Asymmetric Reissert-type reaction promoted by bifunctional catalyst [21]

Full text of DOI:10.1021/ja0010352

J. Am. Chem. Soc. 2000, 122, 6327-6328
Table 1. Catalytic Asymmetric Reissert-type Reactiona
6327
Asymmetric Reissert-type Reaction Promoted by
Bifunctional Catalyst
Masahiro Takamura, Ken Funabashi, Motomu Kanai, and
Masakatsu Shibasaki*
Graduate School of Pharmaceutical Sciences
The UniVersity of Tokyo,
Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
ReceiVed March 24, 2000
The addition of cyanide to quinoline or isoquinoline derivatives
(the Reissert-type reaction1) has been widely used as a key step
for the synthesis of various heterocyclic compounds, especially
for the synthesis of biologically important alkaloids.2 Moreover,
the Reissert reaction has been applied to solid-phase synthesis,
opening the way to utilize this reaction for synthesizing hetero-
cyclic compounds by a combinatorial strategy.3 It is also
specifically noteworthy that Reissert compounds could offer direct
entries for synthesizing the chiral tetrahydroquinoline-2-carboxy-
late (9) derivative, which has been recently identified as a
pharmacophore of the glycine-site antagonist of a N-methyl-D-
aspartate (NMDA) receptor on neurons.4,5 Despite the importance
of Reissert compounds as a versatile chiral building block, no
asymmetric Reissert-type reaction of cyanide has been reported
so far, even by using a stoichiometric amount of a chiral
promoter.6 Developing a catalytic asymmetric Reissert-type
reaction is a great challenge, mainly due to the following two
reasons. First, strong electrophiles such as an acid halide or
TMSCl (generated during the reaction) could decompose the
catalyst by acylating and/or silylating the ligand. Second, the
conformation (the s-trans/s-cis isomers of the amide bond, see
11 and 12) of the reactive acyl quinolinium or isoquinolinium
ion7,8 is rather flexible. The two conformers would produce
opposing enantiomers even if TMSCN attacks the reactive
intermediate from a defined side. So, these two conformers should
be strictly differentiated by the catalyst. Despite these formidable
difficulties, we expected that using the bifunctional catalyst9 5, 6
should be advantageous for developing a catalytic asymmetric
Reissert-type reaction. The bifunctional catalyst has been reported
to promote the cyanosilylation reaction of aldehydes and imines
with high enantioselectivity, simultaneously activating both the
substrate and TMSCN by the Lewis acid and the Lewis base
moieties of the catalyst. In this paper, we disclose the first catalytic
asymmetric Reissert-type reaction promoted by the Lewis acid-
Lewis base bifunctional catalyst.
a For the representative procedure, see ref 14. b For entries 1-10
and 18, 1.1 equiv of acid chlorides and 2 equiv of TMSCN were used.
For entries 12-15, 2 equiv of 2-furoyl chloride and TMSCN were used.
For entries 11, 16 and 17, 4 equiv of 2-furoyl chloride and TMSCN
were used. c Isolated yield. d Determined by chiral HPLC analyses (see
Supporting Information). e TMSCN was added slowly over 12 h. f -60
°C. g 1:1. h 1:5. i Absolute configurations were determined to be R (see
Supporting Information).
the catalyst and quinoline 1a as the substrate at -40 °C. The
substituent (R) should have the predominant effect in controlling
the distribution of s-trans/s-cis amide conformers of the acyl
quinolinium intermediate. As shown in Table 1 (entries 1-6), it
was found that benzoyl chloride (entry 1) and 2-furoyl chloride
(entry 2) afforded higher enantiomeric excesses than aliphatic or
substituted aromatic acid chlorides.10 The more electron-rich and
(4) (a) Leeson, P. D.; Carling, R. W.; Moore, K. W.; Moseley, A. M.;
Smith, J. D.; Stevenson, G.; Chan, T.; Baker, R.; Foster, A. C.; Grimwood,
S.; Kemp, J. A.; Marshall, G. R.; Hoogsteen, K. J. Med. Chem. 1992, 35,
1954-1968. (b) Nagata, R.; Tanno, N.; Kodo, T.; Ae, N.; Yamaguchi, H.;
Tamiki, N.; Antoku, F.; Tatsuno, T.; Kato, T.; Tanaka, Y.; Nakamura, M. J.
