetic ent-3 was opposite
to the curves of both, natural and synthetic ancistrocladinium
A (3), confirming the absolute configuration at the N,C-axis of
ent-3 to be P in its major atropo-diastereomer (see Figure 3).
The total synthesis of both, 3 and ent-3, fully corroborates the
previous structural assignment9 of the isolated natural product.
Synthesis of Ancistrocladinium B (4). With this first synthetic
route to an N,C-coupled naphthyldihydroisoquinoline alkaloid,
ancistrocladinium A (3), elaborated, we focused on the con-
struction of other members of this novel subclass of alkaloids,
such as the antileishmanial11 alkaloid ancistrocladinium B (4).9
Because 3 and 4 just differ by their coupling positions (N,8′ vs
N,6′) and their O-methylation patterns (OH instead of OMe at
C-5′), only the naphthalene module (i.e., 9 for the synthesis of
3 above) had to be replaced by the 6-bromonaphthalene 16.
The required naphthalene precursor 15 (Scheme 5) was
obtained in analogy to a literature procedure.52-54 Introduction
In a similar way, by applying the same strategy as above,
yet starting from ent-8 instead of 8, the synthesis of the unnatural
(51) For further, more recent Buchwald-Hartwig aminations on likewise
electron-rich substrates, see: Organ, M. G.; Abdel-Hadi, M.; Avola,
S.; Dubovyk, I.; Hadei, N.; Kantchev, E. A. B.; O’Brien, C. J.; Sayah,
M.; Valente, C. Chem.-Eur. J. 2008, 14, 2443. Saelinger, D.;
Brueckner, R. Synlett 2009, 109.
9
1154 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

Naphthylisoquinoline Alkaloids Ancistrocladinium A and B
A R T I C L E S
Scheme 5. Preparation of Ancistrocladinium B (4)
of the bromine at C-6 was attained by applying the “directed
ortho-metalation” (DoM) strategy,55,56 assisted by the meth-
oxymethyl (MOM) ether group at C-5. Thus, treatment of
15 with n-butyllithium/N,N,N′,N′-tetramethylethylendiamine
(TMEDA) at -10 °C, followed by careful addition of 1,2-
dibromotetrachloroethane to the C-6 metalated intermediate,
gave the desired building block 16 exclusively (Scheme 5). 2D-
NOESY investigations confirmed the bromo substituent of 16
to be located at the 6-position of the naphthalene, without any
regioisomeric or overhalogenated byproduct. At lower temper-
atures, the reaction proceeded very slowly, while inseparable
mixtures of the differently halogenated products were obtained
at room temperature.
Nitrogen-carbon bond formation of the primary amine 8 with
the 6-bromonaphthalene 16 utilizing Pd2(dba)3/rac-BINAP as
the catalytic system and KOtBu as the base gave amine 17 in
a satisfying yield (49%, Scheme 5). N-Acetylation of 17 and
Lewis-acid mediated ring closure with concomitant cleavage
of the MOM directing group in the naphthalene portion yielded
ancistrocladinium B (4) in 82% yield and as a 46:54 mixture
(see Figure 4) of atropo-diastereomers. NMR investigations and
comparison of the chromatographic and physical data of the
obtained product with those of the natural alkaloid 4 revealed
the major diastereomer to be M-configured at its iminium-aryl
axis. The slight preference for 4a is in agreement with the
thermodynamically controlled equilibrium of the two configu-
rationally semistable atropo-diastereomers of 4, which slowly
interconvert at room temperature, by rotation about the N,C-
axis.9
The assignment of the absolute axial configuration of the two
atropo-diastereomers of ancistrocladinium B (4) was further
corroborated by their HPLC resolution in hyphenation with CD
spectroscopy (Figure 4a). As already described for the authentic
natural product,9 this separation succeeds easily. The CD spectra
of peaks I (rapid) and II (slow, Figure 4b), which were recorded
in the stopped-flow mode,57 were almost mirror-imaged,
although derived from diastereomers and not from enantiomers,
showing that the iminium-carbon linkage and thus the orientation
of the major aryl chromophores dominates the CD curve over
Figure 4. Stereochemical assignment of the two diastereomers of synthetic
ancistrocladinium B (4) by LC-CD coupling and comparison of the
obtained CD spectra with those of the isolated alkaloid 4.
