Regioselective nucleophilic addition to pyridinium salts: A new route to substituted dihydropyridones

Article abstract of DOI:10.1021/ol902402v

"Chemical Equation Presented" The regioselective addition of nucleophiles to pyridinium salts generates an intermediate enol-ether, which can be hydrolyzed in situ to provide a range of dihydropyridones. Certain Grignards have shown inherent differences i

Full text of DOI:10.1021/ol902402v

ORGANIC
LETTERS
2009
Vol. 11, No. 23
5562-5565
Regioselective Nucleophilic Addition to
Pyridinium Salts: A New Route to
Substituted Dihydropyridones
Timothy J. Donohoe,*,† Matthew J. Connolly,† and Lesley Walton‡
Chemistry Research Laboratory, Department of Chemistry, UniVersity of Oxford, Mansfield
Road, Oxford OX1 3TA, U.K. and Lilly Research Centre, Eli Lilly and Company, Ltd., Erl
Wood Manor, Sunninghill Road, Windlesham, Surrey GU20 6PH, U.K.
timothy.donohoe@chem.ox.ac.uk
Received October 19, 2009
ABSTRACT
The regioselective addition of nucleophiles to pyridinium salts generates an intermediate enol-ether, which can be hydrolyzed in situ to
provide a range of dihydropyridones. Certain Grignards have shown inherent differences in the regioselectivity of addition to these salts, and
this difference can be tuned to give single regioisomeric addition products. The dihydropyridone products can be further manipulated in many
ways using standard transformations.
We have previously reported on the ammonia-free Birch
reduction of substituted pyridinium salts 1 (Scheme 1).1
Scheme 1
Addition of two electrons to the aromatic system gives an
anionic intermediate that is capable of reaction with elec-
trophiles. After in situ hydrolysis, the resultant dihydropy-
ridones 2 that are formed have potential applications in
natural product synthesis and medicinal chemistry.2 However,
during the Birch reduction problems have been encountered
concerning the reproducibility of some results with pyri-
dinium salts. In particular, the low solubility of salts such
as 1 can lead to variable results and has hampered attempts
to scale up the reduction.
Recently, it has been demonstrated that similar unsubsti-
tuted dihydropyridones are accessible Via nucleophilic ad-
dition to 4-methoxy substituted pyridinium salts.3-6 Encour-
† University of Oxford.
aged by these results, we proposed that addition of Grignard
reagents to pyridinium salts such as 1, followed by acidic
‡ Eli Lilly and Company, Ltd.
(1) (a) Donohoe, T. J.; Johnson, D. J.; Mace, L. H.; Bamford, M. J.;
Ichihara, O. Org. Lett. 2005, 7, 435. (b) Donohoe, T. J.; Johnson, D. J.;
Mace, L. H.; Thomas, R. E.; Chiu, J. Y. K.; Rodrigues, J. S.; Compton,
R. G.; Banks, C. E.; Tomcik, P.; Bamford, M. J.; Ichihara, O. Org. Biomol.
Chem. 2006, 4, 1071.
(3) (a) Comins, D. L.; Joseph, S. P.; Goehring, R. R. J. Am. Chem. Soc.
1994, 116, 4719. (b) Comins, D. L.; Brown, J. D. Tetrahedron Lett. 1986,
27, 4549
(4) Focken, T.; Charette, A. B. Org. Lett. 2006, 8, 2985
(5) Knapp, S.; Yang, C.; Pabbaraja, S.; Rempel, B.; Reid, S.; Withers,
S. G. J. Org. Chem. 2005, 70, 7715
(6) Ege, M.; Wanner, K. T. Org. Lett. 2004, 6, 3553
.
(2) (a) Donohoe, T. J.; McRiner, A. J.; Sheldrake, P. Org. Lett. 2000,
2, 3861. (b) Donohoe, T. J.; McRiner, A. J.; Helliwell, M.; Sheldrake, P.
J. Chem. Soc., Perkin Trans. 1 2001, 1435. (c) Donohoe, T. J.; Mace, L. H.;
Helliwell, M.; Ichihara, O. Synlett 2002, 331.
.
.
.
10.1021/ol902402v CCC: $40.75
Published on Web 11/05/2009
 2009 American Chemical Society

