Stereomodulating effect of remote groups on the NADH-mimetic reduction of alkyl aroylformates with 1,4-dihydronicotinamide-β-lactam amides

Article abstract of DOI:10.1016/j.tet.2010.02.085

Conformationally restricted NADH peptidomimetics 4a-e, characterized by the presence of a (1,4-dihydronicotinamide)-(β-lactam) moiety, have been synthesized and used to study the Mg²⁺ cation-promoted asymmetric reduction of alkyl aroylformates

Full text of DOI:10.1016/j.tet.2010.02.085

Tetrahedron 66 (2010) 3187–3194
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Stereomodulating effect of remote groups on the NADH-mimetic reduction
of alkyl aroylformates with 1,4-dihydronicotinamide-b-lactam amides
*
´
´
Jesus M. Aizpurua , Claudio Palomo, Raluca M. Fratila, Pablo Ferron, Jose I. Miranda
´
´
´
Departamento de Qu´ımica Organica-I, Universidad del Paıs Vasco, Joxe Mari Korta R&D Center, Avda. Tolosa-72, 20018 San Sebastian, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 12 January 2010
Received in revised form 18 February 2010
Accepted 22 February 2010
Available online 25 February 2010
Conformationally restricted NADH peptidomimetics 4a–e, characterized by the presence of a (1,4-di-
hydronicotinamide)-(
promoted asymmetric reduction of alkyl aroylformates in acetonitrile. Increasing the bulkiness of
peripheral substituents at the nitrogen atom of the -lactam ring, at the 1,4-dihydronicotinamide moiety,
b
-lactam) moiety, have been synthesized and used to study the Mg2þ cation-
b
or at the aroylformate ester group, was found to cause weak but clearly detectable variations of the
enantiomeric excess of the reaction. A rational for these observations was consistent with a chelated
NADH/Mg2þ/ArCOCO2R3 ternary complex model, according to DFT calculations computed at a B3LYP/6-
Keywords:
NADH
31G theory level.
*
b
-Lactams
Ó 2010 Elsevier Ltd. All rights reserved.
DFT calculations
Enantioselective reduction
1. Introduction
R
R
H
H
(β
O
H
-LSAD)
O
H
H
H
Nicotinamide adenine dinucleotide (NADþ) and its reduced
form NADH are cofactors of -lactate dehydrogenase, which is
N
R1
6
5
7
N
Add 1C atom
R1
4
8
N
H
N
3
*
*
L
H
α
ψO
φ
O
2
= 120
ψ
involved in the enantioselective reduction of pyruvate during
anaerobic glycolysis. As NADH binds to the enzyme, different
conformational changes occur at the active site,1 resulting in the
controlled orientation of the substrate approach and in the
stereoselective transfer of one of the diastereotopic hydrogen
atoms in NADH to the prochiral pyruvate.2 The 1,4-dihi-
dronicotinamide moiety acts in NADH as an inherently reactive
and recyclable center for the catalytic activity.3 Thus, many or-
ganic and bioorganic chemists have been challenged to mimic
the activity of the NADþ/NADH redox system. Since the first
asymmetric reduction using an NADH mimic reported by Ohno4
in 1975 a large number of approaches to NADH mimicking have
been made.5 Conventional strategies to design an efficient bio-
mimetic system include the synthesis of 1,4-dihydropyridines
with a chiral center at C46 and/or chiral side arms on the car-
bamoyl group.7–10 Generally, the nonenzymatic reduction of ac-
N
N1
Me
Me
2
1
Figure 1. The ‘
formal insertion in the native peptide of one single carbon atom (C
CH2–N) provides pseudopeptides 2 rigidified around the
(z120ꢀ) torsion angle.
Definition of dihedral angles:
: (C2–C3–C4–N5), : (C4–N5–C6–C7), : (N5–C6–C7–N8).
b
-Lactam Scaffold-Assisted Design’ (
b
-LSAD) approach to NADH models:
a
–HþH–N/C
a–
j
a
4
j
represent, not only interesting artificial models to mimic the
chiral field of enzymatic NADH reductions, but also excellent
probes to conduct mechanistic and computational studies on the
origin of the stereoselectivity of Mg2þ-promoted nonenzymatic
reactions. Since they are able to form strongly chelated in-
termediates with Mg2þ cation, they can be used to further un-
derstand the architecture of the NADH/Mg2þ/ArCOCO2R ternary
complexes and the subsequent transition states leading to the
nonracemic reduction products.
tivated carbonyl compounds (e.g.,
a-keto esters) is achieved in
the presence of a suitable divalent metal ion (most often Mg2þ
and Zn2þ), which plays a fundamental role in the activation and
stereodirection of the hydride transfer mechanism.11 Within this
context, 1,4-dihydronicotinamide peptides 112,13 (Fig. 1)
Recently, we have introduced the puckering-free
b-lactam
NADH mimetics 2 to decrease the flexibility around the chelated
reaction intermediates in these reduction reactions.14 Inspired in
the Freidinger’s inter-residual
icotinamide- -lactam peptides 2 were designed to have simulta-
neously a minimal scaffold and a maximal rigid constraint issued
from the ‘
-LSAD’ principle.16 This enabled a more accurate in-
terpretation of computational calculations by minimizing the
Ca
–N bridging,15 1,4-dihydron-
b
b
* Corresponding author. Tel.: þ34 943 015383; fax: þ34 943 015959.
