Synthesis of fused piperidinones through a radical-ionic cascade

Article abstract of DOI:10.1021/jo801308j

(Chemical Equation Presented) Azabicyclo[4.3.0]nonanes were assembled, from chiral allylsilanes possessing an oxime moiety, using a stereocontrolled formal [2 + 2 + 2] radical-ionic process. The cascade involves the addition of an α-iodoester to the less substituted end of the enoxime which is then followed by a 5-exo-trig cyclization onto the aldoxime function, producing an alkoxyaminyl radical species which finally lactamizes to afford the titled piperidinone. High levels of stereoinduction were observed, demonstrating the ability of a silicon group located at the allylic position to efficiently control the stereochemistry of the two newly created stereogenic centers. When the radical cascade was extended to ketoximes, the resulting sterically hindered alkoxyaminyl radical did not react further with the initiator Et₃B to produce the expected nucleophilic amidoborane complex. In sharp contrast, this long-lived radical recombined with the initial α-stabilized ester radical to produce a cyclopentane incorporating two ester fragments.

Full text of DOI:10.1021/jo801308j

Synthesis of Fused Piperidinones through a Radical-Ionic Cascade
Edouard Godineau,† Kurt Schenk,‡ and Yannick Landais*,†
UniVersity Bordeaux 1, Institut des Sciences Mole´culaires, UMR-CNRS 5255, 351, Cours de la Libe´ration,
F-33405 Talence Cedex, France, and UniVersity of Lausanne, Institut de Cristallographie, BCP,
CH-1015 Dorigny-Lausanne, Switzerland
y.landais@ism.u-bordeaux1.fr
ReceiVed June 18, 2008
Azabicyclo[4.3.0]nonanes were assembled, from chiral allylsilanes possessing an oxime moiety, using a
stereocontrolled formal [2 + 2 + 2] radical-ionic process. The cascade involves the addition of an
R-iodoester to the less substituted end of the enoxime which is then followed by a 5-exo-trig cyclization
onto the aldoxime function, producing an alkoxyaminyl radical species which finally lactamizes to afford
the titled piperidinone. High levels of stereoinduction were observed, demonstrating the ability of a silicon
group located at the allylic position to efficiently control the stereochemistry of the two newly created
stereogenic centers. When the radical cascade was extended to ketoximes, the resulting sterically hindered
alkoxyaminyl radical did not react further with the initiator Et3B to produce the expected nucleophilic
amidoborane complex. In sharp contrast, this long-lived radical recombined with the initial R-stabilized
ester radical to produce a cyclopentane incorporating two ester fragments.
SCHEME 1. Piperidinones as Valuable Precursors for
Natural Product Synthesis
Introduction
The piperidine skeleton, as in (ꢀ)-conhydrine 1, is ubiquitous
in Nature1 and is also found as a key structural motif in many
synthetic pharmaceutical leads.2 This common heterocycle may
also be fused with other rings as seen, for example, in
indolizidines (i.e., castanospermine 2) which have been of
widespread interest given their attractive glycosidase inhibitory
activities.3,4 The piperidine skeleton may also be imbedded in
fairly complex tricyclic alkaloids, such as in lepadiformine 3
and related analogues5 (Scheme 1). The control of the relative
configuration at the ring junction as well as the stereochemistry
† University Bordeaux 1.
‡ University of Lausanne.
(1) (a) Buffat, M. G. P. Tetrahedron 2004, 60, 1701–1729. (b) Weintraub,
P. M.; Sabol, J. S.; Kane, J. M.; Borcherding, D. R. Tetrahedron 2003, 59, 2953–
2989. (c) Baliah, V.; Jeyaraman, R.; Chandrasekaran, L. Chem. ReV. 1983, 83,
379–423.
(2) Watson, P. S.; Jiang, B.; Scott, B. Org. Lett. 2000, 23, 3679–3681.
(3) (a) Micheal, J. P. Nat. Prod. Rep. 2004, 21, 625–649. (b) Bailey, P. D.;
Millwood, P. A.; Smith, P. D. Chem. Commun. 1998, 633–640.
(4) (a) Asano, N. Glycobiology 2003, 13, 93R–104R. (b) de Melo, E. B.; da
Silveira Gomes; Carvalho, I. Tetrahedron 2006, 62, 10277–10302.
(5) (a) Weinreb, S. M. Chem. ReV. 2006, 106, 2531–2549, and references
therein. (b) Scha¨r, P.; Renaud, P. Org. Lett. 2006, 8, 1569–1571. (c) Swidorski,
J. J.; Wang, J.; Hsung, R. P. Org. Lett. 2006, 8, 777–780.
of the quaternary center in targets such as 3 constitute two of
the key problems encountered in the syntheses of these alkaloids.
Several elegant solutions to these stereochemical issues have
nevertheless been accomplished during the course of their total
10.1021/jo801308j CCC: $40.75
Published on Web 08/26/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 6983–6993 6983

Godineau et al.
SCHEME 2. A formal [2 + 2 + 2] Process for the Synthesis
of Piperidinones
configuration of the three contiguous stereogenic centers during
the 5-exo-trig cyclization. This would allow the controlled
installation of the ring junction, a problem which, until recently,
had not received satisfying answers using radical-mediated
processes.12 The two-carbon building block, incorporating the
carbonyl unit would be generated from the corresponding
R-iodo, R-bromo, or R-xanthate ester IV, identified as the
electron-poor radical partner.13 Addition of such radicals onto
olefins is known to be an efficient process, providing a new
radical intermediate that can then cyclize in a 5- or 6-exo-trig
mode onto the oxime moiety. Such cyclizations, pioneered by
Corey14 and Bartlett15 and later developed by Naito,16 Marco-
Contelles,17 and others,18 are faster than the corresponding
cyclizations onto olefins.19 We envisaged that the formation of
the ring should be followed by the reaction of the resulting
alkoxyaminyl radical with a suitable radical mediator20 (Et3B
or Et2Zn). In the event, this radical polar crossover reaction
would provide a nucleophilic boron or zinc amide complex
which could then cyclize to afford the desired piperidinone III.
syntheses.5 Piperidinones of type II may be considered as simple
precursors for the construction of the polysubstituted piperidine
skeleton I. Structure II may be easily functionalized at any
position through well-known methods, including alkylation at
C3,6 transformation into an R,ꢀ unsaturated amide,6a partial
reduction of the CdO bond into versatile iminium intermedi-
ates,7 or recently developed radical-mediated hydroxyalkylation
at C6.8 A chiral auxiliary may also be introduced on the nitrogen
atom allowing a ready access to enantiopure analogues.9
Therefore, as virtually all carbon centers may be functionalized,
piperidinones II constitute valuable intermediates en route to
the synthesis of more complex piperidines.
(12) (a) James, P.; Landais, Y. Org. Lett. 2004, 6, 325–328. (b) James, P.;
Felpin, F.-X.; Schenk, K.; Landais, Y. J. Org. Chem. 2005, 70, 7985–7995. (c)
James, P.; Schenk, K.; Landais, Y. J. Org. Chem. 2006, 71, 3630–3633. (d)
d’Antuono, P.; Fritsch, A.; Ducasse, L.; Castet, F.; James, P.; Landais, Y. J.
Phys. Chem. A 2006, 110, 3714–3722. (e) Wang, C.; Russell, G. A. J. Org.
Chem. 1999, 64, 2346–2352. (f) Harvey, I. W.; Philips, E. D.; Whitham, G. H.
Tetrahedron 1997, 53, 6493–6508. (g) De Riggi, I.; Surzur, J.-M.; Bertrand,
M.-P.; Archavlis, A.; Faure, R. Tetrahedron 1990, 46, 5285–5294. (h) Brumwell,
J. E.; Simpkins, N. S. Tetrahedron Lett. 1993, 34 (7), 1219–1222. (i) De Riggi,
I.; Gastaldi, S.; Surzur, J.-M.; Bertrand, M.-P.; Virgili, A. J. Org. Chem. 1992,
57, 6118–6125.
(13) (a) Porter, N. A.; Zhang, G.; Reed, A. D. Tetrahedron Lett. 2000, 41,
5773–5777. (b) Ollivier, C.; Renaud, P. J. Am. Chem. Soc. 2001, 123, 4717. (c)
Renaud, P.; Ollivier, C.; Panchaud, P. Angew. Chem., Int. Ed. 2002, 41, 3460.
(d) Panchaud, P.; Ollivier, C.; Renaud, P.; Zigmantas, S. J. Org. Chem. 2004,
69, 2755. (e) Panchaud, P.; Chabaud, L.; Landais, Y.; Ollivier, C.; Renaud, P.;
Zigmantas, S. Chem.sEur. J. 2004, 10, 3606–3614. (f) Quiclet-Sire, B.; Zard,
S. Z. Top. Curr. Chem. 2006, 264, 201–236.
We recently reported10 on a straightforward one-pot synthesis
of fused piperidinones III via a formal [2 + 2 + 2] radical-
ionic cyclization of enoximes V (Scheme 2).11 We describe here
a full account of these investigations, focusing on various aspects
of our research, including: (1) the stereocontrolled generation
of the azabicyclo[4.3.0]nonane skeleton III through 1,2- and
1,3-induction and (2) an unexpected tandem addition-
cyclization-recombination process. Our piperidinones may be
easily accessible starting from structurally simple enoximes V
having an allylic stereocenter and a simple R-haloester or
xanthate IV. The substituent at the allylic position is central to
the methodology as it is expected to dictate the relative
(14) Corey, E. J.; Pyne, S. G. Tetrahedron Lett. 1983, 24, 2821–2824.
(15) Bartlett, P. A.; McLaren, K. L.; Ting, P. C. J. Am. Chem. Soc. 1988,
110, 1633–1634.
(16) (a) Naito, T.; Tajiri, K.; Harimoto, T.; Ninomiya, I.; Kiguchi, T.
Tetrahedron Lett. 1994, 35, 2205–2206. (b) Naito, T.; Torieda, M.; Tajiri, K.;
Ninomiya, I.; Kiguchi, T. Chem. Pharm. Bull. 1996, 44, 624–626. (c) Kiguchi,
T.; Tajiri, K.; Ninomiya, I.; Naito, T. Tetrahedron 2000, 56, 5819–5833. (d)
Kiguchi, T.; Tajiri, K.; Ninomiya, I.; Naito, T.; Hiramatsu, H. Tetrahedron Lett.
1995, 36, 253–256. (e) Miyata, O.; Muroya, K.; Kobayashi, T.; Yamanaka, R.;
Kajisa, S.; Koide, J.; Naito, T. Tetrahedron 2002, 58, 4459–4479. (f) Miyata,
O.; Takahashi, S.; Tamura, A.; Ueda, M.; Naito, T. Tetrahedron 2008, 64, 1270–
1284.
