Enantioselective total syntheses of welwitindolinone A and fischerindoles I and G

Article abstract of DOI:10.1021/ja056171r

The first total syntheses of welwitindolinone A and the most complex members of the fischerindole family, fischerindoles I and G, are reported. Highlights of these short, protecting-group-free syntheses include the application of the recently developed di

Full text of DOI:10.1021/ja056171r

Published on Web 10/18/2005
Enantioselective Total Syntheses of Welwitindolinone A and Fischerindoles
I and G
Phil S. Baran* and Jeremy M. Richter
Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
Received September 8, 2005; E-mail: pbaran@scripps.edu
Moore and co-workers’ discovery of the hapalindole, fischerin-
dole, ambiguine, and welwitindolinone indole alkaloids from marine
blue-green algae launched an exciting new chapter in natural
products chemistry.1 Missions aimed at their total syntheses are
inspired by their array of promising bioactivities and unprecedented
molecular architectures.2 Welwitindolinone A (1, Figure 1) is a
densely functionalized oxindole harboring three all-carbon quater-
nary centers, a neopentyl chlorine atom, and a striking spiro-fused
cyclobutane that, upon cursory inspection, appears too strained to
exist. Fischerindoles I (2) and G (3, Figure 1) represent the apex
of molecular complexity within the fischerindole alkaloid class.1
To date, no chlorinated members of the fischerindole alkaloid class
have been prepared by total synthesis nor has any member of the wel-
witindolinones.2,3a This Communication describes enantioselective,
protecting-group-free syntheses of 1, 2, and 3 in nine or fewer steps
driven by a unique perspective on their biogenetic relationships
Figure 1. Retrosynthetic analysis of welwitindolinone A (1) and fischer-
and the development of powerfully simplifying transformations.
As illustrated in Figure 1, Moore and co-workers hypothesized1
that 1 might arise from an unusual cationic cyclization of the
hypothetical intermediate 4. Unless there is enzymatic intervention,
this transformation seems unlikely to occur based on thermodynamic
considerations. In addition, despite the dozens of natural products
isolated in this family, none contain architecture similar to the
putative intermediate 4. The planned synthesis of 1 was based on
our proposition that 2 is the actual precursor to 1 by an oxidative
ring contraction4 despite the fact that such ring contractions to form
four-membered carbocyclic rings are rare and, if successful, proceed
under harsh conditions.5,6 In principle, fischerindole I (2) could arise
by dehydrogenation of the cyclohexane ring of fischerindole G (3)
by way of indole oxidation followed by selective tautomerization
to the ∆10,11 alkene isomer. The latter alkaloid should be accessible
from 5 by a Friedel-Crafts alkylation and simple functional group
manipulations. In turn, the union of chloroketone 6 and indole could
lead to 5 in short order by the direct indole coupling recently
developed in this lab.3 The successful implementation of this plan
is outlined in detail below.
It was clear from the outset that the C-12 (fischerindole num-
bering1) quaternary center with adjacent chlorine atom at C-13
would pose significant difficulties (Scheme 1). Indeed, the instal-
lation of such hindered chlorine atoms has historically been a great
challenge.7 After a multitude of strategies were evaluated, a simple
solution emerged as detailed in Scheme 1. Enolization of (R)-
carvone oxide with LiHMDS followed by addition of vinylmag-
nesium bromide furnished neopentyl alcohol 7 as a single isomer
in 30% isolated yield.8 In accord with Wender’s studies,8 the major
byproduct of this reaction was derived from attack at the enolate-
bearing carbon (C-10) via an SN2′ pathway. To the best of our
knowledge, this is the first example of such a reaction installing
an all-carbon quaternary center. Chlorination using NCS/Ph3P
provided key chloroketone 6 in 55% yield. Despite the modest
overall yield of this two-step sequence, it can be used to easily
prepare multigram quantities of 6 within 1 day. By comparison,
indoles I (2) and G (3).