the chiroptical contribution of the stereogenic carbon center. A
similar phenomenon is known for many “normal”, C,C-linked
naphthylisoquinoline alkaloids.58,59 The spectrum of the more
rapidly eluting isomer (peak I) perfectly matched the CD curve
of the P-isomer of the natural alkaloid, 4b, while the curve of
the slower isomer (peak II) was in excellent agreement with
that of the M-atropo-diastereomer of isolated 4. By this
comparison, and by the chromatographic and spectroscopic
identity of 4 from synthetic and natural origins, the structure of
the natural product9 was fully confirmed, including the absolute
configuration.
Synthesis of Ancistrocladinium C (5) and D (6). To validate
the general applicability of the developed first synthetic pathway
to N,C-coupled naphthyldihydroisoquinolines, to even sterically
more congested representatives, the N,3′- and N,1′-linked
analogues 5 and 6 were prepared, too. The synthesis of 5 and
6, which are as yet unknown as natural products, should,
simultaneously, facilitate the directed search for these com-
pounds in nature and permit studies on the impact of the
coupling position in the naphthalene moiety on the anti-infective
activity.
The naphthalene building block 20, with the halogen atom
now at C-3, was obtained from the known60 naphthol 18, which
(52) Hasegawa, T.; Yamamoto, H. Synthesis 2003, 1181.
(53) Bringmann, G.; Gu¨nther, C. Synlett 1999, 216.
(54) Bringmann, G.; Gu¨nther, C.; Peters, E. M.; Peters, K. Tetrahedron
2001, 57, 1253.
(58) Bringmann, G.; Messer, K.; Wolf, K.; Mu¨hlbacher, J.; Gru¨ne, M.;
Brun, R.; Louis, A. M. Phytochemistry 2002, 60, 389.
(59) Bringmann, G.; Gulden, K. P.; Hallock, Y. F.; Manfredi, K. P.;
Cardellina, J. H., II; Boyd, M. R.; Kramer, B.; Fleischhauer, J.
Tetrahedron 1994, 50, 7807.
(55) Snieckus, V. Chem. ReV. 1990, 90, 879.
(56) Snieckus, V. Pure Appl. Chem. 1990, 62, 2047.
(57) Bringmann, G.; Gulder, T. A. M.; Reichert, M.; Gulder, T. Chirality
2008, 20, 628.
(60) Bringmann, G. Liebigs Ann. Chem. 1985, 2126.
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 1155

A R T I C L E S
Bringmann et al.
Scheme 6. Synthesis of the N,3′-Coupled Naphthyldihydro-
isoquinoline Ancistrocladinium C (5), an As Yet Unknown but
Imaginable Natural Product
was accessible by Diels-Alder reaction of 2-bromoanisole and
N,N′-diethyl-3,3-dimethylacrylamide (not shown).61 After for-
mation of the methoxymethylether 19, the bromide was
introduced selectively by a DoM reaction (Scheme 6). By using
THF as a strongly coordinating solvent, the bromonaphthalene
20a was obtained, but only in 42% yield. Replacing the THF
by a nonpolar and thus noncoordinative solvent, such as
n-pentane, the bromination was achieved in 82% yield.
Linkage of the amine 8 and the naphthalene 20a under the
same Buchwald-Hartwig conditions as described for the
synthesis of ancistrocladinium A (3) and B (4) unfortunately
did not give the desired product 21, despite extended variations
of the reaction parameters such as palladium source, ligand,
base, and solvent, but only resulted in the reisolation of starting
materials. This lack of reactivity can be explained by the
substantially higher steric hindrance and the further increased
electron density at the coupling site of the naphthalene substrate
20a as compared to the above reacted arylbromides, 9 and 16.
