Stereoselective synthesis of 2,4,5-trisubstituted piperidines via radical cyclization

Article abstract of DOI:10.1021/jo101631y

A novel approach to 2,4,5-trisubstituted piperidines is reported, involving the 6-exo cyclization of stabilized radicals onto α,β-unsaturated esters. Only two of the four possible diastereoisomers are observed, with diastereomeric ratios ranging from 3:2

Full text of DOI:10.1021/jo101631y

pubs.acs.org/joc
Stereoselective Synthesis of 2,4,5-Trisubstituted Piperidines via
Radical Cyclization
Maria-Eleni Ragoussi,† Stephen M. Walker,† Andrea Piccanello,† Benson M. Kariuki,†,§
Peter N. Horton,‡ Neil Spencer,† and John S. Snaith*,†
‡
†School of Chemistry, The University of Birmingham, Edgbaston, Birmingham B15 2TT, U.K., and EPSRC
National Crystallography Service, School of Chemistry, University of Southampton, Highfield,
Southampton SO17 1BJ, U.K. §Current address: School of Chemistry, Cardiff University, Main Building,
Park Place, Cardiff CF10 3AT, U.K.
j.s.snaith@bham.ac.uk
Received August 18, 2010
A novel approach to 2,4,5-trisubstituted piperidines is reported, involving the 6-exo cyclization of
stabilized radicals onto R,β-unsaturated esters. Only two of the four possible diastereoisomers are
observed, with diastereomeric ratios ranging from 3:2 to 40:1 when the radical stabilizing group is
vinyl or phenyl. Cyclization of a (triethylsilyl)vinyl-stabilized radical gives the corresponding
piperidine radical as a single diastereoisomer that may either be trapped by tributyltin hydride
to afford the 2,4,5-trisubstituted piperidine or undergo a second 5-endo cyclization onto the
(triethylsilyl)vinyl substituent to produce the 3,5,7-trisubstituted octahydro[2]pyrindene as a single
diastereoisomer.
Introduction
Free radical cyclizations are one of the most useful tech-
niques employed in ring construction and have played an
important role in the synthesis of heterocycles6 and natural
products.7 Nevertheless, there are relatively few reports of
radical cyclizations leading to piperidines.8-10
Substituted piperidines occur widely in natural products1
and feature in a number of successful pharmaceuticals, with
the ring system being regarded as a privileged scaffold for
drug discovery.2,3 The biological importance of piperidines
has led to the development of a multitude of synthetic
approaches;4 nevertheless, many piperidine substitution pat-
terns remain difficult to access, a factor that continues to
drive the search for new methodologies.5
(5) (a) Davis, F. A.; Zhang, J.; Li, Y.; Xu, H.; DeBrosse, C. J. Org. Chem.
2005, 70, 5413–5419. (b) Legault, C. Y.; Charette, A. B. J. Am. Chem. Soc.
2005, 127, 8966–8967. (c) Wurz, R. P.; Fu, G. C. J. Am. Chem. Soc. 2005, 127,
12234–12235. (d) Leverett, C. A.; Cassidy, M. P.; Padwa, A. J. Org. Chem.
2006, 71, 8591–8601. (e) Gotchev, D. B.; Comins, D. L. J. Org. Chem. 2006,
71, 9393–9402. (f) Kauffman, G. S.; Watson, P. S.; Nugent, W. A. J. Org.
Chem. 2006, 71, 8975–8977. (g) Cortez, G. A.; Schrock, R. R.; Hoveyda,
A. H. Angew. Chem., Int. Ed. 2006, 46, 4534–4538. (h) Terada, M.;
Machioka, K.; Sorimachi, K. J. Am. Chem. Soc. 2007, 129, 10336–10337.
(i) Heretsch, P.; Rabe, S.; Giannis, A. Org. Lett. 2009, 11, 5410–5412. (j) Zhu,
W.; Mena, M.; Jnoff, E.; Sun, N.; Pasau, P.; Ghosez, L. Angew. Chem., Int.
(1) (a) Findlay, J. A. In The Alkaloids; Brossi, A., Ed.; Academic Press:
London, 1985; Vol. 26, pp 89-18d3. (b) O’Hagan, D. Nat. Prod. Rep. 2000,
17, 435–446.
(2) Watson, P. S.; Jiang, B.; Scott, B. Org. Lett. 2000, 2, 3679–3681.
(3) Constantino, L.; Barlocca, D. Curr. Med. Chem. 2006, 13, 65–85.
(4) For reviews, see: (a) Bailey, P. D.; Millwood, P. A.; Smith, P. D.
Chem. Commun. 1998, 633–640. (b) Laschat, S.; Dickner, T. Synthesis 2000,
1781–1813. (c) Weintraub, P. M.; Sabol, J. S.; Kane, J. M.; Borcherding,
D. R. Tetrahedron 2003, 59, 2953–2989. (d) Buffat, M. G. P. Tetra-
hedron 2004, 60, 1701–1729. (e) Escolano, C.; Amat, M.; Bosch, J.
Chem.;Eur. J. 2006, 12, 8198–8207. (f) Leino, R.; Kallstrom, S. Bioorg.
Med. Chem. 2008, 16, 601–635. (g) De Risi, C.; Fanton, G.; Pollini, G. P.;
Trapella, C.; Valente, F.; Zanirato, V. Tetrahedron: Asymmetry 2008, 19,
131–155.
€
Ed. 2009, 48, 5580–5583. (k) Gnamm, C.; Krauter, C. M.; Brodner, K.;
Helmchen, G. Chem.;Eur. J. 2009, 15, 2050–2054.
(6) For reviews, see: (a) Majumdar, K. C.; Basu, P. K.; Mukhopadhyay,
P. P. Tetrahedron 2005, 61, 10603–10642. (b) Bowman, W. R.; Bridge, C. F.;
Brookes, P. J. Chem. Soc., Perkin Trans. 1 2000, 1–14. (c) Bowman, W. R.;
Fletcher, A. J.; Potts, G. B. S. J. Chem. Soc., Perkin Trans. 1 2002, 2747–
2762. (d) Srikanth, G. S. C.; Castle, S. L. Tetrahedron 2005, 61, 10377–10441.
(7) For reviews, see: (a) Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem.
Rev. 1991, 91, 1237–1286. (b) Bar, G.; Parsons, A. F. Chem. Soc. Rev. 2003,
32, 251–263.
DOI: 10.1021/jo101631y
r
Published on Web 10/11/2010
J. Org. Chem. 2010, 75, 7347–7357 7347
2010 American Chemical Society

Ragoussi et al.
JOCArticle
FIGURE 1. Stereocontrol models for the radical cyclization of 7-substituted-6-aza-8-bromooct-2-enoates to 2,4-disubstituted piperidines.11
We previously reported the radical cyclization of bromides
of the type 1 to give 2,4-disubstituted piperidines that
favored the trans diastereomer 2.11 However, the approach
suffered from an inherently low stereoselectivity (between
3:1 and 6:1), giving significant amounts of the cis diaste-
reomer 3 in which the 2-substituent was equatorial, Figure 1.
In piperidine sulfonamides the 2-substituent shows a
FIGURE 2. Cyclization of the stabilized radical 4, where R0 is a
marked preference to adopt the axial position as a result of
radical-stabilizing group, to afford 2,4,5-trisubstituted piperidine 5.
torsional strain with the sulfonamide.12 We reasoned that the
modest axial:equatorial ratios in this case resulted from the
accordance with the Hammond postulate.14,15 In this
rather early transition states for radical cyclization which
later, more product-like transition state, in which the
allow distortion from the idealized chair.13 This distortion,
forming C-C bond is shorter, the steric interactions
with the forming bond being rather elongated, means that
should be accentuated, hopefully resulting in an enhanced
factors such as the torsional strain which are important in the
stereoselectivity.
product are of less importance in the transition state, and
The cyclization of radicals of the type 4 would generate
other conformers are energetically accessible.
2,4,5-trisubstituted piperidines 5. A number of natural pro-
We hypothesized that a stabilized radical of the type 4,
ducts have 2,4,5-trisubstituted piperidines at their core,
Figure 2, should react through a later transition state, in
where the 5-substituent is a vinyl group or an ethyl group,
including hirsuteine, corynantheine, strychnofoline, and mitra-
gynanine. The quinuclidine ring of quinine may also be
(8) (a) Katritzky, A. R.; Luo, Z.; Fang, Y.; Feng, D.; Ghiviriga, I.
J. Chem. Soc., Perkin Trans. 2 2000, 1375–1380. (b) Koreeda, M.; Wang,
viewed as a piperidine, and in Stork’s 2001 synthesis a 2,4,5-
trisubstituted piperidine was used as a late-stage interme-
diate.16 In addition to natural products, there is pharmaceutical
interest, with a recent report that 2,4,5-trisubstituted piperi-
dines, including those bearing an aryl substituent at C-5, are
highly potent renin inhibitors.17 Stereocontrolled approaches
to these ring systems are therefore of considerable interest.