the C-12 epimer of 6 was previously prepared from (R)-carvone
oxide by a 10-step sequence in 10% overall yield.9
The crucial coupling of 6 with indole proceeded smoothly to
furnish 5 in 55% isolated yield as a single diastereomer (colorless
needles, mp 169-170 °C) despite the presence of an elimination-
prone chlorine atom.3 This effective method for direct C-C bond
formation between sp2 and sp3 centers enables all the necessary
carbon atoms of these complex natural products (1-3) to be
procured in only three steps. Functional group manipulations are
all that remained to complete the synthesis. Notably, this reaction
was routinely carried out on a multigram scale. The stereochemical
course of this reaction, and those preceding it, was verified by X-ray
crystallographic analysis (Figure 2).
The observed reactivity of 5 and later intermediates differed sig-
nificantly to that observed in the des-chloro fischerindole series.3a
Several acid catalysts (see Supporting Information for a list) screen-
ed to elicit Friedel-Crafts cyclization of 5 to 8 routinely furnished
ca. 20% isolated yield of 8 along with a number of side products
(no recovered 5). The use of Montmorillonite K-10 (DCE, 120 °C,
microwave, 6 min) finally provided a workable solution, furnishing
8 in 40% isolated yield (57% brsm) after a single recycling of
recovered starting material.10 Attempted cyclizations of the amine
or alcohol derived from ketone 5 were unsuccessful. Reductive ami-
nation11 of 8 proceeded with complete diastereoselectivity (axial
delivery) in the opposite sense11,3a to that obtained previously in
systems lacking the C-13 chlorine atom to produce amine 9 in 26%
isolated yield (48% brsm). Since the stereochemical outcome of the
reduction prevented access to 3, the reduction was left unoptimized
and an alternative path was charted to access the correct amine (10,
Scheme 1). Thus, ketone 8 was converted to amine 10 by following
a sequence similar to that developed by Fukuyama:9 (1) reduction
from the R-face using NaBH4, (2) mesylation using Ms2O in
pyridine (69% over two steps), (3) displacement with LiN3 (58%
yield), and (4) reduction of the resulting azide using sodium-
9
15394
J. AM. CHEM. SOC. 2005, 127, 15394-15396
10.1021/ja056171r CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1. Short, Enantioselective Total Syntheses of (+)-2 and (-)-3a
a Reagents and conditions: (a) LHMDS (1.2 equiv), THF, - 78 °C, 30 min; -15 °C, CH2CHMgBr (2.0 equiv), 15 min, 30%; (b) THF, PPh3 (1.0 equiv),
NCS (1.0 equiv), 18 h, 55%; (c) indole (2.0 equiv), LHMDS (3.1 equiv), THF, -78 °C, 30 min, then Cu(II)2-ethylhexanoate (1.5 equiv), -78 to 23 °C, 15
min, 55%; (d) DCE, Montmorillonite K-10 clay (10 equiv), microwave irradiation, 120 °C, 6 min, filter, then repeat, 40% + 30% recovered 5; (e) THF,
MeOH, NaCNBH3 (10 equiv), NH4OAc (40 equiv), 7 days, 26% 9 + 46% 8; (f) MeOH, NaBH4 (1.5 equiv), 0 °C, 5 min; then Ms2O (2.0 equiv), py, 23
°C, 30 min, 69% overall; (g) DMF, LiN3 (3.0 equiv), 120 °C, 48 h; then EtOH, Na(Hg) (10 equiv), reflux, 4 h, 38% overall; (h) HCO2H (1.3 equiv), CDMT
(1.4 equiv), DMAP (cat.), NMM (1.4 equiv), CH2Cl2, 23 °C, 30 min, 87%; (i) PhH, Burgess reagent (2.0 equiv), 23 °C, 30 min, 82%; (j) same as (h), 98%;
(k) THF, TEA (1.0 equiv), t-BuOCl (1.5 equiv), 0 °C, 10 min, then SiO2/Et3N (PTLC), then PhH, Burgess reagent (2.0 equiv), 23 °C, 30 min, 47% overall.