To facilitate the oxidative addition of the naphthalene 20, and
thus the successful formation of 21, the more reactive iodonaph-
thalene 20b was used as a substrate, which was also accessible
from 19 using a DoM reaction (45% yield). Amination of the
building block 20b with 8 gave 21 in 42% yield after 3 d
(Scheme 6). This secondary amine was further transformed into
the final product 5 by N-acetylation followed by cyclization of
the resulting acetamide. The isoquinoline 5 was obtained as a
3:2-mixture of, in this case configurationally stable, atropo-
diastereomers (P:M) and in 53% chemical yield.
Figure 5. Assignment of the relative and absolute axial configurations of
the two atropo-diastereomers of ancistrocladinium C (5) by NOE investiga-
tions and quantum chemical CD calculations, respectively (for the respective
offline CD spectra of the pure isolated atropo-diastereomers, see the
Supporting Information).
of the S-configuration at C-3, indicated an absolute P-config-
uration at the N,C-axis for peak I (Figure 5a, left). In a
complementary way, an NOE interaction between the methyl
groups at C-3 and C-2′ revealed an axial M-configuration for
peak II (Figure 5a, right). These assignments were further
corroborated by the individual CD spectra of the two atropo-
diastereomers recorded online, by HPLC-CD in the stopped flow
mode.57 The two spectra were found to be virtually opposite to
each other, again due to the dominant CD effect of the element
of axial chirality (Figure 5b),58,59 and showed a strong analogy
to the spectra recorded for the regioisomeric alkaloids 4a and
4b (Figure 4).
An efficient and reliable method for the assignment of
absolute stereostructures is the combination of experimental CD
investigations with quantum chemical CD calculations.62 With
the stereogenic center at C-3 known to be S-configured,
theoretical investigations concentrated on the absolute config-
uration at the chiral axes, that is, on the two possible atropo-
diastereomers of 5. The analysis of the conformational space
based on DFT (B3LYP/6-31G*)63-66 led to four relevant
In accordance with the known representatives of this class
of isoquinolines, ancistrocladinium A (3) and B (4), and due to
its N,3′-coupling of the two molecular portions, compound 5
was henceforth given the name ancistrocladinium C.
The configuration at the axis relative to the stereogenic centers
at C-3 was deduced from an NOE correlation between the
protons of the methyl group at C-2′ and at H-3, which, in view
(62) Bringmann, G.; Bruhn, T.; Maksimenka, K.; Hemberger, Y. Eur. J.
Org. Chem. 2009, 2717–2727.
(63) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(64) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
(65) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56,
2257.
(61) Watanabe, M.; Hisamatsu, S.; Hotokezaka, H.; Furukawa, S. Chem.
Pharm. Bull. 1986, 34, 2810.
(66) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
9
1156 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

Naphthylisoquinoline Alkaloids Ancistrocladinium A and B
A R T I C L E S
Scheme 7. Synthesis of the N,1′-Coupled
Naphthyldihydroisoquinoline Ancistrocladinium D (6)
conformers (within a range of 3 kcal mol-1) for each of the
isomers. These structures were subjected to RI-SCS-MP2/
TZVP67,68 single-point energy calculations to provide more
reliable values for the enthalpy of formation. On the basis of
these conformers, CD calculations were now carried out with
TDB3LYP using the 6-31G* basis set. The curve calculated
for 5b was found to match with the experimental one of peak
I (Figure 5b, left), while peak II gave a match with the CD
curve predicted for 5a (Figure 5b, right). This led to the clear
attribution of the absolute configurations of the two atropo-
diastereomers of ancistrocladinium C (5), finally confirming the
results of the NOE correlations above.
