Unexpected epimerization at C2 in the Horner-Wadsworth-Emmons reaction of chiral 2-substituted-4-oxopiperidines

Article abstract of DOI:10.1039/b608067g

The observed epimerization at C₂ in the Horner-Wadsworth-Emmons (HWE) reaction of chiral 2-substituted-4-oxopiperidines has been investigated and, on the basis of the experimental results, a mechanism for this unexpected process has been propos

Full text of DOI:10.1039/b608067g

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Unexpected epimerization at C2 in the Horner–Wadsworth–Emmons
reaction of chiral 2-substituted-4-oxopiperidines{
Pablo Etayo, Ramo´n Badorrey, Mar´ıa D. D´ıaz-de-Villegas* and Jose´ A. Ga´lvez*
Received (in Cambridge, UK) 7th June 2006, Accepted 27th June 2006
First published as an Advance Article on the web 19th July 2006
DOI: 10.1039/b608067g
result was observed using the phosphorus ylide generated by
treatment of ethoxycarbonylmethyltriphenylphosphonium chlor-
ide (2.8 eq.) with potassium tert-butoxide (2.8 eq.). In this case the
reaction was carried out for 43 h in toluene under reflux and only
15% of compound 3E was isolated from the reaction mixture.
On the other hand, HWE reaction12 of 4-oxopiperidine 2 with
triethyl phosphonoacetate using DBU as the base only occurred
under the reaction conditions reported by Masamune et al.,13
which involve the use of lithium chloride when an amine is used as
the base (Scheme 1). In addition, it was observed that in this case it
was necessary to use at least 3.5 eq. of phosphonate to achieve
complete conversion. Under these reaction conditions the
corresponding olefins 3 were obtained in moderate to good yields
as a diastereomeric E/Z mixture in which the E alkene was highly
predominant. However, this product was surprisingly contami-
nated with varying amounts of the corresponding E/Z olefins 4
with the opposite configuration at C2.14
The observed epimerization at C2 in the Horner–Wadsworth–
Emmons (HWE) reaction of chiral 2-substituted-4-oxopiper-
idines has been investigated and, on the basis of the
experimental results, a mechanism for this unexpected process
has been proposed.
Functionalized piperidines are among the most common building
blocks in natural products and, more interestingly, in many
biologically active compounds.1 The synthesis of these types of
compounds has been studied extensively and in recent years
thousands of piperidine compounds have been mentioned in
clinical and preclinical studies directed toward the development of
new drugs. Before such compounds can be used as drugs, however,
it is necessary to perform their stereoselective synthesis.
A number of review articles have covered recent progress in the
stereoselective synthesis of substituted piperidines.2 Most of the
synthetic methodologies described in the literature involve a
plethora of approaches specific to the synthesis of a target
compound, whereas the development of general synthetic
methodologies in which preformed chiral non-racemic building
blocks are used for the construction of a wide variety of simple or
complex structures is less common. In this context, it is worth
mentioning the use of chiral non-racemic N-cyanomethyloxazoli-
dines,3 N-alcoxycarbonyl 2,3-dihydro-4-pyridones,4 6-substituted
2,3-didehydropiperidine-2-carboxylates5 and bicyclic piperidones.6
In our effort to develop a novel chiral non-racemic building
block for the synthesis of substituted piperidines we have
synthesized7 enaminone 1 in a hetero Diels–Alder reaction with
double stereodifferentiation. This chiral building block has proven
to be a versatile intermediate in the synthesis of enantiomerically
pure pipecolic acid derivatives such as (R)-4-oxopipecolic acid7
The extent of epimerization at C2 was dependent on the amount
of DBU and the reaction time. On using 1.5 eq. of base and short
reaction times (8 h) the formation of epimeric compounds was
almost negligible and a 94/6 mixture of 3E/3Z was obtained in 82%
yield. The use of a large excess of DBU (10 eq.) and long reaction
times (7 d) led to the highly preferential formation of olefins with
the S configuration at C2 in which compound 4E was the major
product (Table 1, entry 1). Olefins 4 were isolated from the
reaction mixture in 78% yield.
and
(2R,4S)-N-tert-butoxycarbonyl-4-hydroxypipecolic
acid
tert-butylamide.8
Previous results in this field9 encouraged us to use Wittig-type
reactions10 as a way to introduce hydrocarbon substituents at C4 in
carbonylic compound 2, which in turn was obtained by reduction
of enaminone 1 with L-selectride1 (Scheme 1).