(17) (a) Marco-Contelles, J.; Gallego, P.; Rodriguez-Fernandez, M.; Khiar,
N.; Destabel, C.; Bernabe, M.; Martinez-Grau, A.; Chiara, J. L. J. Org. Chem.
1997, 62, 7397–7412. (b) Chiara, J. L.; Marco-Contelles, J.; Khiar, N.; Gallego,
P.; Destabel, C.; Bernabe, M. J. Org. Chem. 1995, 60, 6010–6011. (c) Marco-
Contelles, J.; Destabel, C.; Gallego, P.; Chiara, J. L.; Bernabe, M. J. Org. Chem.
1996, 61, 1354–1362. (d) Marco-Contelles, J.; Destael, C.; Chiara, J. L.; Bernabe,
M. Tetrahedron: Asymm. 1995, 6, 1547–1550.
(6) (a) Johnson, T. A.; Jang, D. O.; Slafer, B. W.; Curtis, M. D.; Beak, P.
J. Am. Chem. Soc. 2002, 124, 11689–11698. (b) Okitsu, O.; Suzuki, R.;
Kobayashi, S. J. Org. Chem. 2001, 66, 809–823. (c) Takacs, J. M.; Weidner,
J. J. J. Org. Chem. 1994, 59, 6480–3. (d) Yamaguchi, M.; Hirao, I. Tetrahedron
Lett. 1983, 24, 1719–22.
(7) (a) Matsumura, Y.; Aoyagi, S.; Kibayashi, C. Org. Lett. 2003, 5, 3249–
3252. (b) Vink, M. K. S.; Schortinghuis, C. A.; Luten, J.; van Maarseveen, J. H.;
Schoemaker, H. E.; Hiemstra, H.; Rutjes, F. P. J. T. J. Org. Chem. 2002, 67,
7869–7871. (c) Hayashi, R.; Ohmori, E.; Isogaya, M.; Moriwaki, M.; Kumagai,
H. Bioorg. Med. Chem. Lett. 2006, 16, 4045–4047. (d) Igarashi, M.; Fuchikami,
T. Tetrahedron Lett. 2001, 42, 1945–1947. (e) Kano, S.; Tanaka, Y.; Sugino,
E.; Hibino, S. Synthesis 1980, 695–697.
(8) Yoshimitsu, T.; Arano, Y.; Nagaoka, H. J. Am. Chem. Soc. 2005, 127,
11610–11611.
(9) (a) Groaning, M. D.; Meyers, A. I. Tetrahedron 2000, 56, 9843–9873.
(b) Amat, M.; Bosch, J.; Hidalgo, J.; Canto, M.; Perez, M.; Llor, N.; Molins, E.;
Miravitlles, C.; Orozco, M.; Luque, J. J. Org. Chem. 2000, 65, 3074–3084. (c)
Amat, M.; Llor, N.; Escolano, C.; Huguet, M.; Perez, M.; Molins, E.; Bosch, J.
Tetrahedron: Asymmetry 2003, 14, 293–295.
(18) (a) Takahashi, S.; Terayama, H.; Koshino, H.; Kuzuhara, H. Tetrahedron
1999, 55, 14871–14884. (b) Gartenmann Dickson, L.; Leroy, E.; Reymond, J.-
L. Org. Biol. Chem. 2004, 2, 1217–1226. (c) Cronje, J. J.; Holzapfel, C. W.
Tetrahedron Lett. 1997, 38, 7429–7432. (d) Enholm, E. J.; Burroff, J. A.;
Jaramillo, L. M. Tetrahedron Lett. 1990, 31, 3727–30. (e) Fernandez, M.; Alonso,
R. Org. Lett. 2005, 7, 11–14. (f) Fernandez-Gonzalez, M.; Alonso, R. J. Org.
Chem. 2006, 71, 6767–6775. (g) Hatem, J.; Henriet-Bernard, C.; Grimaldi, J.;
Maurin, R. Tetrahedron Lett. 1992, 33, 1057–1058. (h) Marco-Contelles, J.;
Balme, G.; Bouyssi, D.; Destabel, C.; Henriet-Bernard, C. D.; Grimaldi, J.;
Hatem, J. M. J. Org. Chem. 1997, 62, 1202–1209.
(19) Rate constants k > 107 M-1s-1 for a 5-exo-trig cyclization onto an oxime
vs 105 M-1 s-1 for the related cyclization onto an olefin have been reported;
see: (a) Fallis, A. G.; Brinza, I. M. Tetrahedron 1997, 53, 17543–17594. (b)
Sturino, C. F.; Fallis, A. G. J. Org. Chem. 1994, 59, 6514–6516. (c) Kim, S.;
Yoon, K. S.; Kim, Y. S. Tetrahedron 1997, 53, 73–80. (d) Griller, D.; Schmid,
P.; Ingold, K. U. Can. J. Chem. 1979, 57, 831–834.
(10) Godineau, E.; Scha¨fer, C.; Landais, Y. Org. Lett. 2006, 8, 4871–4874.
(11) For recent reports on two-step radical-ionic annulations, see: (a)
Korapala, C. S.; Qin, J.; Friestad, G. K. Org. Lett. 2007, 9, 4243–4246. (b)
Friestad, G. K.; Marie´, J. C.; Suh, Y.; Qin, J. J. Org. Chem. 2006, 71, 7016–
7027.
(20) (a) Nozaki, K.; Oshima, K.; Utimoto, K. Tetrahedron Lett. 1988, 29,
1041–1044. (b) Bazin, S.; Feray, L.; Siri, D.; Naubron, J.-V.; Bertrand, M. P.
Chem. Commun. 2002, 2506–2507. (c) Ueda, M.; Miyabe, H.; Sugino, H.; Miyata,
O.; Naito, T. Angew. Chem., Int. Ed. 2005, 44, 6190–6193. (d) Miyabe, H.;
Ueda, M.; Naito, T. Synlett 2004, 1140–1157.
6984 J. Org. Chem. Vol. 73, No. 18, 2008

Synthesis of Fused Piperidinones
SCHEME 3. Enoximes 4-10
group on the reactivity of the olefin and on the stereochemical
course of the reaction. The cyclization was first performed using
AIBN as an initiator at 80 °C in benzene. Under these
conditions, the simplest model 8, lacking an allylic stereogenic
center, led to a modest stereocontrol in favor of the cis isomer
11a, as previously described by Naito et al. (entry 1).16e,21 When
the same reaction was performed with allylsilane 6, the desired
cyclopentane 12a was generated with better yield and stereo-
control (entry 2), the trans,cis stereoisomer being formed as
the major product along with a minute amount of the trans,trans
isomer which could be separated by chromatography. The
structure of trans,cis 12a was unambiguously determined
through X-ray diffraction studies (see the Supporting Informa-
tion). No traces of other stereoisomers were detected on the 1H
NMR of the crude reaction mixture. The higher reactivity of
allylsilanes toward electrophilic radical species has already been
observed by us10,12a-c and others22 and may be ascribed to the
enhanced nucleophilic character of allylsilanes as compared to
simple olefins mainly due to σ-donation of electrons from the
C-Si bond to the π-system of the olefin.23 Attack of a radical
entity on the allylsilane moiety also generates a radical ꢀ to
silicon (C2) which is known to be stabilized by 2-3 kcal/mol.24
The excellent 1,2- and good 1,3-stereocontrol occurring during
5-exo-trig cyclization of allylsilane-oxime 6 is in line with those
reported under similar conditions with 1,6-diene analogues.12a-c
Finally, photochemical activation allowed the reaction to be
carried out at room temperature or lower (entries 3-7), which
led to improved diastereocontrol in the case of allylsilane 6.
Complete trans,cis selectivity was thus observed at -40 °C, at
the expense of the yield (entry 6). A satisfying 92:8 ratio was,
however, obtained at -10 °C at 69% conversion (entry 5). It is
worth noting that although conversions were not complete, all
reactions were extremely clean, the only other component of
the crude mixture being unconsumed starting material. In sharp
contrast, at room temperature, precursor 8 under irradiation led
to a poor conversion to the corresponding aminocyclopentane
(entry 3), showing the remarkable “accelerating effect” of an
allylic silicon group toward the addition of radicals having an
electrophilic character.12a,b
TABLE 1. Influence of an Allylic Silicon Group on the Course of
the Thiyl Addition-5-Exo-Trig Cyclization onto Enoxime 6 and 8
entry
oxime
T (°C)
cis/trans a:ba
yieldd (%)
1
2
3
4
5
6
7
8
6
8
6
6
6
6
80a
77:23
89:11
nd
89:11
92:8
>95:5
>95:5
49
77
<10
95
69 (99)
32 (99)
22 (92)
80a
25b
25b
-10b
-40b
-78b
a Initiation with AIBN. b Initiation with hν (sunlamp). c Estimated
from 1H NMR of the crude mixture. d Isolated yields (yield based on
recovered starting material).
Results and Discussion
Ene-Aldoxime Precursors. Our laboratory has previously
shown for related 5-exo-trig cyclization of 1,6-dienes that allylic
silicon groups efficiently control the relative configuration both
at the ring junction and at proximal stereogenic centers.12a-c It
was reasoned that a similar behavior might be expected with
enoximes. We therefore embarked on the preparation of a series
of chiral allylsilanes 4-7 possessing an oxime moiety at C6.
In order to evaluate the impact of the silicon substituent on
stereoselectivity, we also prepared enoxime 8 devoid of any
allylic chiral center and enoximes 9 and 10 which possess
respectively a methyl and a sterically demanding tert-butyl group
at C3 (Scheme 3) (see the Supporting Information for prepara-
tion of these substrates).10
Thiyl Radical Addition onto Chiral Allylsilanes Oximes. Prior
to embarking on the design of a formal [2 + 2 + 2] cascade as
depicted in Scheme 2, we first investigated the tandem addition-
5-exo-trig cyclization of thiyl radicals on enoxime V (see
Scheme 2), in order to test the feasibility of the first two steps
of the process (Table 1). Electrophilic thiyl radicals are known
to add onto enoximes to produce the corresponding five-
membered ring through a 5-exo-trig cyclization, with generally
modest levels of stereocontrol.16e,21 These addition-cyclization
sequences were performed on allylsilane 6 and its nonsubstituted
counterpart 8 in order to evaluate the effect of an allylic silicon
[2 + 2 + 2] Process.25 Study of the Stereocontrol. The
cascade was then extended to the addition of radical centers
located R to an ester group (i.e., IV, Scheme 2). Addition of
these ambiphilic radicals26 onto olefins is a useful process which
produces a new radical that can be trapped intra- or intermo-
lecularly in a stereocontrolled manner. Such additions were
shown to be particularly efficient on electron-rich allylsilanes.