E-mail address: jesusmaria.aizpurua@ehu.es (J.M. Aizpurua).
0040-4020/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.02.085

3188
J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
energy contributions arising from the conformational mobility
around the and dihedral angles. Owing to this particular design,
the transition state TS-I (Fig. 2) could be characterized at a B3LYP/6-
31G computation level as the preferred pathway for the reduction
of methyl benzoylformate to the (S)-methyl mandelate major
product.14 The intrinsic stereodirecting effect of the
-lactam ring
was also studied experimentally from N-[bis(trimethyl-
silyl)methyl]- -lactams 3 and it was shown to be superior to the
bis[trimethylsilyl]methyl substituent of 4a. Compound 4e (R2¼Me)
was chosen as 2-methyl-substituted dihydropyridines are known
to reduce carbonyl compounds in very high enantioselectivi-
ties.6d,10d,17 Among the reaction substrates, we selected the aroyl-
formate esters 5–9 containing differently sized R3 ester groups and
aromatic rings with both electron-donating and electron-attracting
groups.
4
j
*
b
b
The synthesis of NADH models 4a–e involved the preparation of
their nicotinamide precursors 11a–e as disclosed in Scheme 1. Ac-
nonbridged NADH peptides 1. Accordingly, it was found that the
enantiomeric excess of the reduction increased with the bulkiness
cordingly, the N-[bis(trimethylsilyl)methyl]-b-lactams 11a and 11e
of the
a
-substituent attached to the stereogenic center of the
were first obtained from 1018 upon hydrogenolytic N-deprotection
b-lactam ring.
of the diphenyloxazolidinone moiety, followed by acylation of the
intermediate
a-amino-b-lactam with nicotinic acid or 2-methyl-
nicotinic acid in the presence of EDC. Then, 11a was submitted to
partial or total chemoselective desilylation18c with cesium fluoride
or tetra-n-butylammonium fluoride to provide, respectively, the
R
R1
R
N
L
Mg
O
O
O
H
H
mono-silylated derivative 11b and the N-methyl-
fair yields. Conversely, the synthesis of the N-tert-butyl-
b
-lactam 11c in
-lactam
*
L
*
N
H
O
b
N
H
S
N
H
SiMe3
SiMe3
O
11d was accomplished following a procedure developed in our
laboratory,19 which used the (R)-2-benzyl-serine methyl ester 12 as
starting material.20 After treating this compound with nosyl chlo-
ride and KHCO3 in dry acetonitrile, the resulting N-nosylaziridine
intermediate was reacted successively with tert-butylamine and
lithium bis(trimethylsilyl)amide21 to give the desired N-nosyl
R3
O
R2
N
O
R
H
Me
N
Me
Ar
3
TS-I
Figure 2. Lower-energy structure of the NADH/Mg2þ/ArCOCO2R3 ternary transition
state predicted by B3LYP/6-31G computation for the enantioselective reduction of
*
b
-lactam 13 in 65% overall yield. Further deprotection of the nosyl
alkyl aroylformates with the
gray color.
b
-lactam mimetics 3. Peripheral groups are highlighted in
group and coupling of the intermediate -amino- -lactam with
nicotinyl chloride hydrochloride in the presence of DIPEA afforded
a
b
the nicotinamide 11d.
Despite this general stereodirecting trend, our chelating model
should also explain the yet unknown modulation of the enantio-
meric excess by peripheral groups distant from the asymmetric
stereocenter (e.g., R1, R2, R3, and Ar in Fig. 2). Herein we report our
investigation concerning the synthesis of diversely substituted
N
R2
O
O
Bn
O
Bn
Ph
H
N
1) H2, Pd/C, EtOH
2)
N
R2
Ph
N
N
SiMe3
SiMe3
N
R1
enantiopure 1,4-dihydronicotinamide-b-lactam mimetics 4 (see
O
O
,
below) and the experimental and computational exploration of
their reactivity as enantioselective reducing agents of various alkyl
aroylformate esters 5–9. Our new experimental results are reflected
in the computations, confirming the previously proposed chelated
ternary entity TS-I as a plausible model to account for the stereo-
chemical behavior observed in these nonenzymatic NADH
reduction reactions.
