Rational design of novel axially chiral NADH models based on configurational control of atropisomeric lactams

Article abstract of DOI:10.1016/S0040-4039(01)00555-X

The preparation of a new class of axially chiral NADH models based on configurational control of atropisomeric lactams is described. The configurational control of this chiral axis is achieved via the presence of a second chirality element installed on th

Full text of DOI:10.1016/S0040-4039(01)00555-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 3713–3716
Rational design of novel axially chiral NADH models based on
configurational control of atropisomeric lactams
Jean-Luc Vasse, Georges Dupas, Jack Duflos, Guy Que´guiner, Jean Bourguignon and
Vincent Levacher*
Laboratoire de Chimie Organique Fine et He´te´rocyclique associe´ au CNRS, IRCOF-INSA, rue Tenie`res BP 08,
F-76131 Mont Saint Aignan Cedex, France
Received 27 March 2001; revised 30 March 2001; accepted 3 April 2001
Abstract—The preparation of a new class of axially chiral NADH models based on configurational control of atropisomeric
lactams is described. The configurational control of this chiral axis is achieved via the presence of a second chirality element
installed on the lactam moiety of the reagent 2. Reduction of methyl benzoylformate led to methyl mandelate in 89% yield and
92% e.e. This result suggests that the chiral axis about C3ꢀCꢁO bond is the main configurational element responsible for the high
enantioselectivity observed with this biomimetic model. © 2001 Elsevier Science Ltd. All rights reserved.
The out-of-plane orientation of the nicotinamide amide
adopted in the coenzyme seems to be an important
conformational feature in the stereochemical outcome
of the hydride transfer. It has been suggested that the
amide carbonyl dipole would be syn oriented with
respect to the transferring hydrogen (Fig. 1).1 Conse-
quently it was of interest to mimic this conformational
effect by considering the out-of-plane orientation of the
CꢁO as a novel chiral element in the design of axially
chiral NADH models.
enantiopure preparation of such models seriously limits
their potential as asymmetric reducing agents.
We previously reported the high performance of models
1 during the reduction of methyl benzoylformate.3 Our
interpretation for the high stereoselectivity observed is
founded on the following considerations. The cyclic
structure sets the CꢁO lactam out-of-plane with respect
to the dihydropyridine ring with a dihedral angle of
about 50°. If one assumes that the stereoselective depar-
ture of the syn oriented hydrogen is favoured, an
out-of-plane orientation of the carbonyl lactam raises
questions about the stereocontrol of this masked chiral
axis. Given the good performance of model 1, it may be
assumed that the stereocontrol of this new chiral axis
would be ensured via the presence of the chiral auxil-
iary at the nitrogen lactam which would give rise to a
dynamic resolution process of both conformational
atropodiastereoisomers (aR,S)-1 and (aS,S)-1 (Fig. 3).
To argue for this hypothesis and to design new axially
A number of research groups2 have been interested in
the use of atropisomeric amides as a chiral relay for the
preparation of enantiopure NADH models chiral at
C-4 (Fig. 2). Indeed, it allows the stereoselective reduc-
tion of quinolinium A to provide enantiopure model B.
This former model was shown to be highly enantio-
selective during the reduction of methyl benzoylformate.
In the course of this reduction process, axial chirality of
quinolinium A is restored with a high level of stereo-
control. Although, it has been suggested an out-of-
plane orientation of the carbonyl, the presence of a
chiral centre at C-4 prevents us from appreciating the
role played by this conformational effect and its degree
of participation in the stereochemical outcome of the
reduction (Fig. 2). From a practical point of view, the
Substrate
Hsyn
O
H
NH2
N
Keywords: NADH; atropisomeric lactam; conformational atropiso-
mers; asymmetric reduction.
* Corresponding author. Fax: +33 (0) 2 35 52 29 62; e-mail: vincent.
levacher@insa-rouen.fr
Figure 1. Out-of-plane orientation of nicotinamide amide
moiety in the active conformation of the coenzyme NADH.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00555-X

3714
J.-L. Vasse et al. / Tetrahedron Letters 42 (2001) 3713–3716
O
Me
Me
H
OH
COOMe
Ph
N
aR
Ph
PNAH
N
+
Me
(S)-methyl mandelate
e.e. =99%
A
Mg(ClO4)2
Me
O
Me
H
O
Ph
N
Me
S
PNA+
Ph
COOMe
N
B
Figure 2. Use of atropisomeric lactams as chiral relay in the design of chiral NADH models. Intramolecular transfer of an axial
chiral element to a central chiral element and vice versa.
