6-Exo-spiro (alkoxycarbonylamino)methyl radical cyclization: highly regio- and stereoselective synthesis of (-)-sibirine.

Article abstract of DOI:10.1021/ol026671e

[reaction: see text] The (methoxycarbonylamino)methyl radical can be readily generated from its PhSe precursor and undergoes preferential 6-exo-spiro cyclization when PhSO(2) is attached at the distal alkene carbon. This property was applied to the synthe

Full text of DOI:10.1021/ol026671e

ORGANIC
LETTERS
2002
Vol. 4, No. 19
3329-3332
6-Exo-spiro
(Alkoxycarbonylamino)methyl Radical
Cyclization: Highly Regio- and
Stereoselective Synthesis of (−)-Sibirine
Masato Koreeda,*,†,‡ Yamin Wang,† and Liming Zhang‡
Departments of Chemistry and Medicinal Chemistry, UniVersity of Michigan,
Ann Arbor, Michigan 48109-1055
koreeda@umich.edu
Received August 5, 2002
ABSTRACT
The (methoxycarbonylamino)methyl radical can be readily generated from its PhSe precursor and undergoes preferential 6-exo-spiro cyclization
when PhSO is attached at the distal alkene carbon. This property was applied to the synthesis of the racemic and optically active spirocyclic
2
alkaloid sibirine.
Regio- and stereocontrolled generation of quaternary carbon
centers continues to remain a formidable challenge in organic
synthesis.1 Among a number of such methods currently
available, those based on radical cyclization are often quite
effective due to the exceptional predilection of most of the
reactive radical species to form bonds at the proximately
disposed unsaturated center.2 In connection with our con-
tinued interest in developing synthetic methods through the
use of radicals next to a heteroatom such as 2, the synthesis
of the spirocyclic alkaloid (-)-sibirine (3)3 was investigated.
† Department of Chemistry.
‡ Department of Medicinal Chemistry.
(1) For reviews on the synthesis of quaternary carbon centers, see: (a)
Martin, S. F. Tetrahedron 1980, 36, 419-460. (b) Fuji, K. Chem. ReV.
1993, 93, 2037-2066. For reviews on the synthesis of spiro systems, see:
(c) Corey, E. J. Angew. Chem., Ed. Engl. 1998, 53, 6031-6035. (d)
Sannigrahi, M. Tetrahedron 1999, 55, 9007-9071.
(2) For some leading references, see: (a) Set, L.; Cheshire, D. R.; Clive,
D. L. J. J. Chem. Soc., Chem. Commun. 1985, 1205-1207. (b) Bennett, S.
M.; Clive, D. L. J. J. Chem. Soc., Chem. Commun. 1986, 878-880. (c)
Harling, J. D.; Motherwell, W. B. J. Chem. Soc., Chem. Commun. 1988,
1380-1382. (d) Hart, D. J.; Wu, S. C. Tetrahedron Lett. 1991, 32, 4099-
4102. (e) Cossy, J.; Bouzide, A.; Leblanc, C. Synlett 1993, 202-204. (f)
Srikrishna, A.; Nagaraju, S.; Raghava Sharma, G. V. J. Chem. Soc., Chem.
Commun. 1993, 285-288. (g) Back, T. G.; Gladstone, P. L.; Parvez, M. J.
Org. Chem. 1996, 61, 3806-3814. For a review, see: (h) pp 9062-9066
in ref 1d.
Notwithstanding a substantial literature on a radical R to
a nitrogen atom,4 its application in synthesis is by no means
straightforward. Although the generation of the radical 2 from
a precursor such as 1 could readily be achieved due to the
(3) (a) Ibragimov, A. A.; Osmanov, Z.; Tashkhodzhaev, B.; Abdullaev,
N. D.; Yagudaev, M. R.; Yunusov, S. Y. Chem. Nat. Compd. 1981, 17,
458-463. (b) Tashkhodzhaev, G. Chem. Nat. Compd. 1982, 18, 70-74.
(c) Osmanov, Z.; Ibragimov, A. A.; Yunusov, S. Y. Chem. Nat. Compd.
1982, 18, 206-208. (d) Omsmanov, Z.; Ibragimov, A. A.; Yunusov, S. Y.
Chem. Nat. Compd. 1982, 18, 206-208. (e) Ibragimov, A. A.; Yunusov,
S. Y. Chem. Nat. Compd. 1988, 24, 71-78.
(4) Review: Renaud, P.; Giraud, L. Synthesis 1996, 913-926.
