Asymmetric total syntheses of ecteinascidin 597 and ecteinascidin 583

Article abstract of DOI:10.1002/anie.200603179

(Chemical Equation Presented) The limited availability of bioactive natural products for pharmaceutical studies and drug development is a problem that can be solved, in principle, by total synthesis. The potent antitumor marine natural products ecteinasci

Full text of DOI:10.1002/anie.200603179

Communications
Natural Products Synthesis
DOI: 10.1002/anie.200603179
Asymmetric Total Syntheses of Ecteinascidin 597
and Ecteinascidin 583**
Jinchun Chen, Xiaochuan Chen, Matthieu Willot, and
Jieping Zhu*
Dedicated to Professor Yulin Li
on the occasion of his retirement
The ecteinascidins, a family of tetrahydroisoquinoline alka-
loids isolated from the Caribbean tunicate Ecteinascidia
turbinata,[1] display a wide range of antitumor and antimicro-
bial activities.[2] One member of this family, ecteinascidin743
(Et743, 1; Scheme 1), is currently in late phase II/III clinical
trials against ovarian, endometrium, and breast cancer, and
several other types of sarcoma. The restricted natural
availability of the ecteinascidins (1 g of Et743 from 1 ton of
tunicate) in conjunction with their potent antiproliferative
activities and complex molecular architecture has made them
attractive synthetic targets.[3] Since the landmark synthesis of
Et743 by Corey and co-workers in 1996,[4] Fukuyama and co-
workers[5] and our research group[6] have also completed total
syntheses of this molecule, and Danishefsky and co-workers[7]
very recently reported a formal total synthesis of Et743. A
semisynthesis of Et743 from cyanosafracin B was developed
by Cuevas, Manzanares, and co-workers at PharmaMar,[8] and
further synthetic approaches have been reported by a number
of research groups.[9–14]
Scheme 1. Structure of representative ecteinascidins.
[*] Dr. J. Chen, X. Chen, M. Willot, Dr. J. Zhu
Institut de Chimie des Substances Naturelles
CNRS, 91198 Gif-sur-Yvette Cedex (France)
Fax: (+33)1-69077247
E-mail: zhu@icsn.cnrs-gif.fr
[**] Financial support from the CNRS and the Institut de Chimie des
Substances Naturelles are gratefully acknowledged.
Supporting information for this article (experimental procedures,
product characterization, and copies of 1H and 13C NMR spectra of
synthetic Et 597 (3) and Et 583 (4)) is available on the WWW under
http://www.angewandte.org or from the author.
8028
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 8028 –8032

Angewandte
Chemie
Ecteinascidins597 (Et597, 3) and 583 (Et583, 4;
Scheme 1) are putative biosynthetic precursors of other
ecteinascidins.[1d,15] Although the cytotoxicity of Et597 (3),
which lacks the third tetrahydroisoquinoline unit, is generally
2.5–10 times less potent than that of Et743 against P388,
A549, HT29, and CV-1 cell lines, its antiproliferative activity
is even greater than that of taxol and camptothecin.[1d] In
continuation of our research towards the development of an
efficient synthetic route to this family of marine natural
products and on the structure–activity relationships (SARs)
of these compounds, we became interested in the synthesis of
Et597 (3) and Et583 (4). In a different approach to our
previous strategy,[6] we envisaged that we could take advant-
age of the presence of two free hydroxy groups in ring A of 3
and 4, as illustrated in Scheme 2. We planned to construct the
Scheme 3. Synthesis of the A-ring unit 7: a) TBSCl, imidazole, DMF,
RT, 98%; b) mCPBA, CHCl3, 458C; then Na2CO3, MeOH, RT, 85%;
c) MOMCl, DIPEA, CH2Cl2, 08C!reflux, 96%; d) nBuLi, THF, ꢀ108C;
then MeI, ꢀ788C!RT, 92%; e) TMSBr, CH2Cl2, ꢀ20!08C, 90%.
