Enantioselective total syntheses of ditryptophenaline and ent-WIN 64821 [12]

Full text of DOI:10.1021/ja0166141

J. Am. Chem. Soc. 2001, 123, 9465-9467
9465
Scheme 1
Enantioselective Total Syntheses of Ditryptophenaline
and ent-WIN 64821
Larry E. Overman* and Daniel V. Paone
Department of Chemistry, 516 Rowland Hall
UniVersity of California, IrVine, California 92697-2025
ReceiVed July 13, 2001
Many complex bioactive indole alkaloids are derived from two
tryptophan units and incorporate two additional R-amino acids
in diketopiperazine motifs.1 The vast majority of these alkaloids
have a 3a,3a′-bispyrrolidinoindoline core.2 Ditryptophenaline (1)3
and WIN 648214 are typical of most of these natural products in
having their contiguous stereogenic quaternary carbons related
by C2 symmetry. Ditryptophenaline has been isolated from several
Aspergillus species and was the first member of this alkaloid
family to be structurally characterized.3 The gross structure and
relative configuration of 1 were secured by X-ray crystallographic
studies;3 its absolute configuration was defined by the formation
of 1 in low yield from oxidative dimerization of cyclo-(N-methyl-
(S)-phenylalanyl-(S)-tryptophyl).5 Like most members of this
natural products family, WIN 64821 (also called Q20547-A)6 was
identified by bioactivity-guided investigations of various fungal
strains in the pharmaceutical industry.3,6 WIN 64821 is a
competitive substance P antagonist with submicromolar potency
against the human NK1 receptor7 and also an antagonist of the
cholecystokinin type-B receptor.6 We recently disclosed a practical
method for asymmetric construction of contiguous stereogenic
quaternary carbon centers.8 Herein we describe use of this
chemistry to prepare ditryptophenaline (1) and ent-WIN 64821
(2), representative members of the two families of C2-symmetric
bispyrrolidinoindoline diketopiperazine alkaloids that differ in
relative orientation between their bispyrrolidinoindoline and
diketopiperazine subunits.
Scheme 2
Our approach to 1 and 2 exploits the ready availability of
bisoxindole cyclohexanediol 58 and is outlined in retrosynthetic
format in Scheme 1. Logical precursors of these alkaloids are
the bisoxindole diamines 3 and 4. The initial challenge would be
elaborating the dialdehyde derived from oxidative cleavage of
cyclohexanediol 5 to these C2-symmetric tetracyclic intermediates.
Although introducing the two new stereogenic centers might be
achieved selectively using reagent or catalyst control, we sought
(1) For reviews, see: (a) Hibino, S.; Choshi, T. Nat. Prod. Rep. 2001, 18,
66-87 and earlier reviews in this series. (b) Anthoni, U.; Christophersen C.;
Nielsen, P. H. In Alkaloids Chemical and Biological PerspectiVes; Pelletier,
S. W., Ed.; Pergamon: London, 1999; Vol. 13, pp163-236.
(2) Asperazine is one exception. For its inaugural total synthesis, see:
Govek, S. P.; Overman, L. E. in following paper J. Am. Chem. Soc. 2001,
123, 9468-9469.
(3) Springer, J. P.; Bu¨chi, G.; Kobbe, B.; Demain, A. L.; Clardy, J.
Tetrahedron Lett. 1977, 2403-2406.
(4) Barrow, C. J.; Cai, P.; Snyder, J. K.; Sedlock, D. M.; Sun, H. H.; Cooper,
R. J. Org. Chem. 1993, 58, 6016-6021.
in this inaugural endeavor to define the extent of stereoselection
that could be realized from substrate control alone.
(5) (a) Nakagawa, M.; Sugumi, H.; Kodato, S.; Hino, T. Tetrahedron Lett.
1981, 5323-5326. (b) A related simpler C2-symmetric product of unknown
stereochemistry was recently described as a byproduct of a free radical coupling
reaction, see: Depew, K. M.; Marsden, S. P.; Zatorska, D.; Zatorski, A.;
Bornmann, W. G.; Danishefsky, S. J. J. Am. Chem. Soc. 1999, 121, 11953-
11963.
