Total synthesis of gelsemoxonine

Article abstract of DOI:10.1021/ja208617c

The first total synthesis of gelsemoxonine (1) has been accomplished. Divinylcyclopropane-cycloheptadiene rearrangement of the highly functionalized substrate was successfully applied to assemble the spiro-quaternary carbon center connected to the bicyclic seven-membered core structure. A one-pot isomerization reaction of the α,β-unsaturated aldehyde to the saturated ester via the TMSCN-DBU reagent combination allowed a facile diastereoselective introduction of the latent nitrogen functionality of the unique azetidine moiety.

Full text of DOI:10.1021/ja208617c

COMMUNICATION
pubs.acs.org/JACS
Total Synthesis of Gelsemoxonine
Jun Shimokawa, Takaaki Harada, Satoshi Yokoshima, and Tohru Fukuyama*
Graduate School of Pharmaceutical Sciences, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
S Supporting Information
b
Scheme 1. Retrosynthetic Analysis
ABSTRACT: The first total synthesis of gelsemoxonine (1)
has been accomplished. DivinylcyclopropaneÀcycloheptadiene
rearrangement of the highly functionalized substrate was
successfully applied to assemble the spiro-quaternary carbon
center connected to the bicyclic seven-membered core struc-
ture. A one-pot isomerization reaction of the α,β-unsatu-
rated aldehyde to the saturated ester via the TMSCNÀ
DBU reagent combination allowed a facile diastereoselec-
tive introduction of the latent nitrogen functionality of the
unique azetidine moiety.
mong the vast array of monoterpenoid indole alkaloids,
gelsemium alkaloids constitute a relatively large family.
A
These alkaloids have provided a benchmark for state-of-the-art
synthetic strategies due to their highly complex, compact, and
strained structures. Gelsemine, the most well-known member of
this family, has long been a challenging target, and a variety of its
total syntheses have been reported to date.1 A relatively new
member of these attractive gelsemium alkaloids, gelsemoxonine
(1), was isolated in 1991 from the leaves of Gelsemium elegans.2
The structure of gelsemoxonine was originally misassigned, but
was later revised by Aimi in 2003 on the basis of an X-ray
crystallographical analysis.3 The corrected structure features the
unique azetidine substructure bearing a tetra-substituted carbon
center, which is atypical among indole alkaloids. Combined with
the quaternary center of the spiro-N-methoxy indolinone and the
intricate cyclic system, 1 posed a formidable challenge to
synthetic chemistry. Inspired by the unique and complex struc-
ture, along with the intriguing cytotoxic potency of the closely
related natural products,4 we embarked on a synthetic study of 1.
As shown in our retrosynthetic analysis (Scheme 1), formation
of the azetidine moiety and introduction of a three-carbon unit
would be performed in the late stages of the synthesis. The
requisite amine moiety on the carbon skeleton of 2 is to be
introduced via a process involving a Curtius rearrangement from
3. We planned the construction of the oxabicyclo[3.2.2]nonane
skeleton of 3 through a process involving the divinylcyclopro-
paneÀcycloheptadiene rearrangement.5 Substrate 4, bearing
a densely functionalized cyclopropane moiety, was envisioned to
form through an aldol condensation.
lipase AK. To our delight, the enantiopurity of the S-isomer
improved through the use of lipase CR from Candida rugosa.8 7
was reproducibly obtained with greater than 99% ee on a 10-g
scale. With the optically pure material in hand, 7 was treated with
diethyl bromomalonate (8) and DBU. Conjugate addition of the
bromomalonate anion from the less hindered α-face of the
enone, followed by cyclization, delivered cyclopropane 9 as a
single isomer. Chiral HPLC analysis of the product indicated no
signs of base-induced racemization. Therefore, stereochemical
control of the cyclopropane moiety was successfully achieved.
With the construction of the quaternary stereocenter secured, we
next turned our attention to the removal of the acetoxy group.
Despite the fact that 9 was found to be unstable under acidic
conditions, extensive examination of the reaction conditions
eventually led to finding that the combination of Et3SiH and
TMSOTf in acetonitrile effected the efficient reduction of 9 to
10. Notably, the use of acetonitrile was indispensable to suppress
the decomposition of 9.