CO2H
11
10
EDC,HOBT,Et3N,
CH2Cl2
11a R1= CH(SiMe3)2 R2= H (85%)
11b R1= CH2SiMe3 R2= H (95%)
11c R1= Me R2= H (64%)
CsF,
MeCN
OH
Bn
TBAF,
THF
H2N
11e R1= CH(SiMe3)2 R2 Me (42%)
=
MeO2C
12
2. Results and discussion
1) NsCl, KHCO3, MeCN
2) tBuNH2, MeCN
3) LHMDS, THF
Ref. 19
2.1. Synthesis of NADH peptide b-lactam models
N
O2N
Bn
The 1,4-dihydronicotinamides 4a–e bearing R1 and R2 groups of
different size (Fig. 3) were synthesized to be used as constrained
H
N
Bn
1) PhSH, K2CO3,
O
O
MeCN
S
N
H
N
N
O
2)
, DIPEA,
MeCN
H
NADH peptide models.
b-Lactams 4a–c were readily accessible
tBu
Cl
O
tBu
O
N
through the total or partial desilylation of the large
13 (65%)
Scheme 1. Synthesis of
11d (55%)
COCl
-nicotinylamido-
a
b
-lactams 11. aKey: EDC¼1-[3-(dimethyl-
O
Bn
H
H
O
amino)propyl]-3-ethyl carbodiimide hydrochloride; HOBT¼1-hydroxybenzotriazole;
*
DIPEA¼N,N-diisopropylethylamine;
LHMDS¼lithium hexamethyldisilazide.
TBAF¼tetra-n-butylammonium
fluoride;
OR3
N
H
Ar
N
O
O
N
R2
R1
Me
Next, nicotinamides 11 were quantitatively N-alkylated with
methyl iodide and the resulting pyridinium salts 14a–e were sub-
mitted to regioselective reduction with a mixture of sodium
dithionite and sodium carbonate.22 The desired 1,4-dihydron-
icotinamides 4 were obtained in good overall yields with the ex-
ception of 2-methyl-1,4-dihydronicotinamide 4e, which afforded
complex mixtures under such conditions. Fortunately, a 49% yield
of pure 4e product could be attained when the reduction of 14e was
carried out using sodium bicarbonate as a base (Table 1).23
4a-e
5-9
R3= Me
R3= Et
R3= tBu
R3= Me
R3= Et
a R1= CH(SiMe3)2
R1= CH2SiMe3
c R1= Me
R1= tBu
e R1= CH(SiMe3)2
R2= H
R2= H
R2= H
R2= H
R2= Me
5 Ar= Ph
6 Ar= Ph
7 Ar= Ph
b
8 Ar= 4-MeOC6H4
9 Ar= 4-NO2C6H4
d
Figure 3.
reduced.
b-Lactam 1,4-dihydronicotinamides prepared and aroyl benzoylformates

J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
3189
found between the size of the R2 group and the enantioselectivity
of the reaction.
The stereoreductive ability of NADH model 4a was further tested
for different a-keto esters (5–9). Ethyl and tert-butyl benzoylfor-
Table 1
Preparation of 1,4-dihydronicotinamide-
b
-lactams 4 from nicotinamides 11
O
O
Bn
Bn
Na2S2O4,
Na2CO3
MeI
N
N
11
H
H
N
mates 6 and 7 were reduced in slightly higher enantiomeric excess
(82 and 92%, respectively) than methyl benzoylformate 5 (78%),
suggesting that an increase of the R3 ester group size might also
favor higher enantioselectivities. Finally (entries 8 and 9), the
stereomodulating effect of electron-donating (OMe, 8) and
electron-withdrawing (NO2, 9) groups placed at the aromatic ring
of the aroylformate esters was also tested. As expected from their
different chelating ability with Mg2þ cation, electron-rich sub-
strates favored a slight increase in enantiomeric excess values
(compare entries 1 and 8), whereas electron-poor substrates
resulted in lower enantiomeric excess (compare entries 6 and 9).
N
MeOH,
CH2Cl2
MeCN
R1
R1
R2
O
O
N
R2
N
I
Me
Me
4
14
Entry
R1
R2
Product
Yield (%)
Product
Yield (%)
1
2
3
4
5
CH(SiMe3)2
CH2SiMe3
Me
H
H
H
H
14a
14b
14c
14d
14e
100
96
98
87
72
4a
4b
4c
64
84
82
74
47
tBu
4d
CH(SiMe3)2
Me
4ea
a
NaHCO3 was used as a base instead of Na2CO3.
2.2. Reduction of alkyl aroylformates with
models 4
b-lactam NADH
2.3. Computational calculations
Reduction of alkyl aroylformates 5–9 with NADH models 4a–e
was studied at 300 K in the presence of an equivalent of magnesium
perchlorate. The reactions were conducted in deuterated acetoni-
trile and followed by monitoring the NMR spectra of the mixtures.