The resulting ꢀ-silyl radical intermediate can then react further
(22) (a) Sakurai, H.; Hosomi, A.; Kumada, M. J. Org. Chem. 1969, 34, 1764–
1768. (b) Jarvie, A. W. P.; Rowley, R. J. J. Chem. Soc. B 1971, 2439–2442. (c)
Gozdz, A. S.; Maslak, P. J. Org. Chem. 1991, 56, 2179–2189.
(23) (a) Kahn, S. D.; Pau, C. F.; Chamberlin, A. R.; Hehre, W. J. J. Am.
Chem. Soc. 1987, 109, 650–663. (b) Panek, J. S. In Silicon Stabilization,
ComprehensiVe Organic Chemistry; Trost, B. M., Fleming, I., Eds; Pergamon
Press: Oxford, 1989; p 579. (c) Jarvie, A. W. P. Organomet. Chem. ReV. Sec. A
1970, 6, 153–207. (d) White, J. M. Aust. J. Chem. 2004, 15, 1227–1251.
(24) (a) Kawamura, T.; Kochi, J. K. J. Am. Chem. Soc. 1972, 94, 648–650.
(b) Griller, D.; Ingold, K. U. J. Am. Chem. Soc. 1974, 96, 6715–6720. (c) Jackson,
R. A.; Ingold, K. U.; Griller, D.; Nazran, A. S. J. Am. Chem. Soc. 1985, 107,
208–211. (d) Auner, N.; Walsh, R.; Westrup, J. J. Chem. Soc., Chem. Commun.
1986, 207–207.
(25) The term [2 + 2 + 2] refers here to a process involving 2 atoms + 2
atoms + 2 atoms and not to a concerted process.
(26) (a) Curran, D. P. In ComprehensiVe Organic Synthesis; Semmelhack,
M. F., Trost, B. M., Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 4,
pp 715-830. (b) DeVleeschouwer, F.; VanSpeybroeck, V.; Waroquier, M.;
Geerlings, P.; DeProft, F. Org. Lett. 2007, 9, 2721–2724. (c) Giese, B.; He, J.;
Mehl, W. Chem. Ber. 1988, 121, 2063–2066.
(21) (a) Miyata, O.; Muroya, K.; Koide, J.; Naito, T. Synlett 1998, 271–272.
(b) Miyabe, H.; Tanaka, H.; Naito, T. Chem. Pharm. Bull. 2004, 52, 74–78.
J. Org. Chem. Vol. 73, No. 18, 2008 6985

Godineau et al.
SCHEME 4. Preliminary Studies on the Radical Additions of r-Iodoester 13 to Enoximes 4 and 8
SCHEME 5. Two-Pot Radical-Cascade-Lactamization
Process
with azides, xanthates, and bromoamides, with a high level of
stereocontrol.13,27 Although 5-exo- and 6-exo-trig cyclizations
onto oximes are known to be fast processes,19 halogen atom
transfer may however compete to provide the corresponding
ꢀ-halosilanes.28These species are known to ꢀ-eliminate readily
even at low temperatures to produce an olefin.13a,29 Aware
of this potential drawback, we added R-iodoester 13 to
nonsilylated enoxime 8 in the presence of Et3B as an initiator
(Scheme 4). We were pleased to observe that as with thiyl
radicals, the addition proceeded smoothly and was followed
by the cyclization to provide cyclopentanes 14a,b in good
yields and modest selectivities in the favor of the cis-isomer.
In contrast to the above studies, lowering the temperature had
little effect on the yield and stereocontrol. In none of these
experiments have we observed the formation of the piperidinone
skeleton. The cascade process was then applied to enoxime 4
which provided a mixture of cis- and trans-cyclopentanes 15a,b
as well as the expected [2 + 2 + 2] product 16a as a single
diastereomer (stereochemistry as shown), resulting from a
spontaneous lactamization which occurred only with the cis-
cyclopentane isomer 15a. These preliminary results indicated
that the 5-exo-trig cyclization was faster than the iodine atom
transfer since no olefin formation was observed.
The process thus occurred with a very satisfying 91:9
stereoselectivity. As before with sulfur models, the trans,cis
stereoisomer 15a (and 16a) was formed predominantly. In order
to obtain the exclusive formation of 16a, the lactamization of
amino esters 14a,b and 15a,b was then studied using various
conditions, including acidic (TFA), basic (K2CO3, EtOH), as
well as Weinreb conditions (Me3Al) (Scheme 5).30 Treatment
with TFA led to the desired piperidinones in good yield and
excellent stereocontrol. Et3N in ethanol at reflux, and also Me3Al
in toluene at 100 °C, led to the desired piperidinones 16a,b
and 17a,b, albeit in lower yield.
with acidic, basic, or organometallic conditions described above
were no longer required. For instance, radical addition of
iodoester 1831 onto enoxime 8, followed by 5-exo-trig cycliza-
tion and lactamization, occurred smoothly to provide a 76:24
mixture of cis- and trans-piperidinones 17a,b in 69% yield
(Table 2, entry 1). After some optimizations, the one-pot process
was extended to the cyclization of enoximes having allylic
stereogenic centers. The results are summarized in Table 2. The
best yields and diastereoselectivities were observed with
enoximes having an allylsilane moiety (Table 2, entries 2-5).
With these precursors, the trans,cis-isomer was obtained
predominantly with only a small amount of trans,trans-isomer
being observed in one case (entry 3). The presence of a
ꢀ-substituent (R′) brought little change in the level of stereo-
control (entries 3-5). Similarly, when tert-butyl precursor 10
was treated under the optimized conditions, the piperidinone
was formed with excellent stereocontrol in favor of the trans,cis
stereoisomer 23a, although the yield of 23a was modest (entry
7). In contrast, application of the radical cascade to precursor 9
(R ) Me) led to a 85:13:2 mixture of three stereoisomers (out
In our quest for a more straightforward “one-pot” [2 + 2 +
2] process, we finally found that exchanging a phenyl ester for
the ethyl ester led directly to the piperidinone. Further treatments
1
of four possible diastereomers) as shown by the H NMR of
(27) (a) Chabaud, L.; Landais, Y.; Renaud, P. Org. Lett. 2002, 4, 4257–
4260. (b) Briggs, M. E.; Zard, S. Z. Synlett 2005, 334–336. (c) Chabaud, L.;
Landais, Y.; Renaud, P.; Robert, F.; Castet, F.; Lucarini, M.; Schenk, K.
Chem.sEur. J. 2008, 14, 2744–2756.
the crude reaction mixture. The relative configuration of the
two major products was assumed to be respectively trans,cis
and trans,trans, in agreement with that of the tert-butyl analogue
which is assumed to lead to a similar stereochemical outcome.
(28) Rate constants for 5-exo-trig cyclizations onto oximes and iodine atom
transfer from R-iodoacetate have been estimated: k ∼ 3. 17 and k ∼ 17 M s-1
,
respectively.
(29) (a) Guindon, Y.; Gue´rin, B.; Chabot, C. C.; Ogilvie, W. W. J. Am. Chem.
Soc. 1996, 118, 12528–12535. (b) Masterson, D. S.; Porter, N. D. Org. Lett.
2002, 4, 4253–4256. (c) Chabaud, L.; Landais, Y. Tetrahedron Lett. 2003, 44,
6995–6998.
(30) Lipton, M. F.; Basha, A.; Weinreb, S. M. Org. Synth. 1988, 50, 492–
495.
(31) Phenyl iodoacetate was prepared following a two-step sequence,
including the reaction of phenol with the R-chloroacetyl chloride followed by
the exchange of chlorine for iodine using NaI in acetone (see the Supporting
Information).
6986 J. Org. Chem. Vol. 73, No. 18, 2008

Synthesis of Fused Piperidinones
TABLE 2. [2 + 2 + 2]-Radical-Ionic Cascade Involving Enoximes
Et3B
(equiv)
yieldc
(%)
entry
oxime
R
R′
product
dr (a:b)a
1
2
3
4
5
6
7
8
4
5
6
7
9
10
H
H
H
2
3
3
3
3
6
6
17a,b
16a,b
19a,b
20a,b
21a,b
22a,c
23a,b
76:24
>95:5
69
76
61
78
85
51
47
SiMe2Ph
SiMe2Ph
SiMe2Ph
SiMe2Ph
Me
CO2Me
OAc
Me
H
H
>90:10
>95:5
>95:5
85:13:2b
>95:5
t-Bu
a
Estimated ratio through H NMR of the crude reaction mixture. b Ratio estimated through GC. c Isolated yield of both isomers.
1
SCHEME 6. Study on the Effect of the Nature of the
r-Iodoester
The poor efficiency of radical additions32 onto olefins bearing
an alkyl group on the allylic stereogenic center was already
noticed in related sulfonyl process onto 1,6-dienes.12a-d Such
a result further demonstrates the beneficial effect of an allylic
silicon group on the efficiency of the addition step, which can
be attributed to the σ-donating effect of this silicon substituent
(polar effect, vide infra).23
The effect of the structure of the R-iodo ester component was
further studied. As anticipated, the addition of R-iodo tert-butyl
ester 24 did not affect the stereochemical course of the process;
it, however, considerably slowed down the lactamization step,
such that cyclopentanes 25a,b, 26, and 27 were now isolated
without traces of the corresponding piperidinones (Scheme 6).
As above, excellent diastereoselectivities were observed with
silylated precursors 4 and 6 and a lower cis/trans ratio with
simpler model 8. Functionalized R-bromo-R,R-difluoroester 2833
was also tested and led to fluorinated piperidinones 29 and 30
in yields and selectivities similar to those obtained previously.
Attempts to use R-bromo esters or R-iodolactone however met
with little success. Weinreb R-iodoamides behave similarly to
the corresponding ethyl esters, leading to mixtures of the
corresponding aminoamides and piperidinones and were not
studied further. As a summary, the addition-5-exo-trig cycliza-
tion sequence using various esters and ene-aldoxime precursors
can be performed with good chemical efficiency and excellent
level of stereoinduction, provided that a silicon group is present
at the allylic position of the ene-aldoxime component. In
addition, we have also demonstrated that the final lactamization
can be favored or disfavored simply by tuning the structure of
the R-iodoester substituent.