Our investigations began by preparing 5 in 30% overall yield
(five linear steps) from unnatural (S)-tartaric acid.8 Sodium
periodate oxidation of 5 then provided dialdehyde 6 in essentially
quantitative yield (Scheme 2). Initial survey experiments showed
that this intermediate reacted most cleanly with Grignard reagents.
Thus, sequential reaction of stannane 79 with n-butyllithium,
MgBr2‚OEt2,10 and dialdehyde 6 provided two readily separated
products 8 and 9 in nearly equal amounts and 86% combined
yield. It was readily apparent from NMR spectra that 8 was one
(6) Hiramoto, M.; Shibazaki, M.; Miyata, H.; Saita, Y. Tennen Yuki
Kagobutsu Toronkai Koen Yoshishu 1994, 36, 557-569; Chem. Abstr. 123,
131999.
(7) (a) Sedlock, D. M.; Barrow, C. J.; Brownell, J. E.; Hong, A.; Gillum,
A. M.; Houck, D. R. J. Antibiot. 1994, 47, 391-398. (b) Oleynek, J. J.;
Sedlock, D. M.; Barrow, C. J.; Appell, K. C.; Casiano, F.; Haycock, D.; Ward,
S. J.; Kaplita, P.; Gillum, A. M. J. Antibiot. 1994, 47, 399-410. (c) Popp, J.
L.; Musza, L. L.; Barrow, C. J.; Rudewicz, P. J.; Houck, D. R. J. Antibiot.
1994, 47, 411-419.
(9) Stannane 7 was prepared in 91% yield from n-Bu3SnH and SEMCl
following a general procedure: Buchwald, S. L.; Nielsen, R. B.; Dewan, J.
C. Organometallics 1989, 8, 1593-1598.
(8) Overman, L. E.; Larrow, J. F.; Stearns, B. A.; Vance, J. M. Angew.
Chem., Int. Ed. 2000, 39, 213-215.
10.1021/ja0166141 CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/31/2001

9466 J. Am. Chem. Soc., Vol. 123, No. 38, 2001
Communications to the Editor
Scheme 3
of the two possible C2-symmetric products, whereas 9 had C1
symmetry. A variety of alkoxymethyl Grignard reagents contain-
ing various hydroxyl protective groups condensed with 6 in similar
fashion. The seldom used (trimethylsilyl)ethyl (TMSE) group was
chosen because it allowed for facile separation of diastereomers
by standard silica gel chromatography and proved compatible with
later transformations.
Coupling of 11 with N-(9-fluorenylmethoxycarbonyl)-(R)-phen-
ylalanine [FMOC-(R)-Phe-OH] and 1,3-dicyclohexyl carbodiimide
(DCC) provided tetrapeptide 12 in high yield. The TMSE-
protecting groups were then removed with BCl3 at -78 °C to
deliver diol 13 in 80% yield.16 Direct oxidation of this intermediate
to the corresponding acid failed, using all standard oxidants we
surveyed. However, this delicate transformation could be realized
in two steps by first exposing 13 to Dess-Martin periodinane
(DMP) in MeCN17,18 to give the dialdehyde, which upon further
The conversion of 8 to ent-WIN 64821 is summarized in
Scheme 3. The initial challenge is elaborating 8 to form the
3a,3a′-bispyrrolidinoindoline ring system. Such a conversion must
at some point involve lowering the oxidation state of the oxindole
carbonyl groups, an event that places the fragile 3a,3a′ σ-bond
in jeopardy. To the best of our knowledge, conversions of this
type are unknown with bisoxindoles having branched side chains.
That such a conversion would be difficult became immediately
clear when the four-step sequence utilized previously in our
synthesis of (+)- and (-)-chimonanthine8,11 failed in this substi-
tuted system because of cleavage of the labile 3a,3a′ σ-bond and
competing formation of tetrahydrofuranylindolines. As a result,
a shorter sequence was developed involving direct reduction of
a bisoxindole to the generate the 3a,3a′-bispyrrolidinoindoline ring
system.12 Diol 8 was first converted to diazide 10 under standard
Mitsunobu conditions with diphenylphosphoryl azide (DPPA).13
Treatment of 10 with sodium bis(2-methoxyethoxy)aluminum
hydride (Red-Al) at ambient temperature resulted in immediate
reduction of the azides; subsequent heating to 100 °C initiated
cyclization to the bis-amidine,14,15 which then was converted
slowly to bispyrrolidinoindoline 11. Under optimized conditions,
this demanding reduction could be accomplished in 71% yield.
