New entries in Lewis acid-Lewis base bifunctional asymmetric catalyst: Catalytic enantioselective Reissert reaction of pyridine derivatives

Article abstract of DOI:10.1021/ja045966f

The first catalytic enantioselective Reissert reaction of pyridine derivatives that affords products with excellent regio- and enantioselectivity is described. The key for success is the development of new Lewis acid-Lewis base bifunctional asymmetric catalysts containing an aluminum as a Lewis acid and sulfoxides or phosphine sulfides as a Lewis base. These reactions are useful for the synthesis of a variety of chiral piperidine subunits, and catalytic enantioselective formal synthesis of CP-293,019, a selective D₄ receptor antagonist, was achieved. Preliminary mechanistic studies indicated that both sulfoxides and phosphine sulfides can activate TMSCN as a Lewis base. In addition, the sulfoxides with appropriate stereochemistry might stabilize a highly enantioselective bimetallic complex (a presumed active catalyst) through internal coordination to aluminum. Copyright

Full text of DOI:10.1021/ja045966f

Published on Web 09/02/2004
New Entries in Lewis Acid-Lewis Base Bifunctional Asymmetric Catalyst:
Catalytic Enantioselective Reissert Reaction of Pyridine Derivatives
Eiko Ichikawa,†,‡ Masato Suzuki,† Kazuo Yabu,† Matthias Albert,† Motomu Kanai,*,†,‡ and
Masakatsu Shibasaki*,†
Graduate School of Pharmaceutical Sciences, The UniVersity of Tokyo, Tokyo 113-0033, Japan, and
PRESTO, Japan Science and Technology Corporation, Japan
Received July 6, 2004; E-mail: mshibasa@mol.f.u-tokyo.ac.jp
Chiral piperidines are among the most important building blocks
for biologically active molecules and natural products. Many
synthetic methodologies have been developed to access these useful
heterocyclic compounds.1 Among them, nucleophilic asymmetric
addition to activated pyridine derivatives such as N-acyl pyridinium
salts is a direct and attractive method. There are potential difficulties
in this strategy, however, with regard to regio- and enantiocontrol.
Typically, three reaction sites (2-, 4-, and 6-positions) on N-acyl
pyridinium contain close electrophilicities, which generally results
in nonregioselective additions. Moreover, rotation of the acyl
carbon-nitrogen bond of the N-acyl pyridinium intermediate should
be fixed in the transition state to achieve high enantioselectivity.
Previous pioneering work in this field overcame these difficulties
using various methods.2 Although some of these reactions are
practical and useful, stoichiometric amounts of chiral controllers
were required, and the nucleophiles were restricted to reactive
organometallics such as Grignard or organocopper reagents. We
describe herein the first catalytic enantioselective Reissert reaction
of pyridine derivatives, which significantly expands the utility of
this strategy in chiral piperidine synthesis.
We developed a catalytic enantioselective Reissert reaction of
quinolines and isoquinolines using an asymmetric bifunctional
catalyst prepared from Et2AlCl and 1 in a 1:1 ratio (1-Al).3 The
basic concept of this catalysis is that both the electrophile (N-acyl
(iso)quinolinium) and nucleophile (TMSCN) are activated at defined
positions by the Lewis acid (aluminum) and the Lewis base
(phosphine oxide) moieties of the asymmetric catalyst, respectively,
which controls the access of the nucleophile to the substrate.4 We
planned to extend this concept to a catalytic enantioselective Reissert
reaction of pyridine derivatives. A survey of the literature, however,
revealed that this type of reaction is very challenging; none of the
achiral catalysts promote the reaction at a synthetically useful level,5
probably due to both an attenuated nucleophilicity of cyanide
compared to more reactive organometallic reagents and the lesser
polarizability of pyridines compared to isoquinolines and quinolines.