hydrolysis, could provide a useful alternative to the Birch
reduction, but this time using the aromatic heterocycle as
an electrophile.
Our efforts began with the THF-soluble pyridinium salt
4, which was prepared in quantitative yield from the known
disubstituted pyridine 3 and methyl triflate (Scheme 2).7
recovery of starting material or decomposition. This limita-
tion would impose a significant barrier on the synthetic utility
of this methodology, so we decided to investigate an alternate
N-protecting group. Corey has demonstrated that reaction
of substituted pyridines with allyl triflate (generated in situ
from allyl alcohol, triflic anhydride, and a tertiary amine
base) produces N-allyl pyridinium salts in good yield.11
Application of Corey’s protocol to pyridine 3 furnished
pyridinium salt 11 in excellent yield, and treatment of 11
with methylmagnesium bromide furnished dihydropyridone
12 in 74% yield (Scheme 3). It was subsequently found that
Scheme 2
Scheme 3
a Compound 3 was prepared in one step from commercially available
picolinic acid.
Pleasingly, treatment of 4 with methylmagnesium bromide,
followed by in situ hydrolysis, gave dihydropyridone 5 as a
single regioisomer in 89% yield. This is a rare example of
nucleophilic addition to a C-2 substituted pyridinium salt,
generating a tertiary center with complete control of regi-
oselectivity.8 With this success in hand, dihydropyridones
6-10 were prepared in excellent yield from their corre-
sponding Grignard reagent.
a Reaction performed on a 2.5 g (7 mmol) scale.
Interestingly, we found that alkenyl and alkynyl Grignards
gave another regioisomeric product, resulting from nucleo-
philic attack at C-6 rather than attack at C-2, as observed
for alkyl Grignards (Scheme 2, entries 5 and 6). We attributed
this discrepancy to the fact that harder nucleophiles would
tend to attack at the harder (more electron-deficient) C-2
center, adjacent to the electron-withdrawing ester group, and
softer nucleophiles would attack the softer C-6 center.9 The
importance of hard and soft factors in additions to pyridinium
salts has previously been described by Yamaguchi et al.10
We were disappointed to find that attempts at N-demethy-
lation of the dihydropyridones 5-8, under nucleophilic,
electrophilic, or oxidative conditions, resulted in either
conducting the Grignard addition at -60 °C improved the
yield of 12 to 86% and, pleasingly, performing the addition
on a 2.5 g scale did not result in any significant reduction in
yield, scale-up difficulties being a major limitation of Birch
reduction methodology with these substrates. Other alkyl
Grignards gave their corresponding dihydropyridones in
similarly excellent yield. While we found that the yield of
dihydropyridone was reduced as the bulk of the Grignard
reagent was increased, note that dihydropyridones 15 and
16 could not be prepared by the Birch reduction route owing
to the poor SN2 substitution reactions of the corresponding
secondary and tertiary bromide electrophiles. The synthesis
of dihydropyridones 15 and 16 highlights the greater versatil-
ity of this methodology over the Birch reduction (Scheme
3, entries 6 and 7). Unfortunately, addition of aryl and vinyl
(7) Sundberg, R. J.; Jiang, S. Org. Prep. Proced. Int. 1997, 29, 117.
(8) Barbe, G.; Pelletier, G.; Charette, A. B. Org. Lett. 2009, 11, 3398.
(9) (a) Trost, B. M.; Thaisrivongs, D. A. J. Am. Chem. Soc. 2008, 130,
14092. (b) Keinan, E.; Roth, Z. J. Org. Chem. 1983, 48, 1769.
(10) Yamaguchi, R.; Nakazono, Y.; Matsuki, T.; Hata, E.; Kawanisi,
M. Bull. Chem. Soc. Jpn. 1987, 60, 215.
(11) Corey, E. J.; Helal, C. J. Tetrahedron Lett. 1996, 37, 5675.
Org. Lett., Vol. 11, No. 23, 2009
5563

