Hydroformylation of homoallylic azides: A rapid approach toward alkaloids

Article abstract of DOI:10.1021/ol802314g

(Chemical Equation Presented) Unprecedented hydroformylation of homoallylic azides combined with useful one-pot operations provides an expeditive access to alkaloids.

Full text of DOI:10.1021/ol802314g

ORGANIC
LETTERS
2009
Vol. 11, No. 2
261-264
Hydroformylation of Homoallylic Azides:
A Rapid Approach toward Alkaloids
Thomas Spangenberg,†,‡ Bernhard Breit,*,‡ and Andre´ Mann*,†
De´partement de Pharmacochimie de la Communication Cellulaire, UMR 7175-LC-1
ULP/CNRS, Faculte´ de Pharmacie, 74 route du Rhin BP 60024,
F-67401 Illkirch, France, and Institut fu¨r Organische Chemie and Biochemie, Freiburg
Institute of AdVanced Studies (FRIAS), Albert-Ludwigs-UniVersita¨t Freiburg,
Albertstrasse 21, D-79104 Freiburg, Germany
andre.mann@pharma.u-strasbg.fr; bernhard.breit@chemie-freiburg.de
Received October 6, 2008
ABSTRACT
Unprecedented hydroformylation of homoallylic azides combined with useful one-pot operations provides an expeditive access to alkaloids.
In natural products and pharmaceuticals, the piperidine core
is one of the most encountered and has been recognized as
a priviledged structure in medicinal chemistry.1 Indeed over
12,000 discrete piperidine entities have been mentioned in
clinical or preclinical studies over the past decade.2 To date,
a variety of approaches to piperidines have been developed,3
but general and efficient methods meeting the criteria of atom
economy are still desirable. Following our interest in the
preparation of piperidine-related alkaloids,4 we envisioned
using homoallylazides as direct precursors for the piperidine
core. Indeed homoallylazides have already been used for the
synthesis of pyrrolidines5 by means of a hydroboration-
cycloalkylation reaction (Figure 1). However, the extension
† De´partement de Pharmacochimie de la Communication Cellulaire, UMR
7175-LC-1 ULP/CNRS.
Figure 1. Strategy to access pyrrolidine and piperidine alkaloids.
‡ Albert-Ludwigs-Universita¨t Freiburg.
(1) Evans, B. E.; Rittle, K. E.; Bock, M. G.; DiPardo, R. M.; Freidinger,
R. M.; Whitter, W. L.; Lundell, G. F.; Veber, D. F.; Anderson, P. S.; Chang,
R. S. L.; Lotti, V. J.; Cerino, D. J.; Chen, T. B.; Kling, P. J.; Kunkel, K. A.;
Springer, J. P.; Hirshfield, J. J. Med. Chem. 1988, 31, 2235.
(2) Watson, P. S.; Jiang, B.; Scott, B. Org. Lett. 2000, 2, 3679.
(3) (a) Laschat, S.; Dickner, T. Synthesis 2000, 1781. (b) Felpin, F.-X.;
Lebreton, J. Eur. J. Org. Chem. 2003, 3693. (c) Buffat, M. G. P. Tetrahedron
2004, 60, 1701.
of this strategy for the preparation of piperidine rings has
proven to be troublesome because the corresponding azido
olefins were prone to [3 + 2] cycloadditions.6 Conversely,
(4) (a) Spangenberg, T.; Airiau, E.; Bui The Thuong, M.; Donnard, M.;
Billet, M.; Mann, A. Synlett 2008, 2859. (b) Airiau, E.; Spangenberg, T.;
Girard, N.; Schoenfelder, A.; Salvadori, J.; Taddei, M.; Mann, A.
Chem.-Eur. J. 2008, 14, 10938.
(5) (a) Evans, D. A.; Weber, A. E. J. Am. Chem. Soc. 1987, 109, 7151.
(b) Jego, J. M.; Carboni, B.; Vaultier, M.; Carrie´, R. Chem. Commun. 1989,
142. (c) Jego, J. M.; Carboni, B.; Vaultier, M. Bull. Soc. Chim. Fr. 1992,
129, 554.
10.1021/ol802314g CCC: $40.75
Published on Web 12/15/2008
2009 American Chemical Society