Y.; Zhang, L. Org. Lett. 2002, 4, 3329–3332. (c) Yoo, S. E.; Yi, K. Y.; Lee,
S. H.; Jeong, N. Synlett 1990, 575–576. (d) Parsons, A. F.; Pettifer, R. M.
J. Chem. Soc., Perkin Trans. 1 1998, 651–660. (e) Ihara, M.; Setsu, F.; Shoda,
M.; Taniguchi, N.; Tokunaga, Y.; Fukumoto, K. J. Org. Chem. 1994, 59,
5317–5323. (f) Lee, E.; Kang, T. S.; Joo, B. J.; Tae, J. S.; Li, K. S.; Chung,
C. K. Tetrahedron Lett. 1995, 36, 417–420. (g) Lee, E.; Jeong, E. J.; Min, S. J.;
Hong, S.; Lim, J.; Kim, S. K.; Kim, H. J.; Choi, B. G.; Koo, K. C. Org. Lett.
2000, 2, 2169–2171. (h) Prevost, N.; Shipman, M. Org. Lett. 2001, 3, 2383–
2385. (i) Takasu, K.; Ohsato, H.; Kuroyanagi, J.; Ihara, M. J. Org. Chem.
2002, 67, 6001–6007.
(9) For examples of formation of piperidines within polycyclic natural
products by radical ring closure, see: (a) Takayama, H.; Watanabe, F.;
Kitajima, M.; Aimi, N. Tetrahedron Lett. 1997, 38, 5307–5310. (b) Kaoudi,
T.; Miranda, L. D.; Zard, S. Z. Org. Lett. 2001, 3, 3125–3127. (c) Quirante, J.;
Escolano, C.; Bosch, J.; Bonjoch, J. J. Chem. Soc., Chem. Commun. 1995,
2141–2142. (d) Ishibashi, H.; Inomata, M.; Ohba, M.; Ikeda, M. Tetrahedron
Lett. 1999, 40, 1149–1152. (e) Kuehne, M. E.; Wang, T.; Seaton, P. J. J. Org.
Chem. 1996, 61, 6001–6008. (f) Quirante, J.; Escolano, C.; Merino, A.;
Bonjoch, J. J. Org. Chem. 1998, 63, 968–976. (g) Kuehne, M. E.; Bandarage,
U. K.; Hammach, A.; Li, Y.-L.; Wang, T. J. Org. Chem. 1998, 63, 2172–2183.
(h) Eichberg, M. J.; Dorta, R. L.; Grotjahn, D. B.; Lamottke, K.; Schmidt,
M.; Vollhardt, K. P. C. J. Am. Chem. Soc. 2001, 123, 9324–9337.
(10) Zard has reported radical Smiles-type rearrangements to assemble
cyclization precursors for piperidine synthesis: Gheorghe, A.; Quiclet-Sire,
B.; Vila, X.; Zard, S. Z. Org. Lett. 2005, 7, 1653–1656.
Results and Discussion
Our investigation started by exploring the use of a phenyl
radical-stabilizing group in our system, which would become
the substituent at position 5 of the piperidine after cyclization
(Figure 3). Our starting materials were the amino alcohols
(S)-alaninol 6a, (S)-phenylalaninol 6b, (S)-valinol 6c, and
(S)-phenylglycinol 6d, affording cyclization precursors with a
range of steric demands at the 2-position.
(11) Gandon, L. A.; Russell, A. G.; Guveli, T.; Brodwolf, A. E.; Kariuki,
B. M.; Spencer, N.; Snaith, J. S. J. Org. Chem. 2006, 71, 5198–5207.
(12) For examples of the phenomenon in N-acyl, N-sulfonyl, and related
piperidines, see: (a) Beak, P.; Zajdel, W. J. J. Am. Chem. Soc. 1984, 106,
1010–1018. (b) Brown, J. D.; Foley, M. A.; Comins, D. L. J. Am. Chem. Soc.
1988, 110, 7445–7447. (c) Heintzelman, G. R.; Weinreb, S. M.; Parvez, M.
J. Org. Chem. 1996, 61, 4594–4599. (d) Lemire, A.; Charette, A. B. Org. Lett.
2005, 7, 2747–2750. (e) Cariou, C. A. M.; Snaith, J. S. Org. Biomol. Chem.
2006, 4, 51–53.
FIGURE 3. Retrosynthetic analysis.
(13) Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry of Radical
Reactions; VCH: Weinheim, 1996; pp 77-78.
(14) Hammond, G. S. J. Am. Chem. Soc. 1955, 77, 334–338.
(15) (a) Fueno, T.; Kamachi, M. Macromolecules 1988, 21, 908–912. (b)
Wong, M. W.; Pross, A.; Radom, L. J. Am. Chem. Soc. 1994, 116, 6284–6292.
(c) Fischer, H.; Radom, L. Angew. Chem., Int. Ed. 2001, 40, 1340–1371.
(16) Stork, G.; Niu, D.; Fujimoto, A; Koft, E. R.; Balkovec, J. M.; Tata,
J. R.; Dake, G. R. J. Am. Chem. Soc. 2001, 123, 3239–3242.
(17) Herold, P.; Mah, R.; Tschinke, V.; Marti, C.; Stutz, S.; Jelakovic, S.;
Konteatis, Z. D.; Ludington, J.; Quirmbach, M.; Stojanovic, A.; Behnke, D.;
Hollinger, F. U.S. Patent 7,687,495, 2010.
7348 J. Org. Chem. Vol. 75, No. 21, 2010

Ragoussi et al.
JOCArticle
SCHEME 1. Synthesis of Phenyl Bromide Precursors
TABLE 1. Radical Cyclizations of Phenyl-Stabilized Precursorsa
The desired bromide cyclization precursors were synthe-
sized in a 6-step procedure. Ditosylation of the amino
alcohols was followed by cyanide displacement of the
O-tosyl group to form amino nitriles 7a-d (Scheme 1).18
N-Alkylation with 2-bromoacetophenone and Cs2CO3 as the
base, giving nitriles 8a-d, was followed by a DIBAL-H reduc-
tion to afford lactols 9a-d as a mixture of diastereoisomers.
These were directly subjected to Wittig olefination to give the
corresponding R,β-unsaturated esters 10a-d in good yields,
chiefly or exclusively as the (E)-diastereoisomer after column
chromatography. In the case of lactol 9d the olefination
proceeded with poor E/Z selectivity; however, the isomers were
separable and were taken forward separately. Subsequent
bromination, using phosphorus tribromide, gave the target
bromides 11a-d in good yields.
entry
bromide
R
12:13b
yield (%)c
1
2
3
4
11a
11b
11c
11d
Me
Bn
i-Pr
Ph
7:3
10:1
40:1
2.5:1
85
68
73
85
aReactions performed by syringe pump addition of benzene or
toluene solutions of Bu3SnH and AIBN to a solution of the bromide
in toluene heated at 90 °C, final concentration typically 0.01-0.015 M.
bRatio determined by HPLC of the crude reaction mixture and/or H
1
With the bromide precursors in hand, we proceeded to
study the radical cyclization. The reactions were performed
in toluene, heated at 90 °C, with slow addition of tributyltin
hydride and AIBN by syringe pump; the results are summar-
ized in Table 1.
NMR after chromatography to remove tin residues. cIsolated yields
following chromatography.
In all cases only two of the four possible diastereoisomers
were produced. The major diastereoisomer was 12, in which
the 2-substituent is axial and the 4- and 5-substituents are
equatorial (Figure 4). While separation could be achieved for
all compounds on an analytical HPLC column, only 12d and
13d were separable by preparative HPLC, and the major
diastereoisomer 12d proved to be crystalline, allowing us to
confirm the stereochemical assignment by single-crystal
X-ray analysis (Figure S1 in Supporting Information). The
other compounds were oils, but in the case of 12b and 12c the
structure of the major diastereoisomer was confirmed by a
combination of NOE experiments and analysis of the cou-
pling constants in the 1H NMR spectra of the diastereoiso-
meric mixtures.
The structure of the separated minor diastereoisomer 13d
was likewise secured from 1H NMR coupling constants and
NOE experiments, although 13d was inseparable from a
small amount of another piperidine that could not be
identified. The structure of the minor isomer from the
cyclization of 11b was more difficult to determine as it was
present in only small amounts, inseparable from the major
FIGURE 4. Structures of major and minor diastereoisomers.
isomer, with a number of resonances overlapping in the 1H
NMR spectrum. However, the 2-substituent was clearly
identified as being axial,19 and it is therefore reasonable to
conclude that the 4-substituent is also axial, as in 13d, since a
significant energetic penalty would be incurred by an axial
phenyl substituent at the 5-position. The same is assumed to
hold true for 13c, although the 40:1 product ratio meant that
resonances from 13c were very small in the 1H NMR
spectrum of the two diastereoisomers. For the cyclization
of 11a the situation was less clear-cut, since resonances from
the two diastereoisomers were heavily overlapped in the 1H
NMR spectrum. Once again it was clear from the 1H NMR
spectrum that the 2-substituent was axial in the major and
minor diastereoisomers. It is not unreasonable to assume
(19) The equatorial proton at position 2 experiences significant deshield-
ing by the tosyl group compared with an axial proton at the same site,
resulting in the equatorial proton coming into resonance up to 1 ppm
downfield of the axial proton. For other examples see ref 11.
(18) Cariou, C. A. M.; Kariuki, B. M.; Snaith, J. S. Org. Biomol. Chem.
2008, 6, 3337–3348.
J. Org. Chem. Vol. 75, No. 21, 2010 7349

Ragoussi et al.
JOCArticle
that this compound would behave in a similar fashion to the
cyclizations by Hanessian,20 the geometry of the acceptor
double bond was found to be an important feature in
controlling the stereoselectivity of the cyclization. The (Z)-
ester experienced greater 1,3-diaxial interactions when it
adopted a pseudo axial position, resulting in preferential
cyclization to the trans product (Figure 6).
