Highly reactive and stereospecific reaction of quinoline-type NADH model compounds with methyl benzoylformate

Article abstract of DOI:10.1016/S0040-4039(99)02227-3

NADH model compounds with a dihydroquinoline ring can reduce methyl benzoylformate with very high reactivity and stereospecificity. Chiral column HPLC is effective in the separation and analysis of enantiomers of NADH and NAD model compounds, respectively. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02227-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1035–1038
Highly reactive and stereospecific reaction of quinoline-type
NADH model compounds with methyl benzoylformate
a,∗
a
a
a
a
Yuji Mikata, Kayo Hayashi, Kiyoko Mizukami, Sawako Matsumoto, Shigenobu Yano,
b
b
Norimasa Yamazaki and Atsuyoshi Ohno
a
Department of Chemistry, Faculty of Science, Nara Women’s University, Nara 630-8506, Japan
b
Institute for Chemical Research, Kyoto University, Uji 611-0011, Japan
Received 27 August 1999; revised 25 October 1999; accepted 26 November 1999
Abstract
NADH model compounds with a dihydroquinoline ring can reduce methyl benzoylformate with very high
reactivity and stereospecificity. Chiral column HPLC is effective in the separation and analysis of enantiomers
of NADH and NAD model compounds, respectively. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: NADH model; NAD model; axial chirality; HPLC; optical resolution.
The elucidation of the origin of A/B face specificity of the nicotinamide coenzyme in dehydrogenase
is an exciting field in biological chemistry.1,2 In one interesting explanation of this specificity, it has been
demonstrated that the orientation of the carbamoyl oxygen in the nicotinamide coenzyme governs the
3
reactivity of the hydrogen atom in the hydride transfer reaction. We and other chemists have attempted
4–8
to account for this reactivity through calculation or model studies. Vekemans et al. have synthesized
,2,4,N,N-pentamethyl-1,4-dihydronicotinamide as a simple NADH model and studied its reaction with
1
7
methyl benzoylformate. However, this model compound is too reactive at room temperature to exert sat-
isfactory stereospecificity. Here we describe preliminary studies using novel quinoline-type NADH mo-
del compounds with simple structures: 3-N,N-dimethylcarbamoyl-1,2,4-trimethyl-1,4-dihydroquinoline
(1a) and 3-piperidinylcarbonyl-1,2,4-trimethyl-1,4-dihydroquinoline (1b) (Fig. 1). These compounds
undergo highly stereospecific reaction at room temperature.
NAD model compounds, 3-N,N-dialkylcarbamoyl-1,2,4-trimethylquinolinium iodide (2) were prepar-
4
ed from 2,4-dimethylquinoline-3-carboxylic acid by the procedure shown in Scheme 1 (WSC: 1-ethyl-
3
-(3-dimethylaminopropyl)carbodiimide, HOBt: 1-hydroxy-1H-benzotriazole). Chiral HPLC analysis
®
(column: Daicel Chiralcel OD , eluent: n-hexane/i-PrOH=7/3 with 0.1% diethylamine) revealed that
these compounds have axial chirality with respect to the carbamoyl orientation, and the enantiomer pairs
exist with considerable stability. The racemic compounds 3a and 3b could be preparatively separated by
∗
Corresponding author. E-mail: mikata@cc.nara-wu.ac.jp (Y. Mikata)
0
040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02227-3
tetl 16155

1
036
Fig. 1. Structures of NAD and NADH model compounds
HPLC into their enantiomers, but no preparative separation was achieved for the racemic pairs 2a and
b.
2
Scheme 1.
The 1,4-reductions of compounds 2a and 2b were performed in 80–90% yield by sodium hydrosulfite
9
in an aqueous sodium bicarbonate/dichloromethane biphasic mixture (Scheme 1). The corresponding
pyridine compounds resisted this reduction, although a similar reduction had been reported earlier.7
The separations of enantiomers 1a and 1b were successfully performed by HPLC (column: Daicel
,10
®
Chiralcel OD (2 cmφ×25 cm), eluent: n-hexane/i-PrOH=20/1 for 1a; 9/1 for 1b).
In our earlier studies, the chiral separation of the 3–N,N–dialkylcarbamoyl-1,2,4-trimethyl-1,4-
dihydroquinoline systems modified at the 4-position had been performed by recrystallization from a
4
diastereomer mixture of the crystalline precursor using a chiral auxiliary in the side chain. The present
compounds are the first quinoline-type NADH models in which enantiomer pairs are well resolved. The
absolute configurations of these NADH model compounds were assigned by the comparison of their CD
spectra (Fig. 2) with those of previous compounds.7
,11
The asymmetric reduction of methyl benzoylformate with these chiral NADH model compounds
was performed. When an equimolar amount of methyl benzoylformate was added to an acetonitrile-
d solution of 1 and 1 equiv. of magnesium perchlorate, the reaction was completed within 5 min as
3
1
evidenced by H NMR measurement. The reaction proceeds only in the presence of magnesium ion.
Judging from GC analysis of the product (column: CP-cyclodextrin-B-236-M-19 (Chrompack ), 25
®
m×0.25 mm i.d.), methyl mandelate (80–93% isolated yield) was obtained in excellent stereospecificity
Table 1), and the configuration of the product was identical with previous studies.4 (R)-1 Affords
,7,9
(
(
R)-mandelate and vice versa. The moderate reactivity of the quinoline ring system generated a stable
ternary complex in the transition state even in room temperature reactions.
The NAD model compounds obtained from these reactions are also stereospecific. The determination
of absolute configurations of these compounds was not straightforward because CD measurements are not