Table 1. Optimization of the Hydroformylation Conditions
entry
loadinga
L
solvent/temp
time (h)
l/b
iso %
yield %c (convn)b
1
2
3
4
5
6
1/2/50
1/2/50
1/2/50
1/3/100
1/3/100
1/3/100
A
B
B
B
B
B
THF/65
THF/65
Tol/65
THF/rt
THF/65
MeOH/50
10
6
86/14
>95/5
>95/5
>95/5
>95/5
>95/5
5
<5
<5
10
<5
<5
37 (62)d
87 (99)
87 (99)
83 (99)
87 (99)
87 (99)
6
24e
4
4
a Rh(CO)2(acac)/L/1a. b Conversion and percentage of isomerized product as the l/b ratio were determined by crude 1H NMR. c Isolated after silical gel
chromatography. d Total oxidation of the phosphorus was observed by crude 31P NMR. e 1 atm of syngas.
a carbon homologation of the terminal double bond could
provide a solution. Indeed if the azide function would be
compatible with the conditions of a linear regioselective
hydroformylation, the transformation of the azide to a basic
amine would allow the formation of a cyclic imine, a direct
precursor of aza-heterocycles. This sequence would represent
a new synthetic pathway to piperidines (Figure 1). Hydro-
formylation is a prototype of an atom-economic7 reaction
and an attractive synthetic valorisation of alkenes. A new
carbon-carbon bond is formed by addition of H2 and CO
with concomitant formation of an aldehyde function under
mild conditions allowing, for instance, further transformations
in a one-pot processes.8 In the present case this strategy is
of particular interest as azides are relatively inert to classical
chemical transformations, but if properly activated they can
offer an exploding diversity of chemical reactivities.9 Ad-
ditionally they represent an inexpensive orthogonal protection
of the amine function to a broad range of reaction conditions
and functional groups. Herein we report our first results using
hydroformylation of homoallylic azides in the syntheses of
piperidinyl alkaloids or analogues.
bis-phosphite ligand, greatly favors the linear/branched ratio
(l/b) with a high turnover frequency.10 In addition, phosphites
are more electronically deficient than phosphines, thus
reacting very slowly with azides. Initial experiments were
performed on a model homoallylazide 1a, and the results
are depicted in Table 1. Indeed, the hydroformylation was
operating at low pressure of syngas (5 bar) in THF, toluene,
and MeOH (entries 2-6), with Biphephos as a ligand and
Rh(CO)2(acac) as a precatalyst. The chemical conversion of
the terminal olefin and the regioselectivity in favor of the
linear aldehyde 2a were excellent, but more interestingly,
the azido function remained intact. Even the room temper-
ature atmospheric pressure protocol gave good results (entry
4) albeit demanding a longer reaction time.11 With a low
catalyst/ligand loading good yields were obtained with
excellent regioselectivity. The reaction time could be reduced
to 4 h for reaching complete conversion (entry 5), and
different solvents (THF, toluene, MeOH) can be used. Finally
we achieved our first goal, the regioselective generation of
an aldehyde from an alkene, in the presence of an azide via
hydroformylation under mild conditions.
From the literature data no report was available on the
elusive stability of the azide group under hydroformylation
conditions. Therefore two main questions had to be ad-
dressed: (i) the reduction of the azide to an amine in the
presence of Rh(I) and hydrogen (one partner in the syngas)
and (ii) the oxidation of usual phosphine ligands (e.g.,
Xantphos A) for Rh(I) during the hydroformylation via a
Staudinger reaction at the expense of the azide. A preliminary
experiment with Xantphos revealed a poor regioselectivity
and conversion (Table 1, entry 1).
We next examined the scope of the reaction by subjecting
to hydroformylation a broad range of azido alkenes 1a-h
and 4, carrying various substituents such as alkyl, allyl, aryl,
heteroaryl, and substituted aryl groups (Table 2). Preparations
of 1a-1h were uneventful using standard methods.12,13
Using our optimized conditions, the desired δ-azido-
aldehydes 2a-h could be isolated in good to excellent yields
(entries 1-8). The use of methanol as solvent in the presence
of a catalytic amount of pTSA·H2O (10 mol %) allowed the
direct transformation of 1a into the corresponding ketal 3 in
90% yield (entry 9). Finally, subjecting 4, a 1,2-disubstituted
olefin, to hydroformylation in the presence of triphenylphos-
Proscribing phosphine ligands, it is well-known that
phosphite ligands such as Biphephos, B, a sterically hindered
262
Org. Lett., Vol. 11, No. 2, 2009