Thus, we began by identifying an appropriate substrate for achiral
catalyst-promoted reactions. We found that nicotinic amide 7 gave
the products in 91% yield using 10 mol % Et2AlCl, 2 equiv of
TMSCN, and 1.4 equiv of EtOCOCl at -78 °C in CH2Cl2, although
the regioselectivity (the ratio of 1,6- (8a) and 1,2-adduct (9a)) was
1:1.6,7
sulfoxide as a Lewis base (4-6),9 because such bifunctional cata-
lysts contain additional chiralities on the sulfur atoms, which
might enhance the enantioselectivity if it is matched with the
axial chirality of the BINOL core. The metal/ligand ratio was also
screened to generate a new chiral polymetallic complex of high
regio- and enantioselectivity, stabilized by Lewis base coordination
to the metal.10
Studies based on the above ideas led to the finding that a catalyst
prepared from Et2AlCl and ligand 6, which is not C2 symmetric,
in a 1:2 ratio (6-Al: 5 mol %) gave the product 8a with sig-
nificantly higher regio- (11:1) and enantioselectvity (75% ee: 98%
total yield) than catalysts prepared from other ligands.7,11 With this
promising catalyst, the chloroformate and the amide part of the
substrate were optimized in the presence of 6-Al. Excellent
regioselectivity (12:1∼50:1) and enantioselectivity (up to 96% ee)
were obtained using several chloroformates (Table 1; entries 1-4).
Bulkier amide 10 gave slightly higher enantioselectivity than 7.
An ester analogue 12 can be also used as a substrate, although the
enantioselectivity was moderate (entry 5). Enantiomerically pure
product 13 was obtained, however, in a synthetically useful yield
through a single recrystallization, taking advantage of the highly
crystalline Fmoc-protected amidonitrile.
Next, the optimized conditions were applied to other substrates
(14 and 16), which might be useful for further derivatization. These
substrates, however, produced only moderate regio- (5:1) and
enantioselectivity (61 and 52% ee, respectively) using 6-Al as the
catalyst. Thus, we resurveyed the reaction conditions, including the
catalyst structure. The catalyst prepared from Et2AlCl and 3 in a
1:1 ratio (3-Al: 10 mol %), containing a tuned phosphine sulfide
as a Lewis base, afforded optimum results using neopentyl
chloroformate as an acylating reagent:7,12 products 15 and 17 were
obtained in 92 and 89% yields with 91 and 86% ee from 14 and
16, respectively (entries 6 and 7, regioselectivity ) 12:1 and 8:1).
Thus, combined with the 6-Al-catalyzed reaction, these are the
first examples of a catalytic enantioselective Reissert reaction of
pyridine derivatives. Specifically, these enantioselective catalysts
facilitate one specific reaction pathway (1,6-addition from the
Re-face) out of six possible pathways, thus giving the products with
high regio- and enantioselectivity.
On the basis of this finding, the catalytic enantioselective reaction
was investigated. Use of 1-Al (10 mol %) as the catalyst allowed
the reaction of 7 to proceed at -60 °C; however, the products were
obtained with low regio- (8a:9a ) 2.3:1) and enantioselectivity
(<9% ee: 91% total yield). To improve the results, we examined
the effect of a Lewis base in the catalyst.8 Specifically, we used a
† The University of Tokyo.
‡ PRESTO.
9
11808
J. AM. CHEM. SOC. 2004, 126, 11808-11809
10.1021/ja045966f CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Catalytic Enantioselective Reissert Reaction of Pyridine
of aluminum and the ligand in the reactions promoted by sulfoxide-
containing catalysts (catalysts derived from 4-6).7 Thus, a bimetal-
lic complex might be a highly enantioselective catalyst for substrate
7.16 These results suggest that sulfoxides of Al-6 might have dual
roles: one is activation of TMSCN, and the other is stabilization
of a highly enantioselective bimetallic complex through internal
coordination to aluminum.
Derivativesa
In conclusion, we achieved the first catalytic enantioselective
Reissert reaction of pyridine derivatives through the development
of new Lewis acid-Lewis base asymmetric bifunctional catalysts.
