Concise, asymmetric, stereocontrolled total synthesis of stephacidins A, B and notoamide B

Article abstract of DOI:10.1021/ja070259i

Concise asymmetric total syntheses of the fungal metabolites (-)-stephacidin A, (+)-stephacidin B, and (+)-notoamide B are described. Key features of these total syntheses include (1) a facile synthesis of (R)-allyl proline methyl ester, (2) a revised rou

Full text of DOI:10.1021/ja070259i

Published on Web 04/25/2007
Concise, Asymmetric, Stereocontrolled Total Synthesis of
Stephacidins A, B and Notoamide B
Gerald D. Artman III, Alan W. Grubbs, and Robert M. Williams*
Contribution from the Department of Chemistry, Colorado State UniVersity, Fort Collins,
Colorado 80523, and The UniVersity of Colorado Cancer Center, Aurora, Colorado 80045
Received February 6, 2007; E-mail: rmw@lamar.colostate.edu
Abstract: Concise asymmetric total syntheses of the fungal metabolites (-)-stephacidin A, (+)-stephacidin
B, and (+)-notoamide B are described. Key features of these total syntheses include (1) a facile synthesis
of (R)-allyl proline methyl ester, (2) a revised route toward the pyranoindole ring system, (3) a novel cross-
metathesis strategy for the introduction of important functional groups, and (4) an SN2′ cyclization to form
the [2.2.2] bridged bicyclic ring system. Furthermore, our synthesis has taken advantage of microwave
heating to shorten reaction times as well as increase yields for the preparation of vital intermediates.
Introduction
alkaloid, (+)-avrainvillamide (3), were in evidence, suggesting
a simple dimerization-based biogenesis of 2 from 3.3 The unique
Fungi have proven to be a rich source of densely function-
alized secondary metabolite alkaloids derived from proline,
tryptophan, and isoprene. In 2002, Bristol-Myers Squibb
reported the biologically active metabolites isolated from a
fermentation broth of Aspergillius ochraceus.1 The stephacidins
A (1) and B (2) were identified as potent inhibitors of several
human tumor cell lines with the complex alkaloid (-)-
stephacidin B (2) exhibiting a high cytotoxic potency against
testosterone-dependent prostate LNCaP lymphoma (Figure 1).
An investigation into their mode of action determined that they
inhibit cell growth via a novel mechanism possibly resulting in
a new, as yet undetermined, target for treating cancer. In addition
to the biological activity of 1 and 2, their structures represent a
new degree of complexity of prenylated indole alkaloids from
fungi. Both are built around the [2.2.2] diazaoctane ring system
common to the brevianamides, paraherquamides, marcfortines,
asperparalines, and related alkaloids. The structural framework
of (+)-stephacidin A was readily determined through NMR
experiments. However, the skeletal connectivity of (-)-stepha-
cidin B could not be elucidated using only NMR experiments.
X-ray crystallography was required to reveal its unprecedented
structure. Containing two [2.2.2]diazaoctane bridged bicycles,
a nitrone, a N-hydroxyindole, and nine stereogenic centers, five
of which are quaternary, prompted von Nussbaum2 to comment
that (-)-stephacidin B “proVides a new leVel of complexity
within prenylated indole alkaloids from fungi.”
dimeric nature of (-)-stephacidin B coupled with its potent
biological activity has resulted in several groups initiating
programs toward its total synthesis. The Baran group reported
the first total synthesis of stephacidin A (1) using a novel
oxidative enolate coupling to form the [2.2.2] bridged bicyclic
ring system followed by an unusual cascade reaction for the
formation of the natural product in 29 total steps from
commercially available starting materials.4a Baran’s initial report
was followed shortly after by Herzon and Myers and co-workers
first total synthesis of (+)-avrainvillamide (3) in 26 total steps
from commercially available starting materials using a very
clever radical-based strategy for the synthesis of the bridged
bicyclic core.5-7 Following a late stage installation of the
chromene ring system and formation of the vinyl nitrone, they
demonstrated that 3 spontaneously dimerizes to (-)-stephacidin
B (2) in the presence of the weak base Et3N in greater than
95% yield by 1H NMR analysis. Alternatively, the Baran group
completed a synthesis of avrainvillamide (3) from synthetic
stephacidin A (1).4b Reduction of the indole 2,3-double-bond
followed by oxidation of the intermediate dihydroindole with
SeO2 afforded 3 in 23% yield. Using the conditions reported
by Myers, the Baran group dimerized 3 to stephacidin B (2)
under basic conditions as well as under acidic conditions.