Wittig reaction11 of 4-oxopiperidine 2 with ethoxycarbonyltri-
phenylphosphonium methylide either did not occur or conversion
was extremely low depending on the reaction conditions. The best
Departamento de Qu´ımica Orga´nica, Instituto de Ciencia de Materiales
de Arago´n, Instituto Universitario de Cata´lisis Homoge´nea, Universidad
de Zaragoza-CSIC, E-50009 Zaragoza, Spain.
E-mail: loladiaz@unizar.es; jagl@unizar.es; Fax: + 34 976761202;
Tel: + 34 976762274
{ Electronic supplementary information (ESI) available: Experimental
details and full characterization data for all compounds. See DOI: 10.1039/
b608067g
Scheme 1 Reagents and conditions: (a) L-selectride1, THF, 278 uC,
85%; (b) (EtO)2P(O)CH2CO2Et, LiCl, DBU, CH3CN, r.t.
3420 | Chem. Commun., 2006, 3420–3422
This journal is ß The Royal Society of Chemistry 2006

View Article Online
3E + 3Z in acetonitrile in the absence of DBU and lithium chloride
remained unaltered after several days – even under reflux
conditions.
Table 1 Epimerization of compounds 3E and 3Z
Entry
Epimerization procedurea
4/3
4E/4Z
Yieldb (%)
1
2
3
4
5
a
A
B
C
D
E
92/8
91/9
93/7
92/8
92/8
88/12
81/19
90/10
85/15
76/24
78
51
50
59
74
On the basis of these experimental results we can state that in
the presence of DBU compounds obtained in the HWE reaction
suffered an epimerization reaction that does not affect the starting
piperidone.
Having established the essential role of the base for epimeriza-
tion, the extent of this reaction with a range of different bases was
investigated. Epimerization was not detected after several days
either in the presence of LDA or KOtBu, stronger bases than
DBU, or in the presence of DIPEA, TMG, Et3N or DABCO,
which have similar pKa values to DBU (11–13). On the other
hand, DBN acted as mediator for epimerization and a mixture
(4E + 4Z)/(3E + 3Z) = 92/8 was obtained after 7 days at room
temperature in the HWE reaction or from 3E + 3Z (Table 1,
entries 4, 5).
A, HWE reaction of 2 with triethyl phosphonoacetate for 7 d using
10 eq. of DBU as base. B, treatment for 7 d of the mixture 3E + 3Z
of 2R configuration with 10 eq. of DBU in the presence of LiCl. C,
treatment for 7 d of the mixture 3E + 3Z of 2R configuration with
10 eq. of DBU in the absence of LiCl. D, HWE reaction of 2 with
triethyl phosphonoacetate for 7 d using 10 eq. of DBN as base. E,
treatment for 7 d of the mixture 3E + 3Z of 2R configuration with
b
10 eq. of DBN in the presence of LiCl. 4E + 4Z isolated yield.
Epimerization at C2 was quite unexpected and additional
experiments were performed in an effort to determine a plausible
pathway for this reaction.
Treatment of an 83/17 mixture of 4E/4Z with LiCl (3.5 eq.) and
DBU (10 eq.) for 7 days at room temperature resulted in the
minoritary formation of olefins with the R configuration at C2
and, once again, an (4E + 4Z)/(3E + 3Z) = 92/8 mixture of
compounds with a 93/7 ratio for 4E/4Z was obtained after 7 days.
The ratio (4E + 4Z)/(3E + 3Z) observed after 7 days (ca. 90/10)
seemed to be independent of the epimerization conditions and on
the configuration at C2, which suggests that this ratio corresponds
to a thermodynamic equilibrium between alkenes 4 and 3.