10.1021/ol026671e CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/22/2002

stabilizing interaction between the incipient radical and the
N lone pair,5 this stability seems to adversely influence the
reactivity of the ensuing radical reaction.4,6 It appears that
for a successful outcome to a radical cyclization reaction,
the rates of these two processes need to be fine-tuned through
highly system-dependent choices of groups Y and Z. A
literature survey seemed to point to the notion that in general,
a radical having an electron-withdrawing group attached to
the nitrogen atom is effective in both.4,6 In this context, a
carbamate7 was chosen as Z (1) in our study since it can
readily be converted to either CH3 or H. The results of our
preliminary study indicated that while the generation of
radical 2 from the sulfide precursor [1: Y ) SPh; Z ) C(d
O)OMe] is sluggish, that of the corresponding selenide8 [1:
Y ) SePh; Z ) C(dO)OMe] is in a fine balance with the
subsequent radical cyclization reaction in a number of
systems we examined.9
Scheme 1
study was the regio- and stereochemical outcome of the
cyclization of the radical intermediate 6 to be generated from
4. The transition state for the 6-exo mode of cyclization with
the radical center approaching from the opposite face of the
axially disposed OR group appeared clearly less strained over
that of the 7-endo. However, the effect of a group X on the
extent of selectivity between the two modes of cyclization
was of particular interest.
On the basis of these results, the study toward the synthesis
of (-)-sibirine10 was initiated predicated upon the retrosyn-
thesis shown in Scheme 1. Of particular significance in this
(5) (a) Griller, D.; Howard, J. A.; Marriott, P. R.; Scaiano, J. C. J. Am.
Chem. Soc. 1981, 103, 619-623. (b) Burkey, T. J.; Castelhano, A. L.;
Griller, D.; Lossing, F. P. J. Am. Chem. Soc. 1983, 105, 4701-4703. (c)
Pasto, D. J.; Krasnansky, R.; Zercher, C. J. Org. Chem. 1987, 52, 3062-
3072. (d) Coolidge, M. B.; Borden, W. T. J. Am. Chem. Soc. 1988, 110,
2298-2299.
(6) For insightful discussions on this point, see: (a) Padwa, A.;
Nimmesgern, H.; Wong, G. S. K. J. Org. Chem. 1985, 50, 5620-5627. (b)
Katritzky, A. R.; Feng, D.; Qi, M.; Aurrecoechea, J. M.; Suero, R.;
Aurrekoetxea, N. J. Org. Chem. 1999, 64, 3335-3338.
(7) For the generation of a radical R to a cyclic carbamate, see: Kano,
S.; Yuasa, K.; Asami, K.; Shibuya, S. Heterocycles 1988, 27, 1437-1447.
(8) For the use of the PhSe group for the generation of a radical R to a
lactam nitrogen, see: (a) Bachi, M. D.; Hoornaert, C. Tetrahedron Lett.
1981, 2693-2694. (b) Bachi, M.; Frolow, F.; Hoornaert, C. J. Org. Chem.
1983, 48, 1841-1849. (c) Burnett, D. A.; Choi, J.-K.; Hart, D. J.; Tsai,
Y.-M. J. Am. Chem. Soc. 1984, 106, 8201-8209. (d) Kametani, T.; Chu,
S.-D.; Itoh, A.; Maeda, S.; Honda, T. J. Org. Chem. 1988, 53, 2683-2687.
(e) Dener, J. M.; Hart, D. J.; Ramesh, S. J. Org. Chem. 1988, 53, 6022-
6030.
The synthesis of the radical precursor 4 (R ) TBDMS)
with X ) H is summarized in Scheme 2. Thus, the
introduction of the carbamate-protected side chain amine 11
Scheme 2a
(9) Wang, Y.; Zhang, L.; Koreeda, M. To be published.
(10) For the synthesis of sibirine and other Nitraria alkaloids, see: (a)
Snider, B. B.; Cartaya-Marin, C. J. Org. Chem. 1984, 49, 1688-1691. (b)
Kozikowski, A. P.; Yuen, P.-W. J. Chem. Soc., Chem. Commun. 1985, 847-
848. (c) Hellberg, L. H.; Beeson, C.; Somanathan, R. Tetrahedron Lett.