Furthermore, the Baran group was able to establish the absolute
stereochemistry of 1-3 through their synthetic efforts.8
Our laboratory has a rich history synthesizing and probing
The isolation of (-)-stephacidin B (2) also poses an interest-
ing series of biosynthetic questions. The researchers at Bristol-
Myers Squibb recognized that if the bonds between C20-C51
and C21-N55 of 2 are broken, two molecules of the related
the biosynthesis of fungal metabolites containing a [2.2.2]
(3) Fenical, W.; Jensen, P.; Cheng, X. C. U.S. Patent 6,066,635, 2000; Chem.
Abstr. 2000, 132, 346709.
(4) (a) Baran, P. S.; Guerrero, C. A.; Ambhaikar, N. B.; Hafensteiner, B. J.
Angew. Chem., Int. Ed. 2005, 44, 606-609. (b) Baran, P. S.; Guerrero, C.
A.; Ambhaikar, N. B.; Hafensteiner, B. J. Angew. Chem., Int. Ed. 2005,
44, 3892-3895.
(1) (a) Qian-Cutrone, J.; Haung, S.; Shu, Y.-Z.; Vyas, D.; Fairchild, C.;
Menendez, A.; Krampitz, K.; Dalterio, R.; Klohr, S. E.; Gao, Q. J. Am.
Chem. Soc. 2002, 124, 14556-14557. (b) Qian-Cutrone, J.; Krampitz, K.
D.; Shu, Y.-Z.; Chang, L. P. U.S. Patent 6,291,461, 2001.
(2) For a short review concerning the proposed mechanism of dimerization,
see: Nussbaum, F. Angew. Chem., Int. Ed. 2003, 42, 2068-3071.
(5) Herzon, S. B.; Myers, A. G. J. Am. Chem. Soc. 2005, 127, 5342-5344.
(6) For a preliminary account of Myers’ work, see: Myers, A. G.; Herzon, S.
B. J. Am. Chem. Soc. 2003, 125, 12080-12081.
(7) Myers, A. G.; Herzon, S. B.; Wulff, J. E.; Siegrist, R.; Svenda, J.; Zajac,
M. A. WO Patent 2006102097, 2006.
9
6336
J. AM. CHEM. SOC. 2007, 129, 6336-6342
10.1021/ja070259i CCC: $37.00 © 2007 American Chemical Society

Synthesis of Stephacidins A, B and Notoamide B
A R T I C L E S
Figure 1. Stephacidin alkaloids and avrainvillamide.
Scheme 1. Proposed Biosynthesis of Related Alkaloids from (+)-Stephacidin A (1)
diazaoctane ring system.9 In particular, we are astonished by
the diverse series of natural products that appear to be
biosynthetically derived from (+)-stephacidin A (1, Scheme 1).
Oxidation of 1 via a pinacol rearrangement generates the spiro-
oxindole ring system found in the recently isolated (-)-
notoamide A (4) and (-)-notoamide B (5).10,11 A different
oxidation sequence can transform 1 into aspergamide B (6).12
Oxidation of the vinyl imine of aspergamide B (6) yields the
unusual vinyl nitrone moiety of (-)-avrainvillamide (3), which
can dimerize to afford (+)-stephacidin B (2). To further explore
these fascinating biosynthetic possibilities, we require access
to isotopically labeled stephacidin A (1), which must be prepared
through total synthesis. Recently, we were able to achieve a
biomimetic total synthesis of d,l-stephacidin A (1) using a
biogenetically inspired intramolecular Diels-Alder reaction for
the preparation of the bridged [2.2.2] diazaoctane ring system.11b,13
However, we also desired an enantioselective synthesis of 1.
