A new, more efficient, and effective process for the synthesis of a key pentacyclic intermediate for production of ecteinascidin and phthalascidin antitumor agents.

Article abstract of DOI:10.1021/ol0056729

An efficient process is described for the synthesis of 5, a key intermediate for the synthesis of the potent antitumor agents ecteinascidin 743 (1) and phthalascidin (2) from the readily available building blocks 3b and 4.

Full text of DOI:10.1021/ol0056729

ORGANIC
LETTERS
2000
Vol. 2, No. 7
993-996
A New, More Efficient, and Effective
Process for the Synthesis of a Key
Pentacyclic Intermediate for Production
of Ecteinascidin and Phthalascidin
Antitumor Agents
Eduardo J. Martinez and E. J. Corey*
Department of Chemistry and Chemical Biology HarVard UniVersity, 12 Oxford Street,
Cambridge, Massachusetts 02138
corey@chemistry.harVard.edu
Received February 15, 2000
ABSTRACT
An efficient process is described for the synthesis of 5, a key intermediate for the synthesis of the potent antitumor agents ecteinascidin 743
(1) and phthalascidin (2) from the readily available building blocks 3b and 4.
Ecteinascidin 743 (1, Et 743) is an exceedingly potent
marine-derived antitumor agent1 which is now being studied
in various clinics with human patients.2 Because this
in 1996.3 More recently, a structural analogue of Et 743,
compound 2 (phthalascidin, Pt 650), has been found to
exhibit antitumor activity essentially indistinguishable from
that of 1.4 Both 1 and 2 are synthesized from building blocks
(1) The pioneering research in this area is due to Prof. Kenneth L.
Rinehart and his group. See: (a) Rinehart, K. L.; Shield, L. S. In Topics in
Pharmaceutical Sciences; Breimer, D. D., Crommelin, D. J. A., Midha, K.
K., Ed.; Amsterdam Medical Press: Noordwijk, The Netherlands, 1989; p
613. (b) Rinehart, K. L.; Holt, T. G.; Fregeau, N. L.; Keifer, P. A.; Wilson,
G. R.; Perun, T. J., Jr.; Sakai, R.; Thompson, A. G.; Stroh, J. G.; Shield, L.
S.; Seigler, D. S.; Li, L. H.; Martin, D. G.; Grimmelikhuijzen, C. J. P.;
Ga¨de, G. J. Nat. Prod. 1990, 53, 771. (c) Rinehart, K. L.; Sakai, R.; Holt,
T. G.; Fregeau, N. L.; Perun, T. J., Jr.; Seigler, D. S.; Wilson, G. R.; Shield,
L. S. Pure Appl. Chem. 1990, 62, 1277. (d) Rinehart, K. L.; Holt, T. G.;
Fregeau, N. L.; Stroh, J. G.; Keifer, P. A.; Sun, F.; Li, L. H.; Martin, D. G.
J. Org. Chem. 1990, 55, 4512. (e) Wright, A. E.; Forleo, D. A.;
Gunawardana, G. P.; Gunasekera, S. P.; Koehn, F. E.; McConnell, O. J. J.
Org. Chem. 1990, 55, 4508. (f) Sakai, R.; Rinehart, K. L.; Guan, Y.; Wang,
H.-J. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 11456.
(2) (a) Business Week September 13, 1999, 22. (b) Science 1994, 266,
1324.
(3) Corey, E. J.; Gin, D. Y.; Kania, R. J. Am. Chem. Soc. 1996, 118,
9202.
compound is not sufficiently available from the natural
source, the tunicate Ecteinascidia turbinate, it is being
produced industrially by the totally synthetic route described
10.1021/ol0056729 CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/15/2000

Scheme 1
3 and 4 via a common pentacyclic intermediate, 5. The
syntheses of 1 and/or 2 would eventually have to be produced
economically on a multikilogram scale, we sought to find a
more efficient and reproducible alternative route from 3 and
4 to 5. Reported herein is a new process which is simpler to
carry out than the original and which proceeds from 3b + 4
to 5 in six steps with an overall yield of 57% (average yield
per step of nearly 92%). The pathway of synthesis of
pentacycle 5 is summarized in Scheme 1.5
synthesis of 5 was accomplished originally3 from building
blocks 3a and 4 in six steps with an overall yield of 35%
(average yield per step of ca. 84%). Because the industrial
A solution of azeotropically dried (C7H8-THF) amino
lactone 43 in THF at 0 °C was treated dropwise with an
acylating reagent prepared from acid 3b3 (1.03 equiv),
1-hydroxy-7-azabenzotriazole (HOAT, 1.08 equiv), 2-chloro-
1,3-dimethylimidazolidinium hexafluorophosphate (CIP, 1.03
equiv), and triethylamine (2.06 equiv) in CH2Cl2 solution at
0 °C.6 The coupling product 6, which was obtained by
extractive workup, was allylated without further purification
by treatment in DMF solution at 23 °C with excess allyl
(4) Martinez, E. J.; Owa, T.; Schreiber, S. L.; Corey, E. J. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 3496.
(5) See: Myers, A. G.; Kung, D. W. J. Am. Chem. Soc. 1999, 121, 10828
for a different approach to the synthesis of structures such as 5.
994
Org. Lett., Vol. 2, No. 7, 2000

