Total synthesis of (±)-alopecuridine and its biomimetic transformation into (±)-sieboldine A

Article abstract of DOI:10.1002/anie.201008147

The features you need: The first total synthesis of lycopodium alkaloid alopecuridine has been achieved in 13 steps in the longest linear sequence, and its biomimetic conversion into sieboldine A has been validated through a two-step oxidation. The synthesis features a semipinacol rearrangement of a functionalized medium-sized ring and an intramolecular pinacol coupling mediated by SmI₂ (see scheme; Boc=tert-butoxycarbonyl).

Full text of DOI:10.1002/anie.201008147

Communications
DOI: 10.1002/anie.201008147
Alkaloid Synthesis
Total Synthesis of (ꢀ )-Alopecuridine and Its Biomimetic
Transformation into (ꢀ )-Sieboldine A**
Xiao-Ming Zhang, Yong-Qiang Tu,* Fu-Min Zhang, Hui Shao, and Xing Meng
Lycopodium alkaloids often possess unusual polycyclic
structures and wide-ranging biological activities.[1] Their
challenging skeletons and potential therapeutic applica-
tions have provoked broad interests in their total
synthesis.[1a,2] Among the lycopodium alkaloids, siebol-
dine A (1) and alopecuridine (2) are two appealing
molecules.[3,4] In particular, sieboldine A inhibits acetyl-
cholinesterase (AChE) significantly (IC50 = 2.0 mm) and
is cytotoxic against murine lymphoma L1210 cells
(IC50 = 5.1 mgmLꢁ1). Both molecules contain two con-
tiguous quarternary stereocenters and sieboldine A Scheme 1. Retrosynthetic analysis. Boc=tert-butoxycarbonyl.
even possesses an unprecedented skeleton with an N-
hydroxyazacyclononane ring bridged to a tetrahydro-
furan ring. Despite their unique structures and significant
biological activity, few reports on their total synthesis have
appeared. Recently, the Overman research group disclosed an
elegant total synthesis of (+)-sieboldine A in 20 steps, in
which an efficient gold(I)-catalyzed cyclization/pinacol
sequence was used to construct the important cis-hydrinda-
none intermediate.[2g] However, the total synthesis of alope-
curidine and its biomimetic conversion into sieboldine A have
not been achieved to date. Herein, we report the first total
synthesis of (ꢀ )-alopecuridine and its biomimetic trans-
formation into (ꢀ )-sieboldine A.
Our retrosynthetic analysis is presented in Scheme 1.
Inspired by Kobayashiꢀs proposed biogenetic pathway,[4] we
expected that a two-step oxidation of alopecuridine (2) would
introduce the N-hydroxy group and construct the tetrahy-
drofuran ring in sieboldine A (1). Alopecuridine may exist in
either a carbinolamine form 2 or an aminoketone form 2’.[3a]
Unlike most of the synthesis performed on fawcettimine-type
alkaloids, our strategy to obtain the tricyclic core of 2’ leaves
the formation of the five-membered B ring until a late stage.
A SmI2-mediated pinacol coupling[5] of compound 4 might
form ring B and simultaneously establish the oxa-quarternary
stereocenter at C4. We further envisioned that the other all-
carbon quarternary center at C12 and the aza-cyclononane
ring could be constructed through a challenging semipinacol
ring expansion[6] of an eight-membered nitrogen-containing
ring from hydroxy epoxide 5. The precursor 5 could be readily
prepared from iodoalkene 6 and carbamate 7 by coupling and
epoxidation.
[*] X.-M. Zhang, Prof. Dr. Y.-Q. Tu, Prof. Dr. F.-M. Zhang, H. Shao,
X. Meng
Department of Chemistry
State Key Laboratory of Applied Organic Chemistry
Lanzhou University, Lanzhou 730000 (China)
Fax: (+86)931-891-5557
E-mail: tuyq@lzu.edu.cn
As depicted in Scheme 2, we began our synthesis by the
preparation of fragment 6. Luche reduction of known
iodide 8[2f,7] afforded cis-allylic alcohol 9 quantitatively
(d.r. > 20:1).[8] After transforming the hydroxy group of 9
into its acetyl ester 10, we attempted to introduce the allyl
group by iodine-catalyzed allylation.[9] However, this method
failed to generate iodoalkene 6; only a small amount of iodo-
substituted products were isolated and large quantities of
starting materials were recovered. To solve this problem, we
attempted to replace the iodine with a Lewis acid. Fortu-
[**] This work was supported by the NSFC (Nos. 20921120404,
20672085, 20732002, and 20972059), the National Basic Research
Program of China “973” program (2010CB833200), the “111”
program of MOE, and the fundamental research funds for the
central universities (lzujbky-2010-k09, lzujbky-2009-76, lzujbky-
2009-158). We thank Prof. Yuan-Jiang Pan of Zhejiang University for
high-resolution mass spectrometry analysis of compounds 9, 10,
and 6.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201008147.