To this end, the SN2′ approach for the formation of the bridged
[2.2.2] diazaoctane ring system previously employed in the
asymmetric total syntheses of brevianamide B14 and paraher-
quamides A15 and B16 was envisioned. However, during our
efforts toward stephacidin A, we have endeavored to employ
several new chemical technologies in order to significantly
condense our approach and render this technology for ongoing
biosynthetic studies. Herein, we wish to detail our recent total
synthesis of the antipodal series, (-)-stephacidin A (1) and
subsequent syntheses of (-)-avrainvillamide (3), (+)-stephacidin
B (2), and (+)-notoamide B (4).
Results and Discussion
In past syntheses employing the SN2′ cyclization strategy,
(R)-allyl proline was prepared using the methodology developed
by Seebach and co-workers.17 Our syntheses commenced with
the condensation of commercially available (S)-proline (7) with
(8) For a full account of experiments conducted by the Baran, laboratories
towards these alkaloids, see: Baran, P. S.; Hafensteiner, B. D.; Ambhaikar,
N. B.; Guerrero, C. A.; Gallagher, J. D. J. Am. Chem. Soc. 2006, 128,
8678-8693.
(13) For preliminary studies towards the synthesis of the stephacidins using a
biomimetic approach, see: Adams, L. A.; Gray, C. R.; Williams, R. M.
Tetrahedron Lett. 2004, 45, 4489-4493.
(9) For reviews of our previous synthetic and biosynthetic work concerning
this class of alkaloids, see: (a) Williams, R. M.; Cox, R. J. Acc. Chem.
Res. 2003, 36, 127-139. (b) Williams, R. M. Chem. Pharm. Bull. 2002,
50, 711-740.
(14) (a) Williams, R. M.; Glinka, T. Tetrahedron Lett. 1986, 27, 3581-3584.
(b) Williams, R. M.; Glinka, T.; Kwast, E. J. Am. Chem. Soc. 1988, 110,
5927-5929. (c) Williams, R. M.; Glinka, T.; Kwast, E.; Coffman, H.; Stille,
J. K. J. Am. Chem. Soc. 1990, 112, 808-821.
(10) Kato, H.; Yoshida, T.; Tokue, T.; Nojiri, Y.; Hirota, H.; Ohta, T.; Williams,
R. M.; Tsukamoto, S. Angew. Chem., Int. Ed. 2006, 46, 2254-2256.
(11) For recent biomimetic total syntheses of notoamides B, C, D and stephacidin
A from our laboratory, see: (a) Grubbs, A. W.; Artman, G. D.; Tsukamoto,
T.; Williams, R. M. Angew. Chem., Int. Ed. 2006, 46, 2257-2261. (b)
Grubbs, A. W.; Greshock, T. J.; Tsukamoto, S.; Williams, R. M. Angew.
Chem., Int. Ed. 2006, 46, 2262-2265.
(15) Williams, R. M.; Cao, J.; Tsujishima, H. J. Am. Chem. Soc. 2003, 125,
12172-12178.
(16) (a) Cushing, T. D.; Sanz-Cervera, J. F.; Williams, R. M. J. Am. Chem.
Soc. 1993, 115, 9323-9324. (b) Cushing, T. D.; Sanz-Cervera, J. F.;
Williams, R. M. J. Am. Chem. Soc. 1996, 118, 557-579.
(17) (a) Seebach, D.; Boes, M.; Naef, R.; Schweizer, W. B. J. Am. Chem. Soc.
1983, 105, 5390. (b) Beck, A. K.; Blank, S.; Job, K.; Seebach, D.;
Sommerfield, T. Org. Synth. 1995, 72, 62.
(12) Fuchser, J. Ph.D. Dissertation, University of Go¨ttingen, 1995.
9
J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007 6337

A R T I C L E S
Artman et al.