Moreover, the E/Z ratio changed significantly during epimeriza-
tion, indicating that this equilibrium must be reached through an
Firstly, the behavior of piperidone 2 and the olefin mixture 3E +
3Z was screened under the reaction conditions that led to a high
level of epimerization. The aim of this experiment was to elucidate
which compound underwent this epimerization process. Under
these conditions it was found that ketone 2 remained unaltered
after several days whereas the mixture 3E + 3Z evolved to give,
after 7 days, a mixture of olefins (4E + 4Z)/(3E + 3Z) = 91/9
(Table 1, entry 2).
Similar behavior was observed under the same conditions
without lithium chloride (Table 1, entry 3), whereas a solution of
Scheme 2 Proposed epimerization mechanism.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 3420–3422 | 3421

View Article Online
enolate intermediate in which rotation around the C4–C49 exocyclic
Notes and references
bond is possible (structures I9 and V9, Scheme 2).
1 See for example: (a) G. Lagler, Adv. Carbohydr. Chem. Biochem., 1990,
48, 319–384; (b) B. Winchester and G. W. J. Fleet, Glycobiology, 1992, 2,
199–210; (c) M. J. Schneider, in Alkaloids: Chemical and Biological
Perspectives, ed. S. W. Pelletier, Pergamon: Oxford, 1996; vol. 10,
pp. 155–299; (d) D. O’Hagan, Nat. Prod. Rep., 1997, 14, 637–651; (e)
N. Asano, R. J. Nash, R. J. Molyneux and G. W. J. Fleet, Tetrahedron:
Asymmetry, 2000, 11, 1645–1680; (f) A. A. Watson, G. W. J. Fleet,
N. Asano, R. J. Molyneux and R. J. Nash, Phytochemistry, 2001, 56,
265–295.
2 Recent reviews: (a) P. D. Bailey, P. A. Millwood and P. D. Smith,
Chem. Commun., 1998, 633–640; (b) A. Mitchinson and A. Nadin,
J. Chem. Soc., Perkin Trans. 1, 2000, 2862–2892; (c) S. Laschat and
T. Dickner, Synthesis, 2000, 1781–1813; (d) P. M. Weintraub, J. S. Sabol,
J. M. Kane and D. R. Borcherding, Tetrahedron, 2003, 59, 2953–2989;
(e) M. G. P. Buffat, Tetrahedron, 2004, 60, 1701–1729.
With this background information in hand,
a possible
mechanism for the formation of compounds with the opposite
configuration at C2 is outlined in Scheme 2. The initially formed
dienolate I/I9 is re-protonated at the piperidine nitrogen to afford
zwitterion II, in which epimerization at C2 is possible through a
retro 1,6/1,6 addition sequence in which there is a 1,5 functionality
distance between the carbonyl group and the amino group.
Deprotonation of zwitterion IV and subsequent protonation of
dienolate V/V9 would lead to the formation of compounds 4E
and 4Z. This process is mediated by a base strong enough to
accept a proton from C3 with a conjugate acid that is acidic
enough to transfer a proton to the piperidine nitrogen. Of the
bases tested this requirement is only fulfilled by the amidine type
bases DBU and DBN. Initial deprotonation at C5 is also possible
and the enolate generated would evolve to yield the starting
3 For a review on this subject see: H. P. Husson and J. Royer, Chem. Soc.
Rev., 1999, 28, 383–394.
4 Leading references: (a) D. L. Comins, S. P. Joseph and R. R. Goehring,
J. Am. Chem. Soc., 1994, 116, 4719–4728; (b) J. Kuethe and
D. L. Comins, Org. Lett., 1999, 1, 1031–1033 and references therein.
5 For leading references see: N. Toyoka, K. Tanaka, T. Momose,
J. W. Daly and H. M. Garraffo, Tetrahedron, 1997, 53, 9553–9574 and
references therein.
6 (a) A. G. Brewster, S. Broady, C. E. Davies (ne´e Mills), T. D. Heightman,
S. A. Hermitage, M. Hughes, M. G. Moloney and G. Woods, Org.
Biomol. Chem., 2004, 2, 1031–1043; (b) A. G. Brewster, S. Broady,
M. Hughes, M. G. Moloney and G. Woods, Org. Biomol. Chem., 2004,
2, 1800–1841.