1
others, and this, coupled with similarities in the H NMR
spectra from all of the cyclizations, led us to assign the major
and minor diastereoisomers for all cyclizations as 12 and 13,
respectively (Figure 4).
FIGURE 6. Improved selectivity for (Z)-esters in Hanessian’s
system.
To test this idea two (Z)-esters, 14b-c, were prepared
using the Still-Gennari-modified Horner-Wadsworth-
Emmons reaction (Scheme 2).21 The yield for both substrates
was disappointingly low at around 20%, increasing to up to
50% based on recovered starting material. Elimination from
the β-amino aldehyde was a significant competing pathway,
and attempts to optimize the reaction conditions did not lead
to any improvement. In addition, the (Z)-ester 14d was
available from the poorly E/Z-selective Wittig olefination
of lactol 9d (vide supra). Bromination of all compounds with
PBr3 gave the required (Z)-bromides 15b-d.
FIGURE 5. Proposed transition states leading to major and minor
diastereoisomers.
Pleasingly, and in contrast to our earlier work on the 2,4-
disubstituted systems, the 2-substituent is axial in both the
major and minor diastereoisomers, suggesting that the sta-
bilized radicals derived from 11a-d react through later,
more product-like transition states (TS1, Figure 5) in accor-
dance with our original hypothesis. Also, in contrast to our
earlier work, is the observation that the 4-substituent lies
axial in the minor product, something that is likely to result
from a gauche interaction between the R,β-unsaturated ester
and the phenyl substituent, which presumably can be
relieved either by the ester lying axial (TS2) or by cycliza-
tion through a twist boat (TS3). If formation of the minor
diastereoisomer is through TS2, then the increasing 1,3-
diaxial interaction between the ester and the axial 2-sub-
stituent on going from TS2a to TS2c would explain the
increasing major:minor ratio. If the minor isomer is formed
by cyclization through a twist boat (TS3), this is likely to
introduce an element of torsional strain between the 2-sub-
stituent and the sulfonamide, increasing with the size of the
2-substituent from TS3a to TS3c and so disfavoring this
TS, in line with the stereoselectivity trend. Rather more
surprising is the very poor major:minor ratio in the cycliza-
tion of 11d, which on the basis of the steric demand of the
2-substituent would be expected to display the highest
stereoselectivity. It seems unlikely that this product ratio
could result from cyclization through chairlike transition
states, leading us to speculate that the exceptional steric
demands of the phenyl substituent at the 2-position results
in both the major and minor diastereoisomers being
formed through boat-like transition states.
The esters were cyclized under conditions identical to
those used for the (E)-esters; the results are summarized in
Table 2.
Surprisingly, 15b and 15c cyclized with stereoselectivities
essentially identical to those of the corresponding (E)-esters,
something which, in the light of Hanessian’s results, argues
against production of the minor diastereoisomer through
TS2 and adds weight to the twist boat TS3 (Figure 5). For
15d the major:minor ratio decreased slightly (but repro-
ducibly), a result which is incompatible with chairlike transi-
tion states, where the ratio might be expected to increase.
Encouraged by the results of the phenyl-stabilized radical
precursors, our attentions turned to the cyclization of vinyl-
stabilized radicals. The synthesis of this family of com-
pounds resembled the route to the phenyl series described
previously, using nitriles 7a-d as a starting point. Initially,
nitrile 7a was alkylated using allylbromide 16 bearing a
benzoate ester, affording 17. It was hoped that a DIBAL-H
reduction of the nitrile to the aldehyde would result in con-
comitant ester removal, but disappointingly, none of the
desired compound could be isolated from a complex mixture.
Instead the benzoate was cleaved by transesterification with
sodium methoxide and the resulting alcohol reprotected as the
If the minor diastereoisomer results from cyclization
through TS2 it should be possible to disfavor this pathway
by switching to the (Z)-alkene. In earlier work on 6-exo-trig
(20) Hanessian, S.; Dhanoa, D. S.; Beaulieu, P. L. Can. J. Chem. 1987, 65,
1859–1866.
(21) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405–4408.
7350 J. Org. Chem. Vol. 75, No. 21, 2010

Ragoussi et al.
JOCArticle
SCHEME 2. Preparation of Z-Cyclization Precursors 15b,c
SCHEME 3. Synthesis of Vinyl Bromide Precursors
TABLE 2. Cyclizations of (Z)-Esters 15b-da
77% yield as a separable 14:1 E/Z mixture. Concomitant silyl
ether removal and bromination occurred on treatment with
triphenylphosphine and bromine,22 giving the target bromide
19a in 95% yield as an inconsequential 1:1 mixture of E/Z
diastereomers at the allyl double bond (Scheme 3).
The failure of the DIBAL-H reduction of the benzoate
ester led us to examine the corresponding acetate. The
required alkylating agent 20, prepared in two steps from (E)-
but-2-ene-1,4-diol, smoothly alkylated amino nitriles 7b-c to
afford 21b-c. These compounds reacted cleanly with DIBAL-H
to give the aldehyde alcohols 22b-c; Wittig reaction and
bromination proceeded uneventfully to yield the required bro-
mides 19b-c. A Still-Gennari-modified Horner-Wadsworth-
Emmons olefination gave the (Z)-esters in a satisfactory
63-70% yield, and these were brominated as before to give
23b-c. Attempts to alkylate 7d were very low yielding, appar-
ently as a result of the N-alkylated aminonitrile undergoing a
facile β-elimination of the amine to give the fully conjugated
cinnamonitrile, and so this compound was not pursued further.
entry
bromide
R
12:13b
yield (%)c
1
2
3
15b
15c
15d
Bn
i-Pr
Ph
10:1
40:1
1.7:1
94
73
84
aReactions performed by syringe pump addition of toluene solutions
of Bu3SnH and AIBN to a solution of the bromide in toluene heated at
90 °C, final concentration typically 0.01-0.015 M. bRatio determined by
HPLC of the crude reaction mixture and/or 1H NMR after chromatog-
raphy to remove tin residues. cIsolated yields following chromatog-
raphy.
tert-butyldimethylsilyl ether in 80% yield over 2 steps. DIBAL-H
reduction proceeded smoothly to give aldehyde 18 in 90% yield,
and Wittig olefination introduced the R,β-unsaturated ester in
(22) Aizpurua, J. M.; Cossio, F. P.; Palomo, C. J. Org. Chem. 1986, 51,
4941–4943.
J. Org. Chem. Vol. 75, No. 21, 2010 7351

Ragoussi et al.
JOCArticle
TABLE 3. Cyclizations of Vinyl-Stabilized Precursorsa
entry
bromide
R
E or Z
24:25b
yield (%)c
1
2
3
4
5
19a
19b
23b
19c
23c
Me
Bn
Bn
i-Pr
i-Pr
E
E
Z
E
Z
3:2
3:1
3:1
10:1
10:1
76d
72
69
70
71
aReactions performed by syringe pump addition of toluene solutions of Bu3SnH and AIBN to a solution of the bromide in toluene heated at 90 °C,
b
1
final concentration typically 0.01-0.015 M. Ratio determined by HPLC of the crude reaction mixture and/or H NMR after chromatography to
remove tin residues. cIsolated yields following chromatography. dCombined yield of 24a, 25a, and 26.
The radical cyclization was performed under the previously
optimized conditions; the results are summarized in Table 3.
As before, only two of the possible four diastereoisomers
were formed. In the cyclization of 19a, a significant amount
of the inseparable reduction product 26 was also formed.
This was surprising and contrasted with 19b-c and 23b-c
where direct reduction was not observed, but given the poor
stereoselectivity of this substrate, optimization to favor
cyclization over reduction was not attempted.
In each case the two diastereoisomers had such similar
retention times on HPLC that preparative separation was
not possible. Nevertheless, the major and minor diastereoi-
somers were identified by NOE experiments and analysis
of the 1H NMR spectrum of the 3:1 mixture of 24b and 25b.
In the major diastereoisomer the 2-substituent occupied the
expected axial position, while the 4- and 5-substituents were
equatorial; the minor diastereomer differed by having the
5-substitent axial (Figure 7).
chair, forcing the sulfonamide closer to the axial vinyl
substituent and so disfavoring this minor diastereoisomer.
It was apparent that the small size of the vinyl 5-substituent
resulted in only a modest energetic penalty when it occupied
the axial position, thereby eroding the stereoselectivity of the
cyclization. In an attempt to overcome this, we decided to
study cyclization of the corresponding (triethylsilyl)vinyl-
stabilized radical. The (triethylsilyl)vinyl group has signifi-
cantly greater steric demands yet can be viewed as synthetically
equivalent to the vinyl group via protodesilylation.