Figure 6. Stereochemical assignment of the axial configurations of the
two atropo-diastereomers of ancistrocladinium D (6) by NOE investigations
and LC-CD coupling in combination with quantum chemical CD calcula-
tions. Similar CD spectra were obtained offline, after preparative resolution
of 6a and 6b (see the Supporting Information).
For the synthesis of the sterically even more hindered N,1′-
linked naphthylisoquinoline 6, a regioselective introduction of
the bromine substituent at C-1 was required after O-methylation
of 18 (Scheme 7). In this case, the method of choice was an
electrophilic bromination reaction with Br2, directly on 22, which
occurred predominantly at C-1, leading to the desired bro-
monaphthalene 23. Optimum reaction conditions were found
by using a slight excess of bromine at -40 °C in the presence
of sodium acetate (1.2 equiv), giving 23 together with a small
amount of dibrominated product. The latter was easily removed
by column chromatography, yielding pure 23 in 87% yield.
Amination of the building block 23 with the primary amine
8 was achieved under the same conditions as described for the
synthesis of ancistrocladinium A (3) and B (4), giving the amine
24 in an initially unsatisfactory 31% yield after 2 d (Scheme
7). The low conversion can be explained by the substantially
higher steric hindrance and the further increased electron density
at the coupling site of the naphthalene substrate 23 as compared
to the above reacted arylbromides 9 and 16. By addition of a
second portion of catalyst after 48 h (1 mol % Pd2(dba)3 and 2
mol % BINAP) and refluxing for another 24 h, the yield was
improved to 56%. Conversion of 24 into the N,1′-naphthyldihy-
droisoquinoline 6 succeeded by N-acetylation and Bischler-
Napieralski cyclization in 47% chemical yield to furnish the
target molecule 6 as a 2:1-mixture of its, configurationally stable,
atropo-diastereomers, 6a and 6b. Remarkably, and in contrast
to the closely related regioisomeric natural product ancistro-
cladinium A (3a/b), resolution of 6a and 6b succeeded on a
Waters Symmetry C18 column. Because of its coupling type, 6
was named ancistrocladinium D.
The axial configurations of the two atropo-diastereomeric
products relative to the stereocenters were established by 2D-
NOESY investigations on the pure isolated isomers. Interactions
between H-8′ and the proton at C-3 (Figure 6b, bottom left) of
the major diastereomer of 6 revealed that these spin systems
were on the same side of the isoquinoline “plane”, while the
interactions between the methyl groups at C-2′ and C-3 indicated
that these groups were both located on the other side. This, in
combination with the known absolute S-configuration of the
stereogenic center at C-3, permitted assignment of the absolute
configuration at the axis of the major diastereomer of ancistro-
cladinium D, 6a, as M. In a similar way, NOESY correlations
between H-3 and the Me-2′ and between the methyl substituent
at C-3 and the proton at H-8′ for the minor diastereomer, 6b
(Figure 6b, bottom right), again in conjunction with the known
S-configuration at C-3, revealed a P-configuration at its iminium-
aryl axis, thus also corroborating the above assignment for 6a.
To further confirm the absolute axial configuration of
ancistrocladinium D (6), LC-CD investigations were carried
out in the stopped-flow mode (Figure 6b). Because 6 constitutes
the first representative of an N,1′-coupled naphthyldihydroiso-
quinoline, again quantum chemical CD calculations were
performed. Given the known absolute configuration at the
stereogenic center at C-3, only the two possible 3S-diastereomers
of 6 were investigated. As for 5a and 5b (see above), the same
conformational investigations, DFT geometry optimizations, and
(67) Grimme, S. J. Chem. Phys. 2003, 118, 9095.
(68) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829.