The densely substituted cyclopropane moiety in 4 was con-
structed from 10 by way of the triol 11. While 10 could be readily
reduced by treatment with LiAlH4, isolation of the resultant triol
11 proved troublesome due to its high chelating ability and water
solubility. After extensive experimentation, we were pleased to
find that the reduction could be achieved by NaBH4 in refluxing
THF with a slow addition of methanol.9 Quenching of the
reaction with acetic acid and repeated evaporationÀaddition
cycles from methanol to remove trimethyl borate gave the mixture
of sodium acetate and the triol 11, which could be purified by silica
gel chromatography after filtration. With access to the desired
Our synthesis began with the m-CPBA-mediated Achmato-
wicz reaction6 of furfuryl alcohol (5) to afford the hydroxy enone
6 (Scheme 2). According to the report by Feringa,7 we employed
lipase AK for the enantioselective preparation of (S)-7 via
dynamic kinetic resolution (DKR), albeit with rather unsatisfac-
tory enantiopurity (74% ee). Solvolysis of the DKR product was
thus examined for the selective removal of the undesired
R-isomer using a lipase that exhibited a selectivity opposite that of
Received: September 13, 2011
Published: October 08, 2011
r
2011 American Chemical Society
17634
dx.doi.org/10.1021/ja208617c J. Am. Chem. Soc. 2011, 133, 17634–17637
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 2. DivinylcyclopropaneÀCycloheptadiene
Scheme 3. Pivotal, Stereoselective Introduction of a Nitro-
gen Atoma
Rearrangementa
a Reagents and conditions: (a) NaH, MeOH, 0 °C to rt, 88%; (b)
TEMPO, PhI(OAc)2, CH2Cl2, rt, 90%; (c) TMSCN, DBU, THF, 0 °C;
allyl alcohol, 0 °C, 78% (21:epi-21 = 4:1); (d) Pd(PPh3)4, pyrrolidine,
MeCN, 0 °C to rt, 96%; (e) (COCl)2, CH2Cl2, rt; (f) NaN3, toluene
H2O, 0 °C; (g) benzyl alcohol, toluene, 80 °C, 82% (3 steps).
3
a Reagents and conditions: (a) m-CPBA, CH2Cl2, 0 °C to rt, 73%; (b)
lipase AK, vinyl acetate, rt, 78%, 74% ee; (c) lipase CR, n-BuOH/n-
hexane, rt, 65%, >99% ee; (d) 8, DBU, THF, 0 °C, 78%; (e) TMSOTf,
Et3SiH, MeCN, 0 °C to rt, 82%; (f) NaBH4 (10 equiv), MeOH (12
equiv, slow addition), THF, reflux; AcOH, 0 °C, 84%; (g) Piv2O,
pyridine, DMAP, CH2Cl2, reflux, 86%; (h) IBX, DMSO, 50 °C, 93%; (i)
13, n-Bu2BOTf, i-Pr2NEt, CH2Cl2, À78 °C; (j) MsCl, TMEDA,
CH2Cl2, À78 °C, 88% (2 steps); (k) TMSCl, THF; LHMDS, À78
°C; (l) toluene, 70 °C, 30 min; TBAF, AcOH, rt, 89% (2 steps).
molecules, manipulation of the carbonÀcarbon double bond on
the bicyclic structure 17 proved extremely difficult. After ex-
tensive studies, we shifted our focus on the redox isomerization
of the unsaturated aldehyde 18 (Scheme 3).13 Thus, solvolysis of
17 followed by TEMPO oxidation provided the α,β-unsaturated
aldehyde 18. Upon treatment of 18 with TMSCN and DBU, the
α,β-unsaturated nitrile 19 arose from the DBU-mediated iso-
merization of the incipient cyanohydrin trimethylsilyl ether.14
Interestingly, ensuing treatment of the reaction mixture with allyl
alcohol resulted in the selective protonation of the silyl enol ether
19 to afford the acyl cyanide 20, which was further transformed
to the corresponding allyl ester 21 (78%) in one pot15 as a 4:1
diastereomeric mixture with the thermodynamically more stable
epi-21.16 The ester 21 thus prepared could be transformed to the
Cbz-protected amine 23 in a three-step sequence involving
deallylation, Curtius rearrangement of the resulting carboxylic acid
22, and treatment with benzyl alcohol. Stereoselective introduction
of the nitrogen atom on the bicyclic skeleton was thus achieved.