Results summarized in Table 2 showed that a uniform stereo-
inducting bias for (S) enantiomers prevailed in all instances, irre-
spective of the nature of the peripheral R1, R2 and R3 groups.
However, the enantiomeric excess of the reactions was sensitive to
the size of such groups.
To investigate the correlation of the stereochemical results
collected in Table 2 with our biomimetic reaction model based on
chelated ternary complexes NADH/Mg2þ/ArCOCO2R and transition
states TS-I (see Fig. 2), a computational study was carried out using
the functional B3LYP24 and the 6-31G
Gaussian 03.25
*
basis sets as implemented in
Reaction free energy differences (DDGz) expected for enantio-
meric excess values below 90% are in the range of 2 kcal/mol.
Therefore, it was essential to limit the conformational energy of the
calculated reaction intermediates and transition states in order to
make a reliable interpretation of the computational data. Accord-
We first checked the stereomodulating effect of the R1 sub-
stituent attached to the b-lactam nitrogen of NADH models 4a–
d on the reduction of methyl benzoylformate 5. Accordingly,
replacement of the bulky bis(trimethylsilyl) methyl group in
model 4a by the less hindered trimethylsilylmethyl and methyl
groups (models 4b and 4c, respectively) resulted in an enantio-
meric excess decay from 78% to 60% (entries 1–3). Changing to the
bulkier model 4d with a tert-butyl group, restored the enantio-
meric excess to 78% (entry 4). These data strongly suggested the
existence of a direct relationship between the size of R1 and the
enantioselectivity of the reactions. On the other hand, the re-
duction of methyl benzoylformate 5 with the 2-methyl-NADH
ingly, we designed the simplified b-lactam NADH structures 20–22
(Table 3) with the partially flexible benzyl substituent of com-
pounds 4a–e (Fig. 3) replaced by the rigid methyl group at the
a
position of the b-lactam ring. To gauge the steric effect of
the peripheral R1, R2, and R3 groups on the enantioselectivity of the
alkyl aroylformate reduction reaction, we designed the N-methyl-
and N-tert-butyl-substituted models 20 and 21 to represent, re-
spectively, small and large R1 groups. Similarly, the model 22 in-
corporating a 2-methyl-dihydropyridine moiety was selected to put
into evidence the stereomodulating effect of the R2 group. Finally,
the
b
-lactam NADH models 20–22 and the substrates 5 (R3¼Me
b
-lactam model 4e (entry 5) proceeded with poor enantiomeric
small group) and 7 (R3¼tert-Bu large group) were combined in four
computational reactions (A–D, see Table 3) and the corresponding
DDGz values were calculated.
excess (30%), in contrast to previous reports anticipating very high
enantioselectivities for other 2-methyl-substituted NADH
models.17 Therefore, an (unexpected) inverse relationship was
Table 2
Reduction of alkyl aroylformates 5–9 with b-lactam NADH models 4a–e
O
OH
4a-e
CO2R3
Ar
CO2R3
Ar
*
15-19
Mg(ClO4)2, CD3CN, r.t.
5-9
Entry
NADH model
R1
R2
Substrate
Ar
R3
Product
Reaction time
(h)a
Yield (%)b
ee (%)c
Config.
1
2
3
4
5
6
7
8
9
4a
4b
4c
4d
4e
4a
4a
4a
4a
CH(SiMe3)2
CH2SiMe3
Me
H
H
H
H
Me
H
H
H
H
5
5
5
5
5
6
7
8
9
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Me
Me
Et
15
15
15
15
15
16
17
18
19
14
72
24
16
18
18
22
40
22
66
48
60
70
30
70
43
35
63
78
65
60
78
30
92
88
90
82
S
S
S
S
S
S
S
tBu
CH(SiMe3)2
CH(SiMe3)2
CH(SiMe3)2
CH(SiMe3)2
CH(SiMe3)2
tBu
Me
Et
4-MeOC6H4
4-NO2C6H4
S
S d
Reactions were quenched when complete consumption of NADH model was observed by 1H NMR.
Calculated from 1H NMR spectra on the basis of NADþ formation.
Determined by HPLC.
a
b
c
d
Configuration assigned on the basis of the relative chiral HPLC retention time.