Reagent
O
H
Ph
Ph
H
Et
O
O
MeO
MeO
H
H
N
H
H
H
H
OH
OH
H
N
N
(S)
(S)
aS
aR
N
N
N
Stabilisation
(aR,S)-1
(aS,S)-1
Stereocontrol
Model 2
Figure 3. (a) Configurational analysis of model 1 and (b) design of a new axially chiral NADH model 2.
chiral NADH models based on configurational control
of atropisomeric lactams, we planned to prevent this
interconversion between the two atropodiastereoiso-
mers by means of a chiral auxiliary carried out on the
lactam moiety as depicted in Fig. 3. We report herein
the stereoselective synthesis of such a model and its
potential as an enantioselective reducing agent in the
reduction of methyl benzoylformate. The high stability
of annelated models,3b,c incited us to undertake the
stereoselective synthesis of model 2 in benzazepino[4,5-
b]quinoline series (Fig. 3).
obtained by acetylation followed by reduction in 98%
overall yield. Treatment of 7 with the acid chloride 85
derived from ethyl malonate afforded the requisite
amide ester 9 in 97% yield which was subsequently
hydrolysed to furnish the expected amide acid 10 in
82% yield. The crux of this reaction scheme is the
cyclisation step of 10 via an intramolecular Friedel–
Crafts acylation. Investigation of the literature indi-
cated that similar SEAr cyclisation have been achieved
to effect seven-membered rings closure in high yields
with an acid chloride in the presence of AlCl3 in
CH2Cl2.6 Compound 10 was thus converted into the
corresponding acid chloride prior to cyclisation. The
desired b-keto lactam 4 was obtained under the former
conditions, albeit in a modest yield of 30% (Scheme 2).
The synthesis of model 2 is based on a Friedlander type
condensation4 of the amino imine 3 and the tetrahydro-
napht[1,8-bc]azepine-2,4-dione 4 to afford the quinoline
derivative 5 followed by the classical steps of quater-
nization and regioselective reduction of the correspond-
ing quinolinium salt 6 (Scheme 1).
The known amino imine 37 was then involved in the
Friedlander reaction under Borsche conditions with the
b-keto lactam
4 to afford the benzazepino[5,4-
The preparation of the desired tetrahydro-napht[1,8-
bc]azepine-2,4-dione 4 was achieved in five steps start-
ing from the commercially available (S)-1,2,3,4-tetra-
hydronapht-1-yl-amine. The N-ethyl derivative 6 was
b]quinoline 5 in 76% yield. Treatment of the former
with methyl trifluoromethanesulfonate gave rise to the
benzazepino[5,4-b]quinolinium salt 6 in 95% yield. The
regioselective reduction was then investigated under
O
O
O
H H
N
Et
Et
Et
MeO
MeO
MeO
MeO
N
MeO
MeO
Tol
+
N
N
N
H
H
H
N
O
NH2
5
3
6
quaternization
reduction
2
4
Friedlander
Scheme 1. Retrosynthetic analysis of model 2.

J.-L. Vasse et al. / Tetrahedron Letters 42 (2001) 3713–3716
3715
this conformational control, the absolute configuration
of the chiral axis about the C3ꢀCꢁO bond is maintained
and subordinated to the absolute configuration of this
cyclic chiral auxiliary. Thus, starting from (S)-tetra-
hydro-napht[1,8-bc]azepine-2,4-dione 4 as chiral auxil-
iary, the resulting masked chiral axis of model 2
possesses an (aR) absolute configuration. The confor-
mational analysis also indicated that the C(4)ꢀC(3)ꢀCꢁ0
dihedral angle is about 46° (Fig. 4).
NH2
NHEt
c
a,b
97%
98%
7
O
O
O
Et
Et
N
RO
N
H
e,f
O
30%
4
With model (aR,S)-2 in hand, we investigated the
reduction of methyl benzoylformate in the presence of
magnesium perchlorate in acetonitrile (Scheme 4).11
Under those conditions, model 2 is shown to be highly
enantioselective, affording methyl mandelate in up to
92% e.e. (S).