19
oxidation with buffered NaClO2 provided diacid 14. Removal
of the FMOC groups with piperidine, followed by DCC-mediated
cyclization, then furnished octacyclic diketopiperazine 15 in 62%
overall yield from 13.20 Finally removal of the aniline benzyl
groups by hydrogenolysis delivered ent-WIN 64821 (2) in 70%
yield. This product was identical in all respects to a sample of
the natural product, save optical rotation: [R]D -200 (lit.4 [R]D
+200).
The related total synthesis of ditryptophenaline began with
oxidation of C1-symmetric diol 9 to dione 16 with pyridinium
dichromate (Scheme 4). After some optimization, we found that
reduction of this intermediate with NaBH4 in methanol at -78
°C resulted in the formation of only one C2-symmetric diol
product 17 (90% yield), together with trace amounts of 9.21 Diol
17 was readily transformed to diazide 18. However, conversion
of this intermediate to 3a,3a′-bispyrrolidinoindoline 19 was
extremely challenging, undoubtedly because the alkoxymethyl side
chain in this series emerges on the same face as the bulky angular
pyrrolidinoindoline substituent. After much experimentation, we
found that competing fragmentation of the 3a,3a′ σ-bond as well
as debenzylation were minimized by incremental heating of the
(10) For a representative transmetalation of an alkyllithium to the Grignard
reagent, see: Lau, P. W. K.; Chan, T. H. Tetrahedron Lett. 1978, 2383-
2386.
(16) More common conditions for this deprotection (BF3‚OEt2 or n-Bu4-
NF) resulted in concomitant loss of the FMOC groups.
(11) Overman, L. E.; Paone, D. V.; Stearns, B. A. J. Am. Chem. Soc. 1999,
121, 7702-7703.
(12) For related reductions to form unsubstituted pyrrolidinoindolines, see,
inter alia: (a) Pei, X.-F.; Bi, S. Heterocycles 1994, 39, 357-360. (b) Fang,
C.-L.; Horne, S.; Taylor, N.; Rodrigo, R. J. Am. Chem. Soc. 1994, 116, 9480-
9486. (c) Hendrickson, J. B.; Go¨schke, R.; Rees, R. Tetrahedron 1964, 20,
565-579.
(13) Lal, B.; Pramanik, B.; Manhas, M. S.; Bose, A. K. Tetrahedron Lett.
1977, 1977-1980.
(14) Intermediates in this sequence were identified by mass spectroscopy
(ESI) and in some cases also by NMR.
(15) The formation of tricyclic amidines from the reduction of simple
substituted oxindoles has precedent, see: Kawasaki, T.; Terashima, R.;
Sakaguchi, K.; Sekiguchi, H.; Sakamoto, M. Tetrahedron Lett. 1996, 37,
7525-7528.
(17) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155-4156.
(18) Use of CH2Cl2 as solvent in this oxidation gave a complex mixture of
products containing largely six-membered ring hemiaminal functionalities.
These intermediates could not be oxidized further.
(19) (a) Bal, B. S.; Childers, W. E., Jr.; Pinnick, H. W. Tetrahedron 1981,
37, 2091-2096. (b) Lindgren, B. O.; Nilsson, T. Acta Chem. Scand. 1973,
27, 888-890.
(20) For abbreviations not defined in J. Org. Chem. 2001, 66, 24A, see
Supporting Information.
(21) Although this issue has not received specific study, the high stereo-
selectivity observed in this reduction would be consistent with external delivery
of hydride to what appears from modeling to be a favorable chelate between
the carbonyl groups of the ketone and distal oxindole. A related assembly in
the first Grignard addition to 6 would be consistent with the generation of
diol products 8 and 9.

