Synthesis of (-)-sedinine by allene cyclization and iminium Ion chemistry

Article abstract of DOI:10.1021/ol1016492

A synthesis of the sedum alkaloid sedinine has been achieved employing silver(I)-catalyzed allenic hydroxylamine cyclization and ring-closing metathesis to form a bicyclic N,O-acetal. Ring opening of this acetal with a silyl enol ether under Lewis acidic conditions is exclusively trans selective, leading to the natural product after reduction. On the other hand, conversion of the bicyclic N,O-acetal to a semicyclic N,O-acetal results in no stereoselectivity during such a reaction. The contrasting results can be rationalized by consideration of the conformation of the iminium ions.

Full text of DOI:10.1021/ol1016492

ORGANIC
LETTERS
2010
Vol. 12, No. 17
3938-3941
Synthesis of (-)-Sedinine by Allene
Cyclization and Iminium Ion Chemistry
Roderick W. Bates* and Yongna Lu
DiVision of Chemistry and Biological Chemistry, School of Physical and Mathematical
Sciences, Nanyang Technological UniVersity, 21 Nanyang Link, Singapore 637371
roderick@ntu.edu.sg
Received July 15, 2010
ABSTRACT
A synthesis of the sedum alkaloid sedinine has been achieved employing silver(I)-catalyzed allenic hydroxylamine cyclization and ring-closing metathesis
to form a bicyclic N,O-acetal. Ring opening of this acetal with a silyl enol ether under Lewis acidic conditions is exclusively trans selective, leading
to the natural product after reduction. On the other hand, conversion of the bicyclic N,O-acetal to a semicyclic N,O-acetal results in no stereoselectivity
during such a reaction. The contrasting results can be rationalized by consideration of the conformation of the iminium ions.
The sedum alkaloids have held the interest of synthetic chemists
over many decades.1-3 In addition to some interesting
biological activity, they offer a range of targets for the
application of synthetic methodology. The sedum alkaloids
may be considered to be either “one-armed”, having a 2′-
hydroxyalkyl substituent at the 2-position of a piperidine,
or “two-armed”, having an additional oxygenated substituent
at the 6-position. Sedinine 1 is a “two-armed” sedum alkaloid
with the two substituents trans to each other. Sedinine also
possesses a double bond within the heterocycle. It was
originally reported by Franck,4 and the structure was revised
by Hootele´ on the basis of X-ray analysis of both the alkaloid
and its hydrochloride.5
Retrosynthetically, the sedum alkaloids may be considered
to be 1,3-amino alcohols. Recently, we reported syntheses of
sedamine 22a and porantheridine 36 (Figure 1) employing
(1) For a review, see: Bates, R. W.; Sa-Ei, K. Tetrahedron 2002, 58,
5957
.
(2) For some recent contributions, see: (a) Bates, R. W.; Nemeth, J.;
Snell, R. Synthesis 2008, 1033, and references therein. (b) Lebrun, S.;
Couture, A.; Deniau, E.; Grandclaudon, P. Chirality 2010, 22, 212. (c)
Davies, S. G.; Fletcher, A. M.; Roberts, P. M.; Smith, A. D. Tetrahedron
2009, 65, 10192. (e) Spangenberg, T.; Airiau, E.; Thuong, M. B. T.;
Dommard, M.; Billet, M.; Mann, A. Synlett 2008, 2859. (f) Chen, L.-J.;
Hou, D.-R. Tetrahedron: Asymmetry 2008, 19, 715. (g) Krishnan, S.;
Bagdanoff, J. T.; Ebner, D. C.; Ramtohul, Y. K.; Tambar, U. K.; Stoltz,
B. M. J. Am. Chem. Soc. 2008, 130, 13745. (h) Wee, A. G. H.; Fan, G.-J.
Org. Lett. 2008, 10, 3869. (i) Bisai, A.; Singh, V. K. Tetrahedron Lett.
2007, 48, 1907. (j) Nagata, K.; Nishimura, K.; Yokoya, M.; Itoh, T.
Heterocycles 2006, 70, 335. (k) Passarella, D.; Barilli, A.; Belinghieri, F.;
Fassi, P.; Riva, S.; Sacchetti, A.; Silvani, A.; Danieli, B. Tetrahedron:
Asymmetry 2005, 16, 2225. (l) Lesma, G.; Crippa, S.; Danieli, B.; Passarella,
D.; Sacchetti, A.; Silvani, A.; Virdis, A. Tetrahedron 2004, 60, 6437. (m)
Kang, B.; Chang, S. Tetrahedron 2004, 60, 7353. (n) Angoli, M.; Barilli,
A.; Lesma, G.; Passarella, D.; Riva, S.; Silvani, A.; Danieli, B. J. Org.