To test whether the additional bulk of the radical stabiliz-
ing group would have an effect, we prepared the four cyclization
precurors 29b-c and 30b-c, incorporating differing steric
demands at the 2-position, as well as E- and Z-acceptor
double bonds. Alkylation of amino nitriles 7b-c was perfor-
med using allyl chloride 27;23 subsequent DIBAL-H reduc-
tion, Wittig olefination, and bromination proceeded as before
to give the (E)-bromides 29b-c (Scheme 4). The (Z)-bromides
30b-c were prepared, as before, by the Still-Gennari-
modified Horner-Wadsworth-Emmons procedure in a dis-
appointing 27-32% yield, with a 4:1 ratio in favor of the
desired (Z)-olefin. A major side reaction, which could not be
suppressed, was base-induced elimination of the amine from
the β-aminoaldehyde. The Ando-modified Horner-Wadsworth-
Emmons procedure24 fared even worse; under the standard
conditions, using KHMDS as base, a yield of around 20% of the
desired olefins was obtained, in a very disappointing 2:1 E/Z ratio,
while using the less basic DBU and LiCl conditions developed for
base-sensitive aldehydes in the Horner-Wadsworth-Emmons
procedure by Masamune and Roush25 increased the yield a little
to around 30% but left the E/Z ratio unchanged.
FIGURE 7. Transition state models for formation of major and
minor diastereoiosmers 24 and 25.
Cyclization of all four compounds afforded two products
in each case, separable by HPLC, both of which had the same
relative stereochemistry around the piperidine ring (Table 4).
The major products from the cyclizations of 29b and 29c
were the expected piperidines 31b and 31c, Figure 8.
Although the compounds were not crystalline, analysis of
The observation that the 2-substituent was axial in both
the major and minor diastereoisomers suggested that, like
before, cyclization was occurring through a later, more pro-
duct-like transition state. However, stereocontrol at the
5-position was modest. By adopting an axial position the
relatively sterically undemanding vinyl 5-substituent is able
to relieve the gauche interaction with the R,β-unsaturated
ester, at the expense of a 1,3-diaxial interaction with the
3-hydrogen and a steric clash with the aromatic ring of the
sulfonamide. As the 2-substituent increases in size, it is
expected that the conformation will distort from a regular
1
the H NMR spectra provided unambiguous proof of the
relative stereochemistry. The minor products, formed by
(23) Thorimbert, S.; Giambastiani, G.; Commandeur, C.; Vitale, M.;
Poli, G.; Malacria, M. Eur. J. Org. Chem. 2003, 2702–2708.
(24) Ando, K. J. Org. Chem. 1999, 64, 8406–8408.
(25) Blanchette, M. A.; Choy, W.;Davis, J. T.; Essenfeld, A. P.; Masamune,
S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25, 2183–2186.
7352 J. Org. Chem. Vol. 75, No. 21, 2010

Ragoussi et al.
JOCArticle
SCHEME 4. Synthesis of the (Triethylsilyl)vinyl Bromide Precursors
TABLE 4. Cyclizations of the (Triethylsilyl)vinyl-Stabilized Precursorsa
entry
bromide
R
E or Z
31:32b
yield (%)c
1
2
3
4
29b
30b
29c
30c
Bn
Bn
i-Pr
i-Pr
E
Z
E
Z
5:4
1:4
2:1
1:4
70
71
68
72
aReactions performed by syringe pump addition of toluene solutions of Bu3SnH and AIBN to a solution of the bromide in toluene heated at 90 °C,
b
1
final concentration typically 0.01-0.015 M. Ratio determined by HPLC of the crude reaction mixture and/or H NMR after chromatography to
remove tin residues. cIsolated yields of the mixture following chromatography. The compounds were separable by HPLC.
FIGURE 8. Products from the cyclization of the (triethylsilyl)vinyl-stabilized radical precursors, and the proposed tandem 6-exo-trig and
5-endo-trig cyclizations leading to their formation.
tandem 6-exo-trig and 5-endo-trig cyclizations that give
complete control of the four new stereogenic centers, were
the octahydropyrindenes 32b and 32c, the structures of
which were confirmed by NOE studies. Octahydropyrin-
denes are interesting synthetic targets and have been exploited,
among other things, as HIV-1 protease inhibitors.26
have to rotate through 180°, a process which is likely to be
slow relative to radical trapping by tributyltin hydride.
Interestingly, the (Z)-esters gave significantly more of the
bicyclic product than the (E)-esters. It is tempting to suggest
that the steric clash between the ester and the triethylsilyl
group favors TS1 over TS2, Figure 9. The (triethylsilyl)vinyl
group in TS1 is correctly oriented to undergo 5-endo-trig
attack, while in TS2 the bulky (triethylsilyl)vinyl group will
FIGURE 9. Proposed transition states for cyclization of the radi-
cals derived from 30b-c.
Intrigued by the differences between the vinyl- and
(triethylsilyl)vinyl-stabilized radicals, we decided to compare
them in the cyclization of radicals lacking the substituent at
the 2-position to see if simpler systems would also undergo
stereoselective tandem 6-exo-trig and 5-endo-trig cyclizations.
(26) Ghosh, A. K.; Lee, H. Y.; Thompson, W. J.; Culberson, C.; Holloway,
K. M.;McKee, S. P.;Munson, P. M.; Duong, T. T.; Smith, A. M.; Darke, P. L.;
Zugay, J. A.; Emini, E. A.; Schleif, W. A.; Huff, J. R.; Anderson, P. S. J. Med.
Chem. 1994, 37, 1177–1188.
J. Org. Chem. Vol. 75, No. 21, 2010 7353

Ragoussi et al.
JOCArticle
SCHEME 5. Synthesis of Achiral Vinyl-Stabilized and (Triethylsilyl)vinyl-Stabilized Radical Precursors 35 and 37
The required cyclization precursors were readily prepared
from 3-(tosylamino)propan-1-ol 33, Scheme 5. Alkylation
with allyl chlorides 20 or 27 gave 34 and 36, respectively. PCC
oxidation followed by Wittig olefination gave exclusively the
(E)-esters after chromatography. Hydrolysis of the acetate
groups proved troublesome. Use of K2CO3 in methanol27 or
KCN in methanol28 led to extensive fragmentation by removal
of the ester γ-proton and elimination of the sulfonamide.
Fortunately the acetate could be cleanly removed by the use
of guanidine in MeOH or EtOH.29 The reaction in EtOH was
preferred as it gave a higher yield, although under these
conditions the R,β-unsaturated ester also underwent transes-
terification to the ethyl ester. Finally, bromination was straight-
forwardly carried out using carbon tetrabromide and triphe-
nylphosphine to afford bromides 35 and 37 in excellent yields.
Cyclization of 35 and 37 was performed under the condi-
tions used previously. In both cases the reaction took signif-
icantly longer to reach completion than had been the case for
the 2-substituted examples, likely the result of the additional
substituent reducing the entropy of activation.30 Cyclization
of 35 gave a 3:2 inseparable mixture of the two expected
diastereoisomers 38 and 39, Figure 10.
The larger fraction was the cis-fused octahydropyrindene 40
which was crystalline, and single-crystal X-ray analysis con-
firmed the structure (Figure S2, Supporting Information).
The smaller fraction comprised a 1:1 mixture of piperidine 41
and trans-fused octahydropyrindene 42, Figure 11.
FIGURE 11. Products isolated from cyclization of 37.
Thus, in the absence of a 2-substituent, the cyclizations
display an inherently low stereoselectivity, with a significant
fraction cyclizing through transition states that give rise to
the cis products 39 and 40.
In conclusion, we have demonstrated a diastereoselective
radical route to 2,4,5-trisubstituted piperidines. All stabi-
lized radicals underwent 6-exo-trig cyclization to afford the
corresponding piperidines in good yields, with the acyclic
reduction product isolated in only one instance. The major
diastereoisomer in all cases could be accounted for by
cyclization through a chairlike transition state in which the
2-substituent is pseudo axial, thereby minimizing torsional
strain with the sulfonamide, and the 4- and 5-substituents are
pseudo equatorial, although cyclization through a boat-like
transition state is likely in at least some cases. Model reac-
tions with cyclization precursors lacking the 2-substituent
suggested that the inherent stereocontrol between the 4- and
5-positions is low. Inclusion of the 2-substituent gave better
stereocontrol, with the minor diastereoisomer disfavored
either as a result of a 1,3-diaxial interaction between the
2- and 4-substituents or from a repulsive interaction between
the axial 5-substituent and the sulfonamide, the latter
experiencing torsional strain with the 2-substituent. In the
case of the 5-phenyl examples, a minor isomer was obtained
which had an axial 4-susbtituent, whereas for the 5-vinyl
examples the relatively sterically undemanding vinyl group
FIGURE 10. The two piperidines isolated from cyclization of 35.
On the other hand, cyclization of 37 gave a 4:1:1 mixture of
three compounds, inseparable by column chromatography.
Purification of a portion by HPLC afforded two fractions.
(27) Robarge, M. J.; Husbands, S. M.; Kieltyka, A.; Brodbeck, R.;
Thurkauf, A.; Newman, A. H. J. Med. Chem. 2001, 44, 3175–3186.
(28) Herzig, J.; Nudelman, A.; Gottlieb, H. E.; Fischer, B. J. Org. Chem.
1986, 51, 727–730.
(29) Kunesch, N.; Miet, C.; Poisson, J. Tetrahedron Lett. 1987, 28, 3569–
3572.
(30) Ingold, C. K. J. Chem. Soc. 1921, 119, 305–329.
7354 J. Org. Chem. Vol. 75, No. 21, 2010

Ragoussi et al.
JOCArticle
occupied the axial position in the minor diastereomer.
Exchanging the vinyl group for the more bulky (triethysilyl)-
vinyl substituent led solely to isolation of products in which
the 2-substituent was axial and the 4- and 5-substituents were
equatorial. As well as the expected piperidine products,
octahydropyrindenes were also isolated as single stereoi-
somers, resulting from an unexpected 5-endo cyclization of
the piperidine radical onto the (triethysilyl)vinyl substituent.