Scheme 2. Gram-Scale Synthesis of (R)-Allyl Proline Methyl Ester (11)
pivaldehyde (8) in the presence of TFA under azeotropic
conditions for ∼7-10 days to afford 9 (Scheme 2).17b Following
isolation of 9, allylation can be achieved in a diastereoselective
manner to afford 10 and subsequently the methyl ester 11
following removal of the auxiliary. However, the sensitive nature
of 9 coupled with the cost of pivaldehyde (∼$400/100 mL),
which is required in 7-fold molar excess, has made the synthesis
of 11 using the Seebach protocol less than desirable. Interest-
ingly, Wang and Germanas have reported an alternative to 11
that can be prepared from the inexpensive starting materials of
trichloroacetaldehyde and (S)-proline.18 The trichloro oxazoli-
none 12 is an air- and moisture-stable, commercially available
crystalline solid that can be stored at room temperature with
no decomposition observed after several weeks. In a similar
manner to the Seebach compound 9, alkylation of the oxazoli-
none 12 with allyl bromide using LDA readily affords the allyl
lactone 13 in high yield and as a single diastereoisomer.
Cleavage of the chloral auxiliary from 13 to the amino ester
salt 11 under the reported conditions of refluxing HCl/MeOH
for 1 h only provided <10% of the desired product.18 A search
of the literature revealed that other groups that have employed
this oxazolinone required greater than 24 h of reflux in HCl/
MeOH to obtain modest yields of the desired product.19
Interestingly, cleavage of the auxiliary to the N-formyl methyl
ester using NaOMe is achieved in less than 30 min.18 Recogniz-
ing that the slow step for the cleavage of the oxazolinone 13
under acidic conditions must be the formation of the methyl
ester, we developed a one-pot process to rapidly cleave the
auxiliary and in high yield. Exposure of the allyl lactone 13 to
sodium in methanol followed by the addition of AcCl to the
solution and heating to reflux readily removes the trichloroac-
etaldehyde auxiliary to produce the desired methyl ester
hydrochloride salt 11 in 85% on a 20 g scale.20
the benzyl ether removed by hydrogenation to afford the
intermediate phenol 15 (Scheme 3). Without isolation, alkylation
of the phenol 15 using commercially available 3-chloro-3-
methylbut-1-yne 16 (X ) Cl) in the presence of CuCl2 and DBU
afforded the propargyl ether 17 in 58% yield over the three steps.
Since our original communication, we have found that the yield
of this alkylation can be improved using the methyl carbonate
derivative22 of 16 (X ) OCO2Me) to 92% yield over the three
steps. Aromatic Claisen cyclization of 17 to introduce the pyran
ring can be achieved via two sets of conditions. Under thermal
heating of 17 in o-dichlorobenzene at 180 °C, the pyranoindole
18 can be prepared in 82% yield after ∼2 h. Due to the difficulty
of removing the solvent for the thermal conditions from the
desired product, we have explored alternative conditions to
synthesize 18. To this end, we have taken advantage of
microwave heating in order to facilitate the desired aromatic
Claisen and N-Boc deprotection.23 With the use of a microwave
reactor, the propargyl ether 17 in MeCN was heated at 180 °C
for 20 min to afford the desired product 18 in 95% yield. Use
of microwave heating for this reaction not only greatly reduced
the reaction time but also increased the yield. Furthermore,
removal of the reaction solvent (MeCN) can easily be achieved
at moderate reduced pressure and temperature.
With multiple grams of the pyranoindole 18 rapidly acces-
sible, formation of the tryptophan derivative was explored.
Conversion of 18 to the gramine 19 was conducted under
standard conditions and in high yield. Coupling of the gramine
19 to the commercially available benzophenone imine of glycine
20 under standard Somei-Kametani conditions24 of catalytic
n-Bu3P only afforded 65% yield of the desired protected
tryptophan derivative 21 after 24 h of reflux. Once again, we
found microwave technology to be superior to standard reflux
conditions in that heating of the gramine 19 and 20 with n-Bu3P
in MeCN at 140 °C for 20 min cleanly afforded the coupled
product 21, which after removal of the benzophenone protecting
group with 1 N HCl in THF afforded the amino ester 22 in
90% over the two steps. Chemoselective introduction of the Boc
protecting group onto the primary amine of 22 followed by
The synthesis of the tryptophan derivative was achieved from
the gramine derivative 19, which we previously reported by an
efficient six-step protocol.21 Since our initial communication,
we have been able to improve upon the overall yield of this
key piece through several subtle modifications of the route.