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 38, 2001 9467
Scheme 4
Red-Al reaction to 100 °C over 54 h.22 Under these conditions,
the conversion of 18 to bispyrrolidinoindoline 19 could be realized
in 52% yield. We were completely unsuccessful in attempts to
couple 19 with FMOC-(S)-MePhe-OH, presumably again reflect-
ing the increased steric crowding in this series. To decrease steric
congestion in the vicinity of the pyrrolidine nitrogen, the benzyl
groups were removed with sodium and liquid ammonia at -78
°C.23 Selective acylation of the resulting tetraamine with FMOC-
(S)-MePhe-OH (DCC and 1-hydroxy-7-azabenzotriazole, HOAT)
gave tetrapeptide 20 in 82% yield over two steps.24 Cleavage of
the (trimethylsilyl)ethyl groups of 20 was also challenging, and
again the conditions used in the diastereomeric series led to
extensive fragmentation of the fragile 3a,3a′ σ-bond. Fortunately,
when the solvent was changed to toluene and 3 equiv of 2,6-di-
tert-butyl-4-methylpyridine (DTBMP) were added, BCl3-promoted
cleavage of 20 provided diol 21 in 87% yield. Finally the two-
step oxidation procedure proceeded cleanly in the presence of
the free anilines to deliver 22, which upon deprotection with
piperidine in THF and cyclization with DCC gave ditrypto-
In summary, concise enantioselective total syntheses of di-
tryptophenaline (1) and ent-WIN 64821 (2) have been completed
from a common, readily available8 precursor 5. The synthesis of
2 confirms the structure proposed for WIN 64821;4 more
importantly, it is the first total synthesis of a member of the more
bioactive3,26 family of C2-symmetric bispyrrolidinoindoline dike-
topiperazine alkaloids that have a cis relationship of the angular
hydrogens flanking the pyrrolidine nitrogens. The chemistry
described herein should for the first time allow the bispyrroli-
dinoindoline diketopiperazine structural motif to be rationally
modified by chemical synthesis.27,28
Acknowledgment. We thank NIH (HL-25854) for financial support,
Professor Collin Barrow (University of Melbourne) for a sample of natural
WIN-64821, Dr. Jeremy Scott for helpful discussions, and Dr. John
Greaves for assistance with mass spectrometric analyses. NMR and mass
spectra were determined at UCI using instruments purchased with the
assistance of NSF and NIH.
1
phenaline (1) in 54% overall yield from 21. High field H and
Supporting Information Available: Experimental procedures for key
transformations (preparation of 8, 9, 11, 15, 2, 19, 21, and 1) and copies
of 1H and 13C NMR spectra for these compounds (PDF). This material is
available free of charge via the Internet at http://pubs.acs.org.
13C NMR spectra of synthetic ditryptophenaline agreed perfectly
with those reported for the natural product25 as did optical
rotation: [R]D -317 (lit.3 [R]D -318).
(22) Reduction of amidine intermediates was much slower in this series.
As a result, debenzylation became competitive with reduction of the second
amidine group. Under no conditions surveyed could the resulting debenzylated
amidine be reduced further.
(23) Attempted removal of the benzyl groups from 19 by hydrogenolysis
resulted in decomposition. We have found that if the pyrrolidine nitrogens of
a bispyrrolidinoindoline are acylated, debenzylation by dissolving metal
reduction fails, although hydrogenolysis is successful. The reverse holds true
for the NH or N-alkyl analogues.
(24) Prior studies in the ent-WIN 64821 series had shown that the
phenylalanine residue could not be introduced after the TMSE groups had
been removed, because acylation of the primary alcohols and pyrrolidine
nitrogens occurred at similar rates.
JA0166141
(25) Maes, C. M.; Potgieter, M.; Steyn, P. S. J. Chem. Soc., Perkin Trans.
1 1986, 861-866.
(26) Barrow, C. J.; Sedlock, D. M. J. Nat. Prod. 1994, 57, 1239-1244.
(27) Simple analogues of WIN 64821 lacking the 3a,3a′-bispyrrolodino-
indoline moiety are reported to be poor substance P antagonists, see: Barrow,
C. J.; Musza, L. L.; Cooper, R. Bioorg. Med. Chem. Lett. 1995, 5, 377-380
and ref 3.
(28) Changing the starting material to natural tartaric acid would allow 2,
related alkaloids, and their analogues to be prepared in the natural enantiomeric
series.