The cyclization of stabilized radicals appears to proceed
through later, more piperidine-like transition states, giving
rise to improved diastereoselectivities over the analogous
unstabilized examples that we have previously reported.
8-hydroxy-5-isopropyl-8-phenyl-6-(p-toluenesulfonyl)oct-2-en-
oate as a mixture of two epimers in a 10:1 ratio as a colorless oil
(0.25 g, 60%): Rf = 0.22; IR (film) 3464, 3028, 2962, 2926, 1718,
1646, 1598, 1494 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.77 (d,
J = 6.6, 3H), 0.98-1.00 (m, 3H major, 6H minor), 1.78-1.92
(m, 1H major, 1H minor), 2.23-2.50 (m, 5H major, 5H minor),
3.14-3.32 (m, 2H major, 2H minor), 3.62-3.70 (m, 4H major,
4H minor), 3.90 (br s, 1H major, 1H minor), 4.93-4.96 (m, 1H
major), 5.04-5.08 (m, 1H minor), 5.63 (d, J = 15.6, 1H major),
5.73 (d, J = 15.6, 1H minor), 6.65-6.75 (m, 1H major),
6.80-6.90 (m, 1H minor), 7.26-7.40 (m, 7H major, 7H minor),
7.71 (d, J = 8.2, 2H major, 2H minor); 13C NMR (75 MHz,
CDCl3, mixture of epimers) δ 20.4, 20.7, 21.4, 31.9, 32.9, 34.5,
35.6, 51.3, 53.0, 64.7, 73.1, 73.4, 122.7, 125.7, 127.4, 127.8, 128.5,
129.6, 129.8, 136.7, 142.0, 143.7, 146.0, 146.2, 166.2; MS (elec-
trospray) m/z 468.2 (100%, [M þ Na]þ). HRMS (electrospray)
calcd for C24H31NO5NaS: 468.1821, found 468.1809.
Experimental Section
(R)-4-Aza-3-isopropyl-6-oxo-6-phenyl-4-(p-toluenesulfonyl)-
hexanenitrile (8c). Cesium carbonate (1.81 g, 5.56 mmol) was
added to a solution of (3R)-(p-toluenesulfonyl)amino-4-methyl-
pentanenitrile (1.14 g, 4.28 mmol) in anhydrous MeCN (15 mL).
The reaction mixture was stirred at ambient temperature for
30 min before a solution of 2-bromoacetophenone (1.02 g, 5.12
mmol) in MeCN (7 mL) was added dropwise. The resulting
mixture was stirred for 18 h before the solvent was removed
in vacuo. The residue was dissolved in ethyl acetate and poured
into water. The aqueous phase was extracted with ethyl acetate,
and the combined organic extracts were washed with brine, dried
over MgSO4, and concentrated to give a pale yellow oil that
was purified by flash column chromatography (toluene/Et2O,
15:1) to afford (R)-4-aza-3-isopropyl-6-oxo-6-phenyl-4-(p-toluene-
sulfonyl)hexanenitrile as a white solid (1.23 g, 75%): Rf = 0.16; mp
159-162 °C; [R]20D = þ112.3 (c 1.0, CHCl3); IR (film) 3020, 2972,
2249, 1706, 1598 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.73 (d, J =
6.6, 2H), 0.95 (d, J=6.6, 2H), 1.92-2.04 (m, 1H), 2.42 (s, 3H), 2.67
(dd, J = 5.1, 17.3, 1H), 2.87 (dd, J = 3.3, 17.3, 1H), 3.52-3.59 (m,
1H), 4.61 (d, J = 18.7, 1H), 5.10 (d, J = 18.7, 1H), 7.31 (d, J = 8.3,
2H), 7.44-7.49 (m, 2H), 7.56-7.61 (m, 1H), 7.77 (d, J = 8.3, 2H),
7.94 (d, J = 8.3, 2H); 13C NMR (75 MHz, CDCl3) δ 19.6, 21.5,
21.8, 29.3, 49.6, 59.0, 117.6, 127.8, 127.9, 128.7, 129.5, 133.6, 134.7,
136.9, 143.8, 194.4; MS (electrospray) m/z 407.1 (100%, [M þ
Na]þ). HRMS (electrospray) calcd for C21H24N2O3NaS: 407.1405,
found 407.1425.
(6R)-5-Aza-1-hydroxy-6-isopropyl-2-oxa-3-phenyl-4-(p-toluene-
sulfonyl)cycloheptane (9c). DIBAL (1.5 M solution in toluene, 1.46
mL, 2.18 mmol) was added dropwise to a solution of nitrile 8c
(0.35 g, 0.91 mmol) in CH2Cl2 (15 mL) under argon at -78 °C.
After stirring at -78 °C for 1 h, MeOH (1 mL) was added, and the
reaction allowed to warm to ambient temperature. H2SO4 (1 M
solution, 10 mL) was added, and the mixture was stirred vigor-
ously for 15 min before the aqueous phase was extracted with
CH2Cl2. The combined organic extracts were washed with brine,
dried over MgSO4, and concentrated to afford the crude lactol
(6R)-5-aza-1-hydroxy-6-isopropyl-2-oxa-3-phenyl-4-(p-toluene-
sulfonyl)cycloheptane as an orange oil (0.32 g, 90%), which was
used without any further purification: IR (film) 3465, 3027, 2967,
2929, 1598, 1494 cm-1; MS (electrospray) m/z 412.1 (100%, [M þ
Na]þ), 444.1 (25%, [M þ Na þ MeOH]þ). HRMS (electrospray)
calcd for C21H27NO4NaS: 412.1559, found 412.1567. The 1H
NMR and the 13C NMR spectra of the sample were complex
due to the presence of four diastereoisomers as well as a small
amount of the open-chain aldehyde.
Methyl (5R,E)-6-Aza-8-bromo-5-isopropyl-8-phenyl-6-(p-
toluenesulfonyl)oct-2-enoate (11c). PBr3 (1 M solution in
CH2Cl2, 0.35 mL, 0.35 mmol) was added dropwise to a
solution of ester 10c (0.13 g, 0.29 mmol) in CH2Cl2 (8 mL)
at 0 °C. The mixture was allowed to warm to room tempera-
ture and was stirred for 1 h, after which it was poured into
water and extracted with CH2Cl2. The combined organic
extracts were washed with brine, dried over MgSO4, and
concentrated before being purified by column chromatogra-
phy (hexane/EtOAc, 4:1) to afford bromide methyl (5R,E)-6-
aza-8-bromo-5-isopropyl-8-phenyl-6-(p-toluenesulfonyl)oct-
2-enoate as a mixture of epimers in a 5:1 ratio as a colorless oil
(0.12 g, 80%): Rf = 0.34; IR (film) 3063, 3028, 2961, 2874,
1721, 1656, 1598, 1494 cm-1; 1H NMR (300 MHz, CDCl3) δ
0.40 (d, J = 6.6, 3H major), 0.68 (d, J = 6.6, 3H minor), 0.73 (d,
J = 6.6, 3H major), 0.86 (d, J = 6.6, 3H minor), 2.04-2.15 (m,
1H major, 1H minor), 2.29-2.43 (m, 5H major, 5H minor),
3.24-3.31 (m, 1H major), 3.34-3.40 (m, 1H minor), 3.58-
3.82 (m, 5H major, 5H minor), 5.36-5.41 (m, 1H minor),
5.47-5.61 (m, 2H major, 1H minor), 6.50-6.60 (m, 1H major,
1H minor), 7.28-7.37 (m, 5H major, 5H minor), 7.44-7.47
(m, 2H major, 2H minor), 7.68 (d, J = 8.3, 2H major, 2H
minor); 13C NMR (75 MHz, CDCl3, mixture of epimers) δ
20.1, 20.3, 20.6, 21.1, 21.5, 32.2, 32.7, 35.1, 51.38, 51.41, 52.4,
52.7, 53.9, 65.3, 122.1, 122.5, 127.7, 128.2, 128.5, 128.7, 128.8,
128.9, 129.6, 129.7, 136.1, 139.4, 143.9, 146.1, 146.3, 166.2;
MS (electrospray) m/z 530.2 (40%, [M(79Br) þ Na]þ), 532.3
(35%, [M (81Br) þ Na]þ), 450.3 (100%, [M þ Na - HBr]þ).
HRMS (electrospray) calcd for C24H30NO4NaS79Br: 530.0977,
found 530.0970.