1986, 27, 3955-3956. (d) McCloskey, P. J.; Schultz, A. G. Heterocycles
1987, 25, 437. (e) Carruthers, W.; Moses, R. C. J. Chem. Soc., Chem.
Commun. 1987, 509-510. (f) Quirion, J.-C.; Grierson, D. S.; Royer, J.;
Husson, H.-P. Tetrahedron Lett. 1988, 29, 3311-3314. (g) Tanner, D.;
Ming, H. H.; Bergdahl, M. Tetrahedron Lett. 1988, 29, 6493-6496. (h)
Wanner, M. J.; Koomen, G. J. Tetrahedron Lett. 1989, 30, 2301-2304. (i)
Tanner, D.; He, H. M. Tetrahedron 1989, 45, 4309-4316. (j) Kim, D.;
Kim, H. S.; Yoo, J. Y. Tetrahedron Lett. 1991, 32, 1577-1578. (k) Imanishi,
T.; Kurumada, T.; Maezaki, N.; Sugiyama, K.; Iwata, C. J. Chem. Soc.,
Chem. Commun. 1991, 1409-1411. (l) Belvisi, L.; Gennari, C.; Poli, G.;
Scolastico, C.; Salom, B.; Vassallo, M. Tetrahedron 1992, 48, 3945-3960.
(m) Fujii, M.; Kawaguchi, K.; Nakamura, K.; Ohno, A. Chem. Lett. 1992,
1493-1496. (n) Westermann, B.; Scharmann, H. G.; Kortmann, I.
Tetrahedron: Asymmetry 1993, 4, 2119-2122. (o) Keppens, M.; De Kimpe,
N. Synlett 1994, 285-286. (p) Keppens, M.; De Kimpe, N. J. Org. Chem.
1995, 60, 3916-3918. (q) Yamane, T.; Ogasawara, K. Synlett 1996, 925-
926. (r) Kim, D.; Choi, W. J.; Hong, J. Y.; Park, I. Y.; Kim, Y. B.
Tetrahedron Lett. 1996, 37, 1433-1434. (s) Wanner, M. J.; Koomen, G. J.
Pure Appl. Chem. 1996, 68, 2051-2056. (t) Trost, B. M.; Radinov, R.;
Grenzer, E. M. J. Am. Chem. Soc. 1997, 119, 7879-7880. (u) Franc¸ois,
D.; Lallemand, M.-C.; Selkti, M.; Tomas, A.; Kunesch, N.; Husson, H.-P.
Angew. Chem., Int. Ed. 1998, 37, 104-105. Reviews: (v) Imanishi, T.;
Iwata, C. In Studies in Natural Products Chemistry: StereoselectiVe
Synthesis; Atta-ur-Rahman, Ed.; Elsevier: Amsterdam, 1994; Vol 14, Part
I, pp 517-550. (w) Wanner, M. J.; Koomen, G. J. In Studies in Natural
Products Chemistry: StereoselectiVe Synthesis; Atta-ur-Rahman, Ed.;
Elsevier: Amsterdam, 1994; Vol 14, Part I, pp 731-768. (x) Husson, H.-
P.; Royer, J. Chem. Soc. ReV. 1999, 28, 383-394.
a Reagents and conditions: (a) 9-BBN (1.5 mol equiv)/THF, rt,
3 h; (b) Pd(PPh3)4 (3 mol %), NaOH/THF/H2O, reflux, 1.5 h; (c)
NaH (1.2 mol equiv)/THF, rt, 15 min; then ICH2Sn(n-Bu)3 (1.06
mol equiv), rt, 12 h (85%); (d) n-BuLi (1.2 mol equiv)/THF, -78
°C; (PhSe)2 (1.02 mol equiv)/THF, -78 °C, 0.5 h (49%); (e) (n-
Bu)3SnH (1.5 mol equiv), AIBN (cat.)/toluene, reflux, 13 h.
3330
Org. Lett., Vol. 4, No. 19, 2002

was smoothly achieved by the Suzuki coupling reaction of
the tert-butyldimethylsilyl (TBS)-protected bromide 9 with
the borane 1011 generated from allylamine methyl carbamate.