Commencing with commercially available 6-benzyloxyindole
(14), the indole nitrogen was protected with a t-Boc group and
(22) Tisdale, E. J.; Vong, B. G.; Li, H.; Kim, S. H.; Chowdhury, C.; Theodorakis,
E. A. Tetrahedron 2003, 59, 6873-6887.
(23) For reviews concerning microwave-assisted organic synthesis, see: (a)
Lidstrom, P.; Tierney, J.; Wathey, B.; Westman, J. Tetrahedron 2001, 57,
9225-9283. (b) Bose, A. K.; Manhas, M. S.; Ganguly, S. N.; Sharma, A.
H.; Banik, B. K. Synthesis 2002, 1578-1591. (c) Kuhnert, N. Angew.
Chem., Int. Ed. 2002, 41, 1863-1866. (d) The thermal lability of the N-t-
Boc protecting group has documented, see: Krakowiak, K. E.; Bradshaw,
J. S. Synth. Commun. 1996, 26, 3999-4004.
(24) (a) Somei, M.; Karasawa, Y.; Kaneko, C. Heterocycles 1981, 16, 941-
949. (b) Kametani, T.; Kanaya, N.; Ihara, M. J. Chem. Soc., Perkin Trans.
1 1981, 959-963.
(18) Wang, H.; Germanas, J. P. Synlett 1999, 33-36.
(19) (a) Hoffman, T.; Lanig, H.; Waibel, R.; Gmeiner, P. Agnew Chem., Int.
Ed. 2001, 40, 3361-3365. (b) Bittermann, H.; Einsiedel, J.; Hu¨bner, H.;
Gmeiner, P. J. Med. Chem. 2004, 47, 5587-5590. (c) Bittermann, H.;
Gmeiner, P. J. Org. Chem. 2006, 71, 97-102.
(20) Artman, G. D.; Williams, R. M. Org. Synth., submitted for publication
2007.
(21) Grubbs, A. W.; Artman, G. D.; Williams, R. M. Tetrahedron Lett. 2005,
46, 9013-9016.
9
6338 J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007

Synthesis of Stephacidins A, B and Notoamide B
A R T I C L E S
Scheme 3. Synthesis of the Pyranoindole Tryptophan Derivative 23
Scheme 4. Formation of the Diketopiperazine and Cross-Metathesis Results
saponification using LiOH produced the N-Boc acid 23 in
excellent yield over the two steps.
the Boc protecting group employed, we found that heating of
the crude reaction mixture for the coupling product 24 under
microwave heating for 30 min at 150 °C produced the desired
diketopiperazine 25a/b as a 1:1 mixture of diastereomers that
were separable by flash silica gel chromatography in 70% yield;
both diastereomers were carried forward separately. Introduction
of the lactim ether protecting group onto the amides 25a/b using
Coupling of the allyl proline 11 to the tryptophan acid 23
was accomplished using HATU in the presence of DIPEA in
MeCN (Scheme 4). Previously, the Baran group reported a
similar intermediate en route to stephacidin A, wherein ring
closure to the diketopiperazine ring system can be facilitated
under thermal conditions.4a Recognizing the thermal lability of
-
Me3O+BF4 in the presence of Cs2CO3 followed by Boc
9
J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007 6339

A R T I C L E S
Artman et al.
Scheme 5. Formation of the [2.2.2] Bridged Bicycle
alcohol in CH2Cl2 followed by slowly warming to room
temperature and stirring for >12 h in 71% yield over the two
steps. The addition of external chloride sources such as LiCl or
Bu4NCl to help increase the rate of reaction for the formation
of 32 often led to lower yields and additional byproducts.
Cyclization of the allyl chloride 32 under our standard SN2′
conditions for the formation of the [2.2.2] bridged bicycle were
then explored. Exposure of 32 to 20 equiv of NaH in benzene
followed by refluxing for 30 h afforded the desired bridged
bicycle 33 in 60% yield and as a single diastereoisomer, which
presumably arises through a tight ion-pair-driven closed-
transition state.9a However, the length of the reaction time was
less than desirable (∼30 h).