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 1157

A R T I C L E S
Bringmann et al.
single-point energy calculations were performed on the two
atropisomers of 6, again leading to four relevant conformers
each. For these conformers, CD calculations by TDB3LYP
permitted a first assignment of the absolute configurations, but
the agreement between the theoretical spectra and the experi-
mental ones, especially in the case of 6b, was not perfect over
the whole range of wavelengths. Slightly improved results were
achieved when taking into account solvent effects, by using the
COSMO approach, but still the shorter-wavelength region
(210-230 nm) in the experimental curve of the more slowly
eluting atropo-diastereomer (peak II) provisionally assigned as
6b (see above) was not reproduced. Therefore, a combined
approach of DFT (B3LYP/SVP) and MRCI methods69 was
pursued, resulting in much more accurate excitation energies
and a nearly perfect match between peak I and the CD curve
predicted for 6a (Figure 6b, left), on the one hand, and between
peak II and the curve computated for 6b (Figure 6b, right), on
the other. This permitted an unambiguous assignment of the
absolute configurations of the two atropo-diastereomers of
ancistrocladinium D (6), thus leading to the same results as
deduced from the observed NOE correlations for the two signal
were synthesized, via the key precursor 8, in only eight linear
steps and good yields, for example, in 43% overall yield in the
case of ancistrocladinium A (3), and without the need of
protective groups. The coupling of the two molecular portions
at a late stage of the synthesis by Buchwald-Hartwig amination,
followed by N-acetylation and cyclization to the N-aryldihy-
droisoquinoline, furthermore makes this convergent approach
a versatile strategy for generating structurally diverse analogues
of these natural products. This was demonstrated in the likewise
successful first total synthesis of the related alkaloid ancistro-
cladinium B (4) and the as yet unknown, sterically even more
hindered regioisomers ancistrocladinium C (5) and D (6). The
presented work paves the way for ongoing studies on this
promising class of anti-infective lead compounds, which will
help to gain deeper insight into their structure-activity relation-
ships, and, thus, to improve their activity and selectivity, and
to decrease their cytotoxicity against mammalian cells. For the
future, more detailed investigations on the stereochemical course
of the Bischler-Napieralski ring closure and the improvement
of the stereoselectivity, or even a directed, diastereo-divergent
preparation of any of the respected atropoisomers, will be of
high interest. This work is currently in progress.
1
sets in H NMR (see Figure 6, bottom).
Conclusion
Acknowledgment. This work was supported by the Deutsche
Forschungsgemeinschaft (SFB 630: Recognition, Preparation, and
Functional Analysis of Agents against Infectious Diseases, project
A2), the Fonds der Chemischen Industrie (fellowship to T.G. and
funds), and the Hochschul- und Wissenschaftsprogramm of the
University of Wu¨rzburg (fellowship to T.G.). We thank Dr. Inga
Kajahn for providing isolated ancistrocladinium A (3) and B (4)
for comparison and Melanie Pavlov for technical assistance.
Because of their unprecedented molecular architectures, their
potent activities against several pathogens of infectious diseases,
and their difficult availability from plant resources, the N,C-
linked naphthyldihydroisoquinolinium alkaloids ancistrocla-
dinium A (3) and B (4) are valuable synthetic targets. We have
therefore developed effective and convergent synthetic pathways
to prepare these plant metabolites, and the not (yet) isolated
N,3′- and N,1′-coupled analogues 5 and 6, accessible in larger
quantities, including the possibility to produce a plethora of
structurally diverse analogues and derivatives. Starting from
L-alanine (12), the first N-naphthyldihydroisoquinoline alkaloids
Supporting Information Available: Full characterization,
including copies of 1H and 13C NMR spectra, and experimental
procedures for selected compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(69) Neese, F. J. Inorg. Biochem. 2006, 100, 716.
(70) Bringmann, G.; Mutanyatta-Comar, J.; Greb, M.; Ru¨denauer, S.; Noll,
T. F.; Irmer, A. Tetrahedron 2007, 63, 1755.
JA9097687
9
1158 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010