7 (a) R. Badorrey, C. Cativiela, M. D. D´ıaz-de-Villegas and J. A. Ga´lvez,
Tetrahedron Lett., 1997, 38, 2547–2550; (b) R. Badorrey, C. Cativiela,
M. D. D´ıaz-de-Villegas and J. A. Ga´lvez, Tetrahedron, 1999, 55,
7601–7612.
8 R. Badorrey, C. Cativiela, M. D. D´ıaz-de-Villegas and J. A. Ga´lvez,
Tetrahedron, 1999, 58, 341–354.
9 (a) J. W. Skiles, P. P. Giannoussis and K. R. Fales, Bioorg. Med. Chem.
Lett., 1996, 6, 963–966; (b) P. S. Watson, B. Jiang and B. Scott, Org.
Lett., 2000, 2, 3679–3681.
10 (a) B. E. Maryanoff and A. B. Reitz, Chem. Rev., 1989, 89, 863–927; (b)
S. E. Kelly, in Comprehensive Organic Synthesis, ed. B. M. Trost and
I. Fleming, Pergamon Press: Oxford, 1991, vol. 1, chap. 3.1.9; (c) Ylides
and Imines of Phosphorus, ed. A. W. Johnson, John Wiley & Sons, Inc.:
New York, 1993; (d) Phosphorus Ylides: Chemistry and Applications in
Organic Synthesis; Wiley-VCH: New York, 1999.
11 G. Wittig and G. Geissler, Liebigs Ann. Chem., 1953, 580, 44–57.
12 W. S. Wadsworth and W. D. Emmons, J. Am. Chem. Soc., 1961, 83,
1733–1738.
13 M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S. Masamune,
W. R. Roush and T. Sakai, Tetrahedron Lett., 1984, 25, 2183–2186.
14 E/Z configurations for compounds 3E, 4E and 4Z were clearly
determined by a series of selective 1D gradient enhanced nuclear
Overhauser enhancement spectroscopy (ge-1D NOESY) experiments
and the Z configuration of compound 3Z was established by
elimination.
compounds.
A retro-conjugate addition/conjugate addition
sequence has been reported previously in 2,5-disubstituted
pyrrolidine alkylations.15
Isolation or trapping of intermediates was not possible but the
reaction was followed by 1H NMR spectroscopy and the
appearance of signals at ca. 6.0 and 6.5 ppm was observed.
These signals can be attributed to protons bonded to the highly
conjugated disubstituted double bond of compound III.
On the basis of these results we reasoned that to avoid
epimerization at C2 the HWE reaction must be conducted using
LDA as the base. Indeed, reaction of compound 2 with triethyl
phosphonoacetate (3.0 eq.) in the presence of LDA (3.5 eq.) led to
the exclusive formation of compounds 3E and 3Z in a 97/3 ratio in
nearly quantitative yield (97%) after 14 h. This approach
constitutes a valuable synthetic methodology for the synthesis of
compounds with the R configuration at C2.
In conclusion, a mechanism that accounts for the unexpected
epimerization at C2 in the Horner–Wadsworth–Emmons reaction
of 2-substituted-4-oxopiperidines has been proposed and is
supported by experimental results. This study has established the
optimal conditions for the diastereodivergent synthesis of E/Z
mixtures of 2-[(S)-1,2-dibenzyloxyethyl]-4-ethoxycarbonylmethy-
lene-N-[(S)-1-phenylethyl]piperidine of R and S configuration at
C2 starting from 4-oxopiperidine 2.
The authors wish to thank the Spanish MCYT and FEDER
(Project CTQ2004-05358) and the Diputacio´n General de Arago´n
for financial support and the Spanish MCYT for a predoctoral
fellowship (P. E.).
15 (a) S. R. Hussaini, M. G. Moloney and G. Woods, Org. Biomol. Chem.,
2003, 1, 1838–1841; (b) S. R. Hussaini, M. G. Moloney and G. Woods,
Tetrahedron Lett., 2004, 45, 1125–1127.
3422 | Chem. Commun., 2006, 3420–3422
This journal is ß The Royal Society of Chemistry 2006