Building off our success in previous steps employing
microwave heating, we attempted our key cyclization under
microwave-assisted conditions. Initial cyclizations of 32 in
benzene at 120 °C for 30 min on a small scale (<75 mg) gave
similar results to the thermal conditions. However, on slightly
larger scales (>150 mg), microwave heating lead to the bursting
of the glass microwave tube and loss of the valuable starting
material and product. Recognizing that part of the so-called
“microwave effect” may be the influence of pressure on the
reaction, we made a second thermal attempt using a sealed tube.
Heating of 32 with NaH in benzene at 130 °C for 9 h readily
produced the desired product 33 in an improved 71% yield with
10% recovered starting allyl chloride 32.
With formation of the [2.2.2] bridged bicycle completed, we
were left with the task of forming the heptacyclic ring system
as well as removing the lactim ether and Boc protecting groups.
Closure to the heptacycle was achieved using a one-pot, two-
step procedure previously developed by Trost and Fortunak29
and employed in the paraherquamides A15 and B16 syntheses.
Exposure of 33 to 5 equiv of Pd(TFA)2 with 100 equiv of
propylene oxide in acetonitrile at room temperature rapidly
forms the alkyl palladium intermediate 34 (Scheme 6). The
reaction mixture is diluted with EtOH, and the alkyl-Pd
intermediate is reduced using NaBH4 to afford 35 in 71% yield.
After extensive investigation, we found that we were not able
to cleave the lactim ether or the Boc protecting group to afford
stephacidin A (1) under a wide range of acidic conditions often
resulting in decomposition or multiple products.30 As an
alternative, we discovered that if the propylene oxide is not
added to the palladium-mediated cyclization of 33 two products
(36/37) can be cleanly formed as an inseparable mixture
following quenching with acid. When 1 N HCl is employed,
the allylic alcohol 36 and the heptacycle lacking the lactim ether
(37) in a 1.6:1 ratio was obtained. The ratio between 36 and 37
can be perturbed by the acid strength employed during the
workup of this reaction. By decreasing the concentration from
1 to 0.5 N HCl, the ratio moves in favor of 37 with a 0.7:1
ratio of 36:37. Further dilution of the acid strength to 0.1 N
HCl results in only the formation of the desired 37. Heating of
the crude product 37 in acetonitrile using the microwave reactor
at 180 °C23 for 15 min afforded (-)-stephacidin A (1) as an
amorphous white powder, which displayed identical spectro-
scopic characteristics to the reported literature values. This last
sequence has been performed several times with good
protection of the indole nitrogen prepares our key substrate 26
for the desired olefin cross-metathesis reaction.25,26 Attempts
to directly convert the terminal olefin of 26 to the requisite allyl
chloride 31 (Y ) -CH2Cl) using Grubbs’ second-generation
catalyst (27) or Hoveyda-Grubbs catalyst 28 and commercially
available 3-choro-2-methyl-2-propene (29) failed to produce the
desired product with most of the starting material 26 being
recovered unchanged.27 Alternatively, cross-metathesis of 26
with methacrolein (30) using catalytic amounts of 27 in refluxing
CH2Cl2 for 24 h readily affords the aldehyde 31 (Y ) CHO) in
65% yield with 15% recovered starting material. Unfortunately,
additional quantities of the catalyst 27 had to be added during
the reaction raising the catalyst loading from the initial 5 to 20
mol %. Switching to the Hoveyda derivative of the Grubbs
second-generation catalyst (28), we were able to reduce the
catalyst loading to 5 mol % and increased the yield for 31 (Y
) CHO) to 70% with 10% recovered 26 after 48 h of reflux.
Once again, we desired to reduce the time required for this
reaction and turned to the microwave to facilitate the heating
of this metathesis reaction.28 In the event, heating of the olefin
26 with methacrolein (30) in the presence of 5 mol % of
28 in CH2Cl2 at 100 °C for 20 min generated the aldehyde 31
(Y ) CHO) in 73% with 10% recovered starting material.