TBS=tert-butyldimethylsilyl, mCPBA=m-chloroperbenzoic acid,
MOM=methoxymethyl, DIPEA=N,N-diisopropylethylamine,
DMF=N,N-dimethylformamide.
hydrolysis of the oxazolidine moiety in 12 was more difficult
than expected. Eventually, reaction conditions that were
previously developed for the cleavage of acetonides
(CeCl3·7H2O, oxalic acid, acetonitrile, room temperature)
afforded alcohol 13 in 91% yield.[19] Swern oxidation[20] of the
primary alcohol furnished the corresponding aminoaldehyde
8, which without purification underwent the stereoselective
phenolic aldol condensation with the magnesium phenolate of
7[21,22] to provide the syn aminoalcohol 14 in 74% yield. The
anti aminoalcohol was neither isolated nor detected. The
existence of rotamers made NMR spectroscopic analysis of 14
difficult, and it was hard to determine if this product was a
mixture of two diastereomers with respect to the stereogenic
silicon center.[23] Nevertheless, the question of diastereomers
was of no consequence, as the silyl protecting group was due
to be removed in the next step. Compound 14 was trans-
formed into aminoalcohol 15 in excellent overall yield by a
three-step sequence: 1) protection of the phenol and secon-
dary alcohol as the corresponding methoxymethyl ethers,
2) simultaneous removal of the N-Boc and O-silyl protecting
groups according to the procedure of Sakaitani and
Ohfune,[24] and 3) hydrolysis of the acetate. The Pictet–
Spengler reaction[25] of 15 and TrocOCH2CHO (16; prepared
in two steps from ethyleneglycol)[26] was the key step of our
synthesis. To our pleasure, the desired transformation pro-
ceeded efficiently in dichloromethane in the presence of
acetic acid and 3-Š molecular sieves to provide 17 as a single
diastereomer in 90% yield. Swern oxidation of the amino-
alcohol 17 followed by a zinc chloride catalyzed intramolec-
ular Strecker reaction provided aminonitrile 18 as a single
stereoisomer to complete the highly efficient construction of
the pentacyclic ring system.
Scheme 2. Retrosynthetic analysis of Et597 (3) and Et583 (4).
Alloc=allyloxycarbonyl, Boc=tert-butoxycarbonyl.
highly oxygenated A–B ring system by starting from phenol 7
and tetrahydroisoqinoline 8 and carrying out a sequence of
phenolic aldol condensations followed by a Pictet–Spengler
reaction. An intramolecular Strecker reaction would then
afford the entire A–B–C–D–E pentacycle, whereupon the
closure of the 10-membered lactone by formation of the
carbon–sulfur bond would lead to the natural products.
The synthesis of the aromatic segment 7 is summarized in
Scheme 3. 3-Methoxy-4-hydroxybenzaldehyde (9) was con-
verted into 10 through a well-established three-step sequence.
Interestingly, ortho lithiation of 10 followed by the addition of
methyl iodide[16] gave a compound in which both the aromatic
ring and the TBS protecting group had been methylated.
Under optimized conditions (3 equivalents of nBuLi, 4 equiv-
alents of MeI), the dual-methylation product 11 was isolated
in 92% yield. The MOM group could be removed without
touching the silyl ether by treatment with TMSBr to provide
phenol 7 in excellent yield.[17]
The relative configuration of compound 17 was deter-
mined upon its conversion into 19. The characteristic NOEs
observed between H1/H3, H3/H4, and H11/H13 (ecteinasci-
din numbering) indicated that the configuration of 19, and
hence that of 17, is 1R, 3R, 4R, 11R, 13S. The configuration at
C21 was determined to be R by detailed NMR spectroscopic
Our synthetic route to the pentacyclic compound 18 is
depicted in Scheme 4. The synthesis of the tetrahydroiso-
quinoline 12 featured a highly diastereoselective Pictet–
Spengler condensation of the (S)-Garner aldehyde with (S)-3-
hydroxy-4-methoxy-5-methylphenylalanol.[6,18] The selective
Angew. Chem. Int. Ed. 2006, 45, 8028 –8032
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 8029

Communications
Scheme 4. Synthesis of the pentacyclic compound 18: a) CeCl3·7H2O, oxalic acid, acetonitrile, RT, 91%; b) oxalyl chloride, DMSO, CH2Cl2, ꢀ608C,
then Et3N; c) MeMgCl, THF, 7; then 8, CH2Cl2, RT, 74%; d) MOMCl, DIPEA, CHCl3, 08C!reflux, 88%; e) TBSOTf, 2,6-lutidine, CH2Cl2, ꢀ788C!