For the construction of the azetidine moiety, an epoxide
functionality with the proper stereochemistry must be intro-
duced to the bicyclic structure. This should require a prior
introduction of a 1-propanoyl group at the α-position of the
ketone in 23. Since introduction of a three-carbon unit to 23 met
with failure, we opted to employ a stepwise approach. Fortu-
nately, introduction of a one-carbon unit using Bredereck’s
reagent17 (24) proceeded smoothly to give 25, in spite of the
presence of the nearby carbamate (Scheme 4). The vinylogous
amide 25 was then treated with Vilsmeier’s reagent to give the
β-chloro unsaturated aldehyde 26.18 Dechlorination was subse-
quently achieved by treatment with Pd(PPh3)4 and Et3SiH19 to
furnish the aldehyde 27. Addition of EtMgBr followed by IBX
oxidation afforded the ethyl ketone 28 and provided the com-
plete carbon framework of gelsemoxonine (1). To our delight,
triol 11 assured, the less hindered primary alcohol was selectively
protected as the pivalate. Subsequent oxidation of the resulting
diol with IBX afforded the keto-aldehyde 12 in 93% yield.
Since the conventional Knoevenagel condensation between the
aldehyde 12 and indolinone 1310 proved unsuccessful, a two-step
aldol condensation by means of boron enolate11 was chosen instead.
Gratifyingly, the aldol reaction followed by dehydration, mediated
by MsCl and TMEDA12 at À78 °C, successfully afforded 14 as
a single diastereomer. Subsequent treatment with LHMDS in the
presence of TMSCl furnished the divinylcyclopropane 15, the
substrate for the key divinylcyclopropaneÀcycloheptadiene re-
arrangement. In accordance with our retrosynthetic analysis, the
silyl enol ether 15 rearranged smoothly, upon heating at 70 °C
for 30 min. Removal of the TMS group with TBAF in the
presence of acetic acid furnished the desired bicyclic ketone 17 in
89% yield from 14. The stereochemistry of 17 was consistent
with that predicted from the stereochemistry of the transition
state 16, thereby affording the bicyclic structure with the correct
stereochemistry of the quaternary spirocenter.
With the core seven-membered structure of gelsemoxonine
constructed, the next task was to introduce a nitrogen atom to the
seven-membered carbon skeleton with the desired configuration.
As is often the case for densely functionalized and crowded
17635
dx.doi.org/10.1021/ja208617c |J. Am. Chem. Soc. 2011, 133, 17634–17637

Journal of the American Chemical Society
COMMUNICATION
Scheme 4. End Game of the Total Synthesis of
Gelsemoxoninea
’ ACKNOWLEDGMENT
We thank Prof. Hiromitsu Takayama (Chiba Univ.) for kindly
providing the natural sample of gelsemoxonine. Financial sup-
port for this research was provided by Grants-in-Aid (21790009
and 20002004) from the Ministry of Education, Culture, Sports,
Science and Technology of Japan.
’ REFERENCES
(1) (a) Sheikh, Z.; Steel, R.; Tasker, A.; Johnson, A. J. Chem. Soc.,
Chem. Commun. 1994, 763. (b) Dutton, J.; Steel, R.; Tasker, A.;
Popsavin, V.; Johnson, A. J. Chem. Soc., Chem. Commun. 1994, 765.
(c) Newcombe, N. J.; Ya, F.; Vijn, R. J.; Hiemstra, H.; Speckamp, W. N. J.
Chem. Soc., Chem. Commun. 1994, 767. (d) Speckamp, W. N.; New-
combe, N. J.; Hiemstra, H.; Ya, F.; Vijn, R. J.; Koot, W.-J. Pure Appl.
Chem. 1994, 66, 2163. (e) Dijkink, J.; Cintrat, J.-C.; Speckamp, W. N.;
Hiemstra, H. Tetrahedron Lett. 1999, 40, 5919. (f) Fukuyama, T.; Liu, G.
J. Am. Chem. Soc. 1996, 118, 7426. (g) Fukuyama, T.; Liu, G. Pure Appl.
Chem. 1997, 69, 501. (h) Yokoshima, S.; Tokuyama, H.; Fukuyama, T.
Angew. Chem., Int. Ed. 2000, 39, 4073. (i) Atarashi, S.; Choi, J.-K.; Ha, D.-
C.; Hart, D. J.; Kuzmich, D.; Lee, C.-S.; Ramesh, S.; Wu, S. C. J. Am.