Table 2. Scope of the Reaction
Scheme 1. Synthesis of (()-Sedamine
Next, the azido group was considered as an orthogonal
protection of the amino function with respect to the aldehyde
in compounds 1a-h. Indeed the aldehyde function produced
by hydroformylation can be employed for performing a
Wittig reaction in a one-pot manner.15 For instance the
generation of an R,ꢀ-unsaturated ester can be amenable to
an in situ aza-Michael addition via the nitrene generated from
the azido group. With this sequence, the formation of 2,6-
disubstituted piperidine rings could be expected.
To illustrate this strategy, we first investigated a one-pot
hydroformylation/Wittig olefination using 1a as azido-olefin
and ethyl-(triphenylphosphoranylidene)-acetate as the Wittig
reagent. Thus the hydroformylation reaction was performed
on 1a (entry 1, Table 2), the gas was released, and then the
Wittig reagent was added into the crude reaction mixture,
which produced enoate 8 in 80% isolated yield. When enoate
8 was directly submitted to the classical Staudinger condi-
tions, piperidine 9 (cis/trans ) 96/4) was obtained in a one-
pot fashion with an overall yield of 60%. To our knowledge
this sequence represents, the first one-pot hydroformylation/
Wittig olefination/Staudinger reaction/Michael addition
(Scheme 2).
Scheme 2. One-Pot Access to 2,6-Disubstituted Piperidine 9
a Isolated yield after silica gel chromatography. b Rh(CO)2(acac) 1 mol
%, biphephos 3 mol %, pTSA·H2O 10 mol %, H2/CO (1:1) 5 bar, MeOH
[0.04 M], 50 °C, 4 h. c Rh(CO)2(acac) 2 mol %, triphenylphosphite 8 mol
%, H2/CO (1:1) 20 bar, THF [0.04 M], 65 °C, 24 h.
phite as coligand at high pressure of syngas (20 bar) led to
the branched azido-aldehyde 5 in a moderate yield with
incomplete conversion (entry 10).
With this practical method in hand, we first embarked on
the synthesis of a well-known alkaloid, (()-sedamine (7),
starting from 2g.4,14 The N-methyl piperidine 6 was obtained
with the following one-pot reduction protocol (Scheme 1):
azido-aldehyde 2g was reduced to the primary amine with
Pearlman’s catalyst in presence of aqueous formaldehyde.
After TBS deprotection in 6, (()-sedamine was obtained in
34% overall yield in eight steps from benzaldehyde. This
new procedure represents an alternative to the piperidine
syntheses by ring-closing metatheses from 1,3-hydroxy
homoallyl-amines, if atom- and step-economy are
considered.3a,15
Finally the chiral pyridinyl-homoallylazide 1e16 was
recognized as an ideal substrate for a hydroformylative
skeletal diversity oriented synthesis of two major alkaloids
from Nicotiana tabacum, (S)-anabasine (10) and (S)-nicotine
(12). Thus after hydrofromylation of azido-alkene 1e,
replacement of the syngas by hydrogen allowed reduction
of the azide followed by an intramolecular domino reductive
amination in the presence of Pearlman’s catalyst. This one-
pot sequence allowed us to obtain (S)-anabasine (10) in 81%
yield and 94% ee (four steps from 3-pyridinecarboxaldehyde,
61% overall yield; Scheme 3).
Org. Lett., Vol. 11, No. 2, 2009
263