RT; then KF, MeOH, RT, 86%; f) K2CO3, MeOH, RT, 94%; g) AcOH, TrocOCH2CHO (16), 3-Š MS, CH2Cl2, RT, 90%; h) oxalyl chloride, DMSO,
CH2Cl2, ꢀ608C; then TMSCN, ZnCl2, CH2Cl2, RT, 87%. DMSO=dimethyl sulfoxide, Tf=trifluoromethanesulfonyl, TMS=trimethylsilyl,
Troc=2,2,2-trichloroethoxycarbonyl.
studies of the N-de-Alloc derivative 20 (NOEs observed for
H21/H22 and H14/H21; Scheme 5).
phenol group in ring A is essential to the success of the
reaction; the condensation of amine 21 with 2-benzyloxy-
acetaldehyde (or ethyl glyoxylate) failed to provide the
desired tetrahydroisoquinoline.
The high diastereoselectivity observed in the condensa-
tion of 15 and 16 could be explained by assuming that the
iminium intermediate had a trans configuration and that the
substituents at C3 and C4 were both pseudoequatorial (A,
Scheme 5).[27] However, at the present stage an alternative
sequence involving a phenolic aldol condensation and b elim-
ination to give the orthoquinone methide intermediate B
(Scheme 5) followed by intramolecular Michael addition of
the tethered amine can not be eliminated as a possibility.[28] It
is nevertheless interesting to note that the presence of the free
The total synthesis of Et597 (3) and Et583 (4) was
completed as shown in Scheme 6. Unmasking of the Troc-
protected primary alcohol under reductive conditions fol-
lowed by chemoselective allylation of the phenol group
provided compound 22, which was coupled with (R)-N-Troc-
S-4,4’,4’’-trimethoxytritylcysteine to afford the corresponding
ester 23 in excellent yield. The cyclization of 23 was examined
under a variety of reaction conditions; the acid (p-toluene-
sulfonic acid, TFA, MeSO3H, TMSBr), the solvent (CH2Cl2,
toluene, 2,2,2-trifluoroethanol, MeCN), and the temperature
were varied, and the reaction was carried out in the presence
or absence of molecular sieves. Unfortunately, no combina-
tion provided the desired 10-membered lactone 25. The
attempted cyclization of phenol 28 also met with failure. We
then decided to separate the S-deprotection and cyclization
steps. Removal of the S-4,4’,4’’-trimethoxytrityl group from 23
with Et3SiH/TFA afforded the stable thiol 24 in 88% yield
after flash column chromatography.[29] Gratifyingly, the treat-
ment of thiol 24 with TMSBr afforded the bridged macrocycle
25, after the phenol had been masked as the corresponding
acetate, in 60% yield.[30] In this simple experiment, a complex
reaction sequence involving O-MOM deprotection, 1,4-
elimination to give the orthoquinone methide,[31] and macro-
cyclization through an intramolecular Michael addition
occurred in a highly ordered manner to bring about the key
ꢀ
C S bond formation.
Simultaneous removal of the N-Alloc and O-allyl func-
tionalities as described by GuibØ and co-workers[32] provided
amine 26 in 85% yield. A sequence of reductive N methyla-
tion, removal of the N-Troc group (zinc/AcOH), and con-
version of the aminonitrile functionality into a hemiaminal
(AgNO3 in a mixture of acetonitrile and water) afforded
Scheme 5. Structures for the discussion of configuration and possible
reaction pathways.