3190
J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
Table 3
Relative energies (in kcal/molꢁ1) and main torsion angles of the [
b
-lactam NADH (20–22)/Mg2þ/PhCOCO2R3 (5, 7)] ternary complexes 23–26 and transition structures thereof
computed at B3LYP/6-31G
*
level for the reactions (A)–(D)
O
R1
Me
O
2+
Me
O
N
Me
φ
HN
N
H
Mg2+
MeCN
Ph
C
Ph
OR3
+
*
OR3
N
N
R1
O
O
N
R2
[23] (R1= Me; R2= H; R3= Me)
[24] (R1= tBu; R2= H; R3= Me)
O
(A)
(B)
OH
R2
N
O
C
Mg
O
α
Me
OR3
20 (R1= Me; R2= H)
21 (R1= tBu; R2= H)
22 (R1= R2= Me)
15 (R3= Me)
17 (R3= tBu)
5 (R3= Me)
7 (R3= tBu)
N
[25] (R1= Me; R2= Me; R3= Me) (C)
O
Me
[26] (R1= Me; R2= H; R3= tBu)
(D)
Ph
Me
b
e
Entry
Reaction
Ternary complexa
D
Gꢀ
TSa,c
23S-syn
23S-anti
D
Gzd
DDGz
af
23.1
ꢁ87.3
ꢁ20.9
132.0
4f
62.0
60.2
ꢁ10.7
4.3
rel
(R–S)
1
2
3
4
(A)
(A)
(A)
(A)
23S
23S
23R
23R
0.0
*
22.4
25.9
25.5
26.6
0.0
d
0.0
0.6
0.6
23R-syn
*
3.1 (3.7)
d
23R-anti
5
6
7
8
(B)
(B)
(B)
(B)
24S
24S
24R
24R
1.3
1.3
0.0
0.0
24S-syn
*
22.7
22.7
27.4
27.4
0
d
22
-83
130.5
-18
63
60
7.5
ꢁ5.6
24S-anti
24R-anti
d
24R-syn
*
4.7 (4.8)
9
(C)
(C)
(C)
(C)
(C)
(C)
25S-1
25S-2
25S-2
25R-1
25R-1
25R-2
0.0
2.2
2.2
4.0
4.0
1.7
25S-syn1
25S-syn2
25S-anti2
21.3
24.9
20.1
22.0
25.9
25.8
d
29.1
ꢁ13.8
96.3
121.9
28.0
61.4
67.7
60.6
10
11
12
13
14
d
*
0.0
1.9 (2.4)
d
25R-syn1
*
11.1
25R-anti1
25S-syn2
ꢁ6.8
d
ꢁ28.0
ꢁ12.8
15
16
17
18
(D)
(D)
(D)
(D)
26S
26S
26R
26R
0.8
0.8
0.0
0.0
26S-syn
*
25.8
29.8
29.3
29.8
0.0
d
22.6
ꢁ83.5
131.9
ꢁ21.2
60.2
60.2
4.4
ꢁ10.6
26S-anti
26R-anti
26R-syn
*
3.5 (4.6)
d
a
Descriptors R and S denote the configuration of the final mandelate product.; ‘1’ and ‘2’ descriptors denote, respectively, positive and negative
a dihedral angles (see text).
b
c
Relative
Transition states with asterisk identify the lowest energy pathways to the R and S enantiomer products.
D
Gꢀ formation energies calculated from the most stable ternary complexes 23–36 in each reaction (A)–(D).
d
e
Relative
Values in parentheses correspond to activation energy differences after single-point refinement at B3LYP/6-311þþG** level including the IEF-PCM solvation model (MeCN,
D
Gꢀ activation energies calculated from the most stable ternary structures 23–36 in each reaction (A)–(D).
3
¼36.64).
f
Dihedral angles of the transition states as defined in Figure 1.
As outlined in Scheme 2, the reaction mechanism proposed14
the b-lactam NADH models 20–22 and alkyl benzoylformates 5 and
involved the initial formation of the ternary complexes NADH/
Mgþ2/PhCOCO2R3 23–26, which were formally built up from
[Mg(NCMe)6]2þ by displacement of four acetonitrile ligands with
7. In all instances, the bonding patterns were created assuming an
octahedral coordination geometry for the Mg2þ cation.26 The ter-
nary complexes 23S-26S and 23R-26R, leading, respectively to the
Me
Me
Me
R1
Me
N
N
R1
O
S
H
HR
N
L
Mg
O
N
HN
L
Mg
O
HS
O
R2
N
Ph
R2
H
Ph
O
R2
N
O
α
L
L
R O
OR3
O
H
OR3
N
H
H
N
H
O
N
S
R
O
H
O
Me
O
Me
H
O
Me
O
L
O
L
R2
Mg
L
R2
Mg
L
O
OR3
OR3
O
HS
HR
N
N
R1
N
N
R1
Ph
TS-S-anti
Me
Ph
Me
L = MeCN
TS-R-anti
TS-R-syn
TS-S-syn
TS-R
Δ
ΔG
TS-S
(RŠS)
Me
Me
Ph
R1
R1
O
L
N
N
O
O
O
L
OR3
L
H
Mg
O
HN
N
Mg
L
15, 17
R enant.
O
O
O
R2
N
R2
N
15, 17
S
S
Me
OR3
H
H
Me
S enant.