CDMT ) 2-Chloro-4,6-dimethoxy-1,3,5-triazine; DCE ) 1,2-dichloroethane; DMF ) N,N-dimethylformamide; DMAP ) 4-(dimethylamino)pyridine; IBX
) o-iodoxybenzoic acid; LHMDS ) lithium hexamethyldisilazide; Ms ) methanesulfonyl; NCS ) N-chlorosuccinimide; NMM ) N-methylmorpholine.
tion12 of amine 9 with DMT-MM/formic acid furnished 11 in 98%
yield. Remarkably, 2 was produced in 47% overall yield when 11
was treated with t-BuOCl and Et3N at 0 °C followed by treatment
with deactivated (Et3N) silica gel and then exposure to the Burgess
reagent. Synthetic (+)-2 ([R]D +24.4 (CH2Cl2, c 0.09), nat. value
not given1) was spectroscopically identical to that reported. This
compelling transformation likely occurs through 3-chloroindolenine
12,14 elimination to methyleneindolenine 13a, tautomerization to
Figure 2. X-ray crystal structure of 5.
13b, and dehydration to 2. Dehydration of 13a followed by
mercury amalgam (66% yield). Amine 10 was then formylated with
tautomerization to 2 cannot be ruled out. The intermediate forma-
mide 13a/b could be isolated; however, it was unstable and therefore
difficult to characterize. On the basis of the foregoing results and
the stereochemical diversity present in this family of natural
products, we propose that 11-epi-fischerindole G (not yet isolated)
might be the actual biogenetic precursor to 2.
formic acid/DMT-MM12 and dehydrated using the Burgess reagent13
to provide 3 in 71% yield over two steps. Synthetic (-)-3 was
spectroscopically identical to that reported with the exception of
optical rotation [[R]D -50 (CH2Cl2, c 0.02), nat. [R]D +67 (CH2-
Cl2, c 0.09)]. In principle, the naturally occurring enantiomer of 3
could be prepared from (S)-carvone oxide (vide infra).
With chemistry in place to access (+)-2, the S enantiomer of
Despite a number of attempts, 3 and its formamide precursor
carvone oxide was employed to construct (-)-2 ([R]D -14.0 (CH2-
could not be oxidized to 2 by either direct or indirect means. Expo-
Cl2, c 0.05) in preparation for its conversion to (+)-1 (absolute
sure to various agents, such as DDQ, p-chloroanil, MnO2, Pd/C,
configuration not rigorously established by Moore1), as shown in
and IBX, led only to recovered starting material or decomposition
as judged by TLC and 1H NMR. Treatment of 3 or the correspond-
ing formamide with DMDO or t-BuOCl led to stable C(3)-OH or
-Cl species, respectively, that were resistant to elimination/dehy-
dration under a range of conditions. Electrophilic attack from the
top face of 3, away from the C-11 nitrogen functionality, would
explain these failures since a difficult syn elimination would be
required.
Scheme 2. Although the proposed conversion of 2 to 1 might appear
intuitive, practical difficulties were expected due to the sensitivity
of both alkaloids to acidic media and the sheer ring strain of the
resulting product. Our concerns were intensified by the knowledge
that oxidative ring contractions to generate five- and six-membered
rings typically require elevated temperatures, and hence, forming
a strained four-membered ring should be even more difficult.4
Preliminary studies on ketone (+)-8 were conducted in order to
determine the feasibility of the ring contraction. Thus, intermediate
Attention was therefore returned to amine 9 in the hope that
electrophilic attack would occur from the opposite face. Formyla-
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15395

C O M M U N I C A T I O N S
Scheme 2. Short, Enantioselective Total Syntheses of (+)-1 and (-)-2a
a Reagents and conditions: (a) THF, TEA (1.0 equiv), t-BuOCl (1.7 equiv), 0 °C, 10 min; then 40:20:1 MeOH:H2O:AcOH, 5 min, 41%, 1:1 mixture of
14 and 15; (b) THF, TEA (1.0 equiv), t-BuOCl (1.5 equiv), -30 °C, 1 min; then 95:4:1 THF:H2O:TFA, -30 to 0 °C, 5 min, 28%, 10:1 mixture of 1 and
3-epi-1. TFA ) trifluoroacetic acid.