Trans,cis relative configurations were firmly established from
X-ray diffraction studies on cyclopentane 12a and from NOEs
measured on silylated piperidinone 20a (see the Supporting
Information). From these and previous observations from 5-exo-
trig processes carried out on 1,6-dienes,12a-d a transition-state
model was proposed to rationalize the stereochemical outcome
resulting from the cyclization step. Ab initio calculations on
5-exo-trig cyclizations performed by Houk and Schiesser34
established that such cyclizations occur through a chairlike
transition state, in which steric effects and torsional strain are
minimized. This model was recently revisited in sulfonyl radical
mediated 5-exo-trig processes with the implementation of a
bulky silicon group on the carbon chain.12d Application of this
model to the enoxime case (i.e., A, Scheme 7) leads to a
transition state where the silicon group occupies a pseudoequa-
torial position in order to minimize steric interactions with
neighboring substituents. Interestingly, such a model also implies
a quasi linear arrangement of the electron-rich C2-Si bond and
the incipient C1-C5 bond, reminiscent of the well-known
ꢀ-silicon effect,35 which likely stabilizes further this transition
(34) (a) Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron 1985, 41, 3925–
3941. (b) Spellmeyer, D. C.; Houk, K. N. J. Org. Chem. 1987, 52, 959–974. (c)
Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron Lett. 1985, 26, 373–376. (d)
Beckwith, A. L. J.; Zimmerman, J. J. Org. Chem. 1991, 56, 5791–5796.
(35) Lambert, J. B. Tetrahedron 1990, 46, 2677–2689, and references therein.
(32) Larger amounts of Et3B (up to 6 equiv) were always required to allow
the reaction to reach completion.
(33) Moreno, B.; Quehen, C.; Rose-He´le`ne, M.; Leclerc, E.; Quirion, J. C.
Org. Lett. 2007, 9, 2477–2480.
J. Org. Chem. Vol. 73, No. 18, 2008 6987

Godineau et al.
SCHEME 7. Transition-State Model for the 5-Exo-Trig
Cyclization and Oxidation of the PhMe2Si Group
SCHEME 8. Radical Addition-5-Exo-Trig Cyclization on
Ketoxime 32
state. As demonstrated in the studies above, the silicon group
not only activates the olefin toward the addition of electrophilic
radicals but also locks the conformation at the transition state,
allowing high levels of 1,2- and 1,3-stereocontrol. A similar
behavior is observed with the t-Bu group (i.e., 10) which in
turn lacks the polar component of the silicon group, resulting
in a much less efficient addition step (entry 7, Table 2). It was
finally shown that the silicon group could be converted into
the corresponding hydroxyl, using the buffered Fleming-Tamao
oxidation,36 with retention of configuration, thus providing the
alcohol 31 in excellent yield.
in 45% yield along with unreacted 32 (35% recovery), indicating
that lactamization was not competitive with the formation of
the diester 33.
5-Exo-Trig Cyclization onto Ketoximes. A Recombination
Process. The successful approach to piperidinones described
above prompted us to extend this strategy to ketoximes
analogues (Scheme 2, R′ ) alkyl, aryl). Radical addition
followed by a 5-exo-trig cyclization onto a ketoxime function
would provide a cyclopentane ring having a quaternary center
and a carbon chain that may possess functional groups allowing
further elaboration.37 Such an approach would be well suited
to access the tricyclic skeleton of alkaloids such as lepadiformine
3 and its congeners (Scheme 1). Our investigations started with
allylsilane-ketoxime 32 (see Supporting Information for prepa-
ration) (Scheme 8). When submitted to R-iodoester 13 (2 equiv)
and triethylborane, ketoxime 32 did not however furnish the
expected piperidinone. We instead isolated a material incorpo-
rating two ester fragments, assigned as cyclopentane 33, which
was formed as a single stereoisomer. Although all analytical
elements spoke in favor of cyclopentane 33 (1H, 13CNMR,
COSY, HMQC, and HRMS), its structure was unambiguously
proven via X-ray diffraction studies conducted on diester 34.
These studies allowed us to establish the relative stereochemistry
of cyclopentane 33 as being trans,cis, which corroborates the
results obtained in the aldoxime series. The cyclization likely
proceeds through a conformation at the transition state closely
resembling to A (Scheme 7), the ethyl group inducing little steric
effects. This unexpected result is noteworthy as it demonstrates
that the cyclization onto the ketoxime is still favored relative
to the iodine atom transfer. The same reaction carried out using
the corresponding xanthate (Scheme 2) led to the same product
albeit in much lower yield (8%). Interestingly, when only one
equivalent of 13 was employed, cyclopentane 33 was obtained
In an effort to rationalize the [2 + 2 + 2] process as well as
the unexpected formation of diester 33, we proposed the
mechanistic scheme below (Scheme 9). The electrophilic radical
species i is known to be formed through abstraction of iodine
by an ethyl radical issued from the reaction of Et3B with oxygen.
i then regioselectively adds on the less substituted end of the
electron-rich olefinic system to produce nucleophilic radical ii
(ꢀ to silicon).24 The oxime functionality in ii is appropriately
located for the radical species to cyclize in a 5-exo-trig fashion,
providing alkoxyaminyl radical intermediate iii. This nitrogen
centered radical can then react with Et3B and form via a polar
crossover process,20 amido-borane complex iW which is able
to lactamize and produce piperidinone W. While this pathway is
highly efficient with aldoximes (R′ ) H), it is not observed in
the case of ketoximes. Among possible hypotheses that could
be envisioned to rationalize this unexpected behavior, we
reasoned that diester Wi (e.g., 33) could result either from a SN2
displacement of iodide 13 by amido-borane intermediate iW (R′
* H) or through a direct radical recombination of alkoxyaminyl
radical iii (R′ * H) and radical i. Both pathways were thus
investigated. For instance, the radical reaction was carried out
as described above on 32, except that water was used as a
solvent. This led to the unique formation of diester 33. Similar
results were obtained when additives such as AcOH (1-5 equiv)
were employed. Strikingly, the reaction can even be conducted
in 1 N HCl or in concd H2SO4 and still leads to the exclusive
formation of 33 (albeit in lower chemical efficiency). Electro-
philes such as (Boc)2O and trifluoroacetic anhydride were also
tested to trap the putative amidoborane iW, but 33 was invariably
produced as a unique product. These experiments clearly rule
out the formation of the C-N bond and the incorporation of
the second ester fragment through an ionic pathway. Therefore,
we next examined the alternative radical recombination pathway.
We envisioned that significant steric hindrance around the
nitrogen atom might prevent the approach of Et3B and thus
(36) (a) Fleming, I. Chemtracts: Org. Chem. 1996, 9, 1–64. (b) Tamao, K.
In AdVances in Silicon Chemistry; JAI Press Inc.: New York, 1996; Vol. 1, pp
1-62. (c) Jones, G. R.; Landais, Y. Tetrahedron 1996, 52, 7599–7662.
(37) For a recent study on a free-radical addition-6-exo-trig cascade, see:
Fernandez-Gonzalez, M.; Alonso, R. J. Org. Chem. 2006, 71, 6767–6775.
6988 J. Org. Chem. Vol. 73, No. 18, 2008

Synthesis of Fused Piperidinones
SCHEME 9. Mechanistic Proposal for the [2 + 2 + 2] Radical Cascade
SCHEME 10. Persistent Alkoxyaminyl Radicals
SCHEME 11. Unsaturated Ketoximes 37-39
prohibit the formation of amido-borane complex iW. Precedents
from the literature effectively reveal that alkoxyaminyl radicals,
such as 35, are persistent due to steric shielding (Scheme 10).38
In our case, intermediate radical 36 may fit into this category
and therefore possess a long enough lifetime to recombine with
an excess of radical precursor i arising from R-iodoester 13.
Several experiments were designed to reduce the alkoxyaminyl
radical intermediate iii (R′ * H) under free-radical conditions.
For instance, addition of ethyl xanthate in isopropanol as both
a solvent and a reducing agent, according to Zard’s procedure,39
led to 33 in low yield (8%) along with recovered starting
material (78%). No trace of the desired reduced product was
observed. The use of Roberts’s polarity-reversal catalysis40
(Ph3SiH, HSCH2CO2Et (cat.), BrCH2CO2Et) led in turn to a
complex mixture of products in which the desired product was
absent. Finally, substitution of Et3B for indium41 in a MeOH--
H2O mixture led to recovered allylsilane 32 and ethyl acetate
resulting from complete reduction of 13.42
SCHEME 12. Radical Cascade and Recombination Step of
Ketoximes 37 and 38
In parallel, we envisaged being able to trap the alkoxyaminyl
radical (i.e., iii) in an intramolecular fashion through the
introduction of an olefinic appendage. Recent studies by
Jaramillo-Gomez et al.43 effectively showed that a 5-exo-trig
cyclization onto a ketoxime, followed by a second 5-exo-trig
addition of an alkoxyaminyl radical onto an unsaturated Michael
acceptor, was feasible, leading to a good yield of the desired
bicyclic system. Three different precursors 37, 38, and 39 were
thus designed (Scheme 11), possessing electron-poor and
electron-rich olefins that could trap the alkoxyaminyl radical
in a 5-exo-trig and 6-exo-trig fashion, respectively (see the
Supporting Information for preparation).
Ketoximes 37-39 were then submitted to the radical addition
of R-iodoester 13. Under the conditions described above, 37-38
led to the corresponding recombination products 40 and 41 after
radical addition and 5-exo-trig cyclizations. Unfortunately, no
traces of the desired spiro-bicylic systems were detected. Both
40 and 41 were isolated as single diastereomers with the
stereochemistry as shown (Scheme 12). In sharp contrast,
precursor 39 led to a complex mixture where none of the
expected products (recombination or cyclized) could be identi-
fied, presumably due to the instability of the enol ether
functionality under the radical conditions. This last series of
experiments demonstrates that the generation of the
N-CH2CO2Et bond is faster than the intramolecular addition
of the intermediate alkoxyaminyl radical on the Michael acceptor
appendage.
(38) Bella, A. F.; Slawin, A. M. Z.; Walton, J. C. J. Org. Chem. 2004, 69,
5926–5933.
(39) Boiteau, L.; Boivin, J.; Liard, A.; Quiclet-Sire, B.; Zard, S. Z. Angew.
Chem., Int. Ed. 1998, 37, 1128–1131.
(40) Roberts, B. P. Chem. Soc. ReV. 1999, 28, 25–35.
(41) (a) Miyabe, H.; Nishimura, A.; Ueda, M.; Naito, T. Chem. Commun.
2002, 1454–5. (b) Miyabe, H.; Ueda, M.; Nishimura, A.; Naito, T. Tetrahedron
2004, 60, 4227–4235. (c) Miyabe, H.; Ueda, M.; Nishimura, A.; Naito, T. Org.
Lett. 2002, 4, 131–134. (d) Bernardi, L.; Cere, V.; Femoni, C.; Pollicino, S.;
Ricci, A. J. Org. Chem. 2003, 68, 3348–3351. (e) Pitts, M. R.; Harrison, J. R.;
Moody, C. J. J. Chem. Soc., Perkin Trans. 1 2001, 955–977.
