Total syntheses of cytotoxic, naturally occurring kalasinamide, geovanine, and marcanine A

Article abstract of DOI:10.1002/anie.200804388

(Chemical Equation Presented) Two novel synthetic strategies were developed for the natural products geovanine, marcanine A, and kalasinamide. The nine-step synthesis of geovanine marks its first total synthesis. The three compounds are known for their an

Full text of DOI:10.1002/anie.200804388

Angewandte
Chemie
DOI: 10.1002/anie.200804388
Azaanthracenones
Total Syntheses of Cytotoxic, Naturally Occurring Kalasinamide,
Geovanine, and Marcanine A**
Steffen Lang and Ulrich Groth*
Azaanthracenones are a pharmacologically interesting class
of natural compounds. The first representative of an azaan-
thraceneone isolated from natural sources was geovanine (3;
Scheme 1). It was isolated from wood from the trunk of
Scheme 1. Structures of naturally occurring azaanthracenones.
Annona ambotay in 1987 by de Oliveira et al.[1a] Several years
later, Dos Santos et al.[1b] reported the isolation from Annona
dioica collected in Brazil, but no total synthesis has been
published so far. Kalasinamide (1) was discovered by
Tuchinda et al.[2] in 2000 in Polyalthia suberosa collected in
Thailand. Very recently Gandy and Piggott published the first
total synthesis of kalasinamide (1).[3] Marcanine A (2) was
isolated by Soonthornchareonnon et al.[4] in 1999 from
Goniothalamus marcani in Thailand. Perez et al. reported
its synthesis several years before the isolation.[5]
Azaanthracenones show several biological activities.
Soonthornchareonnon et al.[4a] reported on the cytotoxicity
of marcanine A and its derivitives, which displayed IC50
values between 80 nm and 2.1 mm against several human
tumor cell lines. Ichino et al.[4c] found marcanine A to have
in vitro antimalarial activity against the drug-resistant K1
strain of Plasmodium falciparum. Ruomy et al.[4d] reported
the isolation of another azaanthracenone and found activity
against Plasmodium falciparum.
We report here on the first total synthesis of geovanine (3)
along with new syntheses of marcanine A (2) and kalasina-
mide (1). Three key steps were crucial to the new synthetic
approach to kalasinamide and marcanine A (Scheme 2). The
Scheme 2. Reagents and conditions: a) Pd/C, H2, THF; Me2SO4,
reflux; b) HNO3, DCM; c) Ni, H2, 75% over three steps; d) malonic
acid diethyl ester, 1408C, 3 h, 85%; e) 5% NaOH, 608C, 3 h, 93%;
f) POCl3, reflux, 2 h, 85%; g) MeONa, MeOH, RT, 3 d, 98%;
h) MeMgCl, THF, MnCl2, 08C, 2 h, 80%; i) TMSCl, NaI, MeCN, RT 4 h,
95%; j) CAN, MeCN/H2O, RT, 4 h, 71%. DCM=dichloromethane,
CAN=ceric ammonium nitrate.
first is the nitration of dimethoxynaphthalene. This reaction
has been reported several times; however, yields were poor in
almost all cases.[6] Oxidative demethylation usually leads to
significant amounts of naphthoquinone as the side product.[7]
The best yields are obtained by the addition of nitric acid to a
solution of the starting compound in dichloromethane in the
presence of silica gel. A similar method was reported first by
Tapia et al.[8] However, substantial alterations were necessary
for its application to a larger scale reaction. After optimiza-
tion the nitration step leading to 5 was achieved in over 90%
yield.
The second key step is the cyclization of a malonic acid
amide in the presence of phosphorus oxychloride.[9] After
optimization this conversion proceeded smoothly providing 7
in 85% yield. Differentiation between the two chloride
residues at the pyridine ring of 7 was possible by treatment
with sodium methoxide in methanol/THF at room temper-
ature over 24 h. Dimethoxylation was not observed, even at
increased temperatures and reaction times. These results
clearly differ from the analogous reaction of 2,4-dichloro-5,8-
dimethoxyquinoline (10): under similar reaction conditions,
the monomethoxylated products 11 and 12 were obtained in a
1:1 ratio (Scheme 3).[10]
[*] S. Lang, Prof. U. Groth
Fachbereich Chemie and Konstanz Research School Chemical
Biology, Universitꢀt Konstanz
Postfach M-720, 78457 Konstanz (Germany)
Fax: (+49)7531-884-155
E-mail: ulrich.groth@chemie.uni-konstanz.de
[**] S.L. thanks the Studienstiftung des Deutschen Volkes for a doctoral
fellowship and the Konstanz Research School Chemical Biology
(KoRS-CB) for scientific support. We thank Bayer AG, Merck KgaA,
and Wacker AG for generous gifts of reagents.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804388.