(+)-8 was treated with t-BuOCl followed by dilute acid (gently
heated in order to dissolve the substrate) to furnish spirooxindole
14 and enone 15 as a 1:1 mixture in 41% yield (unoptimized). The
stereochemistry of 14 (colorless needles, mp 197-198 °C) was
verified by X-ray crystallographic analysis (Scheme 2). Given the
aforementioned considerations, it is remarkable that this reaction
occurs at low temperature and within minutes. Although ketone
14 represents a potentially viable intermediate to complete the
synthesis of 1, attention was returned to fischerindole I in the hope
that it could be directly converted to 1. After much exploration,
conditions were developed for the conversion of (-)-2 to (+)-1.
Exposure of (-)-2 to 1.5 equiv of freshly prepared t-BuOCl in THF
at -30 °C for 1 min followed by removal of solvent in vacuo,
immediate dissolution in THF/H2O/TFA (95:4:1),15 and warming
to 0 °C led to a 10:1 mixture of (+)-1 and 3-epi-1 (tentatively
assigned) in 28% yield (unoptimized). Synthetic (+)-1 was
spectroscopically identical to that reported [[R]D +160 (CH2Cl2, c
0.02), nat. [R]D + 377 (CH2Cl2, c 0.078)], thus confirming its
unusual structure and absolute configuration. The remarkable ease
with which this ring contraction takes place suggests that Nature
may be employing a similar strategy to access 1.
Eli Lilly, GlaxoSmithKline, Roche, the Searle Scholarship Fund,
and the NSF (predoctoral fellowship to J.M.R.).
Supporting Information Available: Detailed experimental pro-
cedures, copies of all spectral data, and full characterization. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Stratmann, K.; Moore, R. E.; Bonjouklian, R.; Deeter, J. B.; Patterson,
G. M. L.; Shaffer, S.; Smith, C. D.; Smitka, T. A. J. Am. Chem. Soc.
1994, 116, 9935-9942.
(2) Studies toward the welwitindolinones: (a) Wood, J. L.; Holubec, A. A.;
Stoltz, B. M.; Weiss, M. M.; Dixon, J. A.; Doan, B. D.; Shamji, M. F.;
Chen, J. M.; Heffron, T. P. J. Am. Chem. Soc. 1999, 121, 6326-6327.
(b) Holubec, A. A. Ph.D. Thesis, Yale University, 2000. (c) Weiss, M.
M. Ph.D. Thesis, Yale University, 2001. (d) Deng, H. Ph.D. Thesis,
University of California, Santa Cruz, 2001. (e) Huang, S.-T. Ph.D. Thesis,
Brandeis University, 2001. (f) Deng, H.; Konopelski, J. P. Org. Lett. 2001,
3, 3001-3004. (g) Jung, M. E.; Slowinski, F. Tetrahedron Lett. 2001,
42, 6835-6838. (h) Alvarado, P. L.; Garc´ıa-Granda, S.; AÄ lvarez-Ru´a, C.;
Avendan˜o, C. Eur. J. Org. Chem. 2002, 1702-1707. (i) Ready, J. M.;
Reisman, S. E.; Hirata, M.; Weiss, M. M.; Tamaki, K.; Ovaska, T. V.;
Wood, J. L. Angew. Chem., Int. Ed. 2004, 43, 1270-1272. (j) MacKay,
J. A.; Bishop, R. L.; Rawal, V. H. Org. Lett. 2005, 7, 3421-3424. (k)
Baudoux, J.; Blake, A. J.; Simpkins, N. S. Org. Lett. 2005, 7, 4087-4089.