Reduction of the aldehyde 31 using NaBH4 in MeOH afforded
the relatively pure allylic alcohol, which was converted directly
to the allyl chloride 32 (Scheme 5). However, attempts to
perform this relatively simple transformation were often met
with decomposition or low yield using a variety of standard
conditions (MsCl, LiCl, collidine; NCS, Me2S; Ph3P, Cl3-
CCOCCl3). The requisite allyl chloride 32 was finally accessed
by slow addition of MsCl/TEA to a 0 °C solution of the allylic
(25) For a general review on olefin cross-metathesis, see: Chatterjee, A. K.;
Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J. Am. Chem. Soc. 2003, 125,
11360-11370.
(26) For recent total syntheses that employ olefin cross-metathesis, see: (a) Trost,
B. M.; Thiel, O. R.; Tsui, H.-C. J. Am. Chem. Soc. 2003, 125, 13155-
13164. (b) Vyvyan, J. R.; Loitz, C.; Looper, R. E.; Mattingly, C. S.;
Peterson, E. A.; Staben, S. T. J. Org. Chem. 2004, 69, 2461-2468.
(27) Liu, B.; Das, S. K.; Roy, R. Org. Lett. 2002, 4, 2723-2726.
(28) For microwave olefin cross-metathesis, see: Bargiggia, F. C.; Murray, W.
V. J. Org. Chem. 2005, 70, 9636-9639.
(29) Trost, B. M.; Fortunak, J. M. D. Organometallics 1982, 1, 7-13.
(30) Similar results were observed by the Baran group during their synthetic
efforts towards stephacidin A. Please see their full account listed in ref 8
for exact conditions explored.
9
6340 J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007

Synthesis of Stephacidins A, B and Notoamide B
A R T I C L E S
Scheme 6. Completion of the Total Synthesis of (-)-Stephacidin A (1)
Scheme 7. Conversion of Synthetic (-)-1 to (-)-Avrainvillamide
(3), (+)-Stephacidin B (2), and (+)-Notoamide B (4)
in a series of common organic solvents such as CH2Cl2, CHCl3,
THF, Et2O, MeCN, benzene, etc. and is only sparingly soluble
in DMSO. These two factors made it very difficult to determine
if we had actually synthesized the natural product before NMR
analysis after the final step.
The conclusion of the total synthesis of (-)-stephacidin A
(1) has also allowed us to complete the synthesis of (-)-
avrainvillamide (3), (+)-stephacidin B (2), and (+)-notoamide
B (4) from a common intermediate. Following the previously
reported Baran oxidation procedure, 1 was reduced to the 2,3-
dihydroindole 38 using NaCNBH3 in AcOH (Scheme 7).8
Oxidation of the crude dihydroindole 38 using catalytic SeO2
and excess 35% H2O2 for 3 days afforded (-)-avrainvillamide
(3) in 17% yield in addition to 65% recovered starting material.
Synthetic (-)-3 was converted to (+)-stephacidin B (2) using
the Myers’ procedure5 of excess Et3N in CH3CN, which showed
1
95% conversion to the desired alkaloid by H NMR analysis
and HRMS to the previously reported data as well as identity
with an authentic, natural sample kindly provided by Bristol-
Myers Squibb. Finally, synthetic (-)-1 was transformed to (+)-
notoamide B (4) using the procedure established by our group
for the synthesis of d,l-4 by exposure to oxaziridine 39 in 65%
yield.11b
Conclusions
In conclusion, we have completed a very efficient, asymmetric
total synthesis of the fungal metabolite (-)-stephacidin A (1),
in a mere 17 total chemical transformations from commercially
available starting materials and in 6% overall yield. This also
constitutes a formal total synthesis of (+)-stephacidin B
following Baran’s reported conversion of 1 into 3 and then via
Myers’ landmark conversion of 3 into 2 in 19 steps (1% overall
yield). Our synthetic (-)-stephacidin A was subjected to the
Baran-Myers protocol yielding totally synthetic (-)-avrain-
villamide (3) and totally synthetic (+)-stephacidin B (2, 19 steps;
1% overall yield). To place the economy of our synthesis in
context, it is worthwhile comparing the data reported herein to
the Myers’ synthesis of stephacidin B, which was accomplished
in 27 steps from commercially available materials in 0.5%
overall yield and to that of the Baran laboratory, which reported
stephacidin A in 29 steps (1% overall yield) and stephacidin B
reproducibility yielding on the order of 7 mg of synthetic
stephacidin A from 15 mg of 33. This final protocol is a simple
two-step procedure, where compound 37 does not need to be
isolated and purified and the crude material obtained from 33
is subjected to a quick workup and redissolution for the final
N-t-Boc removal.23d
Furthermore, we might note here as an aside, that we were
quite surprised with the physical characteristics of (-)-stepha-
cidin A. First, it has very poor UV activity on TLC and generally
does not develop in numerous TLC stains such as KMnO4,
vanillin, etc., making the product unusually elusive to standard
analytical detection. Second, stephacidin A is quite insoluble
9
J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007 6341

A R T I C L E S
Artman et al.