Med. Chem. 1994, 37, 3956-3968.
(5) Chiral tetrahydroquinoline-2-carboxylate 9 has only been available by
the resolution of the racemic compound: (a) Paradisi, M. P.; Romeo, A. J.
Chem. Soc. Perkin Trans I 1976, 596-600. (b) Katayama, S.; Ae, N.; Nagata,
R. Tetrahedron: Asymmetry 1998, 9, 4295-4299.
(6) To the best of our knowledge, even diastereoselective reactions using
chiral acid chlorides have not been reported.
(7) For determination of the reactive acyl quinolinium ion intermediate of
a Reissert-type reaction, see: (a) Duarte, F. F.; Popp, F. D. J. Heterocycl.
Chem. 1991, 28, 1801-1804. (b) Abushanab, E.; Lee, D.-Y. J. Org. Chem.
1975, 40, 3376-3378. (c) Akiba, K.; Negishi, Y.; Inamoto, N. Synthesis 1979,
55-57.
(8) To exclude that the reaction mechanism involves the direct attack of
the cyanide to the activated quinoline coordinated to Al, followed by the trap
with an acid halide, we performed the following studies. No peaks corre-
sponding to the acyl quinolinium ion 10 were observed in 1H NMR by mixing
acid chlorides (PhCOCl and AcCl) and 1a in CH3CN, even in the presence
of the catalyst 5 or Et2AlCl from -50 °C to rt. Using AcBr, however, we
could confirm the generation of 10 (R ) CH3) by observing new peaks at 9.9
ppm (C-2 proton) and 3.1 ppm (CH3) (see also ref 7c). The ratio of 10 to 1a
depended on temperature (1: 1 at -40 °C and 1: 4 at 0 °C), indicating the
equilibrium between 10 and 1a. By adding only TMSCN to this equilibrium
mixture at 0 °C, peaks corresponding to 10 disappeared, and formation of the
Reissert compound was confirmed. The catalytic asymmetric Reissert reaction
mediated by 5 using AcBr gave 3a in 49% yield with 40% ee in CH2Cl2.
Furthermore, even in the presence of stoichiometric 5 or Et2AlCl, peaks
corresponding to the adduct of TMSCN to 1a were not observed in the absence
of acid halides. In addition, as will be mentioned in ref 13, addition of TMSCN
is not the major rate-determining step, which may not be consistent with the
mechanism involving the direct attack of cyanide to quinoline. Consequently,
it appears, at the moment, that the reaction would proceed via an acyl
quinolinium intermediate.
We started the project by investigating the effect of different
acid chlorides (RCOCl) on the reaction, using 5 (9 mol %) as
(1) (a) Reissert, A. Chem. Ber. 1905, 38, 1603. (b) For Lewis acid-catalyzed
Reissert-type reaction using TMSCN, see: Ruchirawat, S.; Phadungkul, N.;
Chuankamnerdkarn, M.; Thebtaranonth, C. Heterocycles 1977, 6, 43-46.
(2) (a) Popp, F. D. Heterocycles 1973, 1, 165-180. (b) McEwen, W. E.;
Cobb, R. L. Chem. ReV. 1955, 55, 511-549.
(3) (a) Lorsbach, B. A.; Bagdanoff, J. T.; Miller, R. B.; Kurth, M. J. J.
Org. Chem. 1998, 63, 2244-2250. (b) Lorsbach, B. A.; Miller, R. B.; Kurth,
M. J. J. Org. Chem. 1996, 61, 8716-8717.
(9) (a) Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki, M. J. Am. Chem.