Chem. 2003, 68, 9525. (o) Kochi, T.; Tang, T. P.; Ellman, J. A. J. Am.
Chem. Soc. 2003, 125, 11276. (p) Sugiura, M.; Hagio, H.; Hirabayashi, R.;
Figure 1. Sedinine, sedamine, and porantheridine.
silver(I)-catalyzed allenic hydroxylamine cyclization.7 We now
wish to report the use of this chemistry, as well as ring-closing
metathesis and iminium ion trapping,8 for the synthesis of
sedinine.
Kobayashi, S. J. Am. Chem. Soc. 2001, 123, 12510
.
(3) For reviews on the synthesis of piperidines, see: Cossy, J. Chem.
Rec. 2005, 5, 70. Buffat, M. G. P. Tetrahedron 2004, 60, 1701. Laschat,
S.; Dickner, T. Synthesis 2000, 1781. Weintraub, P. M.; Sabol, J. S.; Kane,
(4) Franck, B. Chem. Ber. 1958, 91, 2803. Franck, B. Chem. Ber. 1959,
92, 1001. Franck, B. Chem. Ber. 1960, 93, 2360.
J. M.; Borcherding, D. R. Tetrahedron 2003, 59, 2953
.
10.1021/ol1016492  2010 American Chemical Society
Published on Web 08/12/2010

The desired allenic hydroxylamine was prepared by the
previously reported method (Scheme 1).2a,6 Ring opening
Table 1. AgBF4-Catalyzed Cyclization of Allene 9
entry
AgBF4 loading
isolated yield (%)
cis:trans ratioa
1
2
3
4
5
6
7
8
9
a
100
80
60
50
40
20
10
8
96
95
97
99
95
86
90
63
40
4.5:1
5.5:1
6.5:1
6.0:1
9.0:1
10.0:1
13.0:1
8.0:1
4.0:1
Scheme 1. Allene Cyclization
5
1
Determined by H NMR.
The isoxazolidine 10, as a mixture of isomers, was
converted to the amino alcohol derivative 11 by cleavage of
the N-O bond with molybdenum hexacarbonyl and sodium
borohydride.13 At this point, the two diastereoisomers could
be separated by column chromatography. Although the major
isomer 11a was an oil, the minor isomer 11b proved to be
crystalline. X-ray analysis (Figure 2) demonstrated that it
was, as expected from our earlier work,2a the anti-isomer.14
of the cyclic sulfate 4 derived from (S)-propylene glycol with
lithium acetylide gave (S)-pent-4-yn-1-ol 5.9 The alkyne was
homologated to the corresponding allene 6 following the
Searles-Crabbe´ procedure.10 Neither of these alcohols were
rigorously purified due to their volatility. In each case, the
principle impurity was residual reaction solvent (THF or 1,4-
dioxane) which did not interfere with the subsequent reaction.
After Mitsunobu reaction with N-hydroxyphthalimide,11 the
(R)-hydroxylamine 7 derivative was obtained chemically pure
and with excellent enantioselectivity (>98% ee by chiral
HPLC). Deprotection of the nitrogen atom and reprotection
gave the allenic hydroxylamine substrate 9 for cyclization.
Treatment of this allene with silver(I) tetrafluoroborate in
dichloromethane at room temperature gave the desired
isoxazolidine 10 as an inseparable mixture of diastereoiso-
mers. A significant dependence of the diastereoselectivity
on catalyst loading was observed (Table 1).12 Fortunately,
the diastereoselectivity increased as the catalyst loading was
decreased. The optimum loading was found to be 10 mol
%. At loadings below this, both the dr and the yield dropped.
Figure 2. X-ray structure of amino alcohol derivative 11b.
It was intended to employ the vinyl group generated by
the allene cyclization in a metathesis reaction to form the
alkene present in the piperidine ring of sedinine. To obtain
the required cis-geometry, a ring-closing metathesis, rather
than a cross metathesis, was required. We therefore decided
to attach the second alkene partner via formation of an N,O-
acetal, so that we would also have an iminium ion precursor
without having to resort to redox chemistry.15
(5) (a) Hootele´, C.; Colau, B.; Helin, F.; Declerq, J. P.; Germain, G.;
van Meerssche, M. Tetrahedron Lett. 1980, 52, 5063. (b) For a synthesis
of (()-sedinine, see: Ogawa, M.; Natsume, M. Heterocycles 1985, 23, 831.