(2R,4R,5S)-2-Isopropyl-4-(methoxycarbonylmethyl)-5-phenyl-
1-(p-toluenesulfonyl)piperidine (12c) and (2R,4S,5S)-2-Isopropyl-
4-(methoxycarbonylmethyl)-5-phenyl-1-(p-toluenesulfonyl)piperi-
dine (13c). A solution of bromide 11c (0.10 g, 0.20 mmol) in
toluene (5 mL) was deoxygenated by bubbling nitrogen through
it for 15min. The solution was heated to 90°C, and deoxygenated
solutions of Bu3SnH (0.11 g, 0.10 mL, 0.39 mmol) and AIBN
(0.006 g, 0.04 mmol) in toluene (both in 5 mL) were added
simultaneously via syringe pump over 8 h. Stirring was continued
for a further 8 h. The reaction mixture was cooled to room
temperature, concentrated, and purified by column chromatog-
raphy (hexane 100% to remove the tin residues and then toluene/
Et2O, 20:1 to obtain the product) to afford the mixture of
(2R,4R,5S)-2-isopropyl-4-(methoxycarbonylmethyl)-5-phen-
yl-1-(p-toluenesulfonyl)piperidine and (2R,4S,5S)-2-isopropyl-
4-(methoxycarbonylmethyl)-5-phenyl-1-(p-toluenesulfonyl)-
piperidine in a 40:1 ratio, as determined by analytical HPLC, as a
colorless oil (0.063 g, 73%): analytical HPLC (from water/TFA
99.95:0.05 to MeOH/TFA 99.95:0.05 over 60 min) tR = 53.48
min (13c), 54.20 min (12c); Rf = 0.39; IR (film) 3029, 2963, 1730,
Methyl (5R,E)-6-Aza-8-hydroxy-5-isopropyl-8-phenyl-6-(p-
toluenesulfonyl)oct-2-enoate (10c). Methyl (triphenylphopho-
ranylidene)acetate (0.34 g, 1.02 mmol) was added to a solution
of lactol 9c (0.36 g, 0.92 mmol) in CH2Cl2 (15 mL) at room
temperature and was stirred overnight. The solvent was removed
in vacuo, and the residue was purified by column chromatog-
raphy (hexane/EtOAc, 5:1) to give methyl (5R,E)-6-aza-
1599, 1494 cm-1
;
1H NMR (500 MHz, CDCl3, only major
J. Org. Chem. Vol. 75, No. 21, 2010 7355

Ragoussi et al.
JOCArticle
diastereoisomer reported) δ 0.93 (d, J = 6.6, 3H), 0.99 (d, J =
6.6, 3H), 1.12 (dt, J = 5.1, 13.6, 1H), 1.68 (dd, J = 10.0, 15.9,
1H), 1.95-2.00 (m, 1H), 2.03-2.13 (m, 2H), 2.16-2.39 (m, 2H),
2.43 (s, 3H), 3.02 (dd, J = 12.1, 14.9, 1H), 3.50 (s, 3H), 3.67 (dd,
J = 3.6, 9.9, 1H), 3.78 (dd, J = 4.5, 14.9, 1H), 7.06 (d, J = 7.1,
2H), 7.19-7.32 (m, 5H), 7.74 (d, J = 8.2, 2H); 13C NMR (125
MHz, CDCl3, only major diastereoisomer reported) δ 19.7, 20.3,
21.5, 26.8, 31.4, 32.0, 38.3, 46.8, 47.0, 51.4, 59.5, 127.0, 127.3,
127.7, 128.8, 129.6, 138.6, 140.2, 143.0, 172.7; MS (electrospray)
m/z 452.2 (100%, [M þ Na]þ). HRMS (electrospray) calcd for
C24H31NO4NaS: 452.1872, found 452.1863.
(R,E)-8-Acetoxy-4-aza-3-isopropyl-4-(p-toluenesulfonyl)oct-6-
enenitrile. Cs2CO3 (0.61 g, 1.86 mmol) was added to a solution of
nitrile 7c (0.45 g, 1.69 mmol) in MeCN (15 mL) at room
temperature. The resulting suspension was stirred for 30 min
and cooled to 0 °C, and a solution of chloride 20 (0.30 g, 2.02
mmol) in MeCN (10 mL) wasadded dropwise, followed byTBAI
(0.44 g, 1.18 mmol). The mixture was allowed to warm to room
temperature and was stirred for 12 h, after which it was con-
centrated, and the residue was partitioned between water and
ethyl acetate. The combined organic layers were washed with
brine, dried over MgSO4, and concentrated to give a yellow oil
that was purified by column chromatography (EtOAc/hexane,
1:2) to give (R,E)-8-acetoxy-4-aza-3-isopropyl-4-(p-toluenesul-
fonyl)oct-6-enenitrile as a pale yellow oil (0.51 g, 95%): Rf =
0.56; [R]18D = -35.6 (c 1.0, CHCl3); IR (film) 3021, 2969, 2877,
2246, 1737, 1598, 1494 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.80
(d, J = 6.6, 3H), 0.92 (d, J = 6.6, 3H), 1.86-1.99 (m, 4H), 2.35 (s,
3H), 2.48 (dd, J = 4.4, 17.3, 1H), 2.58 (dd, J = 7.4, 17.3, 1H),
3.65-3.72 (m, 1H), 3.86 (dd, J = 5.1, 16.6, 1H), 3.97 (dd, J = 5.5,
16.6, 1H), 4.47-4.62 (m, 2H), 5.49-5.64 (m, 2H), 7.25 (d, J = 7.7,
2H), 7.68 (d, J = 7.7, 2H); 13C NMR (75 MHz, CDCl3) δ 19.8,
20.3, 20.8, 21.2, 21.4, 30.8, 41.4, 59.4, 60.8, 117.8, 126.1, 127.4,
129.7, 130.9, 137.2, 143.7, 170.5; MS (electrospray) m/z 401.1
(100%, [M þ Na]þ). HRMS (electrospray) calcd for C19H26N2-
O4SNa: 401.1511, found 401.1500.
(R,E)-4-Aza-8-hydroxy-3-isopropyl-4-(p-toluenesulfonyl)oct-6-
enal (22c). To a solution of (R,E)-8-acetoxy-4-aza-3-isopropyl-
4-(p-toluenesulfonyl)oct-6-enenitrile (0.50 g, 1.32 mmol) in CH2-
Cl2 (20 mL) at -78 °C was added DIBAL (1 M solution in
toluene, 3.96 mL, 3.96 mmol), and the reaction was stirred at -78 °C
for 2 h before MeOH (2 mL) was added. After warming to room
temperature, H2SO4 1 M (5 mL) was added, and the mixture was
stirred vigorously for 15 min before being extracted with CH2Cl2,
washed with brine, dried over MgSO4, and concentrated to give
(R,E)-4-aza-8-hydroxy-3-isopropyl-4-(p-toluenesulfonyl)oct-6-enal
as a pale yellow oil, which was used without further purification
(0.43 g, 95%): [R]18D=-7.6 (c 1.0, CHCl3); IR (film) 3485, 2963,
2938, 2882, 1721, 1600, 1470 cm-1; 1H NMR (300 MHz, CDCl3) δ
0.78 (d, J = 6.7, 3H), 0.80 (d, J = 6.7, 3H), 1.69-1.81 (m, 1H), 2.37
(s, 3H), 2.64 (dd, J = 6.4, 17.0, 1H), 3.70-4.24 (m, 6H), 5.43-5.53
(m, 1H), 5.65-5.73 (m, 1H), 7.24 (d, J = 8.0, 2H), 7.65 (d, J = 8.0,
2H), 9.53 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 19.8, 20.3, 21.4,
31.9, 41.4, 46.8, 57.8, 59.0, 127.2, 128.6, 129.5, 131.0, 137.4, 143.4,
199.7; MS (electrospray) m/z 362.1 (100%, [M þ Na]þ), 394.2
(15%, [M þ Na þ MeOH]þ). HRMS (electrospray) calcd for
C17H25NO4NaS: 362.1402, found 362.1408.
5H), 2.64 (br s, 1H), 3.57-3.66 (m, 4H), 3.79 (d, J = 4.9, 2H),
4.17 (d, J = 6.2, 2H), 5.41-5.49 (m, 1H), 5.61-5.69 (m, 2H),
6.65-6.76 (m, 1H), 7.21 (d, J = 7.9, 2H), 7.63 (d, J = 7.9, 2H);
13C NMR (75 MHz, CDCl3) δ 20.2, 20.5, 21.4, 31.5, 34.6, 40.8,
51.4, 57.9, 64.0, 122.9, 127.2, 128.9, 129.4, 130.7, 137.9, 143.1,
146.0, 166.3; MS (electrospray) m/z 418.2 (100%, [M þ Na]þ).
HRMS (electrospray) calcd for C20H29NO5SNa: 418.1664,
found 418.1667. Anal. Calcd for C20H29NO5S: C, 60.73; H,
7.39; N, 3.54. Found: C, 60.75; H, 7.23; N, 3.24.
Methyl (R,E,E)-6-Aza-10-bromo-5-isopropyl-6-(p-toluenesulfonyl)-
deca-2,8-dienoate (19c).To a stirred solution of methyl (R,E,E)-6-aza-
10-hydroxy-5-isopropyl-6-(p-toluenesulfonyl)deca-2,8-dienoate (0.12
g, 0.30 mmol) in CH2Cl2 (10mL) at0°C was added PPh3 (0.12 g, 0.45
mmol) followed by CBr4 (0.15 g, 0.45 mmol). The solution was heated
at reflux for 4 h, concentrated, and purified by column chromatog-
raphy (hexane/EtOAc, 4:1) to give methyl (R,E,E)-6-aza-10-bromo-
5-isopropyl-6-(p-toluenesulfonyl)deca-2,8-dienoate as a colorless oil
(0.13 g, 94%): Rf = 0.62; [R]23D = -31.7 (c 0.7, CHCl3); IR (film)
2972, 2959, 1727, 1659, 1598, 1448, 1336, 1158 cm-1; 1H NMR (300
MHz, CDCl3) δ 0.87 (d, J = 6.7, 3H), 0.94 (d, J = 6.7, 3H), 1.74-
1.86 (m, 1H), 2.22-2.54 (m, 5H), 3.65-3.72 (m, 4H), 3.75-3.87 (m,
2H), 3.95 (d, J = 8.5, 2H), 5.48-5.56 (m, 1H), 5.67-5.78 (m, 2H),
6.68-6.78 (m, 1H), 7.23 (d, J = 8.1, 2H), 7.65 (d, J = 8.1, 2H); 13
C
NMR (75 MHz, CDCl3) δ20.4, 20.6, 21.4, 25.7, 31.7, 34.6, 40.3, 51.4,
64.2, 123.0, 126.6, 127.3, 129.5, 132.2, 137.8, 143.2, 145.9, 166.1; MS
(electrospray) m/z 480.1 (100%, [M(79Br) þ Na]þ), 482.1 (90%,
[M(81Br) þ Na]þ), 400.2 (30%, [M þ Na - HBr]þ).HRMS(electro-
spray) calcd for C20H28NO4NaS79Br: 480.0820, found 480.0830.