3916
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3916 –3919

steric interaction between the allyl group and eight-mem-
bered ring, the conformational isomers 13aa and 13ba would
be more stable than 13ab and 13bb. As for the conforma-
tional isomers 13ac and 13bc with axial methyl groups, 13ac
would have similar energy to 13aa, but 13bc would be less
stable than the corresponding 13ba. Thus, the hydroxy group
should direct epoxidation to occur in the electron-rich alkene
with syn selectivity to the allyl group. After some attempts, we
found that meta-chloroperoxybenzoic acid (m-CPBA) selec-
tively epoxidized the crude coupling products. Epoxide 5 was
then obtained as an inseparable mixture of isomers, which
were considered to be the 4-methyl diastereoisomers (d.r. =
3.5:1).
With 5 in hand, the key semipinacol reaction was
investigated. Although this type of rearrangement of
medium-sized rings has been reported earlier,[6g] this
method has rarely been applied to a total synthesis using a
complex substrate. By examining a range of Lewis acids (e.g.,
TiCl4, SnCl4, ZnBr2, and BF3·Et2O), we found that BF3·Et2O
Scheme 2. Reagents and conditions: a) NaBH4, CeCl3·7H2O, MeOH,
08C (100%), d.r. >20:1; b) Ac2O, Et3N, DMAP, CH2Cl2, 08C to RT
(97%), d.r.>20:1; c) trimethylallylsilane, BF3·Et2O, DCE, 588C (75%),
trans/cis=3:1; d) N2CHCOOEt, BF3·Et2O, Et2O, ꢁ308C (55%); e) aq
K2CO3, THF, reflux (75%). DMAP=4-dimethylaminopyridine,
THF=tetrahydrofuran.
nately, after some optimization, the reaction was found to
proceed in 1,2-dichloroethane (DCE) at 588C in the presence
of excess BF3·Et2O. Iodoalkene 6 was obtained in 75% yield
as an inseparable 3:1 mixture of diastereoisomers. Subsequent
experiments revealed that the major isomer was the trans-
product.
Table 1: Semipinacol rearrangement of epoxide 5.
Ketone 7 was then prepared from commercially available
azepine 11 through a Tiffeneau–Demjanov-type reaction and
subsequent decarboxylation.[10] Compound 12 was obtained
from the rearrangement as the major isomer in 55% yield.
Hydrolysis and decarboxylation of 12 proceeded well in one
pot to give fragment 7 in 75% yield.
Entry
Solvent
Temperature
Products (yield [%][a])
We next attempted to couple fragments 6 and 7 and
achieve epoxidation in a regio- and stereoselective manner
(Scheme 3). The lithium salt of 6 was first transformed into its
cerium salt.[11] Then 7 was added to afford the coupling
product 13. To avoid elimination, the crude coupling product
13 was directly subjected to the next reaction without
purification. As we assumed in Scheme 3, to minimize the
1
2
3
4
5
6
CH2Cl2
CH2Cl2
THF
THF
Et2O
ꢁ78!ꢁ508C
ꢁ408C
trace amount
14a (20), 14b (10)
trace amount
ꢁ40!08C
RT
14a (11), 14b[b]
trace amount
ꢁ78!ꢁ408C
Et2O
ꢁ30!ꢁ158C
14a (45), 14b (16)
[a] Yield of isolated product. [b] Starting materials and 14b were
obtained as an inseparable mixture.