Carbamate 11 was first converted to its tri(n-butyl)stannyl-
methyl derivative and transmetalation of the stannane with
n-BuLi followed by treatment with diphenyl diselenide to
produce phenyl selenide 12 in an unoptimized overall yield
of 42%. Subjection of phenyl selenide 12 to radical reaction
conditions resulted in the effective generation of the expected
radical, which underwent cyclization cleanly with only a
small amount of the reduction product (i.e., H in place of
PhSe in 12) detected. However, the cyclized products
consisted of an inseparable mixture of spirobicycle 13 (one
stereoisomer) and bicyclic sytem 14 (as a 2:1 stereoisomeric
mixture),12 indicating only a meager difference in transition
state energies between the 6-exo- and 7-endo modes of
radical cyclization processes.
Scheme 3a
In an effort to increase the formation of the spirocyclization
product (i.e., pathway a) from radical 6, it was contemplated
that the use of a sterically demanding group for X might
impede the competing 7-endo mode of cyclization. In this
context, a phenylsufonyl group was selected because, in
addition to its steric size, the radical 7 to be generated as a
result of the 6-endo cyclization could be stabilized by this
group.13 The synthesis of the sulfone-containing radical
precursor 19 that commences with 3-phenylthio-2-cyclo-
hexen-1-one (15)14 is outlined in Scheme 3. While the
carbamate-protected side chain could readily be attached by
the Suzuki coupling reaction as above, the introduction of
the PhSeCH2 unit by the two-step sequence [(i) NaH; ICH2-
Sn(n-Bu)3 (43%). (ii) n-BuLi/THF, -78 °C; (PhSe)2 (5%)]
proved to be ineffective. The problem was circumvented by
first oxidizing the PhS to the PhSO2 group (see 17 to 18).
Interestingly, the presence of the sulfone group on the
cyclohexenyl ring appears to exert a considerable influence
in making the NH readily accessible for deprotonation. Thus,
sulfone 18 was converted to the selenide radical precursor
19 via its N-hydroxylmethyl derivative in 53% overall yield.15
Treatment of phenyl selenide 19 with tri(n-butyl)tin
hydride in the presence of a catalytic amount of AIBN in
refluxing toluene provided spirocycle 20 in 60% yield
together with a 1:1 inseparable mixture of 21, the sufone
epimer of 20, and the reduction product 22 (15%).16 The
a Reagents and conditions: (a) NBS (1.5 mol equiv)/CCl4, rt,
24 h (81%); (b) NaBH4 (1.1 mol equiv), CeCl3 (1.1 mol equiv)/
methanol, rt, 0.5 h (97%); (c) TBS-Cl, imidazole, DMF, rt (95%);
(d) 10, Pd(PPh3)4 (3 mol %), NaOH/THF, H2O, reflux, 3 h; (e)
MCPAB (2.4 equiv)/CH2Cl2, rt, 2.5 h; (f) t-BuOK (2.0 mol equiv),
paraformaldehyde (excess)/t-BuOH, rt, 1 h (75%); (g) PhSeH
(excess), p-TsOH (cat.), rt, 2 h (71%); (h) (n-Bu)3SnH (2.0 mol
equiv), AIBN (cat.), toluene, reflux, 7 h; (i) Na (20 mol equiv)/
ethanol (20 mol equiv), from -20 °C to rt, 2 h at rt (92%); (j)
LiAlH4 (10 mol equiv)/THF, rt, 30 min; (k) CH3CN/45% aq HF
(20:1), rt, 1.5 h (85% for steps j and k).
inspection of the H1 NMR spectrum in CDCl3 revealed that
the OTBS and PhSO2 groups in the major product 20 adopt,
respectively, equatorial and axial orientations. The selective
formation of the axial sulfone seems to suggest that the
initially cyclized radical undergoes ring inversion to adopt
a conformation such as 24 and that the observed outcome
may be a reflection of the preferential approach by the (n-
Bu)3SnH from the equatorial face (see 24).
(11) (a) Brosius, A. D.; Overman, L. E. J. Org. Chem. 1997, 62, 440-
441. (b) Kamatani, A.; Overman, L. E. J. Org. Chem. 1999, 64, 8743-
8744. Reviews: (c) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-
2483. (d) Suzuki, A. In Metal-Catalyzed Cross-Coupling Reactions;
Diederich, F., Stang, P. J., Eds.; Wiley-VCH Verlag: Weinheim, Germany,
1998; pp 44-97.
(12) Due to the presence of the rotational isomers of the carbamate group
in each of the three products, analysis of the 1H NMR spectrum of the
mixture of 13 and 14 was severely hampered. Therefore, the mixture was
reduced with LiAlH4 and the resulting, still inseparable mixture of the
N-methyl compounds was analyzed.