Chem. Soc. 1997, 119, 6226. (j) Madin, A.; O’Donnell, C. J.; Oh, T.; Old,
D. W.; Overman, L. E.; Sharp, M. J. Angew. Chem., Int. Ed. 1999,
38, 2934. (k) Madin, A.; O’Donnell, C. J.; Oh, T.; Old, D. W.; Overman,
L. E.; Sharp, M. J. J. Am. Chem. Soc. 2005, 127, 18054. (l) Earley, W. G.;
Jacobsen, J. E.; Madin, A.; Meier, G. P.; O’Donnell, C. J.; Oh, T.; Old,
D. W.; Overman, L. E.; Sharp, M. J. J. Am. Chem. Soc. 2005, 127, 18046.
(m) Lin, H.; Ng, F.; Danishefsky, S. Tetrahedron Lett. 2002, 43, 549. (n)
Ng, F. W.; Lin, H.; Danishefsky, S. J. J. Am. Chem. Soc. 2002, 124, 9812.
(2) Lin, L.-Z.; Cordell, G. A.; Ni, C.-Z.; Clardy, J. Phytochemistry
1991, 30, 1311.
a Reagents and conditions: (a) Bredereck’s reagent (24), toluene, 70 °C,
97%; (b) (COCl)2, DMF, 70 °C, 80%; (c) Pd(PPh3)4, Et3SiH, Et3N,
DMF, 80 °C, quant.; (d) EtMgBr, THF, À78 °C; (e) IBX, DMSO, 50
°C, 65% (2 steps); (f) TBHP, Triton B, THF, À20 °C, 95%; (g) TMSI,
CH2Cl2, 0 °C; (h) EtOH, reflux, 79% (2 steps).
introduction of the desired epoxide moiety to the α,β-unsatu-
rated ketone 28 was accomplished upon exposure to TBHP and
Triton B at À20 °C. The nucleophile was delivered from the
opposite face to the bulky indolinone and selectively provided
the desired diastereomer 29. Since the NÀO bond of the N-
methoxyindolinone moiety in 29 did not survive the hydroge-
nolysis conditions, deprotection of the Cbz group was effected by
trimethylsilyl iodide20 to give the penultimate intermediate 30.
Despite extensive efforts, all attempts to accomplish the requisite
opening of the epoxide by changing protic and Lewis acids as well
as bases in a variety of solvents failed. It was, therefore, entirely
surprising to find that the ring-opening reaction proceeded when
30 was simply heated in boiling ethanol,21 giving almost exclu-
sively the desired gelsemoxonine (1) on over 300-mg scale.
Spectral data of the synthetic gelsemoxonine were identical to
those of the natural product in all respects.
(3) Kitajima, M.; Kogure, N.; Yamaguchi, K.; Takayama, H.; Aimi,
N. Org. Lett. 2003, 5, 2075.
(4) (a) Kitajima, M.; Nakamura, T.; Kogure, N.; Ogawa, M.; Mitsuno,
Y.; Ono, K.; Yano, S.; Aimi, N.; Takayama, H. J. Nat. Prod. 2006, 69, 715.
(b) Kitajima, M.; Nakamura, T.; Kogure, N.; Ogawa, M.; Mitsuno, Y.; Ono,
K.; Yano, S.; Aimi, N.; Takayama, H. J. Nat. Prod. 2007, 70, 142.
(5) (a) For a review, see: Hudlicky, T.; Fan, R.; Reed, J. W.;
Gadamasetti, K. G. In Organic Reactions; Paquette, L. A., Ed.; John Wiley
& Sons, Inc: New York City, 1992; Vol. 41, p 1; (b) Piers, E.; Jung, G. L.;
Moss, N. Tetrahedron Lett. 1984, 25, 3959. (c) Fleming, A.; Sinai-
Zingde, G.; Natchus, M.; Hudlicky, T. Tetrahedron Lett. 1987, 28, 167.
(6) (a) Achmatowicz, O.; Bukowski, P.; Szechner, B.; Zwierzchows-
ka, Z.; Zamojski, A. Tetrahedron 1971, 27, 1973. (b) Constantinou-
Kokotou, V.; Couladouros, E.; Georgiadis, M.; Kokotos, G.; Georgiadis,
T. Carbohydr. Res. 1991, 222, 163.
(7) van den Heuvel, M.; Cuiper, A. D.; van der Deen, H.; Kellogg,
R. M.; Feringa, B. L. Tetrahedron Lett. 1997, 38, 1655.
(8) Candida rugosa is synonymous with Candida cylindracea men-
tioned in reference 7.
(9) Soai, K.; Oyamada, H.; Takase, M.; Ookawa, A. Bull. Chem. Soc.
Jpn. 1984, 57, 1948.
(10) (a) Kikugawa, Y.; Shimada, M. Chem. Lett. 1987, 1771.
(b) Kikugawa, Y.; Shimada, M.; Matsumoto, K. Heterocycles 1994, 37, 293.