(6) Bates, R. W.; Lu, Y. J. Org. Chem. 2009, 74, 9460.
(7) For a review on the use of silver in heterocycle synthesis, see:
´
Alvarez-Corral, M.; Mun˜oz-Dorado, M.; Rodr´ıguez-Garc´ıa, I. Chem. ReV.
2008, 108, 3174. For reviews of allene cyclization, see: Bates, R. W.;
Satchareon, V. Chem. Soc. ReV. 2002, 12. Ma, S. Acc. Chem. Res. 2003,
36, 701.
(8) For recent reviews on iminium ion chemistry: Yazici, A.; Pyne, S. G.
Synthesis 2009, 339. Yazici, A.; Pyne, S. G. Synthesis 2009, 513. Speckamp,
W. N.; Moolenaar, M. J. Tetrahedron 2000, 56, 817.
(9) Bates, R. W.; Maiti, T. B. Synth. Commun. 2003, 633.
(10) Searles, S.; Li, Y.; Nassim, B.; Lopes, M.-T. R.; Tran, P. T.; Crabbe´,
P. J. Chem. Soc., Perkin I 1984, 747.
(13) (a) Cicchi, S.; Got, A.; Brandi, A.; Guarna, A.; Sarlos, F. D.
Tetrahedron Lett. 1990, 31, 3351. (b) Zhang, D.; Su¨ling, C.; Miller, M. J.
J. Org. Chem. 1998, 63, 885. (c) Mulvihill, M. J.; Gage, J. L.; Miller, M. J.
J. Org. Chem. 1998, 63, 3357. (d) Li, F.; Brogan, J. B.; Gage, J. L.; Zhang,
D.; Miller, M. J. J. Org. Chem. 2004, 69, 4538. (e) Yang, Y.-K.; Choi,
J.-H.; Tae, J. J. Org. Chem. 2005, 70, 6995. (f) Calvet, G.; Blanchard, N.;
Kouklovsky, C. Synthesis 2005, 19, 3346.
(11) (a) Grochowski, E.; Jurczak, J. Synthesis 1976, 682. (b) Iwagami,
H.; Yatagai, M.; Nakazawa, M.; Orita, H.; Honda, Y.; Ohnuki, T.; Yukawa,
T. Bull. Chem. Soc. Jpn. 1991, 64, 175.
(14) Details have been deposited with the Cambridge Crystallographic
Database, CCDC 784128, and may be obtained at http://www.ccdc.cam.
ac.uk.
(12) Gallagher has noted such a dependence: Fox, D. N. A.; Gallgher,
T. Tetrahedron 1990, 46, 4697.
Org. Lett., Vol. 12, No. 17, 2010
3939

Treatment of amino alcohol derivative 11a with the
dimethyl acetal of 4-bromobutanal in toluene at 100 °C in
the presence of polymer-supported PPTS,16 followed by
elimination of HBr with potassium t-butoxide, gave the
desired bicyclic N,O-acetal 13 as a 6:1 mixture of isomers
(Scheme 2). The two isomers were separable, and the major
this bicyclic N,O-acetal 14 with a nucleophile under Lewis
acidic conditions would yield the cis isomer under stereo-
electronic control, rather than the required trans isomer.21
We, therefore, employed the same strategy as in our poranthe-
ridine synthesis,6 converting the cyclic N,O-acetal 14 to a
semicyclic N,O-acetal 15a,22 and converting the acyclic car-
bamate to a cyclic carbamate 16. In the case of porantheridine
3, treatment of such an N,O-acetal with a Lewis acid and allyl
trimethylsilane gave exclusively the trans isomer. In the present
case, treatment of 16 with tin(IV) chloride23 and the trimeth-
ylsilyl enol ether of acetophenone24 yielded the desired ketone
17 as a 1:1 diastereoisomeric mixture. This dramatic change
of outcome is attributed to the presence of the ring alkene. The
iminium ion (or dihydropyridinium ion) intermediate 18 is
substantially planar, with no stereoelectronic or steric bias
favoring either face.
Scheme 2. Unstereoselective Iminium Ion Chemistry
Treatment of the bicyclic N,O-acetal 14 under the same
conditions, however, led to ketone 19 as a single isomer
(Scheme 3). As the 1H NMR spectrum showed the typical line
Scheme 3
.