R=Et:9
R=H:10
82%
d
Scheme 2. Reagents and conditions: (a) CH3COCl/NEt3/
CH2Cl2/0°C/2 h; (b) BH3/I2/reflux/12 h; (c) EtOOCCH2COCl
8/NEt3/CH2Cl2/rt/12 h; (d) KOH/EtOH/rt/5 h; (e) (COCl)2/
CH2Cl2/DMF cat/rt/; (f) AlCl3 (2.2 equiv.)/rt/1 h.
As shown in Fig. 4, the conformation adopted by the
model sets the chiral auxiliary in the middle plane of
the dihydroquinoline leaving both diastereotopic hydro-
gen atoms at C-4 accessible to the substrate. It is
reasonable to assume that the stereocontrol of the
reduction arises from the axial chirality based on
classical reduction conditions, i.e. with sodium dithio-
nite.8 Under these conditions, a mixture of unidentified
products was obtained. After screening various reduc-
ing agents, we found out that sodium borohydride lead
in good yield (97%) to the desired model 2 without
formation of by-products (Scheme 3).
OH
COOMe
H
O
a
As it could be anticipated from a conformational analy-
(aR,S)-2
+
1
sis, H NMR spectra of model 29 reveals the presence
Ph
COOMe
of a single conformational diastereoisomer. Indeed,
owing to conformational restraints, the methylene at
the chiral centre of the cyclic chiral auxiliary is con-
strained to adopt an eclipsed conformation with the
neighbouring N-ethyl substituent. As a consequence of
e.e.=92% (S)
yield=89%
Scheme 4. Asymmetric reduction of methyl benzoyformate.
Reagents and conditions: (a) Mg(ClO4)2/CH3CN/rt/24 h.
O
O
O
Et
Et
Et
MeO
MeO
MeO
MeO
Tol
+
N
MeO
MeO
N
N
a
N
b
H
H
H
76%
95%
O
N
NH2
N
+
TfO-
4
3
6
5
O
H
H
Et
MeO
MeO
N
c
H
97%
N
2
Scheme 3. Reagents and conditions: (a) piperidine/EtOH/reflux/12 h; (b) TfOMe/CH2Cl2/rt/2 h; (c) NaBH4/EtOH/rt/2 h.
Figure 4. (a) Conformational and configurational analysis of model 2; (b) molecular modelling (MM2, PCMODEL) of model 2.10

3716
J.-L. Vasse et al. / Tetrahedron Letters 42 (2001) 3713–3716
J.-L.; Levacher, V.; Dupas, G.; Queguiner, G.; Bour-
S
O
guignon, J. Tetrahedron, in press.
-
O
ClO4
4. Borsche modification of the Friedlander procedure pre-
vents the formation of by-products due to autocondensa-
tion reactions, see: (a) Borsche, W.; Ried, W. Liebigs
Ann. Chem. 1943, 554, 269; (b) Borsche, W.; Barthen-
heier, J. Liebigs Ann. Chem. 1941, 548, 50.
Mg2+
O
-
ClO4
Hsyn
O
H
N
Et
N
H
5. Box, V. G. S.; Marinovic, N.; Yiannikouros, G. P. Hete-
rocycles 1991, 32, 245.
6. (a) Mamouni, A.; Da¨ıch, A.; Decroix, B. J. Heterocyclic
Chem. 1996, 33, 1251; (b) Da¨ıch, A.; Decroix, B. J.
Heterocyclic Chem. 1996, 33, 873.
7. Porter, H. K. The Zinin reduction of nitroarenes; Org.
React. 1973, 20, 455–481.
Figure 5. Proposed ternary complex involved in the reduction
of methyl benzoyformate to account for the formation of
(S)-methyl mandelate.
8. Blankenhorm, G.; Moore, E. G. J. Am. Chem. Soc. 1980,
102, 1092.
Cꢀ3ꢀCꢁO bond. As previously proposed in the litera-
ture,2a the sense of the stereoselectivity observed may be
tentatively explained by the establishment of a ternary
complex model/Mg2+/substrate via the complexation of
the CꢁO lactam with magnesium ion to promote the
stereoselective transfer of the syn oriented hydrogen
(Fig. 5).