1
037
Fig. 2. CD spectra of 1a (a) and 1b (b) in EtOH
Table 1
Asymmetric reduction of methyl benzoylformate with NADH models (1)
7
applicable in these compounds. In this study, configurations were determined by HPLC retention time.
As above, the quinolinium racemates separated into two peaks by HPLC when the case eluent contains
0
.1% diethylamine. When diastereomer pairs 4 (Fig. 3) which have known carbonyl configurations12
were subjected to HPLC analysis with same column, the R-isomer eluted earlier than S-isomer in both
cases. Thus, we assigned the configuration of the earlier eluted enantiomer of compound 2 as the R-
isomer. This assignment is reasonable because a similar interaction is expected in quinoline and pyridine
model compounds toward the HPLC column. Also, in light of the reaction mechanism of 1 with methyl
7
,11
benzoylformate,
this assignment should be correct.
Fig. 3. Configuration and HPLC retention time of pyridine derivative
In this study, we synthesized quinoline-type NADH model compounds with simple structures that
have high reactivity and selectivity. In addition, we have determined the enantiomer ratio of NAD model
compounds with axial chirality by HPLC analysis in a precise and convenient manner. This procedure

1
038
would be easily applicable to the other NAD/NADH model compounds. Detailed studies of the reaction
of 1 with other substrates as well as the reduction of 2 with several reducing agents are now under
investigation.
Acknowledgements
We thank Mr. Atsushi Ohnishi of Daicel Company with helpful advice with HPLC measurements, and
Prof. Ryuichi Arakawa of Kansai University for the ESI-MS measurement of 1.
References
1
2
3
4
5
6
7
. You, K. CRC Crit. Rev. Biochem. 1984, 17, 313–451.
. Nambiar, K. P.; Stauffer, D. M.; Kolodziej, P. A.; Benner, S. A. J. Am. Chem. Soc. 1983, 105, 5886–5890.
. Donkersloot, M. C. A.; Buck, H. M. J. Am. Chem. Soc. 1981, 103, 6554–6558.
. Ohno, A.; Kashiwagi, M.; Ishihara, Y.; Ushida, S.; Oka, S. Tetrahedron 1986, 42, 961–973.
. Ohno, A.; Ohara, M.; Oka, S. J. Am. Chem. Soc. 1986, 108, 6438–6440.
. de Kok, P. M. T.; Buck, H. M. J. Chem. Soc., Chem. Commun. 1985, 1009–1010.
. de Kok, P. M. T.; Bastiaansen, L. A. M.; van Lier, P. M.; Vekemans, J. A. J. M.; Buck, H. M. J. Org. Chem. 1989, 54,
1
313–1320.
8
9
. Ohno, A.; Ishikawa, Y.; Yamazaki, N.; Okamura, M.; Kawai, Y. J. Am. Chem. Soc. 1998, 120, 1186–1192.
1
. Selected data for 1a: H NMR (300 MHz, CDCl
N-Me), 3.22 (s, 3H, N-Me), 3.56 (br., 1H, 4-H), 6.84 (br. d, 1H, J=7.8 Hz, ArH), 6.92 (m, 1H, ArH), 7.08 (dd, 1H, J=1.5,
3
): δ (ppm) 1.20 (d, J=6.6 Hz, 3H, 4-Me), 1.99 (s, 3H, 2-Me), 3.02 (s, 6H,
+
+
+
+
+
7
5
6
.8 Hz, ArH), 7.19 (m, 1H, ArH). ESI-MS: m/z 229 [M−Me] , 243 [M−H] , 245 [M+H] , 267 [M+Na] , 490 [2M+H] ,
12 [2M+Na] , 755 [3M+Na] . For 1b: H NMR (300 MHz, CDCl ): δ (ppm) 1.20 (d, J=6.6 Hz, 3H, 4-Me), 1.5–1.8 (m,
H, -CH
-), 2.00 (s, 3H, 2-Me), 3.20 (s, 3H, N-Me), 3.4–3.8 (m, 5H, –CH2−+4-H), 6.80 (br. d, 1H, J=7.8 Hz, ArH), 6.90
+
+
1
3
2
+
+
+
(m, 1H, ArH), 7.05 (br. d, 1H, J=7.8 Hz, ArH), 7.20 (m, 1H, ArH). ESI-MS: m/z 269 [M−Me] , 283 [M−H] , 285 [M+H] ,
+
+
+
+
3
07 [M+Na] , 570 [2M+H] , 592 [2M+Na] , 875 [3M+Na] .
1
0. de Kok, P. M. T.; Donkersloot, M. C. A.; van Lier, P. M.; Meulendijks, G. H. W. M.; Bastiaansen, L. A. M.; van Hooff, H.
J. G.; Kanters, J. A.; Buck, H. M. Tetrahedron 1986, 42, 941–959.
1
1
1. Ohno, A.; Ikeguchi, M.; Kimura, T.; Oka, S. J. Am. Chem. Soc. 1979, 101, 7036–7040.
2. Ohno, A.; Mikata, Y.; Goto, M.; Kashiwagi, T.; Tanaka, T.; Sawada, M. Bull. Chem. Soc. Jpn. 1991, 64, 81–86.