(42) Ranu, B. C.; Dutta, P.; Sarkar, A. J. Chem. Soc., Perkin Trans. 1 1999,
1139–1140.
(43) Jaramillo-Gomez, L. M.; Loaiza, A. E.; Martin, J.; Rios, L. A.; Wang,
P. G. Tetrahedron Lett. 2006, 47, 3909–3912.
J. Org. Chem. Vol. 73, No. 18, 2008 6989

Godineau et al.
a valuable new multicomponent reaction.44 It implies the
generation of two C-C and one C-N bonds and relies on the
interruption of the [2 + 2 + 2]-process by an unexpected
recombination step.
TABLE 3. Radical Cascade and Recombination Step
Conclusion
We described along these lines a stereocontrolled access to
five-membered ring fused piperidinones through a formal [2 +
2 + 2] process involving a cascade of radical steps, followed
by a radical polar crossover process and concluded by an ionic
lactamization. Excellent yields and stereoselectivities were
generally observed starting from enoximes bearing a chiral
allylsilane moiety, thus demonstrating the beneficial role of an
allylic silicon group regarding the chemical reactivity and the
stereocontrol (1,2 and 1,3). Extension of the cascade to the
radical addition onto ketoximes was not followed by lactam-
ization but instead by the recombination of the resulting
alkoxyaminyl radical with the radical species issued from the
starting R-iodoester. We showed that such a recombination could
be used to devise a three-component process where two different
radicals are added on the enoxime precursor, providing a
cyclopentane bearing three contiguous stereocenters and a
benzylamine appendage. Further development of this type of
multicomponent radical reaction is currently actively being
pursued in our laboratory.
entry
13 (equiv)
42 (equiv)
32:33:43a
b
1
2
3
4
5
6
7
0
4
2.5
2.5
2.5
1.5
4
-
2.5
2.5
2.5
2.5
1.5
1.5
50:28:22c
5:46:49d
5:38:57e
25:47:28e
30:14:56e
80:1:19e
10
a Estimated ratio through 1H NMR of the crude reaction mixture. b 13
and 42 in solution in CH2Cl2, Et3B (2 equiv) and air are successively
added. c 42 was added slowly through a syringe pump. d Et3B was added
slowly through a syringe pump. e 13 and 42 were added slowly through
a syringe pump.
In order to firmly establish the occurrence of a radical
recombination pathway and propose a useful alternative to this
process, it was envisioned that introduction in the medium of a
second radical source might enable crossover recombination.
These radicals would, however, be required to (1) be unreactive
toward the olefin fragment (polar effect), (2) possess a sufficient
lifetime for recombination with 36, and (3) be easily generated
from ethyl radicals arising from Et3B. Benzyl radical was thus
elected as a good candidate. Hence, recombination of benzyl
radical and 36 would provide the corresponding benzylamine
which would be a versatile intermediate en route to the desired
piperidinones. The radical reaction was thus conducted in the
presence of benzyl iodide. Results are summarized in Table 3.
We first performed a control experiment and showed that benzyl
radicals alone did not add onto the olefin (Table 3, entry 1). In
the presence of various amounts of benzyl iodide 42, benzyl-
recombined cyclopentane 43 was effectively produced. When
both radical precursors were introduced in equimolar amounts
in the medium, 33 and 43 were generated in a 1:1 ratio (entry
2). Better yields of the benzyl derivative 43 were obtained
through the syringe pump introduction of both 13 and 42 (entries
4-7). In line with this observation, larger amounts of benzyl
amine 43 were formed using 4-10 equiv of benzyl iodide 42
(entries 6 and 7). Recombination of radicals being a fast process,
the competing radical recombination between 36 and i and
between 36 and benzyl radicals is probably a direct reflection
of the relative amounts of radical i and benzyl radical present
in the medium. The linear correlation between 13/42 and 33/
43 ratio (Table 3) argues in this direction. The ionic pathway
was finally excluded on the basis of a last experiment where
highly nucleophilic secondary alkoxyamine 27 prepared above
was treated with an excess of benzyl iodide 42 (2.8 equiv) in
CD2Cl2 at room temperature in the absence of a base. Under
these conditions, which closely match those of the radical
process depicted in Table 3, no trace of the corresponding
benzylamine could be detected even after 20 h (see the
Supporting Information), further discounting a SN2 process.
Taken altogether, these series of experiments clearly suggest
that recombination of radical species 36 and i is the likely
pathway leading to diester 33. Interestingly, benzyl amine 43
can be regarded as the result of a three-component one-pot
radical addition-5-exo-trig cyclization-recombination process.
This radical cascade is noteworthy and might be considered as
Experimental Section
Acetic Acid 2-(Dimethylphenylsilanyl)-4-methoxyamino-3-phe-
nylsulfanylmethylcyclopentyl Ester (12a). To a solution of allyl-
silane oxime 6 (100 mg, 0.31 mmol, 1 equiv) in degassed toluene
(3 mL) was added thiophenol (62 µL, 0.62 mmol, 2 equiv). The
solution was thermostated at 25 °C, and the reaction mixture was
irradiated with a sun lamp for 6 h. TLC then indicated complete
consumption of the starting material. Solvent was then removed in
vacuo. Flash chromatography (silica gel, 80:20 f 75:25 petroleum
pther/EtOAc) afforded the title compound 12a as a yellow solid as
a 88:12 mixture with its diastereomer 12b (125 mg, 95% combined
yield): IR (neat, NaCl) νmax (cm-1) 2938, 1743, 1584, 1427, 1372,
1247, 1112, 1023; 1H NMR (300 MHz, CDCl3) δ (ppm) 7.47-7.51
(m, 2H), 7.35-7.35 (m, 3H), 7.11-7.24 (m, 5H), 5.95 (bs, 1H),
5.41 (dt, J ) 3.0, 6.4 Hz, 1H), 3.68 (q, J ) 6.4 Hz, 1H), 3.50 (s,
3H), 2.85 (dd, J ) 4.5, 12.4 Hz, 1H), 2.68 (dd, J ) 10.6, 12.4 Hz,
1H), 2.40-2.52 (m, 1H), 2.20-2.3 (m, 1H), 1.91 (s, 3H), 1.73-1.87
(m, 2H), 0.33 (s, 3H), 0.30 (s, 3H); 13C NMR (75 MHz, CDCl3) δ
(ppm) 170.2, 138.6, 136.3, 133.6, 129.1, 128.9, 127.9, 126.0, 78.6,
62.3, 61.2, 43.0, 38.8, 36.1, 34.0, 21.2, -2.6, -3.1; HRMS (LSIMS)
calcd for C23H32NO3SSi [M + H+] 430.1872, found 430.1860.
General Procedure for “One Pot” Radical Addition-Ionic
Lactamization Processes. To a solution of oxime (1 equiv) in
CH2Cl2 (0.1 M) (not degassed) under an N2 atmosphere was added
the desired R-iodoester (see text for number of equivalents). The
resulting mixture was cooled to the indicated temperature (generally
-20 °C), and Et3B (1.0 m in hexane, see text for number of
equivalents) was added. A balloon filled with O2 was then adapted
above the flask, and the mixture was stirred at the same temperature
until TLC indicated complete consumption of the starting oxime.
In the case of nonsilylated oxime, completion of the reaction often
required additional Et3B injections (via portions of 2 equiv directly
in the solution to avoid premature reaction with the O2 atmosphere,
1.0 M in hexane, see text for total number of equivalents) every
(44) (a) Multicomponent Reactions; Bienayme´, H., Zhu, J., Eds.; Wiley-VCH
Verlag GmbH & Co. KGaA: Stuttgart, 2005. (b) Bienayme, H.; Hulme, C.;
Oddon, G.; Schmitt, P. Chem.sEur. J. 2000, 6, 3321–3329. (c) Godineau, E.;
Landais, Y. J. Am. Chem. Soc. 2007, 129, 12662–12663.
6990 J. Org. Chem. Vol. 73, No. 18, 2008

Synthesis of Fused Piperidinones
3 h until no starting material could be identified by TLC. Volatile
materials were then removed in vacuo, and the resulting crude oil
was purified by flash chromatography (silica gel).
62.0, 61.9, 40.2, 39.7, 36.6, 31.4, 25.4, 21.3, -3.0, -3.2; HRMS
(LSIMS) calcd for C19H27O4NSiNa [M + Na+] 384.1607, found
384.1602.
5-(Dimethylphenylsilanyl)-1-methoxy-6-methyloctahydro[1]pyri-
din-2-one (16a). Prepared according to the general procedure
described above from allylsilane oxime 4 (0.100 g, 0.38 mmol, 1
equiv), phenyl iodoacetate 18 (0.199 g, 0.76 mmol, 2 equiv), and
triethylborane (1.0 M in hexane 0.76 mL, 0.76 mmol, 2 equiv) in
CH2Cl2 (3.8 mL) at -20 °C. Flash chromatography (silica gel, 80:
20 CH2Cl2/AcOEt) afforded title compound 16a as a yellow oil
(0.087 g, 76%): Rf ) 0.25 (silica gel, 80:20 CH2Cl2/EtOAc); IR
(neat) νmax (cm-1) 2952, 2867, 1673, 1427, 1348, 1249, 1113, 835,
5-(Dimethylphenylsilanyl)-1-methoxy-6-methyloctahydro[1]pyri-
din-2-one (21a). Prepared according to the general procedure
described above from oxime 7 (0.042 g, 0.15 mmol, 1 equiv),
phenyl iodocetate 18 (0.079 g, 0.30 mmol, 2 equiv), and trieth-
ylborane (1.0 M in hexane, 0.30 mL, 0.30 mmol, 2 equiv) in CH2Cl2
(1.5 mL) at -20 °C. Flash chromatography (silica gel, 80:20 f
75:25 CH2Cl2/EtOAc) yielded lactam 21a (0.041 g, 85%) as a
colorless oil: Rf ) 0.27 (silica gel, 75:25 CH2Cl2/EtOAc); IR (neat)
1
νmax (cm-1) 3464, 2956, 1668, 1428, 1250, 1112, 811; H NMR
1
701; H NMR (300 MHz, CDCl3) δ (ppm) 7.46-7.51 (m, 2H),
(300 MHz, CDCl3) δ (ppm) 7.49-7.52 (m, 2H), 7.33-7.35 (m,
3H), 4.10 (q, J ) 7.2 Hz, 1H), 3.73 (s, 3H), 2.13-2.55 (m, 4H),
1.75-1.94 (m, 2H), 1.51-1.62 (m, 1H), 1.30-1.41 (m, 2H), 0.92
(d, J ) 7.2 Hz, 3H), 0.35-0.36 (s, 6H); 13C NMR (75 MHz, CDCl3)
δ (ppm) 168.4, 138.7, 133.5, 129.0, 127.8, 62.2, 61.5, 41.5, 39.1,
36.1, 35.0, 30.8, 25.0, 19.3, -2.6, -2.9; MS (FAB+) m/z calcd for
C18H27NO2Si [M + Na+] 340.17, found 340.18, calcd for
C18H27NO2Si [M + H+] 318.19, found 318.20. HRMS (EI) calcd
for C18H27NO2Si [M + H+] 317.1811, found 317.1808.