9. Spectral data 1H NMR (CDCl3, 200 MHz). Compound
2: l 0.84 (3H, t, J=7.0 Hz), 1.92–2.42 (4H, m), 2.78–2.95
(2H, m), 3.03 (3H, s), 3.32 (1H, six, J=7.0 Hz), 3.58 (1H,
d, J =18.0 Hz), 3.82 (1H, six, J=7.0 Hz), 3.86 (3H, s),
3.90 (3H, s), 4.35 (1H, d, J=18.0 Hz), 4.67 (1H, dd,
J=7.0 Hz, J=3.8 Hz), 6.55 (1H, s), 6.68 (1H, s), 7.12–
7.22 (3H, m). Compound 6: l 1.04 (3H, t, J=7.0 Hz),
2.08 (2H, m), 2.35 (2H, m), 2.86–3.03 (2H, m), 3.57 (1H,
quint, J=7.0 Hz,) 3.84 (1H, quint, J=7.0 Hz), 4.07 (3H,
s), 4.21 (3H, s), 4.41 (3H, s), 4.93 (1H, m), 7.36 (1H, s),
7.37–7.44 (4H, m), 7.61 (1H, s), 9.10 (1H, s). Compound
5: l 1.07 (3H, t, J=7.0 Hz), 1.90–2.15 (3H, m), 2.36 (1H,
m), 2.70–2.82 (1H, m), 2.92–3.05 (1H, dt, J=17.0, 3.5
Hz), 3.55 (1H, six, J=7.0 Hz), 4.00 (1H, six, J=7.0 Hz),
4.02 (3H, s), 4.04 (3H, s), 4.86 (1H, m), 7.13 (1H, s), 7.20
(1H, d, J=7.5 Hz), 7.38 (1H, t, J=7.5 Hz), 7.96 (1H, d,
J=7.5 Hz), 8.62 (1H, s).
In contrast to previous models reported in literature,
absence of chirality at C-4 does not throw discredit on
the role played by this conformational effect in the
stereochemical outcome of the reduction. This configu-
rational control process of atropisomeric lactams by
means of a second chirality element on the lactam
moiety opens up interesting perspectives to design new
axially chiral NADH models and to gain further insight
into the role of the out-of plane orientation of the
carbonyl amide in the coenzyme itself.
10. Molecular mechanics (MM2) and MOPAC (PM3) pro-
gram led to the same conformation.
11. Procedure for the reduction of methyl benzoylformate: In a
flask, flushed with argon, were introduced model (aR,S)-
2 (405 mg, 1 mmol), acetonitrile (3 mL), methyl benzoyl-
formate (142 mL, 1 mmol) and magnesium perchlorate
(220 mg, 1 mmol). The resulting solution was stirred at
room temperature for 24 h in the dark. After addition of
water (10 mL), the organic solvent was evaporated under
reduced pressure and the resulting aqueous phase was
extracted with CH2Cl2 (3×10 mL). After drying (MgSO4)
and evaporation of the solvent, the residue was chro-
matographed on silica gel (eluent Et2O/cyclohexane:2/1).
Yield: 89%. Enantiomeric excesses were determined by
HPLC analysis using a Chiracel OD column (250×4.6
mm, 10 mm). Chromatographic conditions: injection: 20
mL (0.5 mg of methyl mandelate in 10 mL of hexane).
Eluent: hexane/2-propanol: 90/10. Flow rate: 1 mL/min.
Pressure: 300 psi. Temperature: 22°C. UV detection: u=
235 nm. Retention time: 9.2 min [(S)-enantiomer] and
14.8 min [(R)-enantiomer]. Enantiomeric excess: 92% (S).
References
1. (a) Ecklund, H.; Samama, J. P.; Jones, T. A. Biochem-
istry 1984, 23, 5982; (b) Skarzynski, T.; Moody, P. C. E.;
Wonacott, A. J. J. Mol. Biol. 1987, 193, 171; (c)
8
Almarsson, O.; Bruice, T. C. J. Am. Chem. Soc. 1993,
115, 2125–2138; (d) Wu, Y.-D.; Houk, K. N. J. Org.
Chem. 1993, 58, 2043–2045.
2. (a) 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, 1313–1320; (b) Mikita, Y.; Hayashi, K.;
.
Mizukami, K.; Matsumoto, S.; Yano, S.; Yamazaki, N.;
Ohno, A. Tetrahedron Lett. 2000, 41, 1035–1038; (c)
Ohno, A.; Kashiwagi, M.; Ishihara, Y. Tetrahedron 1986,
42, 961–973.
3. (a) Be´dat, J.; Levacher, V.; Dupas, G.; Queguiner, G.;
Bourguignon, J. Chem. Lett. 1996, 359–360; (b) Levacher,
V.; Dupas, G.; Queguiner, G.; Bourguignon, J. Trends
Heterocyclic Chem. 1995, 4, 293–302; (c) Vitry, C.; Vasse,