Stereoselective Iminium Ion Chemistry and Sedinine
Synthesis
isomer 13 was obtained in 53% yield from amino alcohol
11a. For this acetal to be a competent RCM substrate, it is,
of course, necessary for the two alkene moieties to be in
proximity.17 The stereochemistry of the newly formed
stereogenic center of the acetal could not, however, be
1
determined, as the problem of rotamers complicated the H
NMR spectrum. Upon the basis of our previous work with
N,O-acetals,11a and drawing on earlier reports on pip-
eridines,18 including that of Martin,19 we expected that the
vinyl and allyl substituents would both be cis and pseudoaxial
and, hence, in proximity. It was, therefore, gratifying to find
that treatment of N,O-acetal 13 with Grubbs’ first-generation
catalyst yielded the desired bicyclic acetal 14 in 84% yield.
From our recent work on porantheridine,6 and also from the
work of Lhommet,20 it may be expected that treatment of
broadening of such piperidine Boc derivatives, it was converted
to the corresponding cyclic carbamate 17a which proved to be
crystalline. X-ray analysis (Figure 3) showed the trans disposi-
tion of the two piperidine substituents,25 in complete contrast
to the result in the porantheridine case. This is again attributed
to the presence of the ring alkene. The resulting flattening of
the ring of the iminium ion intermediate 20 ensures that neither
(15) (a) Bates, R. W.; Lu, Y.; Cai, M. P. Tetrahedron 2009, 65, 7852.
(b) Bates, R. W.; Boonsombat, J.; Lu, Y.; Nemeth, J. A.; Sa-Ei, K.; Song,
P.; Cai, M. P.; Cranwell, P. B.; Winbush, S. Pure Appl. Chem. 2008, 80,
681.
(16) Li, Z.; Ganesan, A. Synth. Commun. 1998, 28, 3209.
(17) For reviews on the use of RCM in alkaloid synthesis, see: (a)
Compain, P. AdV. Synth. Catal. 2007, 349, 1829. (b) Martin, W. H. C.;
Blechert, S. Curr. Top. Med. Chem. 2005, 5, 1521. (c) Brenneman, J. B.;
Martin, S. F. Curr. Org. Chem. 2005, 9, 1535. (d) Felpin, F.-X.; Lebreton,
J. Eur. J. Org. Chem. 2003, 3693.
(20) David, M.; Dhimane, H.; Vanucci-Bacque´, C.; Lhommet, G. J. Org.
Chem. 1999, 64, 8402.
(18) Johnson, F. Chem. ReV. 1968, 68, 375.
(19) Neipp, C. E.; Martin, S. F. Tetrahedron Lett. 2002, 43, 1779. For
a similar effect in N-sulfonyl piperidines, see: Hedley, S. J.; Moran, W. J.;
Prenzel, A. H.; G, P.; Price, D. A.; Harrity, J. P. A. Synlett 2001, 1596.
(21) For a comprehensive treatment of stereoelectronic effects, see:
Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry; Pergamon
Press: Oxford, 1983.
3940
Org. Lett., Vol. 12, No. 17, 2010

Table 2. Reduction of Ketone 19
entry
reagent and conditions
yield (%)
dr
1
2
3
4
5
DIBAL-H, CH2Cl2, -78 °C
LiAl(Ot-Bu)3H, THF, -78 °C
K-selectride, THF, 0 °C
Red-Al, CH2Cl2, -78 °C
BH3·SMe2, (R)-Me-CBS,
THF -10 to 0 °C
99
99
80
60
92
4.5:1
1.1:1
1.5:1
2.7:1
>99:1
or lithium tri(t-butoxy)aluminum hydride.30 On the other
hand, use of the B-methyl (R)-CBS catalyst31 with borane
dimethyl sulfide gave alcohol 21 as a single diastereoisomer
within the limits of detection.
Finally, reduction of the Boc group to an N-methyl group
to give sedinine 1 was found to proceed best using alane32
as its commercially available dimethylethylamine complex.