7.33-7.38 (m, 3H), 3.75-3.80 (m, 4H), 1.08-2.44 (m, 8H), 0.30
(s, 3H), 0.29 (s, 3H); 13C NMR (75 MHz, CDCl3) δ (ppm) 167.6,
137.5, 133.7, 129.3, 127.9, 63.9, 61.6, 40.0, 32.7, 31.4, 29.6, 26.0,
25.7, -4.5, -4.9; HRMS (LSIMS) calcd for C17H24NOSi [M +
H+] 304.1733, found 304.1736.
1-Methoxyoctahydro[1]pyridin-2-one (17a,b). Prepared accord-
ing to the general procedure described above from oxime 8 (0.100
g, 0.79 mmol, 1 equiv), phenyl iodoacetate 18 (0.414 g, 1.6 mmol,
2 equiv), and triethylborane (1.0 M in hexane, 1.6 mL, 1.6 mmol,
2 equiv) in CH2Cl2 (7.9 mL) at -20 °C. Flash chromatography
(silica gel, 85:15 petroleum ether/EtOAc) afforded minor diaste-
reoisomer 17b (0.022 g, 16%) and major diastereomer 17a (0.070
g, 52%). Major diastereomer 17a: Rf ) 0.17 (silica gel, 60:40
CH2Cl2/EtOAc); IR (neat) νmax (cm-1) 3460 (bs), 2940, 1652, 1455,
1414, 1347, 1192, 1081, 974; 1H NMR (CDCl3, 250 MHz) δ (ppm)
3.97 (q, J ) 5.7 Hz, 1H), 3.77 (s, 3H), 2.29-2.48 (m, 3H),
1.40-2.04 (m, 8H); 13C NMR (75 MHz, CDCl3) δ (ppm) 167.4,
62.8, 61.5, 38.0, 31.7, 31.2, 29.8, 24.0, 23.0; HRMS (EI) calcd for
C9H15NO2 [M+] 169.1103, found 169.1105. Minor diastereomer
17b: Rf ) 0.35 (silica gel, 60:40 CH2Cl2/EtOAc); IR (neat) νmax
(cm-1) 2939, 1660, 1460, 1351, 1200, 1044, 991; 1H NMR (CDCl3,
250 MHz) δ (ppm) 3.75 (s, 3H), 3.26 (dt, J ) 6.7, 11.0 Hz, 1H),
2.50-2.56 (m, 2H), 2.06-2.16 (m, 1H), 1.23-1.95 (m, 8H); 13C
NMR (75 MHz, CDCl3) δ (ppm) 171.0, 67.0, 62.5, 44.8, 33.9, 29.2,
27.4, 25.6, 21.0; HRMS (EI) calcd for C9H15NO2 [M+] 169.1103,
found 169.1105.
1-Methoxy-5-methyloctahydro[1]pyridin-2-one. Major Diaste-
reomer (22a). Prepared according to the general procedure described
above from oxime 9 (0.120 g, 0.85 mmol, 1 equiv), phenyl
iodoacetate 18 (0.442 g, 1.68 mmol, 2 equiv), and triethylborane
(1.0 m in hexane, 1.7 mL, 1.7 mmol, 2 equiv) in CH2Cl2 (3.8 mL)
at -20 °C. Additional triethylborane (1.0 m in hexane, 3.4 mL,
3.4 mmol, 4 equiv) was required to completely consume all starting
materials. Flash chromatography (silica gel, 90:10 f 80:20 CH2Cl2/
EtOAc) afforded major diastereomer 22a (0.068 g, 44%) and a 9:1
mixture of 2 other inseparable diastereomers 22b and 22c (0.012
g, 8%). Major diastereomer 22a: Rf ) 0.75 (silica gel, 80:20 CH2Cl2/
EtOAc); IR (neat) νmax (cm-1) 2952, 1652, 1456, 1418, 1349, 1072,
1
976, 920; H NMR (CDCl3, 250 MHz) δ (ppm) 4.04 (q, J ) 7.0
Hz, 1H), 3.75 (s, 3H), 2.32-2.43 (m, 2H), 2.04-2.19 (m, 1H),
1.59-1.98 (m, 5H), 1.08-1.27 (m, 2H), 1.01 (d, J ) 6.1 Hz, 3H);
13C NMR (CDCl3, 75 MHz) δ (ppm) 167.7, 62.5, 61.6, 46.1, 37.5,
32.5, 31.8, 31.1, 22.7, 19.8; MS (EI) m/z calcd for C10H17NO2 [M+]
183.2, found 183.2; HRMS (EI) calcd for C10H17NO2 [M+]
183.1259, found 183.1239. Minor diastereoisomers 22b and 22c
were isolated as a mixture of two diastereoisomers in a 90:10 ratio
as determined by GC-MS analysis. Relative stereochemistry was
not determined: Rf ) 0.39 (silica gel, 90:10 CH2Cl2/EtOAc); IR
(neat) νmax (cm-1) 2952, 1689, 1462, 1374, 1333, 1203, 1109, 1045;
1H NMR (CDCl3, 250 MHz) δ (ppm) 3.59-3.80 (s, 3H), 3.30-3.43
(m, 1H), 2.48-2.62 (m, 2H), 1.85-2.15 (m, 3H), 1.59-1.80 (m,
2H), 1.35-1.54 (m, 3H), 1.03-1.05 (m, 0.9 × 3H), 0.86-0.89
(m, 0.1 × 3H); 13C NMR (CDCl3, 75 MHz) (only diastereomer
22b signals are reported) δ (ppm) 171.2, 66.8, 63.0, 51.8, 35.7,
34.1, 30.7, 28.1, 24.3, 19.0; MS (EI) m/z calcd for C10H17NO2 [M+]
183.3, found 183.2.
5-(Dimethylphenylsilanyl)-1-methoxy-2-oxooctahydro[1]pyri-
din-6-carboxylic Acid Methyl Ester (19a). Prepared according to
the general procedure described above from oxime 5 (0.029 g, 0.09
mmol, 1 equiv), phenyl iodoacetate 18 (0.047 g, 0.18 mmol, 2
equiv), and triethylborane (1.0 M in hexane, 0.18 mL, 0.18 mmol,
2 equiv) in CH2Cl2 (1.8 mL) at -20 °C. Flash chromatography
(silica gel, 90:10 f 60:40 CH2Cl2/EtOAc) yielded title compound
19a (0.020 g, 61%) as a colorless oil: Rf ) 0.39 (silica gel, 70:30
CH2Cl2/EtOAc); IR (neat) νmax (cm-1) 2956, 1723, 1668, 1428,
1
1113; H NMR (300 MHz, CDCl3) δ (ppm) 7.48-7.51 (m, 2H),
7.35-7.39 (m, 3H), 3.65-3.73 (m, 1H), 3.71 (s, 3H), 3.56 (s, 3H),
2.67 (dt, J ) 6.8, 10.6 Hz, 1H), 2.19-2.47 (m, 3H), 2.02 (q, J )
10.2 Hz, 1H), 1.54-1.71 (m, 4H), 0.32 (s, 6H); 13C NMR (75 MHz,
CDCl3) δ (ppm) 175.5, 167.5, 136.4, 133.9, 129.6, 128.1, 63.3,
62.0, 52.0, 44.1, 40.3, 37.2, 33.5, 31.9, 26.3, -4.5, -4.9; HRMS
(EI) calcd for C19H27NO4Si [M+] 361.1709, found 361.1701.
Acetic Acid 5-(Dimethylphenylsilanyl)-1-methoxy-2-oxoocta-
hydro[1]pyridin-6-yl Ester (20a). Prepared according to the general
method described above from oxime 6 (0.120 g, 0.37 mmol, 1
equiv), phenyl iodoacetate 18 (0.19 g, 0.74 mmol), and triethylbo-
rane (1.0 M in hexane, 0.74 mL, 0.74 mmol) in CH2Cl2 (3.7 mL).
Flash chromatography (silica gel, 80:20 f 60:40 CH2Cl2/EtOAc)
afforded title compound 20a as a yellow oil (0.105 g, 78%): Rf )
0.36 (silica gel, 70:30 CH2Cl2/EtOAc); IR (neat) νmax (cm-1) 2936,
1736, 1672, 1427, 1372, 1234, 1112; 1H NMR (300 MHz, CDCl3)
δ (ppm) 7.47-7.51 (m, 2H), 7.35-7.36 (m, 3H), 5.38 (t, J ) 4.9
Hz, 1H), 4.13 (q, J ) 6.4 Hz, 1H), 3.74 (s, 3H), 2.50-2.61 (m,
1H), 2.21-2.34 (m, 3H), 2.01-2.28 (m, 1H), 1.95 (s, 3H),
1.51-1.55 (m, 3H), 0.39 (s, 3H), 0.35 (s, 3H); 13C NMR (75 MHz,
CDCl3) δ (ppm) 170.1, 167.7, 137.8, 133.6, 129.5, 128.1, 78.0,
5-tert-Butyl-1-methoxyoctahydro[1]pyridin-2-one (23a). Prepared
according to the general procedure described above from oxime
10 (0.098 g, 0.53 mmol, 1 equiv), phenyl iodocetate 18 (0.28 g,
1.09 mmol, 2 equiv), and triethylborane (1.0 m in hexane, 1.09
mL, 1.09 mmol, 2 equiv) in CH2Cl2 (3.8 mL) at -20 °C. Additional
triethylborane (1.0 M in hexane, 2.2 mL, 2.2 mmol, 4 equiv) was
required to completely consume all starting materials. Flash
chromatography (silica gel, 80:20 f 65:35 CH2Cl2/EtOAc) yielded
the lactam 23a (0.055 g, 47%) as a colorless oil: Rf ) 0.32 (silica
gel, 80:20 CH2Cl2/EtOAc); IR (neat) νmax (cm-1) 2956, 1672, 1454,
1365, 1201, 1079, 1048; 1H NMR (CDCl3, 250 MHz) δ (ppm) 3.77
(s, 3H) 3.71-3.77 (m, 1H), 2.26-2.48 (m, 2-H), 2.02-2.20 (m,
2H), 1.53-1.85 (m, 5H), 1.32-1.41 (m, 1H), 0.86 (s, 9H); 13C
NMR (CDCl3, 75 MHz) δ (ppm) 167.3, 63.1, 61.8, 55.0, 39.2, 32.7,
32.4, 31.6, 27.3, 26.7, 25.5; HRMS (EI) calcd C13H23NO2 for [M+]
225.1729, found 225.1717.
tert-Butyl 3-(2-(Methoxyamino)cyclopentyl)propanoate (25a,b).