(13) For the use of the phenylsulfonyl group as a radical acceptor, see,
e.g.: (a) Clive, D. L. J.; Bergstra, R. J. J. Org. Chem. 1990, 55, 1786-
1792. (b) Simpkins, N. S. Chem. Soc. ReV. 1990, 19, 335-354.
(14) Danishefsky, S.; Harayama, T.; Singh, R. K. J. Am. Chem. Soc.
1979, 101, 7008-7012.
The combined yield of the desired spirocycle alkaloids
20 and 21 approached close to 70%. However, due to the
difficulty of removing the contaminant 22 from the mixture
of 21 and 22, the single stereoisomer 20 was subjected to
(15) For some unknown reason, the two-step sequence toward 19 via
the tri(n-butyl)stannylmethyl derivative of 18 proved to be much less
efficient.
(16) In addition, the dealkylated product 18 was also isolated (<10%).
Org. Lett., Vol. 4, No. 19, 2002
3331

the subsequent three-step sequence (78% overall yield) to
complete the synthesis of racemic sibirine (3).
The asymmetric synthesis of (-)-sibirine (3) was achieved
through the use of the chiral allylic alcohol 27 (Scheme 4).
of this alcohol with (S)-O-acetylmandelic acid. Single
recrystallization of the resulting esters 27 from benzene
provided ester 27 with a diastereomeric ratio over 99:1.
Removal of the mandelate portion from 27 proved to be
somewhat problematic. While the use of DIBAL, LiAlH4,
or nucleophilic alkaline conditions resulted in the formation
of a complex mixture of products, that of LiBH4 in THF led
to the clean removal of the mandelate unit, providing allylic
alcohol 26 with >98% ee. The stereochemical integrity of
the resulting allylic alcohol thus obtained was ascertained
by conversion to its (S)-O-acetylmandelate and the 400 MHz
1H NMR analysis of the ester. The enantiomerically pure
alcohol was then converted to its TBS ether 16, and the entire
sequence developed for the racemic series was repeated, thus
achieving the synthesis of (-)-sibirine; [R]23D -29.7 (c 0.63,
CHCl3) [lit.3d [R]D -22.5 (c 0.81, CHCl3)]. While the
synthesis of racemic sibirine (3) was achieved in 11 steps
from 2-phenylthio-2-cyclohexen-1-one (15) in 15.2% overall
yield, that of (-)-sibirine (3) required 14 steps from 15 with
10.9% overall yield.
Scheme 4
In summary, we have shown that a (methoxycarbonylami-
no)methyl radical can readily be generated from the phen-
ylselanyl precursor and the radical undergoes highly 6-exo
stereo- and regioselective spirocyclization to a cyclohexene
unit bearing the sterically demanding and radical stabilizing
phenylsulfonyl group at the distal alkene carbon. This
observation has been applied to the effective synthesis of
racemic and (-)-sibirine.
As in the case of 2-bromo-2-cyclohexen-1-one,17 the CBS-
oxazaborolidine-mediated reduction18 of its 3-PhS analogue,
15, required an ambient temperature for the reaction to
proceed. The enantiomeric excess of the resulting allylic
alcohol 26 was found to be only 88%. Enantiomerically pure
26 was accessed by the formation of diastereomeric esters
Acknowledgment. The work described herein was sup-
ported in part by a grant from the NIH (DK 30025). L.Z.
was a Pfizer Fellowship recipient for 2000-2001 awarded
through the Department of Medicinal Chemistry, University
of Michigan.
Supporting Information Available: Experimental details
as well as analytical and spectroscopic data for new
compounds described. This material is available free of
charge via the Internet at http://pubs.acs.org.
(17) Corey, E. J.; Bakshi, R. K.; Shibata, S.; Chen, C.-P.; Singh, V. K.
J. Am. Chem. Soc. 1987, 109, 7925-7926.
(18) Reviews: (a) Deloux, L.; Srebnik, M. Chem. ReV. 1993, 93, 763-
784. (b) Itsuno, S. In ComprehensiVe Asymmetric Catalysis; Jacobsen, E.
N., Pfaltz, A., Yamamoto, H., Eds.; Springer-Verlag: Berlin, Germany,
1999; Vol. I, pp 289-315.
OL026671E
3332
Org. Lett., Vol. 4, No. 19, 2002