promoted this reaction in CH2Cl2. Further studies
showed that both solvent and temperature were crucial
to the success of this rearrangement (see Table 1) and
the reaction performed using BF3·Et2O in Et2O at
ꢁ30!ꢁ158C gave the best result (entry 6). The two
desired epimers 14a and 14b were readily separated by
column chromatography on silica gel. The relative
configuration of 14b was also assigned by X-ray
analysis.[12]
Having established the all-carbon quarternary
center and the aza-cyclononane ring, we then focused
on the construction of the B ring and the oxa-
quaternary carbon center at C4 (Scheme 4). Protection
of the hydroxy group with methyl chloromethyl ether
(MOMCl), and subsequent ozonolysis, afforded alde-
hyde 4 in 86% overall yield from 14a. Upon treatment
of 4 with SmI2 (0.1 molLꢁ1 in THF) at 08C, stereose-
lective intramolecular pinacol coupling took place to
give cis-diol 3 as a result of a chelation effect in the
Scheme 3. Reagents and conditions: a) tBuLi, anhydrous CeCl3, then 7, ꢁ788C;
b) m-CPBA, NaHCO3, CH2Cl2, 08C (66% over two steps), d.r.=3.5:1.
formation of ketyl radical 15. The structure of 3 was
Angew. Chem. Int. Ed. 2011, 50, 3916 –3919
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3917

Communications
protecting groups.[14] The resulting crude amine was subjected
to N-Boc protection to furnish 16 in 80% overall yield. The
following oxidation with Dess–Martin periodinane, PCC,
PDC, and CrO3·pyridine reagents led to unexpected cleavage
of the glycol unit. However, use of Ley conditions gave the
desired diketone 17 in moderate yield.[15] Finally, removal of
the Boc group with trifluoroacetic acid (TFA) gave alopecur-
idinium trifluoroacetate 18.[16]
To realize our biomimetic transformation to sieboldine A
(1), we investigated a two-step oxidation based on Kobaya-
shiꢀs proposal (Scheme 5).[4] Alopecuridine·TFA 18 would be
probably oxidized to N-oxide 19 by a peroxide agent. Under
suitable conditions, N-oxide 19 might isomerize to N-hydrox-
ide 20 or eliminate to imine 21, both of which would undergo
further oxidation to give nitrone 22. The final tetrahydrofuran
ring in sieboldine A (1) could be formed by nucleophilic
attack of the hydroxy group to the nitrone. To validate this
hypothesis, we first attempted to oxidize alopecuridine·TFA
18 to N-oxide 19 with m-CPBA in CH2Cl2. This transforma-
tion proceeded easily in the presence of NaHCO3. N-oxide 19
was unstable during purification, so was directly subjected to
the next oxidation step. After screening some solvents, it was
found that N-oxide 19 could be efficiently oxidized to
sieboldine A with HgO in MeOH.[17] The spectroscopic data
(1H and 13C NMR, IR, and HRMS analysis) for synthetic
sieboldine A were identical to those reported for the natural
product.
Scheme 4. Reagents and conditions: a) MOMCl, DIPEA, TBAI, CH2Cl2,
RT; b) O3, CH2Cl2, ꢁ788C; then PPh3, RT (86% over 2 steps);
c) HMPA, SmI2 (0.1 molLꢁ1), THF, 08C (60%); d) 6n HCl, THF,
508C; e) (Boc)2O, Et3N, MeOH, RT (80% over 2 steps); f) TPAP,
NMO·H2O, M.S. (4 ꢀ), CH2Cl2, RT (55%); g) TFA, CH2Cl2, RT, then
NaHCO3, CH2Cl2 (96%). DIPEA=N,N-diisopropylethylamine,
HMPA=hexamethylphosphoramide, M.S.=molecular sieves,
NMO·H2O=N-methylmorpholine-N-oxide monohydrate, TBAI=tetra-
butylammonium iodide, TPAP=tetrapropylammonium perruthenate.
In conclusion, we have achieved the first total synthesis of
(ꢀ )-alopecuridine in 13 steps and a biomimetic synthesis of
(ꢀ )-sieboldine A in 15 steps through a common convergent
route from known iodide 8. Key features of this synthesis
include a semipinacol rearrangement of a functionalized
medium-sized ring and a intramolecular pinacol coupling
mediated by SmI2. The biogenetic pathway from alopecur-
idine to sieboldine A is also validated for the first time.
unambiguously confirmed by X-ray crystallographic analy-
sis.[13] At this point, the key tricyclic core and two contiguous
quarternary carbon atoms of alopecuridine (2) has been
established. The only steps that remained were to remove the
protecting groups and oxidize the secondary alcohols. Our
initial attempts to selectively remove the MOM group failed,
so a two-step procedure was adopted. Thus, compound 3 was
treated with 6m HCl in THF at 508C to remove both
Received: December 23, 2010
Revised: February 14, 2011
Published online: March 22, 2011
Keywords: alkaloids · biomimetic synthesis ·
.
semipinacol rearrangement · total synthesis
[1] For recent reviews on lycopodium alkaloids, see: a) Y.