Prepared according to the general procedure described above from
J. Org. Chem. Vol. 73, No. 18, 2008 6991

Godineau et al.
oxime 8 (100 mg, 0.79 mmol, 1 equiv), tert-butyl iodoacetate 24
(573 mg, 2.37 mmol, 3 equiv), and triethylborane (1.0 M in hexane,
2.37 mL, 2.37 mmol, 3 equiv) in CH2Cl2 (7.9 mL) at -20 °C. Flash
chromatography (silica gel, 92:8 f 50:50 petroleum ether/AcOEt)
afforded major diastereomer 25a (26 mg, 14%), a mixture of 25a
and 25b (85 mg, 44%), and minor diastereomer 25b (6 mg, 3%)
(total combined yield, 117 mg, 61%) all as yellow oils. Integration
on the 13C NMR crude spectra showed a 66:34 mixture of 25a/
25b. Major diastereomer 25a: 1H NMR (300 MHz, CDCl3) δ (ppm)
5.42 (bs, 1H), 3.49 (s, 3H), 3.38 (q, J ) 5.5 Hz, 1H), 2.25 (dt, J )
2.0, 8.7 Hz, 2H), 1.65-1.84 (m, 5H), 1.29-1.63 (m, 13H); 13C
NMR (75 MHz, CDCl3) δ (ppm) 173.4, 80.1, 62.9, 61.9, 42.9, 34.8,
29.9, 29.6, 28.2 (3C), 24.5, 22.0. Minor diastereomer 25b: 1H NMR
(300 MHz, CDCl3) δ (ppm) 3.79 (bs, 1H), 3.72 (s, 3H), 3.20-3.26
(m, 1H), 2.28 (t, J ) 7.2 Hz, 2H), 1.40-1.95 (m, 18H); 13C NMR
(75 MHz, CDCl3) δ (ppm) 173.7, 80.7, 67.3, 62.5, 42.3, 34.4, 31.7,
30.0, 28.8, 28.3, 23.8; IR (neat) νmax (cm-1) 2937, 2871, 1729, 1455,
1367, 1256, 1155, 848; HRMS (TOF MS) calcd for C13H25NO3Na
[M + Na+] 266.1726, found 266.1728.
3-[2-(Dimethylphenylsilanyl)-5-methoxyaminocyclopentyl]pro-
pionic Acid tert-Butyl Ester (26). Prepared according to the general
procedure described above from allylsilane oxime 8 (0.100 g, 0.38
mmol, 1 equiv), tert-butyl iodoacetate 24 (0.18 mg, 0.76 mmol, 2
equiv), and triethylborane (1.0 M in hexane, 0.76 mL, 0.76 mmol,
2 equiv) in CH2Cl2 (3.8 mL) at -20 °C. Flash chromatography
(silica gel, 92:8 f 50:50 petroleum ether/AcOEt) afforded the title
compound 26 as a yellow oil (0.144 g, 74%): Rf ) 0.25 (silica gel,
90:10 petroleum ether/EtOAc); IR (neat) νmax (cm-1) 2937, 1727,
1366, 1250, 1149, 830; 1H NMR (300 MHz, CDCl3) δ (ppm)
7.47-7.52 (m, 2H), 7.31-7.35 (m, 3H), 5.85 (bs, 1H), 3.49 (s,
3H), 3.34 (q, J ) 4.9 Hz, 1H), 2.09-2.30 (m, 2H), 1.73-1.96 (m,
2H), 1.40-1.70 (m, 14H), 1.09-1.70 (m, 1H), 0.29 (s, 3H), 0.28
(s, 3H); 13C NMR (75 MHz, CDCl3) δ (ppm) 173.2, 138.5, 133.8,
128.8, 127.6, 79.9, 67.8, 62.9, 62.0, 44.0, 34.3, 29.9, 28.0, 25.2,
24.9, -4.1, -4.4; HRMS (LSIMS) calcd for C21H36NO3Si [M +
H+] 378.2464, found 378.2460.
3-[3-Acetoxy-2-(dimethylphenylsilanyl)-5-methoxyaminocyclo-
pentyl]propionic Acid tert-Butyl Ester (27). Prepared according to
the general procedure described above from allylsilane oxime 6
(0.30 mg, 0.94 mmol, 1 equiv), tert-butyl iodoacetate 24 (0.455 g,
1.88 mmol, 2 equiv), and triethylborane (1.0 M in hexane, 2.7 mL,
2.7 mmol, 6 equiv) in CH2Cl2 (10 mL). Flash chromatography
(silica gel, 85:15 f 30:60 petroleum ether/EtOAc) afforded the
title compound 27 as a yellow oil (0.21 g, 51%) and recovered
starting material (0.112 g, 37%): Rf ) 0.44 (silica gel, 60:40
petroleum ether/EtOAc); IR (neat) νmax (cm-1) 2976, 1732, 1369,
1250, 1151, 1020; 1H NMR (300 MHz, CDCl3) δ (ppm) 7.49-7.51
(m, 2H), 7.26-7.33 (m, 3H), 5.47 (bs, 1H), 5.31 (dt, J ) 3.0, 6.8
Hz, 1H), 3.55 (q, J ) 6.4 Hz, 1H), 3.49 (s, 3H), 2.07-2.25 (m,
4H), 1.85 (s, 3H), 1.74 (dd of ABq, J ) 2.6, 6.4, 14.7 Hz, 1H),
1.48-1.64 (m, 3H), 1.42 (s, 9H), 0.35 (s, 6H); 13C NMR (75 MHz,
CDCl3) δ (ppm) 172.9, 170.2, 139.0, 133.5, 128.9, 127.7, 80.1,
78.2, 61.8, 61.1, 43.5, 39.0, 35.4, 34.1, 28.1, 24.6, 21.1, -2.3, -2.9;
HRMS (TOF ESI) calcd for C23H37NO5SiNa [M + Na+] 458.2339,
found 458.2341.
129.5, 128.0, 112.3 (t, J ) 244.2, 249.2 Hz, 1C), 63.5, 61.8, 35.7,
(t, J ) 22.0 Hz, 1C) 34.0 (dd, J ) 2.7, 6.0 Hz, 1C), 32.1, 30.5,
25.9, -4.7, -5.1; HRMS (TOF ESI) calcd for C17H24F2NO2Si [M
+ H+] 340.1544, found 340.1553.
Acetic Acid 5-(Dimethylphenylsilanyl)-3,3-difluoro-1-methoxy-
2-oxooctahydro[1]pyridin-6-yl Ester (30). Prepared according to the
general procedure described above from allylsilane oxime 6 (0.30
g, 0.94 mmol, 1 equiv), ethyl bromodifluoroacetate 28 (0.24 mL,
1.88 mmol, 2 equiv), and triethylborane (1.0 M in hexane, 2.4 mL,
2.4 mmol, 5 equiv) in CH2Cl2 (10 mL). Flash chromatography
(silica gel, 75:25 f 25:75 petroleum ether/EtOAc) afforded title
compound 30 as a yellow oil (0.24 g, 65%): IR (neat) νmax (cm-1
)
3356, 2958, 1669, 1432, 1253, 1094; 1H NMR (300 MHz, CDCl3)
δ (ppm) 7.47-7.50 (m, 2H), 7.37-7.39 (m, 3H), 5.40 (t, J ) 5.3
Hz, 1H), 4.22 (q, J ) 6.8 Hz, 1H), 3.8 (s, 3H), 2.75 (quint, J ) 7.9
Hz, 1H), 2.40 (d of ABq, J ) 6.4, 13.9 Hz, 1H), 1.91 - 2.09 (m,
6H), 1.56 (t, J ) 6.8 Hz, 1H), 0.42 (s, 3H), 0.39 (s, 3H); 13C NMR
(75 MHz, CDCl3) δ (ppm) 169.8, 158.5 (t, J ) 31.3 Hz, 1C), 136.7,
133.4, 129.7, 128.2, 112.1 (t, J ) 244.4 Hz, 1C), 62.0, 61.4, 39.5,
37.4, 35.3, (t, J ) 22.0 Hz, 1C), 33.5, 21.1, - 3.4, - 3.5; HRMS
(EI) calcd. for C18H22NO4F2Si [M+ - CH3] 382.1286, found
382.1274; HRMS (EI) calcd for C13H20NO4F2Si [M+ - C6H5]
320.1130, found 320.1145.
5-Hydroxy-1-methoxyoctahydro[1]pyridin-2-one (31). To a solu-
tion of silyl lactam 16a (1.66 g, 5.46 mmol, 1 equiv) in AcOH (29
mL) at 5 °C were added sodium acetate (1.34 g, 16.4 mmol, 3
equiv) and potassium bromide (1.17 mg, 9.8 mmol, 1.8 equiv).
Peracetic acid (5.0 mL, 27 mmol, 5 equiv) was then added to the
viscous solution. The resulting orange solution was then allowed
to warm to room temperature. Sodium acetate (4.5 g, 55 mmol, 10
equiv) and peracetic acid (18 mL, 98 mmol, 18 equiv) were then
again added, and the resulting mixture was stirred at room
temperature overnight (ca. 15 h). The orange color disappeared after
a few hours, and saturated aqueous Na2S2O3/NaHCO3 (1:1) was
carefully added. The resulting mixture was concentrated in vacuo.
Purification by flash chromatography (silica gel, 98:2f80:20
CH2Cl2/MeOH) afforded the desired alcohol 31 (0.83 g, 82% yield)
1
as a yellow oil: Rf ) 0.38 (90:10 CH2Cl2/MeOH); H NMR (250
MHz, CDCl3) δ (ppm) 4.23 (q, J ) 7.0 Hz, 1H), 4.11 (q, J ) 3.9
Hz, 1H), 3.74 (s, 3H), 2.01-2.44 (m, 6H), 1.55-1.91 (m, 4H);
13C NMR (250 MHz, CDCl3) δ (ppm) 167.2, 76.3, 61.5, 60.8, 47.4,
37.7, 31.4, 29.4, 21.7; HRMS (EI) calcd for C9H15NO3 [M+]
185.10519, found 185.1052.