(2R,4R,5R)-2-Isopropyl-4-(methoxycarbonylmethyl)-1-(p-tol-
uenesulfonyl)-5-vinylpiperidine (24c) and (2R,4R,5S)-2-Isopro-
pyl-4-(methoxycarbonylmethyl)-1-(p-toluenesulfonyl)-5-vinyl-
piperidine (25c). This compound was prepared in the same way
as 12c using bromide 19c (0.11 g, 0.24 mmol) in toluene (5 mL)
and solutions of Bu3SnH (0.14 g, 0.13 mL, 0.48 mmol) and
AIBN (0.008 g, 0.05 mmol) in toluene (both in 5 mL). Purifica-
tion by flash column chromatography (hexane 100% to remove
the tin residues and then toluene/Et2O, 19:1 to obtain the pro-
duct) afforded a mixture of (2R,4R,5R)-2-isopropyl-4-(meth-
oxycarbonylmethyl)-1-(p-toluenesulfonyl)-5-vinylpiperidine
and (2R,4R,5S)-2-isopropyl-4-(methoxycarbonylmethyl)-1-(p-
toluenesulfonyl)-5-vinylpiperidine in a 10:1 ratio, as determined
1
by H NMR, as a colorless oil (0.064 g, 70%): Rf = 0.35; IR
(film) 2956, 2927, 2877, 1734, 1603, 1456 cm-1; 1H NMR (300
MHz, CDCl3, only major diastereoisomer reported) δ 0.88-
1.02 (m, 7H), 1.60-1.67 (m, 1H), 1.72 (dd, J = 9.6, 15.7, 1H),
1.83-1.91 (m, 2H), 2.00-2.11 (m, 1H), 2.37-2.43 (m, 4H), 2.80
(dd, J = 11.8, 14.9, 1H), 3.60-3.74 (m, 5H), 4.99-5.11 (m, 2H),
5.33-5.45 (m, 1H), 7.29 (d, J = 8.3, 2H), 7.71 (d, J = 8.3, 2H);
13C NMR (75 MHz, CDCl3, only major diastereoisomer
reported) δ 19.8, 20.3, 21.6, 26.6, 30.9, 31.5, 38.5, 45.1, 45.2,
51.5, 59.5, 118.5, 127.1, 129.7, 137.4, 138.7, 143.1, 173.0; MS
(electrospray) m/z 402.1 (100%, [M þ Na]þ). HRMS (elec-
trospray) calcd for C20H29NO4NaS: 402.1715, found 402.1711.
(R,E)-8-Acetoxy-4-aza-3-isopropyl-4-(p-toluenesulfonyl)-7-[(tri-
ethyl)silyl]oct-6-enenitrile. Cs2CO3 (3.62 g, 11.11 mmol) was added
to a solution of nitrile 7c (2.70 g, 10.1 mmol) in MeCN (50 mL) at
room temperature. The resulting suspension was stirred for 30 min
and cooled to 0 °C, and a solution of chloride 27 (2.65 g, 12.12
mmol) in MeCN (50 mL) was added dropwise, followed by TBAI
(2.24 g, 6.06 mmol). The mixture was allowed to warm to room
temperature and was stirred for 12 h, after which it was concen-
trated, and the residue was partitioned between water and ethyl
acetate. The combined organic layers were washed with brine,
dried over MgSO4, and concentrated to give a yellow oil that was
purified by column chromatography (EtOAc/hexane, 1:3) to give
(R,E)-8-acetoxy-4-aza-3-isopropyl-4-(p-toluenesulfonyl)-7-[(triethyl)-
silyl]oct-6-enenitrile as a pale yellow oil (4.95 g, 99%): Rf = 0.35;
[R]23D=-24.8 (c 1.0, CHCl3); IR (film) 3021, 2956, 2876, 2247,
Methyl (R,E,E)-6-Aza-10-hydroxy-5-isopropyl-6-(p-toluene-
sulfonyl)deca-2,8-dienoate. Methyl (triphenylphophoranylidene)-
acetate (0.31 g, 0.93 mmol) was added to a solution of aldehyde
22c (0.29 g, 0.85 mmol) in CH2Cl2 (15 mL), and the resulting
solutionwas stirredfor 12 h. The solutionwas concentratedbefore
being purified by column chromatography (hexane/EtOAc, 1:1)
to give methyl (R,E,E)-6-aza-10-hydroxy-5-isopropyl-6-(p-
toluenesulfonyl)deca-2,8-dienoate as a colorless oil (0.20 g,
60%): Rf = 0.35; IR (film) 3433, 3027, 2962, 2875, 1721, 1657,
1
1598 cm-1; H NMR (300 MHz, CDCl3) δ 0.81 (d, J = 6.5,
3H), 0.90 (d, J = 6.5, 3H), 1.69-1.80 (m, 1H), 2.20-2.49 (m,
7356 J. Org. Chem. Vol. 75, No. 21, 2010

Ragoussi et al.
JOCArticle
1732, 1599 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.53 (q, J = 7.7,
6H), 0.85 (t, J = 7.7, 9H), 0.91 (d, J = 6.6, 3H), 0.98 (d, J = 6.6,
3H), 1.88-2.01 (m, 4H), 2.40 (s, 3H), 2.53 (d, J = 5.3, 2H), 3.80-
3.88 (m, 1H), 4.06 (dd, J = 5.7, 9.9, 2H), 4.58 (d, J=12.8, 1H), 4.69
(d, J = 12.8, 1H), 5.74 (t, J = 5.7, 1H), 7.29 (d, J = 8.1, 2H), 7.76
(d, J = 8.1, 2H); 13C NMR (75 MHz, CDCl3) δ 3.0, 7.3, 20.0, 20.5,
21.0, 21.1, 21.6, 31.0, 42.5, 61.2, 62.2, 117.9, 127.7, 129.8, 135.6,
137.7, 141.8, 143.9, 170.7; MS (electrospray) m/z 515.2 (100%,
[M þ Na]þ). HRMS (electrospray) calcd for C25H40N2O4SSiNa:
515.2376, found 515.2382. Anal. Calcd for C25H40N2O4SSi: C,
60.94; H, 8.18; N, 5.69. Found: C, 60.69; H, 8.48; N, 5.54.
(R,E)-4-Aza-8-hydroxy-3-isopropyl-4-(p-toluenesulfonyl)-7-
[(triethyl)silyl]oct-6-enal (28c). Aldehyde 28c was prepared in the
same way as 22c, from (R,E)-8-acetoxy-4-aza-3-isopropyl-4-(p-
toluenesulfonyl)-7-[(triethyl)silyl]oct-6-enenitrile (1.00 g, 2.03
mmol) and DIBAL (1 M solution in toluene, 6.1 mL, 6.1 mmol),
giving (R,E)-4-aza-8-hydroxy-3-isopropyl-4-(p-toluenesulfonyl)-
7-[(triethyl)silyl]oct-6-enal as a pale yellow oil (0.83 g, 90%) that
was used without further purification: IR (film) 3500.1, 3023.0,
2956.3, 2911.2, 2875.0, 1724.1, 1598.8 cm-1; 1H NMR (300 MHz,
CDCl3) δ 0.54 (q, J = 7.7, 6H), 0.78 (t, J = 7.7, 9H), 0.83 (d, J =
6.6, 3H), 0.86 (d, J = 6.6, 3H), 1.67-1.79 (m, 1H), 2.26-2.34 (m,
4H), 2.59 (dd, J = 6.3, 17.3, 1H), 3.77-3.85 (m, 1H), 3.94-4.04
(m, 2H), 4.19 (m, 2H), 5.70 (t, J = 6.2, 1H), 7.21 (d, J = 8.1, 2H),
7.63 (d, J = 8.1, 2H), 9.51 (s, 1H); 13C NMR (75 MHz, CDCl3) δ
3.0, 7.5, 19.9, 20.5, 21.5, 32.1, 42.8, 46.8, 59.2, 60.1, 127.4, 129.7,
137.7, 140.2, 140.4, 143.5, 199.8; MS (electrospray) m/z 508.2
(100%, [M þ MeOH þ Na]þ), 476.1 (25%, [M þ Na]þ). HRMS
(electrospray) calcd for C23H39NO4SSiNa: 476.2267, found
476.2253.