8030 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 8028 –8032

Angewandte
Chemie
Scheme 6. Synthesis of Et597 (3) and Et583 (4): a) Zn, AcOH, Et2O, RT, 90%; b) allyl bromide, K2CO3, acetonitrile, RT, 94%; c) EDCI, DMAP, (R)-
N-Troc-S-4,4’,4’’-trimethoxytritylcysteine, CH2Cl2, RT, 93%; d) Et3SiH, TFA, CH2Cl2, RT, 87%; e) TMSBr, CH2Cl2, ꢀ208C!108C; f) Ac2O, pyridine,
DMAP, CH2Cl2, RT, 60%; g) [Pd(PPh3)4], nBu3SnH, AcOH, CH2Cl2, RT, 85%; h) CH2O, NaBH3CN, AcOH, MeCN/MeOH, RT, 95%; i) Zn, AcOH,
Et2O, RT, 89% for Et597, 86% for Et583; j) AgNO3, MeCN/H2O, RT, 92% for 3, 88% for 4. DMAP=4-dimethylaminopyridine, TFA=trifluoro-
acetic acid, EDCI=1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride.
ecteinascidin597 (3) in excellent overall yield. Similarly,
amine 26 was converted into ecteinascidin583 (4) in a two-
step sequence. Synthetic Et597 (3) and Et583 (4) exhibited
physical, spectroscopic, and spectrometric characteristics (1H,
13C NMR, IR, [a]D, and HRMS) identical to those reported
for the natural products.
In summary, convergent total syntheses of Et597 (3) and
Et583 (4) have been completed for the first time from readily
accessible starting materials. Notable features of our
approach include: 1) a stereoselective aldol reaction for the
coupling of the A-ring moiety 7 with the D–E unit 8, 2 ) a
highly stereoselective Pictet–Spengler reaction for the con-
struction of the B ring, and 3) TMSBr-promoted macrocycli-
zation of the thiol 24 to give the 1,4-bridged 10-membered
ring. This straightforward synthesis does not require sophis-
ticated reaction conditions and should potentially be ame-
nable to large-scale production. We are currently exploiting
this strategy for the synthesis of ecteinascidin analogues for
detailed SAR studies.[33]
Keywords: alkaloids · antitumor agents · asymmetric synthesis ·
marine natural products · Pictet–Spengler reaction
.
[1] a) A. E. Wright, D. A. Forleo, G. P. Gunawardana, S. P. Guna-
sekera, F. E. Koehn, O. J. McConnell, J. Org. Chem. 1990, 55,
4508 – 4512; b) K. L. Rinehart, T. G. Holt, N. L. Fregeau, J. G.
Stroh, P. A. Keifer, F. Sun, L. H. Li, D. G. Martin, J. Org. Chem.
1990, 55, 4512– 4515; c) R. Sakai, K. L. Rinehart, Y. Guan,
A. H. J. Wang, Proc. Natl. Acad. Sci. USA 1992, 89, 11456 –
11460; d) R. Sakai, E. A. Jares-Erijman, I. Manzanares,
M. V. S. Elipe, K. L. Rinehart, J. Am. Chem. Soc. 1996, 118,
9017 – 9023; e) K. Suwanborirux, K. Charupant, S. Amnuoypol,
S. Pummangura, A. Kubo, N. Saito, J. Nat. Prod. 2002,65, 935 –
937.
[2] K. L. Rinehart, Med. Drug. Rev. 2000, 1 – 2 7.
[3] J. D. Scott, R. M. Williams, Chem. Rev. 2002, 102, 1669 – 1730.
[4] a) E. J. Corey, D. Y. Gin, R. S. Kania, J. Am. Chem. Soc. 1996,
118, 9202 – 9203; b) E. J. Martinez, E. J. Corey, Org. Lett. 2000, 2,
993 – 996.
[5] a) A. Endo, A. Yanagisawa, M. Abe, S. Tohma, T. Kan, T.