3-[5-(Dimethylphenylsilanyl)-2-(ethoxycarbonylmethyl-
methoxyamino)-2-ethylcyclopentyl]propionic Acid Ethyl Ester
(33). Prepared according to the general procedure described above
from allylsilane oxime 32 (580 mg, 2.0 mmol, 1 equiv), ethyl
iodoacetate 13 (0.52 mL, 4.4 mmol, 2.2 equiv), and triethylborane
(1.0 M in hexane, 5.0 mL, 5.0 mmol, 2.5 equiv) in CH2Cl2 (1.5
mL). Flash chromatography (silica gel, 10:1 f 7:1 petroleum ether/
EtOAc) afforded the title compound 33 as a yellow oil (0.927 g,
82%): Rf ) 0.35 (silica gel, 10:1 petroleum ether/EtOAc): IR (neat)
1
νmax (cm-1) 2953, 1732, 1464, 1428, 1373, 1188, 1037; H NMR
(300 MHz, CDCl3) δ (ppm) 7.48 - 7.51 (m, 2H), 7.32-7.34 (m,
3H), 4.18 (q, J ) 7.2 Hz, 2H), 4.07 (q, J ) 7.2 Hz, 2H), 3.55 (m,
5H), 2.22-2.45 (m, 2H), 1.11-1.91 (m, 15H), 0.78-0.90 (m, 4H),
0.30 (s, 3H), 0.69 (s, 3H); 13C NMR (75 MHz, CDCl3) δ (ppm)
173.9, 171.1, 138.5, 133.9, 128.9, 127.7, 75.6, 62.1, 60.6, 60.0,
58.9, 47.6, 32.4, 32.1, 31.8, 28.3, 27.5, 26.2, 14.2, 14.1, 9.3, -3.8;
HRMS (TOF ESI) calcd for C25H41NO5SiNa [M + Na+] 428.2652,
found 428.2656.
5-(Dimethylphenylsilanyl)-3,3-difluoro-1-methoxyoctahydro-
[1]pyridin-2-one (29). Prepared according to the general procedure
described above from allylsilane oxime 4 (0.100 g, 0.38 mmol, 1
equiv), ethyl bromodifluoroacetate 28 (0.154 g, 1.88 mmol, 2 equiv),
and triethylborane (1.0 m in hexane, 2.4 mL, 2.4 mmol, 5 equiv)
in CH2Cl2 (10 mL). Flash chromatography (silica gel, 99:1 f 98:2
CH2Cl2/EtOAc) afforded lactam 29 as a yellow oil (0.069 g, 53%)
and recovered oxime 4 (0.019 g, 71% yield based on recovered
starting material): Rf ) 0.17 (silica gel, 98:2 CH2Cl2/EtOAc); IR
5-[3-(Dimethylphenylsilanyl)-2-(2-ethoxycarbonylethyl)-1-(eth-
oxycarbonylmethylmethoxyamino)cyclopentyl]pent-2-enoic Acid
Ethyl Ester (40). To a solution of oxime 37 (50 mg, 0.13 mmol, 1
equiv) in CH2Cl2 (2 mL) (not degassed) under an N2 atmosphere
was added ethyl iodoacetate 13 (0.046 mL, 0.38 mmol, 3 equiv),
and Et3B (1 M in hexane) was added by portions every 15 min (14
× 0.4 equiv, total of 5.6 equiv, 0.7 mL, 0.7 mmol). The resulting
mixture was then diluted with CH2Cl2 and quenched with saturated
1
(neat) νmax (cm-1) 3398, 1704, 1428, 1250, 1114, 1001; H NMR
(300 MHz, CDCl3) δ (ppm) 7.48-7.51 (m, 2H), 7.36-7.41 (m,
3H), 3.75-3.83 (m, 4H), 2.37-2.48 (m, 1H), 1.68-2.22 (m, 5H),
1.40-1.54 (m, 1H), 1.17-1.26 (m, 1H), 0.33 (s, 6H); 13C NMR
(75 MHz, CDCl3) δ (ppm) 158.4 (t, J ) 30.7 Hz, 1C), 136.6, 133.6,
6992 J. Org. Chem. Vol. 73, No. 18, 2008

Synthesis of Fused Piperidinones
aqueous NaHCO3. Layers were separated, and the organic phase
was washed with saturated aqueous NaCl (1×), dried (MgSO4),
and concentrated in vacuo. Flash chromatography (silica gel, 91:9
f 83:17 petroleum ether/AcOEt) afforded the title compound 40
as a yellow oil (42 mg, 68%): Rf ) 0.15 (91:9 petroleum ether/
EtOAc); 1H NMR (300 MHz, CDCl3) δ (ppm) 7.45-7.48 (m, 2H),
7.31-7.35 (m, 3H), 6.62 (ddd, J ) 6.8, 8.7, 15.5 Hz, 1H), 5.61 (d,
J ) 15.8 Hz, 1H), 4.06-4.23 (m, 6H), 3.54 (s, 3H), 3.51 (s, 2H),
1.15-2.47 (m, 22H), 0.72-0.92 (m, 1H), 0.30 (s, 6H); 13C NMR
(75 MHz, CDCl3) δ (ppm) 173.8, 170.7, 166.7, 149.3, 137.7, 134.0,
129.3, 127.8, 120.8, 74.9, 62.7, 60.6, 60.1, 60.0, 56.6, 48.5, 33.1,
32.6, 32.0, 31.2, 28.1, 27.5, 25.6, 14.3, 14.2, 14.1, - 3.5, - 3.8;
HRMS (TOF ESI) calcd for C30H47NO7SiNa [M + Na+] 584.3020,
found 584.3021.
32 (100 mg, 0.35 mmol, 1 equiv) in CH2Cl2 (3.5 mL) was added
Et3B (1 M in hexane, 3.5 mL, 2.5 mmol, 10 equiv). The septum
was then removed and the flask opened to the air. A solution of
ethyl iodoacetate was added and the mixture stirred for 2 h at room
temperature. TLC analysis indicated incomplete consumption of
the starting material, and additional ethyl iodoacetate 13 (113 mg,
0.52 mmol, 1.5 equiv) and benzyl iodide 42 (381 mg, 1.75 mmol,
5 equiv) were then added slowly via syringe pump over 40 min.
This operation was repeated a third time 2 h later. The mixture
was then concentrated in vacuo. Purification by flash chromatog-
raphy (silica gel, gradient 99:1f80:20 petroleum ether/EtOAc)
yielded recovered starting material (46 mg) and a 1:3 mixture of
silyl cyclopentane 43 and side product PhCH2CH2Ph (bibenzyl) as
a colorless oil. Upon heating of the latter mixture at 70 °C under
vacuum (0.2 mbar) overnight, a pure 43 (46 mg, 28% yield, 58%
yield based on recovered starting material) was obtained as a yellow
oil: Rf ) 0.36 (88:12 petroleum ether/EtOAc): IR (neat, NaCl) νmax
(cm-1) 2957, 1732, 1454, 1428, 1250, 1180, 1040; 1H NMR (300
MHz, CDCl3) δ (ppm) 7.41-7.60 (m, 4H), 7.19-7.40 (m, 6H),
4.12 (q, J ) 7.2 Hz, 2H), 3.90 (s, 2H), 3.10 (s, 3H), 2.24-2.47
(m, 2H), 1.99-2.18 (m, 2H), 1.53-1.86 (m, 5H), 1.19-1.50 (m,
6H), 0.87 (t, J ) 7.5 Hz, 3H), 0.34 (s, 3H), 0.33 (s, 3H); 13C NMR
(75 MHz, CDCl3) δ (ppm) 174.1, 140.0, 138.9, 134.0, 129.5, 129.0,
128.1, 127.8, 126.9, 75.8, 62.9, 60.2, 58.5, 48.0, 37.3, 32.2, 32.1,
28.7, 27.8, 26.5, 14.4, 9.6, -3.5, -3.7; HRMS (TOF) calcd for
C28H41NO3NaSi [M + Na+] 490.2753, found 490.2753.
6-[3-(Dimethylphenylsilanyl)-2-(2-ethoxycarbonylethyl)-1-(ethox-
ycarbonylmethylmethoxyamino)cyclopentyl]-hex-2-enoic Acid Eth-
yl Ester (41). To a solution of oxime 38 (70 mg, 0.174 mmol, 1
equiv) in CH2Cl2 (2 mL) (not degassed) under an N2 atmosphere
was added ethyl iodoacetate 13 (0.054 mL, 0.45 mmol, 4 equiv),
and Et3B (1 M in hexane) was added by portions every 15 min (10
× 0.5 equiv, total of 5 equiv, 0.9 mL, 0.9 mmol). The resulting
mixture was then diluted with CH2Cl2 and quenched with saturated
aqueous NaHCO3. Layers were separated, and the organic phase
was washed with saturated aqueous NaCl (1×), dried (MgSO4),
and concentrated in vacuo. Flash chromatography (silica gel, 91:9
petroleum ether/AcOEt) afforded the title compound 41 as a yellow
1
oil (61 mg, 72%): Rf ) 0.16 (91:9 petroleum ether/EtOAc). H
Acknowledgment. We thank the CNRS and MNERT for
financial support. We gratefully acknowledge J.-C. Lartigue and
C. Vitry for NMR and mass spectrometry experiments and C.
Scha¨fer for preliminary experiments.
NMR (300 MHz, CDCl3) δ (ppm) 7.47-7.49 (m, 2H), 7.32-7.35
(m, 3H), 6.85 (ddd, J ) 6.8, 8.7, 15.5 Hz, 1H), 5.73 (d, J ) 15.5
Hz, 1H), 4.05-4.22 (m, 6H), 3.53 (s, 3H), 3.50 (s, 2H), 2.35-2.46
(m, 1H), 2.13-2.23 (m, 1H), 1.57-1.91 (m, 5H), 1.1-1.6 (m, 17H),
0.85-0.95 (m, 1H), 0.29 (s, 6H); 13C NMR (75 MHz, CDCl3) δ
(ppm) 173.9, 170.8, 166.6, 149.1, 138.1, 134.1, 129.1, 127.8, 121.3,
75.0, 62.5, 60.6, 60.0, 56.7, 48.4, 34.9, 32.9, 32.6, 32.0, 31.2, 28.1,
25.7, 23.2, 14.3, 14.2, 14.1, - 3.5, - 3.7; HRMS (TOF ESI) calcd
for C31H49NO7SiNa [M + Na+] 598.3176, found 598.3179.
3-[2-(Benzylmethoxyamino)-5-(dimethylphenylsilanyl)-2-ethyl-
cyclopentyl]propionic Acid Ethyl Ester (43). To a solution of oxime
Supporting Information Available: Experimental proce-
dures and spectroscopic data for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JO801308J
J. Org. Chem. Vol. 73, No. 18, 2008 6993