R
R
Ph
H
H
23R
20/Mg2+/5
21/Mg2+/5
22/Mg2+/5
20/Mg2+/7
23S
24S
25S
26S
24R
25R
26R
Scheme 2. Structures of productive ternary complexes (NADH/Mgþ2/PhCO2R3) 23–26 and transition states leading to R and S mandelate enantiomers. Rotation of the dihy-
dronicotinamide moiety around the
depicted linearly.
a
torsion angle in each complex prompts the transfer of HS or HR hydride atoms through syn- or anti-transition states. Energy levels are not

J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
3191
major S and minor R enantiomeric products, were also assumed to
participate in a ligand exchange equilibrium. After B3LYP/6-31G
transition states TS-23–26 (see Table 3) and confirmed every pe-
*
ripheral stereomodulating effect calculated for R1, R2, and R3 groups
calculation (Table 3, column 4), only small stability-differences
at B3LYP/6-31G
*
level.
were recorded between the conformers of the ternary complexes
ꢀ
23, 24, and 26 (0.6–1.3 kcal/mol), whereas the
D
Grel formation
3. Conclusions
energies of conformers 25 differed from each other in up to
4.0 kcal/mol. In addition, the number of computationally charac-
terizable complexes 25 for reaction C increased from two to four, as
A
mechanistic model accounting for the stereoinducing
behavior observed during the Mg2þ-promoted nonenzymatic
NADH reduction of alkyl aroylformates has been proposed on the
basis of NADH/Mg2þ/ArCOCO2R3 ternary complexes chelated
a result of the hindered rotation around the
a torsion angle caused
by the methyl group (R2) attached to the dihydronicotinamide ring.
Next, we studied the reaction step determining the rate and
enantioselectivity of the reactions A–D, which involved the con-
certed intramolecular hydrogen atom transfer from the dihy-
dronicotinamide C4 position to the aroylformate ketone. Depending
on the diastereotopic hydrogen atom (HR or HS) transferred to the
carbonyl carbon, the transition states arising from each ternary
complex could adopt two topologies (see Scheme 2) to form the
mandelate products. These conformations were defined as syn/anti
around an octahedral magnesium cation. Using rigidified b-lactam
dihydronicotinamides to minimize conformational energy, we
have predicted not only the stereoinducing sense of the reaction,
but also the stereomodulating trend exerted by remote groups
placed in the peripheral positions of the complex. Despite its
general bias to overweight the activation energies under B3LYP/6-
31G computation level, our model represents a valuable con-
*
ceptual tool to understand important mechanistic aspects of the
stereoselective reduction of alkyl aroylformates with non-
enzymatic NADH models.
on the basis of the
a torsion angle describing the relative disposi-
tion of the dihydronicotinamide ring with respect to the carbonyl
group (see also Fig. 1).
An inspection of the relative free energies and the key dihedral
4. Experimental section
4.1. General experimental
angles (
clearly suggested the existence of a relationship between the steric
size of each peripheral R1, R2, R3 groups, and the DDGz
gov-
a,4) of the transition states TS-23-26 summarized in Table 3
(R–S)
erning the enantiomeric excesses of reactions A–D. For instance,
All reactions were carried out under nitrogen or argon at-
mosphere in oven or flame-dried glassware with magnetic stir-
ring. Solvents were distilled prior to use. Tetrahydrofuran (THF)
was distilled from sodium metal/benzophenone ketyl. Dichloro-
methane (CH2Cl2) was distilled from calcium hydride. Dichloro-
methane, methanol, water, and deuterated acetonitrile (CD3CN)
used for preparation of NADH models and for biomimetic re-
ductions were deoxygenated by nitrogen bubbling for 2–3 h or
by freeze–pump–thaw techniques. Magnesium perchlorate was
dried by heating under vacuum for 24 h, and then deoxygenated
using standard Schlenk techniques. Purification of reaction
products was carried out by flash chromatography using silica gel
60 (230–400 mesh, from Merck 60F PF254). Analytical thin layer
chromatography was performed on 0.25 mm silica gel 60-F
plates. Visualization was accomplished with UV light and phos-
phomolybdic acid/ammonium cerium (IV) nitratesulfuric acid/
water reagent, followed by heating. Melting points were mea-
sured with a Bu¨chi SMP-20 melting point apparatus and are
uncorrected. Infrared spectra were recorded on a Shimadzu IR-
435 spectrophotometer. NMR spectra were recorded on Bruker
Avance500 spectrometers at frequencies of 500 MHz for 1H NMR
and 125 MHz for 13C NMR, at room temperature unless otherwise
stated. Chemical shifts are reported in parts per million (ppm)
relative to the central line of CD2Cl2 (5.33 ppm), CDCl3
(7.28 ppm), and CD3CN (1.94 ppm) for 1H NMR, and to the central
line of CD2Cl2 (54.0 ppm), CDCl3 (77.0 ppm) and CD3CN
(1.39 ppm) for 13C NMR. Combustion analyses were performed
on a Leco CHNS-932 elemental analyzer. Enantiomeric excesses
were determined by analytical high performance liquid chro-
matography (HPLC) on a Hewlett Packard 1050 chromatograph
equipped with a diode array UV detector and on a Waters 600
the activation energy differences (DDGz) computed for the re-
duction of methyl
a
-keto ester 5 with the N-methyl-
b-lactam 20
(reaction A) and the N-tert-butyl-
b-lactam 21 (reaction B) were 3.1
and 4.7 kcal/mol, respectively. Although these absolute values and
their relative increment were clearly overestimated with respect to
the expected enantiomeric excesses, they reflected the enantio-
meric excess increase bias observed experimentally for the re-
duction of methyl benzoylformate 5 with 4c and 4d (entries 3 and 4
in Table 2). Hence, the enantiomeric excess increase recorded ex-
perimentally by increasing the size of the R1 group was fully con-
sistent with our model.