Fukuyama, J. Am. Chem. Soc. 2002, 124, 6552– 6554; b) A.
Endo, T. Kann, T. Fukuyama, Synlett 1999, 1103 – 1105.
Received: August 4, 2006
Published online: November 13, 2006
Angew. Chem. Int. Ed. 2006, 45, 8028 –8032
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8031

Communications
[6] J. Chen, X. Chen, M. Bois-Choussy, J. Zhu, J. Am. Chem. Soc.
1999, 121, 8401 – 8402; d) A. G. Myers, D. W. Kung, J. Am.
Chem. Soc. 1999, 121, 10828 – 10829.
2006, 128, 87 – 89.
[28] One of the referees proposed that we isolate the orthoquinone
methide intermediate by using N-protected starting materials to
probe the reaction sequence. We are investigating in detail the
mechanism of this reaction.
[29] V. T. Ravikumar, A. H. Krotz, D. L. Cole, Tetrahedron Lett.
1995, 36, 6587 – 6590.
[7] a) S. Zheng, C. Chan, T. Furuuchi, B. J. D. Wright, B. Zhou, J.
Guo, S. J. Danishefsky, Angew. Chem. 2006, 118, 1786 – 1791;
Angew. Chem. Int. Ed. 2006, 45, 1754 – 1759; b) C. Chan, S.
Zheng, B. Zhou, J. Guo, R. H. Heid, B. J. D. Wright, S. J.
Danishefsky, Angew. Chem. 2006, 118, 1781 – 1786; Angew.
Chem. Int. Ed. 2006, 45, 1749 – 1754; c) B. Zhou, S. Edmondson,
S. J. Danishefsky, Tetrahedron Lett. 2000, 41, 2039 – 2042; d) B.
Zhou, J. Guo, S. J. Danishefsky, Tetrahedron Lett. 2000, 41,
2043 – 2046; e) B. Zhou, J. Guo, S. J. Danishefsky, Org. Lett.
2002, 4, 43 – 46.
[8] a) C. Cuevas, M. PØrez, M. J. Martin, J. L. Chicharro, C.
Fernꢀndez-Rivas, M. Flores, A. Francesch, P. Gallego, M.
Zarzuelo, F. De La Calle, J. Gracꢁa, C. Polanco, I. Rodriguez, I.
Manzanares, Org. Lett. 2000, 2, 2545 – 2548; b) R. Menchaca, V.
Martꢁnez, A. Rodrꢁguez, N. Rodrꢁguez, M. Flores, P. Gallego, I.
Manzanares, C. Cuevas, J. Org. Chem. 2003, 68, 8859 – 8866.
[9] a) N. Saito, K. Tashiro, Y. Maru, K. Yamaguchi, A. Kubo, J.
Chem. Soc. Perkin Trans. 1 1997, 53 – 69; b) N. Saito, H.
Kamayachi, M. Tachi, A. Kubo, Heterocycles 1999, 51, 9 – 12 ;
c) N. Saito, M. Tachi, R. Seki, H. Kamayachi, A. Kubo, Chem.
Pharm. Bull. 2000, 48, 1549 – 1557.
[30] M. De Paolis, J. Blankenstein, M. Bois-Choussy, J. Zhu, Org.
Lett. 2002, 4, 1235 – 1238.
[31] For a recent comprehensive review, see: R. W. Van De Water,
T. R. R. Pettus, Tetrahedron 2002, 58, 5367 – 5405.
[32] a) F. GuibØ, Y. Saint MꢂLeux, Tetrahedron Lett. 1981, 22, 3591 –
3594; b) P. Four, F. GuibØ, Tetrahedron Lett. 1982, 23, 1825 –
1828; c) for a review, see: F. GuibØ, Tetrahedron 1998, 54, 2967 –
3042.