(3) (a) Baran, P. S.; Richter, J. M. J. Am. Chem. Soc. 2004, 126, 7450-
7451. (b) Baran, P. S.; Richter, J. M.; Lin, D. W. Angew. Chem., Int. Ed.
2005, 44, 609-612.
Aside from the brevity (seven, eight, and nine steps to 2, 1, and
3, respectively, from carvone oxide) and stereochemical control of
the current syntheses, there are a number of prominent features:
(a) application of the powerfully simplifying direct carbonyl-indole
coupling reaction (6 f 5);3 (b) two-step construction of the
challenging quaternary center and adjacent chlorine atom common
to this family of natural products from carvone oxide; (c) a position-
selective tandem dehydrogenation/dehydration sequence (11 f 2);
(d) an exceedingly facile oxidative ring contraction (2 f 1); (e)
experimental evidence for an alternative biogenetic hypothesis for
this family of alkaloids; and (f) absence of protecting groups and
excessive oxidation state manipulations.
Acknowledgment. This work is dedicated to Professor Albert
Eschenmoser on the occasion of his 80th birthday. We thank Dr.
D. H. Huang and Dr. L. Pasternack for NMR spectroscopic
assistance, and Dr. G. Siuzdak and Dr. Raj Chadha for mass
spectrometric and X-ray crystallographic assistance, respectively.
We are grateful to Biotage for a generous donation of microwave
process vials used in this study. Financial support for this work
was provided by The Scripps Research Institute, Amgen, DuPont,
(4) Witkop, B.; Patrick, J. B. J. Am. Chem. Soc. 1953, 75, 2572-2576.
(5) For the sole example of spirocyclobutane formation, see: Wenkert, E.;
Moeller, P. D. R.; Piettre, S. R.; McPhail, A. T. J. Org. Chem. 1987, 52,
3404-3409.
(6) For a mechanistically related oxidative ring contraction that furnishes a
strained â-lactam, see: Baran, P. S.; Shenvi, R. A.; Mitsos, C. A. Angew.
Chem., Int. Ed. 2005, 44, 3714-3717.
(7) For example, see: (a) Gorthey, L. A.; Vairamani, M.; Djerassi, C. J. Org.
Chem. 1985, 50, 4173-4182. (b) Albone, K. S.; Macmillan, J.; Pitt, A.
R.; Willis, C. L. Tetrahedron 1986, 42, 3203-3214.
(8) (a) Wender, P. A.; Erhardt, J. M.; Letendre, L. J. J. Am. Chem. Soc. 1981,
103, 2114-2116. (b) Erhardt, J. M. Ph.D. Thesis, Harvard University,
1982.
(9) Fukuyama, T.; Chen, X. J. Am. Chem. Soc. 1994, 116, 3125-3126.
(10) Smyth, T. P.; Corby, B. W. Org. Proc. Res. DeV. 1997, 1, 264-267 and
references therein.
(11) Vaillancourt, V.; Albizati, K. F. J. Am. Chem. Soc. 1993, 115, 3499-3502.
(12) 4-(4,6-Dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium chloride: De
Luca, L.; Giacomelli, G.; Porcheddu, A.; Salaris, M. Synlett 2004, 14,
2570-2572.
(13) Creedon, S. M.; Crowley, H. K.; McCarthy, D. G. J. Chem. Soc., Perkin
Trans. 1 1998, 1015-1017.
(14) The stereochemistry of the C-3 chlorine is inferred, as shown in Scheme
1, but has not been confirmed.
(15) Cushing, T. D.; Sanz-Cervera J. F.; Williams, R. M. J. Am. Chem. Soc.
1996, 118, 557-579.
JA056171R
9
15396 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005