The significant enantiomeric excess drop of 48% observed for
the reduction of methyl benzoylformate with compound 4a and
the 2-methy-dihydronicotinamide model 4e (compare entries 1
and 5 in Table 2) could also be conveniently interpreted on the
basis of our mechanistic model. Indeed, DDGz
values de-
(R–S)
creased from 3.1 for reaction A to 1.9 kcal/mol for reaction C. A
careful inspection of the geometry of structures TS-25S-anti2 and
TS-25R-syn1 revealed that the methyl group in the NADH ring
caused some steric distorsion around the
a dihedral angle, diffi-
culting a proper O]C–C]C conjugation. This effect could be
p–p
at the origin of the detrimental effect on the enantiomeric excess
caused by an increase of the size of the R2 group (see
Supplementary data for details).
The computational reduction of tert-butyl benzoylformate 7
with the b
-lactam model 20 yielded a DDGz(R–S) value of 3.5 kcal/mol
for reaction D (Table 3), which was slightly higher than the 3.1 kcal/
mol activation energy difference for the reduction of methyl ben-
zoylformate 5 with the same NADH model (reaction A). Again, the
calculated data reflected the experimental enantiomeric excess
increase of 10% observed by comparing the examples of entries 1
and 7 in Table 2 and explained satisfactorily the beneficial effect of
a larger benzoylformate R3 ester group on the reaction enantio-
meric excess.
chromatograph equipped with a Waters 2487 Dual
l absorbance
detector using the chiral columns Chiralcel OD, Chiralcel OB-H,
Chiralcel OJ, and Chiralpak AD with flow rates of 0.7 mL/min
and 0.5 mL/min (mobile phase hexane/isopropanol). Optical ro-
tations were measured at 25ꢂ0.2 ꢀC in methylene chloride unless
otherwise stated. Mass spectra were obtained on a Finnigan GCQ
mass spectrometer (70 eV) using GC–MS coupling (column: fused
silica gel, 15 m, 0.25 mm, 0.25 nm phase SPB-5). Preparation and
physical data of compounds 4a, c, 10, 11a, 11, 12, 14a, and c have
been reported previously.14,19
Finally, to estimate the solvent effect of acetonitrile on the re-
actions studied, single-point calculations with the self-consistent
reaction field (SCRF) based on the IEF-PCM27 solvation model
(MeCN,
3
¼36.64) were performed at B3LYP/6-311þþG** level. This
analysis resulted in an homogeneously biased increase of DDGz
values in 0.1–1.1 kcal/mol for the previously optimized productive

3192
J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
4.2. (3R)-3-Benzyl-1-(1-tert-butyl)-3-
(2-nitrobenzenesulfonylamino)-azetidin-2-one (13)
solution was washed with NaOH 0.1 M (20 mL), NH4Cl (20 mL), and
H2O (20 mL). The organic phase was decanted and dried over
MgSO4. The solvents were evaporated under reduced pressure and
To a stirred solution of -benzyl-serine methyl ester (1 mmol,
a
the crude was purified by flash chromatography, eluent: EtOAc.