Angew. Chem. Int. Ed. 2009, 48, 911 –913
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
911

Communications
hydrobromic acid. The selectivity of the bromination with
NBS was about 88%. As both regioisomers could be used in
the following reaction this did not affect overall yields.
Attachment of ring C was achieved by treating the
bromide with lithium diisopropylamide (LDA) at À788C in
the presence of furan.[19] Similar methods have been applied
to aromatic compounds, but to our knowledge this is the first
report on its application to heteroaromatic compounds.
Acidic cleavage of the oxygen bridge mediated by para-
toluenesulfonic acid[20] provided free phenol 16. When
dichloromethane was used as the solvent, as described in
the literature, the reaction time exceeded several months; it
was therefore replaced by THF. Only very small amounts (less
than 2%) of the other phenolic isomer were obtained.
Treatment of 15 with HClO4[20b] did not lead to any detectable
amount of product. Refluxing in hydrochloric acid or acetic
acid was not an option owing to the lability of the 2-methoxy
moiety.
Scheme 3. Reagents and conditions: MeONa, MeOH, RT, 3 d.
The third key step is the replacement of the remaining
chloride with a methyl group. Manganese(II) chloride cata-
lyzed alkylation[11] in THF with the corresponding Grignard
reagent furnished the desired product in excellent yield.
To prove the selectivity of the
sequence of methoxylation and methyl-
ation, NOE experiments were con-
ducted with 9. The interactions shown
in Scheme 4 were found.
The MnCl2-catalyzed alkylation can
be used to introduce a large variety of
residues. Investigations will be made in
order to determine the scope of this
The two transition-state structures 18a and 18b can be
invoked in the cleavage of the oxygen bridge (Scheme 6).
However, 18a is formed predominantly as a result of the
electron-donating effect of both substituents on the pyridine
moiety. Consequently the desired product 16 is formed with
high selectivity.[21] Traces of the undesired isomer could be
easily removed by recrystallization.
Methylation of phenol 16 could not be achieved with MeI
and K2CO3 but proceeded smoothly in the presence of sodium
hydride.[22] TMS-I mediated demethoxylation[12] at the pyri-
dine residue provided geovanine (3) in good yield
(Scheme 5).
Scheme 4. NOE
interactions
observed for inter-
mediate 9.
method. Demethoxylation of
9 was
achieved by treatment with TMSCl and
NaI in acetonitrile[12] at room temper-
ature and provided 1 in good yield.
Demethylation of the hydroquinone dimethyl ether was
conducted in the presence of ceric ammoniun nitrate
(CAN),[13] yielding mercanine A (2) as a yellow powder.
Geovanine (3) was prepared in only nine steps from 2,5-
dimethoxyaniline (Scheme 5). Intermediate 13 was prepared
in 65% yield over two steps by modification of a procedure
described by Brody et al.[14] Deoxychlorination with phos-
phorus oxychloride[15] and subsequent dehalomethoxyla-
tion[16] yielded 14 in 90% yield. Bromination with 1.05 equiv
NBS[17] proved to be superior to all other methods tested for
the next step (Scheme 5). Bromination with Br2 in chloro-
form,[18] for example, led to extensive demethylation at C2 of
the pyridine moiety as a result of the presence of free
At this stage NOE experiments were conducted in order
to prove the selectivity of the ring opening (Scheme 7).
Scheme 5. Reagents and conditions: a) POCl3, reflux, 3 h; b) MeOK,
MeOH, reflux, 5 h, 89% over 2 steps; c) NBS, DCM, RT 2 h, 90%;
d) LDA, THF, furan, À788C, 3 h, quant.; e) p-toluenesulfonic acid,
THF, reflux, 20 h, 56%; f) MeI, NaH, DMF/THF, 96%; g) TMSCl, NaI,
MeCN, RT 4 h, 85%. NBS=N-bromosuccinimide, LDA=lithium diiso-
propylamide.