[33] Owing to the restricted availability of this family of natural
products, there are very few reports in the literature that deal
with SARs of these compounds, despite their medical impor-
tance; see: E. J. Martinez, T. Owa, S. L. Schreiber, E. J. Corey,
Proc. Natl. Acad. Sci. USA 1999, 96, 3496 – 3501.
[10] a) B. Herberich, M. Kinugawa, A. Vazquez, R. M. Williams,
Tetrahedron Lett. 2001, 42, 543 – 546; b) W. Jin, R. M. Williams,
Tetrahedron Lett. 2003, 44, 4635 – 4639; c) W. Jin, S. Metobo,
R. M. Williams, Org. Lett. 2003, 5, 2095 – 2098; d) D. Fishlock,
R. M. Williams, Org. Lett. 2006, 8, 3299 – 3301.
[11] P. Magnus, K. S. Matthews, V. Lynch, Org. Lett. 2003, 5, 2181 –
2184.
[12] Y.-F. Tang, Z.-Z. Liu, S.-Z. Chen, Tetrahedron Lett. 2003, 44,
7091 – 7094.
[13] S. Aubry, S. Pellet-Rostaing, B. Fenet, M. Lemaire, Tetrahedron
Lett. 2006, 47, 1319 – 1323.
[14] P. A. Ceballos, M. PØres, C. Cuevas, A. Francesch, I. Manza-
nares, A. M. Echavarren, Eur. J. Org. Chem. 2006, 1926 – 1933.
[15] R. G. Kerr, N. F. Miranda, J. Nat. Prod. 1995, 58, 1616 – 1621.
[16] M. R. Winkle, R. C. Ronald, J. Org. Chem. 1982, 47, 2101 – 2108.
[17] S. Hanessian, D. Delorme, Y. Dufresne, Tetrahedron Lett. 1984,
25, 2515 – 2518.
[18] M. De Paolis, X. Chen, J. Zhu, Synlett 2004, 729 – 731.
[19] X. Xiao, D. Bai, Synlett 2001, 535 – 537.
[20] A. J. Mancuso, D. S. Brownfain, D. Swern, J. Org. Chem. 1979,
44, 4148 – 4150.
[21] G. Casiraghi, M. Cornia, G. Rassu, J. Org. Chem. 1988, 53, 4919 –
4922.
[22] a) M. De Paolis, A. Chiaroni, J. Zhu, Chem. Commun. 2003,
2896 – 2897; b) X. Chen, J. Chen, M. De Paolis, J. Zhu, J. Org.
Chem. 2005, 70, 4397 – 4408.
[23] For a recent review, see: S. Rendler, M. Oestreich, Synthesis
2005, 1727 – 1747.
[24] a) M. Sakaitani, Y. Ohfune, J. Org. Chem. 1990, 55, 870 – 876;
b) A. J. Zhang, D. H. Russel, J. Zhu, K. Burgess, Tetrahedron
Lett. 1998, 39, 7439 – 7442.
[25] E. D. Cox, J. M. Cook, Chem. Rev. 1995, 95, 1797 – 1842.
[26] TrocOCH2CHO (16) was prepared in two steps from ethyl-
eneglycol: 1) TrocCl (1 equiv), pyridine (1.1 equiv), ethylene-
glycol (10 equiv), CH2Cl2, 86%; 2) Dess–Martin periodinane
(1.1 equiv), CH2Cl2, 80%. Compound 16 has a high tendency to
undergo trimerization and should be used immediately after
preparation.
[27] For
a 1,3-cis-selective Pictet–Spengler reaction, see: a) G.
Massiot, T. Mulamba, J. Chem. Soc. Chem. Commun. 1983,
1147 – 1149; b) P. D. Bailey, S. P. Hollinshead, N. R. McLay, K.
Morgan, S. J. Palmer, S. N. Prince, C. D. Reynolds, S. D. Wood, J.
Chem. Soc. Perkin Trans. 1 1993, 431 – 439; c) A. G. Myers, D. W.
Kung, B. Zhong, M. Movassaghi, S. Kwon, J. Am. Chem. Soc.
8032 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 8028 –8032