Grignard reagents gave essentially 1:1 mixtures of regioi-
somers, which was disappointing given that addition to
N-methyl salt 4 furnished only the C-6 addition products
(entries 8 and 9). The reason for this is at present unclear,
but it is possible that the N-allyl group adopts a conformation
that hinders the C-6 position more than a simple Me group
and so reduces the rate of nucleophilic attack at that center.
With 1:1 regioisomeric mixtures being of little synthetic
utility, we naturally decided to improve the selectivity of
addition to an acceptable level. Building on our hypothesis
that harder nucleophiles add to the C-2 position and softer
nucleophiles add to the C-6 position, we prepared an
organozinc species by in siu treatment of the Grignard
reagent with zinc chloride prior to addition of the pyridinium
salt 11, hoping that this softer nucleophile would selectively
add at C-6.12 To our delight, the only regioisomer observed
from this protocol was that resulting from C-6 addition, and
these dihydropyridones were obtained in excellent yield
(Scheme 4). Interestingly, reaction of 11 with ethylmagne-
provide piperidones 22 and 23, and gratifyingly, treatment
of these with Guibe´’s conditions furnished the deprotected
products in excellent yield. This success creates the potential
for N-functionalization with a range of electrophiles and
provides a starting point for future synthetic work (Scheme 5).
Scheme 5
Scheme 4
Having developed this simple and efficient procedure, we
began to investigate whether it could be applied to other
classes of substituted pyridinium salts that did not perform
well under Birch reduction conditions. We have previously
reported the ammonia-free Birch reduction of pyridinium salt
28, prepared in excellent yield over two steps from com-
mercially available 6-hydroxypicolinic acid 26.1b Unfortu-
nately, the yield of dihydropyridone 29 was modest under
partial reducing conditions, thus prohibiting any synthetic
application of this work. However, when pyridinium salt 28
was treated to the new nucleophilic conditions reported
herein, we were delighted to find that the corresponding
dihydropyridones could be easily isolated in good yield
(Scheme 6). The olefin functionality in the product dihy-
a Treatment of 11 with EtMgBr under these conditions gave C-2 addition
product 13 in 88% yield.
sium bromide and zinc chloride gave the C-2 addition
product as a single regioisomer in yield similar to that
obtained in the absence of zinc chloride.
To allow elaboration of the ring nitrogen, we focused our
efforts on the removal of the N-allyl group. Inspired by the
work of Guibe´ on the palladium-catalyzed deallylation of
amines, facilitated by 1,3-dimethylbarbituric acid,13 we began
by subjecting dihydropyridone 12 to Guibe´’s conditions.
Unfortunately, only returned starting material was obtained
from the reaction mixture. We reasoned that the ring nitrogen
formed part of a vinylogous amide system and its basicity
would be reduced to the extent that protonation by 1,3-
dimethylbarbituric acid (pKa (H2O) ) 4.7) would not occur.
Conversion of the vinylogous amide to a tertiary amine
would be likely to form part of any further synthetic
manipulations on these products, so carrying out this
transformation prior to N-deallylation would not pose any
significant problems and would increase the basicity of the
nitrogen, circumventing the aforementioned problem. Reduc-
tion of 12 and 14 with L-Selectride proceeded smoothly to
Scheme 6
dropyridones was found to lie exclusively out of conjugation
with the carbonyl group; however, we subsequently found
that treatment with DBU in refluxing THF afforded the R,ꢀ-
unsaturated dihydropyridones in good yield.
In conclusion, we have developed an efficient and practi-
cally simple alternative to the Birch reduction of pyridinium
(12) For examples of organozinc reagents behaving as soft nucleophiles,
see: (a) Dieter, R. K.; Guo, F. J. Org. Chem. 2009, 74, 3843. (b) Komanduri,
V.; Pedraza, F.; Krische, M. J. AdV. Synth. Catal. 2008, 350, 1569.
(13) Garro-Helion, F.; Merzouk, A.; Guibe´, F. J. Org. Chem. 1993, 58,
6109.
5564
Org. Lett., Vol. 11, No. 23, 2009

salts. Nucleophilic addition to 4-methoxy substituted salts
such as 11 yielded a range of dihydropyridones, after in situ
hydrolysis, with the potential for the introduction of a variety
of groups at the C-2 or C-6 position. We have demonstrated
the smooth deallylation of the product dihydropyridones,
enabling further N-functionalization and highlighting the
synthetic utility of this methodology. Furthermore, the
nucleophilic addition to regioisomeric pyridinium salt 28 is
reported allowing access to the corresponding dihydropyri-
dones. The overall yield of dihydropyridone has been
improved for both salts by using this methodology, compared
to Birch reduction.
Acknowledgment. We thank the EPSRC and Eli Lilly
and Company Ltd. for funding this project.
Supporting Information Available: Full experimental
1
details as well as H and 13C NMR spectra for all reported
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL902402V
Org. Lett., Vol. 11, No. 23, 2009
5565