Scheme 6. Possible transition states and products of the acid-cata-
lyzed ring opening of 15.
912
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 911 –913

Angewandte
Chemie
Sophasan, C. Yoosook, T. Santisuk, J. M. Pezzuto, Planta Med.
2001, 67, 572 – 575; d) Y.-C. Wu, F.-R. Chang, C.-Y. Chen, J. Nat.
Prod. 2005, 68, 406 – 408; e) X.-F. He, X.-N. Wang, C.-Q. Fan, L.-
S. Gan, S. Yin, J.-M. Yue, Helv. Chim. Acta 2007, 90, 783 – 791.
[3] M. N. Gandy, M. J. Piggott, J. Nat. Prod. 2008, 71, 866 – 868.
[4] a) N. Soonthornchareonnon, K. Suwanborirux, R. Bavovada, C.
Patarapanich, J. M. Cassady, J. Nat. Prod. 1999, 62, 1390 – 1394;
b) Q. Wang, M. He, J. Liang, Zhongcaoyao 2003, 34, 277 – 280;
c) C. Ichino, N. Soonthornchareonnon, W. Chuakul, H. Kiyo-
hara, A. Ishiyama, H. Sekiguchi, M. Namatame, K. Otoguro, S.
Omura, H. Yamada, Phytother. Res. 2006, 20, 307 – 309; d) V.
Roumy, N. Fabre, F. Souard, S. Massou, G. Bourdy, S. Maurel, A.
Valentin, C. Moulis, Planta Med. 2006, 72, 894 – 898.
[5] J. M. Pꢀrez, L. Vidal, M. T. Grande, J. C. Menendez, C.
Avendano, Tetrahedron 1994, 50, 7923 – 7932.
[6] a) A. Inoue, K. Nakano, N. Kuroki, K. J. Konishi, J. Soc. Org.
Synth. Chem. 1956, 14, 513 – 516; b) A. Inoue, K. Nakano, N.
Kuroki, K. J. Konishi, J. Soc. Org. Synth. Chem. 1956, 14, 622 –
625; c) A. Inoue, N. Kuroki, K. Konishi, Bull. Univ. Osaka
Prefect. Ser. A 1959, 8, 31 – 55; d) D. J. McCaustland, P.-L. Chien,
C. C. Cheng, J. Novotny, W. L. Schreiner, F. W. Starks, J. Med.
Chem. 1973, 16, 1311 – 1314; e) G. Malesani, M. G. Ferlin, J.
Heterocycl. Chem. 1985, 22, 1141 – 1142; f) C. P. Butts, L.
Eberson, M. P. Hartshorn, O. Persson, R. S. Thompson, W. T.
Robinson, Acta Chem. Scand. 1997, 51, 1066 – 1077.
[7] a) For a review see: O. C. Musgrave, Chem. Rev. 1968, 69, 499;
b) B. Errazuriz, R. Tapia, J. A. Valderrama, Tetrahedron Lett.
1985, 26, 819 – 822.
Scheme 7. Anticipated NOE interactions for isomers 3 and 20.
Whereas in isomer 3 one methoxy residue should show long-
range couplings with both the adjacent aromatic H and the
aromatic methyl group, in 20 each of the two central methoxy
groups would display only one of these couplings. The long-
range couplings we obtained proved that the structures of
isomer 3 and our product are identical (for details see the
Supporting Information).
Geovanine has been isolated twice. De Oliveira et al.
1
reported only H NMR data and could not determine which
of the isomers had been isolated. Several years later,
Dos Santos et al. isolated a mixture of the two isomers and
published 1H and 13C NMR data. The two isomers were
distinguished by HBBD, DEPT, COSY and NOE experi-
ments. Our results show that these first assignments were not
entirely accurate (see the Supporting Information). At least
two of the reported carbon signals seem to belong to the
wrong isomer.
Compound 2 showed good results in cytotoxicity tests
against two human cancer cell lines. The IC50 values obtained
with the synthetic marcanine A were comparable to those
reported for the isolated natural product (HeLa S3: 0.75 Æ
0.03 mm; Hep G2: 1.54 Æ 0.78 mm).
[8] R. Tapia, G. Torres, J. A. Valderrama, Synth. Commun. 1986, 16,
681 – 687.
[9] Similar to: S. J. Could, B. Shen, Y. G. Whittle, J. Chem. Soc. 1989,
111, 7932.