Methyl (R,E,E)-6-Aza-10-hydroxy-5-isopropyl-6-(p-toluene-
sulfonyl)-9-[(triethyl)silyl]deca-2,8-dienoate. This compound
was prepared in the same way as methyl (R,E,E)-6-aza-10-
hydroxy-5-isopropyl-6-(p-toluenesulfonyl)deca-2,8-dienoate, from
aldehyde 28c (0.50 g, 1.10 mmol) and methyl (triphenylpho-
phoranylidene)acetate (0.40 g, 1.21 mmol). Purification by column
chromatography (hexane/EtOAc, 5:2) afforded methyl (R,E,E)-6-aza-
10-hydroxy-5-isopropyl-6-(p-toluenesulfonyl)-9-[(triethyl)silyl]-
deca-2,8-dienoate as a colorless oil (0.34 g, 60%): Rf = 0.31;
[R]20D = þ9.2 (c 1.0, CHCl3); IR (film) 3523, 3022, 2954, 2875,
1720, 1658, 1598 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.55 (q,
J = 7.7, 6H), 0.87 (t, J = 7.7, 9H), 0.93 (d, J = 6.6, 6H), 1.71-
1.83 (m, 1H), 2.00 (s, 1H), 2.17-2.27 (m, 1H), 2.30-2.50 (m,
4H), 3.65-3.73 (m, 4H), 3.95 (t, J = 5.9, 2H), 4.22 (s, 2H),
5.66-5.71 (m, 2H), 6.68-6.78 (m, 1H), 7.22 (d, J = 8.1, 2H),
7.66 (d, J = 8.1, 2H); 13C NMR (75 MHz, CDCl3) δ 3.0, 7.5,
20.4, 20.7, 21.6, 31.8, 34.7, 42.2, 51.6, 60.3, 64.1, 123.1, 127.5,
129.7, 138.3, 140.1, 140.6, 143.3, 146.1, 166.4; MS (electrospray)
m/z 532.2 (100%, [M þ Na]þ). HRMS (electrospray) calcd for
C26H43NO5SSiNa: 532.2529, found 532.2512.
Methyl (R,E,E)-6-Aza-10-bromo-5-isopropyl-6-(p-toluenesul-
fonyl)-9-[(triethyl)silyl]deca-2,8-dienoate (29c). Bromide 29c was
prepared in the same way as 19c, from methyl (R,E,E)-6-aza-10-
hydroxy-5-isopropyl-6-(p-toluenesulfonyl)-9-[(triethyl)silyl]-
deca-2,8-dienoate (0.14 g, 0.27 mmol), PPh3 (0.11 g, 0.41 mmol),
and CBr4 (0.13 g, 0.41 mmol). Purification by column chroma-
tography (hexane/EtOAc, 2:1) afforded methyl (R,E,E)-6-aza-
10-bromo-5-isopropyl-6-(p-toluenesulfonyl)-9-[(triethyl)silyl]-
deca-2,8-dienoate as a colorless oil (0.15 g, 95%): Rf = 0.80;
[R]20D = þ30.4 (c 1.0, CHCl3); IR (film) 2954, 2875, 1724, 1659,
1599 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.55 (q, J = 7.7, 6H),
0.87 (t, J = 7.7, 9H), 0.92 (d, J = 6.6, 3H), 0.95 (d, J = 6.6, 3H),
1.73-1.85 (m, 1H), 2.19-2.29 (m, 1H), 2.38 (s, 3H), 2.44-2.53
(m, 1H), 3.69 (s, 3H), 3.73-3.81 (m, 1H), 3.87-4.07 (m, 4H),
5.67-5.76 (m, 2H), 6.68-6.78 (m, 1H), 7.23 (d, J = 8.1, 2H),
7.66 (d, J = 8.1, 2H); 13C NMR (75 MHz, CDCl3) δ 2.7, 7.3,
20.5, 20.8, 21.6, 28.3, 31.8, 34.4, 41.7, 51.5, 64.4, 123.1, 127.5,
129.7, 135.7, 138.2, 143.3, 145.3, 146.1, 166.2; MS (electrospray)
m/z 596.1 (20%, [M(81Br) þ Na]þ), 594.1 (15%, [M(79Br) þ
Na]þ), 514.2 (100%, [M þ Na - HBr]þ). HRMS (electrospray)
calcd for C26H42NO4SSiNa79Br: 594.1685, found 594.1676.
Anal. Calcd for C26H42NO4SSiBr: C, 54.53; H, 7.39; N, 2.45.
Found: C, 54.48; H, 7.54; N, 2.10.
(2R,4R,5R)-2-Isopropyl-4-(methoxycarbonylmethyl)-1-(p-toluene-
sulfonyl)-5-[1-{(triethyl)silyl}vinyl]piperidine (31c) and (3R,4aR,5S,
7S,7aR)-3-Isopropyl-5-methoxycarbonyl-1-(p-toluenesulfonyl)-
7-[(triethyl)silyl]octahydro-1H-[2]pyrindene (32c). This reaction
was performed in the same way as the cyclization of 11c using
bromide 29c (0.12 g, 0.21 mmol) in toluene (5 mL) and solutions
of Bu3SnH (0.12 g, 0.11 mL, 0.42 mmol) and AIBN (0.007 g,
0.04 mmol) in toluene (both in 5 mL). Purification by flash
column chromatography (hexane 100% to remove the tin
residues and then toluene/Et2O 16:1 to obtain the product)
afforded a mixture of (2R,4R,5R)-2-isopropyl-4-(methoxy-
carbonylmethyl)-1-(p-toluenesulfonyl)-5-[1-{(triethyl)silyl}vinyl]-
piperidine and (3R,4aR,5S,7S,7aR)-3-isopropyl-5-methoxycar-
bonyl-1-(p-toluenesulfonyl)-7-[(triethyl)silyl]octahydro-1H-[2]pyri-
ndene in a 2:1 ratio, as determined by analytical HPLC, as a
colorless oil (0.07 g, 68%): Rf = 0.45; analytical HPLC (water/
MeCN/TFA 20:79.95:0.05) tR = 25.38 min (32c) and 27.17 min
(31c). A portion was separated by HPLC.
Data for 31c: [R]18D =þ7.7 (c 1.5, CHCl3); IR (film) 2952,
1
2875, 1735, 1661, 1461 cm-1; H NMR (500 MHz, CDCl3) δ
0.54 (q, J = 7.9, 6H), 0.89-0.97 (m, 15H), 1.07 (dt, J = 5.0, 13.2,
1H), 1.61 (dd, J = 10.8, 15.6, 1H), 1.72 (dt, J = 4.2, 11.5, 1H),
1.93 (d, J = 13.5, 1H), 2.07-2.13 (m, 1H), 2.21-2.26 (m, 1H),
2.43-2.48 (m, 4H), 2.61 (dd, J = 11.8, 14.8, 1H), 3.56-3.62 (m,
5H), 5.52 (s, 1H), 5.76 (s, 1H), 7.27 (d, J = 8.1, 2H), 7.72 (d, J =
8.1, 2H); 13C NMR (125 MHz, CDCl3) δ 2.6, 7.2, 19.7, 20.2,
21.4, 26.7, 31.2, 31.6, 38.1, 44.0, 46.7, 51.4, 59.5, 127.2, 127.5,
129.5, 138.7, 142.9, 148.3, 173.0; MS (electrospray) m/z 516.4
(100%, [M þ Na]þ). HRMS (electrospray) calcd for C26H43-
NO4NaSiS: 516.2580, found 516.2572.
Data for 32c: [R]21D = þ27.9 (c 0.6, CHCl3); IR (film) 2953,
1
2874, 1733, 1157 cm-1; H NMR (500 MHz, CDCl3) δ 0.45-
0.46 (m, 4H), 0.50-0.63 (m, 2H), 0.85-0.99 (m, 17H), 1.05-
1.11 (m, 1H), 1.61-1.68 (m, 1H), 1.76-1.84 (m, 1H), 1.93-2.11
(m, 4H), 2.42 (s, 3H), 2.67-2.81 (m, 1H), 3.67 (s, 3H), 3.77 (dd,
J = 4.7, 10.5, 1H), 3.93 (d, J = 11.3, 1H), 7.27 (d, J = 8.1, 2H),
7.72 (d, J = 8.1, 2H); 13C NMR (125 MHz, CDCl3) δ 2.8, 7.7,
19.8, 20.5, 21.4, 24.6, 26.9, 29.8, 30.1, 45.1, 45.3, 46.9, 48.3, 51.6,
59.7, 127.1, 129.5, 138.9, 142.8, 175.7; MS (electrospray) m/z
516.2 (100%, [M þ Na]þ). HRMS (electrospray) calcd for
C26H43NO4NaSiS: 516.2580, found 516.2589.
Acknowledgment. We thank the Engineering and Physical
Sciences Research Council for studentships to M.-E.R. and
S.M.W., the University of Birmingham for financial support
and Mr. Graham Burns for help with HPLC separations.
The NMR spectrometers used in this research were funded in
part through Birmingham Science City: Innovative Uses for
Advanced Materials in the Modern World (West Midlands
Centre for Advanced Materials Project 2), with support from
Advantage West Midlands (AWM) and part-funded by the
European Regional Development Fund (ERDF).
Supporting Information Available: Experimental proce-
dures, NMR spectra, and full characterization for all com-
pounds, as well as CIF files and ORTEP diagrams for the
X-ray structures. This material is available free of charge via
the Internet at http://pubs.acs.org.
J. Org. Chem. Vol. 75, No. 21, 2010 7357