in 31 total steps and 0.2% overall yield. During the course of
our studies, we have been able to improve upon the synthesis
of (R)-allyl proline using the commercially available oxazoli-
dinone 12 as well as the synthesis of the previously reported
gramine 19. These technologies permit us to install either stable-
or radioisotopes economically into the synthesis which is
currently under study for ongoing biosynthetic investigations
being conducted collaboratively with Professor David Sherman’s
laboratory. Furthermore, our synthesis has extensively taken
advantage of microwave technology, which has reduced reaction
times from hours to minutes as well as increased the yields of
several key transformations. Finally, the synthesis of (-)-
stephacidin A (1) has resulted in the first asymmetric total
syntheses of (+)-notoamide B (4) utilizing an efficient one-
step oxidative pinacol that we recently reported. This work
constitutes the shortest asymmetric route to these four natural
products currently reported in the literature. Current efforts are
being directed to harnessing this technology to prepare numerous
analogs of the stephacidins for biological evaluation and as probe
molecules and provocative biosynthetic intermediates to further
establish the fascinating web of biogenetic relationships within
this family of alkaloids.31,32
Acknowledgment. This paper is dedicated to Professor
William von Eggers Doering of Harvard University on the
occasion of his 90th birthday. The authors would also like to
dedicate this work to Professor Arthur S. Howard on the
occasion of his 70th birthday. We are grateful for the generous
financial support of this work from the National Institutes of
Health (CA70375) and the NSRA Postdoctoral Fellowship for
G.D.A. (GM72296). We are grateful to Bristol-Myers Squibb
for an authentic sample of (-)-stephacidin B and to Pfizer for
an authentic sample of (+)-avrainvillamide. We are further
grateful to Professor Sachiko Tsukamoto of Kanazawa Univer-
sity, Japan, for authenticating our synthetic, racemic samples
of stephacidin A and notoamide B (described previously, ref
11) which were used as authentic standards in the present study
as well as providing us with detailed NMR spectra of these
agents. Mass spectra were obtained on instruments supported
by the NIH Shared Instrument Grant GM49631.
(31) Note added in revision: (a) For a biomimetic synthesis of the related
alkaloid d,l-marcfotine C, see: Greshock, T. J.; Grubbs, A. W.; Williams,
R. M. Tetrahedron 2007, DOI: 10.1016/j.tet.2007.03.016. (b) For a recent
synthesis of d,l-marcfortine B, see: Trost, B. M.; Cramer, N.; Bernsmann,
H. J. Am. Chem. Soc. 2007, 129, 3086-3087.
(32) Note added in proof: Myers, et al., recently reported that stephacidin B
converts to avrainvillamide in cell culture with avrainvillamide being the
biologically active species: Wulff, J. E.; Herzon, S. B.; Siegrist, R.; Myers,
A. G. J. Am. Chem. Soc. 2007, 129, 4898-4899.
Supporting Information Available: Detailed experimental
procedures and spectroscopic data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA070259I
9
6342 J. AM. CHEM. SOC. VOL. 129, NO. 19, 2007