Detailed mechanistic studies to elucidate the origin of the high regio-
and enantioselectivity are currently in progress.
R
′
yield
(%)
ee
entry
substrate
ligand
(R
′
OCOCl)
product
(%)
1f
2f
3f
4f
5f
7:R ) NMe2, Z ) H
7:R ) NMe2, Z ) H
10:R ) NiPr2, Z ) H
10:R ) NiPr2, Z ) H
12:R ) OMe, Z ) H
6
6
6
6
6
Me
8b
8c
98
89
11a 98
11b 98
91d
Fmc
93
96
93
57
Fmc
neopentyl
Acknowledgment. Financial support was provided by PRESTO
of Japan Science and Technology Corporation (JST).
Fmc
13
85
(42)e (>99)e
6g 14:R ) NiPr2, Z ) Cl
7h 16:R ) NiPr2, Z ) Br
3
3
neopentyl
neopentyl
15
17
92
89
91
86
Supporting Information Available: Experimental procedures and
characterization of the products (PDF, CIF). This material is available
free of charge via the Internet at http://pubs.acs.org.
a Yield is isolated yield of the 1,6-adduct. Ee was determined by chiral
HPLC. b Catalyst was prepared from 5 mol % Et2AlCl + 10 mol % 6 or
10 mol % Et2AlCl + 10 mol % 3. c Fm ) fluorenylmethyl. d Absolute
configuration was determined to be (S). e Value observed after recrystal-
lization. f Reaction time ) 5 h. g Reaction time ) 27 h. h Reaction time )
36 h.
References
(1) For recent reviews, see: (a) Buffat, M. G. P. Tetrahedron 2004, 60, 1701.
(b) Laschat, S.; Dickner, T. Synthesis 2000, 1781.
(2) For selected examples, see: (a) Comins, D. L.; Goehring, R. R.; Joseph,
S. P.; O’Connor, S. J. Org. Chem. 1990, 55, 2574. (b) Gosmini, R.;
Mangeney, P.; Alexakis, A.; Commerc¸on, M.; Normant, J.-F. Synlett 1991,
111. (c) Ge´nisson, Y.; Marazano, C.; Das, B. C. J. Org. Chem. 1993, 58,
2052. (d) Charette, A. B.; Grenon, M.; Lemire, A.; Pourashraf, M.; Martel,
J. J. Am. Chem. Soc. 2001, 123, 11829. (e) Yamada, S.; Morita, C. J.
Am. Chem. Soc. 2002, 124, 8184. (f) Legault, C.; Charette, A. B. J. Am.
Chem. Soc. 2003, 125, 6360.
Scheme 1. Catalytic Enantioselective Synthesis of Intermediate
for CP-293,019
(3) (a) Takamura, M.; Funabashi, K.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 2000, 122, 6327. (b) Takamura, M.; Funabashi, K.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 6801. (c) Funabashi, K.;
Ratni, H.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 10784.
(4) Shibasaki, M.; Kanai, M.; Funabashi, K. Chem. Commun. 2002, 1989.
(5) Simple pyridine gave the Reissert product in only 25% yield using NaCN
as a nucleophile: (a) Reuss, R. H.; Smith, N. G.; Winters, L. J. J. Org.
Chem. 1974, 39, 2027. Even more reactive 3-acetylpyridine gave the
product in up to 30-45% yield using TMSCN and benzoyl chloride in
the presence of a catalytic amount of AlCl3: (b) Popp, F. D.; Takeuchi,
I.; Kant, J.; Hamada, Y. Chem. Commun. 1987, 1765.
(6) 1,4-Adduct was obtained in less than trace amounts in all cases.
(7) See Supporting Information for details.
(8) We speculated that there is a Lewis base-catalyzed monoactivation pathway
of low enantioselectivity for reactive N-acyl pyridinium, due to the higher
Lewis basicity of a phosphine oxide than a sulfoxide or a phosphine
sulfide.