Soc. 1999, 121, 2641-2642. (b) Takamura, M.; Hamashima, Y.; Usuda, H.;
Kanai, M.; Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39, 1650-1652.
10.1021/ja0010352 CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/17/2000

6328 J. Am. Chem. Soc., Vol. 122, No. 26, 2000
Communications to the Editor
therefore less reactive 2-furoyl chloride gave slightly better
enantioselectivity than benzoyl chloride. Next, we investigated
the effect of solvent by using the more reactive benzoyl chloride
as the acylating reagent. It was found that, when the polarity of
the solvent was decreased by adding a less polar solvent such as
toluene (entry 7) or pentane (entry 8), the ee increased to 78%
from 71%, although the yield of the product became much lower.11
In contrast, the ee decreased to 37% when the more polar
acetonitrile was used as solvent (entry 9). These experiments
suggested that lowering the concentration of the reactive acyl
quinolinium intermediate (by using 2-furoyl chloride or less polar
solvent) would give better enantioselectivity, although the chemi-
cal yield would be lower. This tendency may arise because the
spontaneous reaction (independent of the catalyst) of the acyl
quinolinium ion with TMSCN that gives the racemic product
could proceed with only a slightly slower rate than mediated by
the chiral catalyst. The spontaneous reaction could become less
negligible if the concentration of the acyl quinolinium becomes
high.
Figure 1. Working model for the catalytic cycle.
Scheme 1. Conversion to Tetrahydroquinoline-2-carboxylate
Thus, we planned to improve the reaction by developing a more
reactive catalyst. Our design for the new catalyst would facilitate
the attack of TMSCN on the acyl quinolinium ion by favoring
the catalyst conformer in which the Lewis acid and the Lewis
base moieties are optimally positioned to assist the reaction.12
By facilitating this step,13 the reaction rate should be increased
with the concentration of the reactive acyl quinolinium ion
maintained sufficiently low. Molecular modeling studies suggested
that if the steric bulkiness of the phosphine oxide’s aryl groups
is increased, the Lewis-basic oxygen atom would be put into a
position close to the acyl quinolinium ion activated by the Lewis
acid. On the basis of this idea, we synthesized a new catalyst 6
containing di-o-tolylphosphine oxide. Because catalyst 6 should
also have higher Lewis basicity than the original catalyst 5, we
expected that 6 would show higher reactivity as well as higher
enantioselectivity. Gratifyingly, by using 6 as the catalyst, both
the chemical yield and ee were improved to 49 and 83% (entry
10), respectively, from 27 and 78% (entry 7). Furthermore, using
4 equiv of TMSCN and 2-furoyl chloride, 3a was obtained in
91% yield with 85% ee (entry 11). For other reactive quinolines,
2 equiv of TMSCN and 2-furoyl chloride were used. By using
the optimized conditions, Reissert compounds were obtained in
up to 91% ee and 99% yield in the case of the reactive, electron-
rich quinolines 1b-1e (entries 12-15).14 The less reactive
substrates 1f and 1g gave less satisfactory results, although still
yielding the products in 67% ee (entries 16, 17). It is also
important that this reaction could be applied to the isoquinoline
derivative 2 (entry 18), giving the product 4 with 71% ee in 99%
yield by using 5 as the catalyst and acetyl chloride as the acylating
reagent.15
purity (Scheme 1). By comparison of the optical rotation of 9
with the literature value,5 the absolute configuration of the Reissert
compounds was determined to be R.