migration is possible in larger ring sizes through the
postulated conformer III.
Scheme 3. Diversity Oriented Synthesis of (S)-Anabasine and
In conclusion, we have shown that azides are compatible
with certain conditions of hydroformylation. Thus, starting
from easily accessible homoallylic azides, we could imple-
ment the hydroformylation reaction in several one-pot
protocols (hydrogenation/Wittig olefination/Staudinger reac-
tion/Michael addition/Schmidt rearrangement) representing
new ways for the expeditive assembly of piperidine- and
pyrrolidine-containing alkaloids in an atom-economic man-
ner. Extension of this methodology toward the synthesis of
other alkaloids is underway.
(S)-Nicotine via a One-Pot Hydroformylation/Hydrogenation and
Hydroformylation/Schmidt Rearrangement
Acknowledgment. This work was supported by the
Ministe`re de´le´gue´ a` l’Enseignement Supe´rieur et a` la
Recherche (T.S.). The authors thank P. Wehrung and P.
Buisine (IFR85) for HRMS analyses, G. Ferenbach (ALU
Freiburg) for chiral HPLC analyses, and Dr. M. Keller (ALU
Freiburg) for helpful NMR experiments.
Supporting Information Available: Detailed experimen-
tal procedures and spectral and analytical data for all of the
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
The synthesis of (S)-nicotine (12) from azido-alkene/
aldehyde 1e/2e required the construction of a pyrrolidine unit
from an open chain structure. Recently Aube´ showed that
γ- ꢁr δ-azidoalkyl aldehydes were amenable to lactams or
formamides via an intramolecular Schmidt rearrangement.17
Therefore homochiral aldehyde 2e, with an aryl residue next
to the azido group, is a good candidate to explore the
intramolecular azido-aldehyde conversion into the corre-
sponding chiral formamide. Indeed (S)-nicotine was prepared
from 1e by performing the hydroformylation reaction in
toluene followed by the addition of TFA upon syngas
removal. Full conversion was observed, formamide 11a was
by far the major adduct over lactam 11b (11a/11b ) 4/1).
Finally pure formamide 11a was treated with formic acid
and aqueous formaldehyde to furnish (S)-nicotine (12) (five
steps from 3-pyridine carboxaldehyde, 35% overall yield,
94% ee).18 As a rationale, the formation of formamide 11a
can be explained by the two favorable conformers I and II
where the equatorial leaving group is antiperiplanar to the
migrating alkyl group. Although the large aryl group is in
the equatorial position in both I and II conformers, the
anomeric effect probably favors conformer I. Noteworthy
and contrary to Aube´’s observations, the lactam product 11b
is formed in traceable quantities, demonstrating that hydride
OL802314G
(6) (a) Pearson, W. H.; Lin, K.-C. Tetrahedron Lett. 1990, 31, 7571.
(b) Martin, M.-T.; Marazano, C. J. Org. Chem. 2002, 67, 1890.
(7) Trost, B. M. Angew. Chem., Int. Ed. 1995, 34, 259.
(8) For a review of one-pot and tandem reactions using hydroformylation,
see: Eilbracht, P.; Ba¨rfacker, L.; Buss, C.; Hollmann, C.; Kitsos-Rzychon,
B. E.; Kranemann, C. L.; Rische, T.; Roggenbick, R.; Schmidt, A. Chem.
ReV. 1999, 99, 3329.
(9) For a review on the chemistry of azides, see: Bra¨se, V.; Gil, C.;
Knepper, K.; Zimmermann, V. Angew. Chem., Int. Ed. 2005, 44, 5188.
(10) Cuny, G. D.; Buchwald, S. L. J. Am. Chem. Soc. 1993, 115, 2066.
(11) Seiche, W.; Schuschkowski, A.; Breit, B. AdV. Synth. Catal. 2005,
347, 1488.
(12) (a) Zallom, J.; Roberts, D. C. J. Org. Chem. 1981, 46, 5173. (b)
Alper, P. B.; Hung, S.-C.; Wong, C.-H. Tetrahedron Lett. 1996, 37, 6029.
(c) Lundquist, J. T., IV; Pelletier, J. C. Org. Lett. 2001, 3, 781.
(13) Homoallylic azides were prepared from their corresponding alcohols
via allylation (1a-1f) or Sakurai reactions (1g) followed by a SN2 reaction
and finally from primary amines via a diazo transfer reaction (1h): see
Supporting Information.
(14) (a) Batey, R. A.; Thadani, A. N.; Smil, D. V.; Lough, A. J. Synthesis
2000, 7, 990. (b) Borah, J. C.; Boruwa, J.; Barua, N. C. Lett. Org. Chem.
2006, 3, 140.
(15) Breit, B.; Zahn, S. K. Angew. Chem., Int. Ed. 1999, 38, 969. Breit,
B.; Zahn, S. K. Angew. Chem., Int. Ed. 2001, 40, 1910. Breit, B.; Zahn,
S. K. Tetrahedron 2005, 61, 6171.
(16) Felpin, F.-X.; Girard, S.; Vo-Thanh, G.; Robins, R. J.; Villie´ras,
J.; Lebreton, J. J. Org. Chem. 2001, 66, 6305.
(17) Milligan, G. L.; Mossman, C. J.; Aube´, J. J. Am. Chem. Soc. 1995,
117, 10449. Lee, H.-L.; Aube´, J. Tetrahedron 2007, 63, 9007.
(18) Mahboobi, S.; Wiegrebe, W. Arch. Pharma. 1988, 321, 175.
264
Org. Lett., Vol. 11, No. 2, 2009