(9) For use of a sulfoxide as a Lewis base activator of TMSCN in a catalytic
asymmetric cyanosilylation of aldehydes, see: (a) Rowlands, G. J. Synlett
2003, 236. For recent examples of chiral sulfoxides as Lewis base
activators of silicon-containing nucleophiles, see: (b) Rowlands, G. J.;
Barnes, W. K. Chem. Commun. 2003, 2712. (c) Kobayashi, S.; Ogawa,
C.; Konishi, H.; Sugiura, M. J. Am. Chem. Soc. 2003, 125, 6610.
(10) For an example in which an internal Lewis base stabilizes a highly
enantioselective chiral polymetallic complex, see: Yabu, K.; Masumoto,
S.; Yamasaki, S.; Hamashima, Y.; Kanai, M.; Du, W.; Curran, D. P.;
Shibasaki, M. J. Am. Chem. Soc. 2001, 123, 9908.
(11) C2-symmetric sulfoxide-containing catalysts 4-Al and 5-Al gave products
in 78% yield with 1.3:1 regioselectivity (15 and 7% ee) and in 56% yield
with 1:1 regioselectivity (14 and 4% ee), respectively.
(12) 3-Al did not give satisfactory results for 7 using neopentyl chlorofor-
mate: 1,6-adduct 8d and 1,2-adduct 9d were obtained in 56% yield with
80% ee and 15% yield with 0% ee, respectively.
(13) Sanner, M. A.; Chappie, T. A.; Dunaiskis, A. R.; Fliri, A. F.; Desai, K.
A.; Zorn, S. H.; Jackson, E. R.; Johnson, C. G.; Morrone, J. M.; Seymour,
P. A.; Majchrzak, M. J.; Faraci, W. S.; Collins, J. L.; Duignan, D. B.; Di
Prete, C. C.; Lee, J. S.; Trozzi, A. Bioorg. Med. Chem. Lett. 1998, 8,
725.
(14) Based on the reaction rate comparison, sulfoxides and phosphine sulfides
are weaker activators of TMSCN than phosphine oxides.
(15) For Lewis base-catalyzed cyanosilylation of aldehydes, see: (a) Evans,
D. A.; Truesdale, L. K. Tetrahedron Lett. 1973, 4929. (b) Kobayashi, S.;
Tsuchiya, Y.; Mukaiyama, T. Chem. Lett. 1991, 537.
(16) Relation between enantioselectivity and the Al:6 ratio in the catalyst
preparation supported this hypothesis: ee values of 11b were 17, 82, 93,
and 89% with Al:6 ratios of 1:1.1, 1:1.3, 1:2, and 1:3, respectively.
The synthetic utility of these reactions was demonstrated by
application to a formal catalytic enantioselective synthesis of the
dopamine D4 receptor-selective antagonist, CP-293,019 (18)13
(Scheme 1). Two-step hydrogenation of 8b (91% ee) followed by
a protection-deprotection protocol gave tetrahydropyridine 19.
Reduction of 19 with NaBH3CN via an iminium cation proceeded
in a 4:1 ratio, and the isolated major isomer was annulated to give
trans-20. The known intermediate 21 was synthesized from 20 in
three steps. Furthermore, the multifunctionality of the Reissert
products allowed for a short-step, stereoselective synthesis of other
various useful chiral building blocks.7
Although the detailed reaction mechanism is under investigation,
the following preliminary information suggests key factors for the
success of the present catalysis. First, on the basis of a reaction
rate comparison of cyanosilylation of hydrocinnamaldehyde in the
presence or absence of the Lewis base, both sulfoxides and
phosphine sulfides can activate TMSCN as a Lewis base.7,14,15
Combined with the previous mechanistic studies of Al-1-catalyzed
reactions,3 the high regio- and enantioselectivity are likely due, at
least partly, to the dual activation of N-acyl pyridinium and TMSCN
at the positions defined by the bifunctional asymmetric catalyst.
Second, ESI-MS studies suggested a relation between the enantio-
selectivity and the amount of a bimetallic 2:3 complex composed
JA045966F
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11809