Hirasawa, J. Kobayashi, H. Morita, Heterocycles 2009,
77, 679; b) J. Kobayashi, H. Morita in The Alkaloids,
Vol. 61 (Eds.: G. A. Cordell), Academic Press, New
York, 2005, p. 1; c) W. A. Ayer, L. S. Trifonov in The
Alkaloids, Vol. 45 (Eds.: G. A. Cordell, A. Brossi),
Academic Press. New York, 1994, p. 233; d) X. Ma,
D. R. Gang, Nat. Prod. Rep. 2004, 21, 752.
[2] For recent reports on the total synthesis of fawcetti-
mine-type lycopodium alkaloids, see: a) J. Ramharter,
H. Weinstabl, J. Mulzer, J. Am. Chem. Soc. 2010, 132,
14338; b) Y. Otsuka, F. Inagaki, C. Mukai, J. Org. Chem.
2010, 75, 3420; c) M. E. Jung, J. J. Chang, Org. Lett.
2010, 12, 2962; d) A. Nakayama, N. Kogure, M.
Kitajima, H. Takayama, Org. Lett. 2009, 11, 5554;
e) J. A. Kozak, G. R. Dake, Angew. Chem. 2008, 120,
4289; Angew. Chem. Int. Ed. 2008, 47, 4221; f) X.
Scheme 5. Reagents and conditions: a) m-CPBA, CH2Cl2, RT; b) HgO, MeOH,
358C (60% over 2 steps).
3918
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 3916 –3919

Linghu, J. J. Kennedy-Smith, F. D. Toste, Angew. Chem. 2007,
119, 7815; Angew. Chem. Int. Ed. 2007, 46, 7671, for the total
synthesis of sieboldine A, see: g) S. M. Canham, D. J. France,
L. E. Overman, J. Am. Chem. Soc. 2010, 132, 7876, for recent
reports on other lycopodium alkaloids, see: h) T. Nishimura,
A. K. Unni, S. Yokoshima, T. Fukuyama, J. Am. Chem. Soc.
2011, 133, 418; i) Y.-R. Yang, Z.-W. Lai, L. Shen, J.-Z. Huang, X.-
D. Wu, J.-L. Yin, K. Wei, Org. Lett. 2010, 12, 3430; j) H. Yang,
R. G. Carter, J. Org. Chem. 2010, 75, 4929; k) K. M. Laemmer-
hold, B. Breit, Angew. Chem. 2010, 122, 2417; Angew. Chem. Int.
Ed. 2010, 49, 2367; l) C. Tsukano, L. Zhao, Y. Takemoto,
Masahiro Hirama, Eur. J. Org. Chem. 2010, 4198; m) D. F.
Fischer, R. Sarpong, J. Am. Chem. Soc. 2010, 132, 5926; n) C.
Yuan, C.-T. Chang, A. Axelrod, D. Siegel, J. Am. Chem. Soc.
2010, 132, 5924; o) V. Bisai, R. Sarpong, Org. Lett. 2010, 12, 2551;
p) B. B. Liau, M. D. Shair, J. Am. Chem. Soc. 2010, 132, 9594;
q) R. A. Altman, B. L. Nilsson, L. E. Overman, J. Read de Ala-
niz, J. M. Rohde, V. Taupin, J. Org. Chem. 2010, 75, 7519; r) X.
Cheng, S. P. Waters, Org. Lett. 2010, 12, 205; s) B. H. Wolfe,
A. H. Libby, R. S. Al-awar, C. J. Foti, D. L. Comins, J. Org.
Chem. 2010, 75, 8564; t) Y. Nakamura, A. M. Burke, S. Kotani,
J. W. Ziller, S. D. Rychnovsky, Org. Lett. 2010, 12, 72.
S. K. Ren, A. S. C. Chan, J. Chem. Soc. Perkin Trans. 1 2000,
3791; g) C. M. Marson, A. Khan, R. A. Porter, A. J. A. Cobb,
Tetrahedron Lett. 2002, 43, 6637.