(11) Kende, A. S.; Luzzio, M. J.; Mendoza, J. S. J. Org. Chem. 1990,
55, 918.
In summary, the work described above constitutes the first
total synthesis of gelsemoxonine, featuring a divinylcyclopropaneÀ
cycloheptadiene rearrangement for the construction of a challenging
quaternary center of the spiro indolinone (15f17) and a redox
isomerization via the TMSCNÀDBU combination (18f21) as
two key transformations.
(12) Yoshida, Y.; Shimonishi, K.; Sakakura, Y.; Okada, S.; Aso, N.;
Tanabe, Y. Synthesis 1999, 1633.
’ ASSOCIATED CONTENT
(13) A similar transformation via the NHC-catalyzed reaction is
known: (a) Burstein, C.; Glorius, F. Angew. Chem., Int. Ed. 2004,
43, 6205. (b) Sohn, S. S.; Rosen, E. L.; Bode, J. W. J. Am. Chem. Soc.
2004, 126, 14370. (c) Sohn, S. S.; Bode, J. W. Org. Lett. 2005, 7, 3873.
(d) Chan, A.; Scheidt, K. A. Org. Lett. 2005, 7, 905.
S
Supporting Information. Experimental procedures, cop-
b
ies of spectral data, and characterization data. This material is
available free of charge via the Internet at http://pubs.acs.org.
(14) (a) Kawabata, H.; Hayashi, M. Tetrahedron Lett. 2002, 43, 5645.
(b) Kobayashi, S.; Tsuchiya, Y.; Mukaiyama, T. Chem. Lett. 1991, 537.
(c) Krepski, L. R.; Jensen, K. M.; Heilmann, S. M.; Rasmussen, J. K.;
Lynch, L. E. Synth. Commun. 1986, 16, 617.
’ AUTHOR INFORMATION
Corresponding Author
fukuyama@mol.f.u-tokyo.ac.jp
17636
dx.doi.org/10.1021/ja208617c |J. Am. Chem. Soc. 2011, 133, 17634–17637

Journal of the American Chemical Society
COMMUNICATION
(15) (a) Ziegler, F. E.; Nelson, R. V.; Wang, T.-F. Tetrahedron Lett.
1980, 21, 2125. (b) Jacobson, R. M.; Lahm, G. P.; Clader, J. W. J. Org.
Chem. 1980, 45, 395. (c) H€unig, S.; Reichelt, H. Chem. Ber. 1986,
119, 1772. (d) Fischer, K.; H€unig, S. Chem. Ber. 1986, 119, 2590.
(16) Treatment with other reagents such as H2O, MeOH, t-BuOH,
or NH3 aq resulted in low diastereoselectivity, probably because of the
unselective protonation of 19 or the epimerization of the acyl cyanide
20.
(17) (a) Bredereck, H.; Simchen, G.; Rebsdat, S.; Willi, K.; Horn, P.;
Wahl, R.; Hoffmann, H.; Peter, G. Chem. Ber. 1968, 101, 41.
(b) Wasserman, H. H.; Ives, J. L. J. Org. Chem. 1985, 50, 3573.
ꢀ
(18) (a) Arnold, Z.; Zemliꢀcka, J. Collect. Czech. Chem. Commun.
ꢀ
1959, 24, 2378. (b) Arnold, Z.; Zemliꢀcka, J. Collect. Czech. Chem.
Commun. 1959, 24, 2385.
(19) (a) Scott, W. J.; Stille, J. K. J. Am. Chem. Soc. 1986, 108, 3033.
(b) Su, Y.; Xu, Y.; Han, J.; Zheng, J.; Qi, J.; Jiang, T.; Pan, X.; She, X. J.
Org. Chem. 2009, 74, 2743.
(20) (a) Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem. Commun.
1978, 315. (b) Olah, G. A.; Narang, S. C.; Gupta, B. G. B.; Malhotra, R.
Synthesis 1979, 61.
(21) For ring-opening reactions of epoxides in protic solvents, see:
(a) Azizi, N.; Saidi, M. R. Org. Lett. 2005, 7, 3649. (b) Wang, Z.; Cui, Y.-
T.; Xu, Z.-B.; Qu, J. J. Org. Chem. 2008, 73, 2270. (c) Vilotijevic, I.;
Jamison, T. F. Science 2007, 317, 1189.
17637
dx.doi.org/10.1021/ja208617c |J. Am. Chem. Soc. 2011, 133, 17634–17637