Scheme 2
bromide and 1.09 equiv of Cs2CO3 to give amide 7 in 81%
OH at 45 °C for 7 h produced pentacyclic product 10 in
89% overall yield from 8. Finally, the lactam function of 10
could be reduced cleanly by treatment with 4 equiv of
LiAlH2(OEt)2 in THF at 0 °C for 35 min to the corresponding
cyclic aminal which upon exposure to HCN provided
pentacyclic amino nitrile 5 in 87% overall yield from 10
after flash chromatography on silica gel.9
The synthesis of 5 which is outlined in Scheme 1 and
described above is advantageous relative to the originally
used synthetic pathway3 not only because of the substantially
greater overall yield (57 vs 35%) but also because of the
simplicity and reproducibility of the individual steps, espe-
cially the amide coupling (3b + 4 f 6) and the internal
Pictet-Spengler cyclization (9 f 10). In addition, no
difficulties have been encountered either in product purifica-
tion or scale-up. A critical element to the success of the
overall yield from 3a and 4 after flash chromatography on
silica gel. Selective reduction of the lactone function of 7 to
the corresponding lactol (8) was effected by reaction with
1.1 equiv of lithium diethoxyaluminum hydride (LiAlH2-
(OEt)2) in ether at -78 °C for 15 min in 95% yield.7,8
Desilylation of 8 to 9 and cyclization of 9 (without
purification) using 0.6 M triflic acid in 3:2 H2O-CF3CH2-
(6) For carboxylic acid-amine coupling methodology using CIP, see:
(a) Akaji, K.; Kuriyama, N.; Kimura, T.; Fujiwara, Y.; Kiso, Y. Tetrahedron
Lett. 1992, 33, 3177. (b) Akaji, K.; Kuriyama, N.; Kiso, Y. Tetrahedron
Lett. 1994, 35, 3315. (c) Akaji, K.; Kuriyama, N.; Kiso, Y. J. Org. Chem.
1996, 61, 3350.
(7) The reagent LiAlH2(OEt)2 was prepared by the addition of 1 equiv
of ethyl acetate to a solution of LiAlH4 in ether solvent at 0 °C and stirring
at 0 °C for 2 h just before use; see: Brown, H. C.; Tsukamoto, A. J. Am.
Chem. Soc. 1964, 86, 1089.
(8) For general reviews on the reduction of lactones, see: (a) Brown, H.
C.; Krishnamurthy, S. Tetrahedron 1979, 35, 567. (b) Cha, J. S. Org. Prep.
Proc. Int. 1989, 21(4), 451. (c) Seyden-Penne, J. Reduction by the Alumino-
and Borohydrides in Organic Synthesis, 2nd ed.; Wiley-VCH: New York,
1997; Section 3.2.5.
(9) For general references on amide reduction by hydride reagents, see
ref 7 and also the following: Myers, A. G.; Yang, B. H.; Chen, H.; Gleason,
J. L. J. Am. Chem. Soc. 1994, 116, 9361.
Org. Lett., Vol. 2, No. 7, 2000
995

sequence shown in Scheme 1 was the high efficiency and
selectivity of LiAlH2(OEt)2 for the two reduction steps 7 f
8 and 10 f 5, which suggest that this reagent can be used
to advantage in synthesis much more frequently than it has
been previously. In this connection it is noteworthy that the
use of diisobutylaluminum hydride in C7H8 and triisobutyl-
aluminum in Et2O for the reduction of 7 produced 8 in yields
of only 41% and 73%, respectively.
The synthetic route from 5 to phthalascidin 2, which
proceeds smoothly and in excellent yield (average yield per
step 90.8%), is outlined in Scheme 2. Pentacyclic triol 5 was
converted to phenolic monotriflate 11 by treatment with 1.1
equiv of PhNTf2 (McMurry reagent), 2 equiv of Et3N, and
0.2 equiv of 4-(dimethylamino)pyridine in CH2Cl2 at -30
°C for 38 h (74%). Conversion of 11 to mono tert-
butyldimethylsilyl (TBS) ether 12 and etherification with
methoxymethyl chloride (MOMCl) produced 13 in high
yield. Cleavage of the N-allyloxycarbonyl and O-allyl groups
in 13 gave secondary amine 14 (94%) which was N-
methylated to 15 and C-methylated to 16. Acetylation of
phenol 16 produced the corresponding acetate 17 which upon
desilylation formed primary alcohol 18. Mitsunobu displace-
ment of the primary hydroxyl of 18 produced phthalimide
19 which upon acid-catalyzed cleavage of the methoxymethyl
ether provided pure phthalascidin 2.
Since the original synthetic route to Et 743 (1) has proved
to be acceptable for large scale synthesis, it is our expectation
that the improved process described herein will be even more
useful, as will the new route to phthalascidin (2).4 Because
phthalascidin is more stable than ecteinascidin 743 and
considerably easier to make, it may prove to be a more
practical therapeutic agent.
Acknowledgment. This research was assisted financially
by grants from Pfizer Inc. and the National Institutes of
Health and a graduate fellowship to E.M. from the Schering
Plough Corporation. Early work in these laboratories on the
application of LiAlH2(OEt)2 to the selective reduction of
lactams to cyclic aminals was carried out by Dr. Victor
Behar, to whom we are most grateful.
OL0056729
996
Org. Lett., Vol. 2, No. 7, 2000