[7] Racemic 8 could be prepared from crotonaldehyde and ethyl
acetoacetate in two steps, see: a) B.-D. Chong, Y.-I. Ji, S.-S. Oh,
J.-D. Yang, W. Baik, S. Koo, J. Org. Chem. 1997, 62, 9323; b) R.
Benhida, P. Blanchard, J.-L. Fourrey, Tetrahedron Lett. 1998, 39,
6849.
[8] a) J.-L. Luche, J. Am. Chem. Soc. 1978, 100, 2226; b) The
diastereomeric ratio was determined by GC-MS (see the
Supporting Information). The major isomer was the cis product
because of the effect of the methyl group in this reaction.
[9] J. S. Yadav, B. V. S. Reddy, K. V. Rao, K. S. Raj, P. P. Rao, A. R.
Prasad, D. Gunasekar, Tetrahedron Lett. 2004, 45, 6505.
[10] For this type of ring expansion of similar substrates, see: B. T.
Shireman, C. A. Dvorak, D. A. Rudolph, P. Bonaventure, D.
Nepomuceno, L. Dvorak, K. L. Miller, T. W. Lovenberg, N. I.
Carruthers, Bioorg. Med. Chem. Lett. 2008, 18, 2103.
[11] G. A. Molander, Chem. Rev. 1992, 92, 29.
[12] CCDC 804335 (14b) contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[13] CCDC 804336 (3) contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via www.
ccdc.cam.ac.uk/data_request/cif.
[14] Y. Kaburagi, H. Tokuyama, T. Fukuyama, J. Am. Chem. Soc.
2004, 126, 10246.
[15] a) S. V. Ley, J. Norman, W. P. Griffith, S. P. Marsden, Synthesis
1994, 639; b) L. A. Paquette, H.-L. Wang, Z. Su, M. Zhao, J. Am.
Chem. Soc. 1998, 120, 5213.
[16] A sample of natural alopecuridine is no longer available. The
NMR spectroscopic data of alopecuridine has not been reported
to date. Our synthetic alopecuridinium trifluoroacetate 18 has
similar NMR spectroscopic features to fawcettimine hydro-
bromide, see Ref. [2f]. Further confirmation is based on its
transformation into sieboldine A.
[3] a) W. A. Ayer, B. Altenkirk, S. Valverde-Lopez, B. Douglas,
R. F. Raffauf, J. A. Weisbach, Can. J. Chem. 1968, 46, 15;
b) W. A. Ayer, B. Altenkirk, Y. Fukazawa, Tetrahedron 1974, 30,
4213.
[4] Y. Hirasawa, H. Morita, M. Shiro, J. Kobayashi, Org. Lett. 2003,
5, 3991.
[5] For a recent review on reactions mediated by SmI2, see: K. C.
Nicolaou, S. P. Ellery, J. S. Chen, Angew. Chem. 2009, 121, 7276;
Angew. Chem. Int. Ed. 2009, 48, 7140.
[6] For Lewis acid promoted semipinacol rearrangement, see: a) K.
Maruoka, M. Hasegawa, H. Yamamoto, K. Suzuki, M. Shima-
zaki, G. Tsuchihashi, J. Am. Chem. Soc. 1986, 108, 3827; b) K.
Suzuki, M. Miyazawa, G. Tsuchihaski, Tetrahedron Lett. 1987,
28, 3515; c) C. M. Marson, A. J. Walker, J. Pickering, A. D.
Hobson, R. Wrigglesworth, J. Campbell, J. Org. Chem. 1993, 58,
5944; d) C. M. Marson, S. Harper, A. J. Walker, J. Picking, J.
Campbell, Tetrahedron 1993, 49, 10339; e) Y. Q. Tu, L. D. Sun,
P. Z. Wang, J. Org. Chem. 1999, 64, 629; f) Y. Q. Tu, C. A. Fan,
[17] A. Goti, S. Cicchi, V. Fedi, L. Nannelli, A. Brandi, J. Org. Chem.
1997, 62, 3119.
Angew. Chem. Int. Ed. 2011, 50, 3916 –3919
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3919