[10] For methoxylation results of related dichloroquinolines see:
a) E. Evertsson, T. Inghardt, J. Lindberg, A. Linusson, F.
Giordanetto, PCT Int. Appl. 2005, pp. 114; b) A. G. Osborne,
G. T. Dimitrova, P. Galbally, D. D. Hughes, C. Jones, A. L.
Lipman, N. Wilstead, J. Chem. Res. Synop. 2002, 4, 124 – 148;
c) J. M. Barker, P. R. Huddleston, D. Holmes, J. Chem. Res.
Synop. 1985, 7, 214 – 215.
[11] M. Rueping, W. Ieawsuwan, Synlett 2007, 247 – 250.
[12] a) G. A. Olah, S. C. Narang, B. Grupta, Org. Chem. 1979, 44,
1247; b) C. Gonzalez, E. Guitian, L. Castedo, Tetrahedron 1999,
55, 5195 – 5206.
[13] a) For an up to date review on CAN see: V. Nair, A. Deepthi,
Chem. Rev. 2007, 107, 1862 – 1891; b) C. Avendaꢁo, E. de La
Cuesta, C. Gesto, Synthesis 1991, 727.
[14] F. Brody, J. J. Leavitt, R. S. Long (American Cyanamid Co.),
US 2754293, 1956.
[15] P. Nickel, E. Fink, Justus Liebigs Ann. Chem. 1976, 367 – 382.
[16] a) T. Lister, R. H. Prager, M. Tsaconas, K. L. Wilkinson, Aust. J.
Chem. 2003, 56, 913 – 916; b) Y.-Q. Fang, R. Karisch, M. Lautens,
J. Org. Chem. 2007, 72, 1341 – 1346.
[17] J. Song, S. Jeong, S. W. Ham, J. Korean Chem. Soc. 2002, 46, 402 –
404.
[18] H. Uno, J. Org. Chem. 1986, 51, 350 – 358.
[19] S. Sꢂrgel, C. Azap, H.-U. Reissig, Eur. J. Org. Chem. 2006, 4405 –
4418.
In conclusion, we have implemented the first total syn-
thesis of geovanine (3). Thereby we were able to prove the
position of the 8-methoxy group and assign all of the signals in
the 13C NMR spectrum. Moreover, we conducted new total
syntheses of the natural compounds marcanine A (2) and
kalasinamide (1). For this purpose we developed two basically
new synthetic approaches to this class of natural compounds.
Both routes offer excellent perspectives for the synthesis of
numerous new azaanthracenones. In the light of the increas-
ing need for new cytostatica we believe that these results may
contribute to substantial progress in this field.
Received: September 5, 2008
Published online: December 29, 2008
Keywords: antitumor agents · heterocycles · natural products ·
.
total synthesis
[20] D. L. J. Clive, A. Khodabocus, P. G. Vernon, A. G. Angoh, L.
Bordeleau, D. S. Middleton, C. Lowe, D. Kellner, J. Chem. Soc.
Perkin Trans. 1 1991, 1433 – 1444.
[21] For similar discussions see: E. Masson, M. Schlosser, Eur. J. Org.
Chem. 2005, 4401 – 4405.
[22] D. H. Hua, M. Tamura, X. Huang, H. A. Stephany, B. A.
Helfrich, E. M. Perchellet, B. J. Sperfslage, J.-P. Perchellet, S.
Jiang, D. E. Kyle, P. K. Chiang, J. Org. Chem. 2002, 67, 2907 –
2912.
[1] a) A. B. De Oliveira, G. G. De Oliveira, F. Carazza, J. Maia, S.
Guilherme, Phytochemistry 1987, 26, 2650 – 2651; b) P. R. D.
Dos Santos, A. Morais, R. Braz-Filho, J. Braz. Chem. Soc. 2003,
14, 396 – 400.
[2] a) P. Tuchinda, M. Pohmakotr, B. Munyoo, V. Reutrakul, T.
Santisuk, Phytochemistry 2000, 53, 1079 – 1082; b) P. Thinapong,
O. Rangsiman, P. Tuchinda, B. Munyoo, M. Pohmakotr, V.
Reutrakul, Acta Crystallogr. Sect. C 2000, 56, e309 – e310; c) P.
Tuchinda, M. Pohmakotr, V. Reutrakul, W. Thanyachareon, S.
Angew. Chem. Int. Ed. 2009, 48, 911 –913
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
913