Heating piperidine 21 with this reagent in THF at reflux gave
sedinine 1 in 70% yield, accompanied by a small amount of
des-methyl sedinine 22 (15%). Reduction of piperidine 21
with lithium aluminum hydride or Red-Al gave lower yields
Figure 3. X-ray structure of piperidine 17a.
of sedinine 1 (40% and 50%, respectively). The 1H and 13
C
of the two transition states is chairlike, hence, again, the absence
of any stereoelectronic bias. However, the oxypropyl substituent
is compelled to adopt a pseudoaxial position to avoid the 1,2-
interaction with the N-Boc group, thus shielding one face of
the iminium ion and compelling the enol ether to approach from
the opposite face. This results in the remarkably high selectivity
in favor of the trans isomer. This trans selectivity is in contrast
to the results of Craig26 and Shipman27 who found cis-selective
additions to dihydropyridinium ions. In those cases, however,
the alkene had an R,ꢀ-relationship to the iminium ion. In our
case, with a ꢀ,γ-relationship, the iminium ion conformation
must be quite different.
With the trans adduct in hand, the synthesis was completed
by reduction of the keto group. Diastereoselective reduction
of side-chain ketones in sedum alkaloid syntheses has been
reported.1 While all reagents tried gave the desired diaste-
reoisomer 21 as the major product, in our hands, the best
diastereoselectivity on reduction of the ketone group of 19
by substrate control was obtained using DIBAL (Table
2).28,2g A much lower dr was obtained using K-selectride29
NMR spectroscopic data of the synthetic sedinine 1 were in
good agreement with those reported by Hootele´.33 In
addition, the optical rotation recorded for our synthetic
20
sample, [R]D ) -97 (c ) 0.57, MeOH) was also in good
20
agreement with that reported by Hootele´, [R]D ) -98 (c
) 1.9, MeOH), as was the melting point of 118-120 °C,
compared to the reported value of 120-121 °C.5a
In conclusion, we have completed a stereoselective
synthesis of sedinine, one of the more complex sedum
alkaloids. All of the stereocenters can be derived from the
readily available starting material (S)-propylene glycol. The
excellent stereoselectivity in the addition to the iminium ion
shows that inclusion of an alkene in the ring is an additional
strategy for switching from cis to trans addition.
Acknowledgment. We thank the Singapore Ministry of
EducationAcademicResearchFundTier2(grantT206B1220RS)
and Nanyang Technological University for generous support
of this work. We thank Professor Steve Collier (Codexis
Laboratories Singapore Pte Ltd.) for suggesting the use of
alane in the final step.
(22) If the reaction is carried out at room temperature, a byproduct
identified as dihydropyridine 15b is also obtained. Formation of this
byproduct was suppressed by maintaining a temperature of 0 to -10°C.
Supporting Information Available: Experimental pro-
cedures for (S)-propylene glycol, compounds 1, 4-17, 19,
21, and 22; 1H and 13C NMR spectra for compounds 1, 6-17,
19, and 21. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL1016492
(23) In contrast, use of TMSOTf, BF3·OEt2, or SnCl2 resulted in
decomposition; starting material was largely recovered when Cp2TiCl2 or
Yb(OTf)3 was used. Use of TiCl4 gave a slightly lower yield and some
formation of dihydropyridine 15b.
(27) Dransfield, P. J.; Gore, P. M.; Prokes, I.; Shipman, M.; Slawin,
A. M. Z. Org. Biomol. Chem. 2003, 1, 2723.
(24) Niwa, H.; Hasegawa, T.; Ban, N.; Yamada, K. Tetrahedron Lett.
1984, 25, 2797. Canan Koch, S. S.; Chamberlain, A. R. J. Org. Chem.
1993, 58, 2725.
(28) Ghiaci, M.; Adibi, M. Org. Prep. Proc. Int. 1996, 28, 474.
(29) Comins, D. L.; Hong, H. J. Org. Chem. 1993, 58, 5035.
(30) Mill, S.; Hootele´, C. Can. J. Chem. 1996, 74, 2434. Driessens, F.;
Hootele´, C. Can. J. Chem. 1991, 69, 211. Mancuso, V.; Hootele´, C.
Tetrahedron Lett. 1988, 29, 5917.
(25) Details have been deposited with the Cambridge Crystallographic
Database, CCDC 784129, and may be obtained at http://www.ccdc.cam.
ac.uk.
(31) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1987.
(32) Yoon, M. N.; Brown, H. C. J. Am. Chem. Soc. 1968, 90, 2927.
(33) Colau, B.; Hootele´, C. Tetrahedron 1984, 40, 2171.
(26) Craig, D.; McCague, R.; Potter, G. A.; Williams, M. R. V. Synlett
1998, 55.
Org. Lett., Vol. 12, No. 17, 2010
3941