Although a detailed reaction mechanism is still not clear at
this moment, the reaction should be promoted by the dual
activation of the acyl quinolinium or isoquinolinium ion and
TMSCN by the Lewis acid (Al) and the Lewis base (oxygen atom
of the phosphine oxide) moieties of the catalyst, respectively. The
results using the control catalyst 7 implied the dual activation
mechanism by 5 and 6. 7 contained diphenylmethyl groups
providing only steric bulkiness without Lewis basicity. Thus,
catalyst 7 afforded 3a (R ) Ph, 24% ee in 73% yield) and 3d (R
) 2-furyl, 46% ee in 95% yield) with the opposite configuration
(S). These results suggest that, in the case of 5 and 6, TMSCN
appears to attack the acyl quinolinium intermediate from the
phosphine oxide’s side. Therefore, we postulated the working
model for the catalytic cycle depicted in Figure 1. The first step
should be the formation of the reactive acyl quinolinium
intermediate 10 by the reaction of quinoline with the acid
chloride.7,8 The acyl quinolinium ion should be activated by
complexation of the amide oxygen to the Lewis acid (Al). The
two conformers of the amide bond 11 and 12 would exist in
equilibrium. However, when TMSCN is activated by the Lewis-
base moiety of the catalyst, the reaction via 11 would be more
favorable than via 12. In the case of 12, the distance between the
activated TMSCN and the electrophilic carbon would be too far
for catalysis. The hypothetical transition state 11 could explain
the absolute configuration of the product was R.
In summary, the first catalytic asymmetric Reissert-type
reaction of quinoline and isoquinoline derivatives has been
achieved by using a Lewis acid-Lewis base bifunctional catalyst.
When quinoline derivatives were used as the substrates, the new
catalyst 6, containing di-o-tolylphosphine oxide, was found to
be a better catalyst than the original catalyst 5, in terms of activity
as well as enantioselectivity. The Reissert compound was suc-
cessfully converted to tetrahydroquinoline-2-carboxylate (9),
without any loss of enantiomeric purity. Further investigations
to increase the generality of substrates and to clarify the precise
reaction mechanism, as well as applications to the catalytic
asymmetric total synthesis of biologically active natural products,
are currently in progress.
The Reissert product was successfully converted to the tetra-
hydroquinoline-2-carboxylate 9 without any loss of enantiomeric
(10) No acylated or silylated ligand was observed by TLC analysis.
(11) Longer reaction time did not improve the chemical yield.
(12) For the strategy to restrict the conformation of the ligand for the dual
activation pathway, see: Kanai, M.; Hamashima, Y.; Shibasaki, M. Tetrahe-
dron Lett. 2000, 41, 2405-2409.
(13) The major rate-determining step would be the formation of the acyl
quinolinium ion. However, the cyanation step was found to have some
contribution to the reaction rate. Kinetic studies indicated that the total reaction
rate was 0.2 order with respect to TMSCN. See Supporting Information.
(14) A representative procedure: To the solution of the ligand (22 mg,
0.029 mmol) in CH2Cl2 (2.5 mL), Et2AlCl (30 µL, 0.029 mmol in hexane)
was added at ambient temperature, and the resulting solution was stirred for
1 h. This catalyst solution of 6 was cooled to -40 °C, and a solution of 1d
(60.5 mg, 0.32 mmol) in CH2Cl2 (0.5 mL) was added, followed by 2-furoyl
chloride (63 µL, 0.64 mmol). After adding toluene (2.5 mL), TMSCN (85
µL, 0.64 mmol) in toluene (0.5 mL) was added slowly over 24 h at -40 °C.
After 40 h, saturated aqueous solution of NaHCO3 was added. Usual workup
and purification by silica gel column chromatography (AcOEt/hexane, 3/7)
gave the product in 99% yield. The ee was determined to be 91% by chiral
Acknowledgment. Financial support was provided by CREST, The
Japan Science and Technology Corporation (JST), and RFTF of Japan
Society for the Promotion of Science.
Supporting Information Available: Experimental procedures and
characterization of the products (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
i
HPLC analysis (DAICEL CHRALPAK AS, PrOH/hexane, 3/7, flow ) 1.0
mL/min, retention time, 15.7 min (major), 20.8 min (minor)).
(15) The absolute configuration was tentatively assigned. Using catalyst 6
with benzoyl chloride, 2-furoyl chloride or CH2Cl2-toluene mixed solvent
with 5 gave less satisfactory results (32, 48, and 59% ee’s, respectively).
JA0010352