25
0.209 g) in dry acetonitrile (20 mL) o-nosyl chloride (2.2 mmol,
0.45 g), and KHCO3 (5 mmol) were added. The mixture was stirred
at reflux for 8–24 h. After checking an aliquot by 1H NMR, freshly
distilled tert-butylamine (0.073 g) dissolved in a minimum volume
of dry CH2Cl2 was added at rt. The mixture was stirred at rt over-
night (followed by 1H NMR), and then was filtered over a Celite pad
Yield: 0.149 g (55%); oil; [
a
]
D
ꢁ11 (c¼1.0, CH2Cl2); IR (cmꢁ1, KBr):
1743.4 (C]O); 1661.2 (C]O); 1248.0 (tert-Bu). HPLC-MS, MeCN/
HCOOH m/z (ion source type: ESI, positive polarity): MSþ1: 338.2,
MS2 (338.2): 253.0, MS3 (188.0): 174.9, 134.0, 106.1. 1H NMR (
d,
ppm, CDCl3): 9.00 (d, 1H, J¼1.7 Hz), 8.75 (dd, 1H, J1¼1.4 Hz,
J2¼4.7 Hz), 8.13–8.10 (m, 1H), 8.38 (br s, 1H), 7.40–7.32 (m, 6H), 7.02
(br s, 1H), 3.60 (d, 1H, J¼5.9 Hz), 3.43 (d, 1H, J¼5.9 Hz), 3.37 (d, 1H,
and the intermediate b-amino ester was purified by flash-column
chromatography (silica gel; eluent: hexane/EtOAc). This product
was dissolved in dry THF (20 mL) and the temperature of the
mixture was cooled down to 0–5 ꢀC. Then, a solution of LiHMDS
(1 M) in THF (2.5 mmol) was slowly added while a deep purple
color developed. The mixture was stirred for 1 h at this temperature
and then it was quenched with NaHCO3 (20 mL). The aqueous
phase was extracted with CH2Cl2 (3ꢃ20 mL), the collected organic
fractions were dried (MgSO4), filtered, and the solvent was re-
moved in vacuo. The product was purified by column chromatog-
J¼13.5 Hz), 3.24 (d, 1H, J¼13.5 Hz), 1.12(s, 9H). 13C NMR (
d, ppm,
CDCl3): 165.9, 165.7, 152.6, 148.8, 135.4, 135.0, 130.6, 127.5, 129.6,
128.7, 127.45, 123.52, 66.03, 53.16, 47.74, 39.00, 27.44. Anal. Calcd
for C20H23N3O2 (337.42): C, 71.19; H, 6.87; N, 12.45. Found: C, 71.16;
H, 6.91; N, 12.85.
4.4. (3R)-3-Benzyl-1-[bis(trimethylsilyl)methyl]-3-(2-methyl-
nicotinamido)-azetidin-2-one (11e)
raphy (silica gel; eluent: hexanes/EtOAc). Yield: 65%, 0.271 g; oil;
A solution of the corresponding amine (1.62 mmol, 0.543 g) in
dry CH2Cl2 (20 mL) under nitrogen atmosphere was cooled to 0 ꢀC.
Then, 2-methylnicotinic acid (1.5 equiv, 2.43 mmol, 0.335 g), NEt3
(1.5 equiv, 2.43 mmol, 0.54 mL), HOBT (1.5 equiv, 2.43 mmol,
1.20 g), and EDC$HCl (1.5 equiv, 2.43 mmol, 0.47 g) were added. The
stirring was continued for an hour at 0 ꢀC, and for 24 h at rt. After
that time, CH2Cl2 (10 mL) was added and the solution was washed
with 1 M aqueous KHSO4 (20 mL) and H2O (20 mL). The organic
phase was decanted and dried over MgSO4. The solvents were
evaporated under reduced pressure and the crude was purified by
25
[
a]
ꢁ47.5 (c¼1.0, Cl2CH2); IR (cmꢁ1, KBr): 1744; 1646 (C]O).
D
HPLC–MS, MeOH/HCOOH m/z (ion source type: ESI, positive po-
larity): MSþ1: 418.1, MS2 (418.1): 390.1, 334.0, 147.0, MS3 (390.0):
334.0, 317.0, 147.0, 130.0. 1H NMR (
d, ppm, CDCl3): 8.20 (dd, 1H,
J1¼1.4 Hz, J2¼7.8 Hz), 7.89 (dd, 1H, J1¼1.4 Hz, J2¼7.8 Hz), 7.74 (m,
2H), 7.29–7.26 (m, 5H), 5.91 (br s,1H), 3.65 (d,1H, J¼5.5 Hz), 3.32 (d,
1H, J¼5.5 Hz), 3.09 (br s, 2H), 1.10 (s, 3H). 13C NMR (
d, ppm, CDCl3):
164.4, 147.7, 135.8, 133.7, 133.1, 131.3, 130.3, 128.8, 127.8, 125.6, 68.0,
53.2, 49.2, 47.6, 40.5, 29.9, 28.8, 27.6. Anal. Calcd for C20H23N3O5S2
(417.48): C, 57.54; H, 5.55; N, 10.07. Found: C, 57.25; H, 5.73; N, 9.70.
flash-column chromatography (eluent EtOAc). Yield: 0.53 g (73%);
25
oil; [a
]
þ4.8 (c¼1.0, CH2Cl2); IR (cmꢁ1, KBr): 1733.5 (C]O); 1664.6
D
4.3. Preparation of
a
-(2-nicotinamido)-
b-lactams 11
(C]O); 848.0 (C–Si). HPLC-MS, MeOH/HCOOH m/z (ion source
type: ESI, positive polarity): MSþ1: 454.2, MS2 (454.2):267.0, 188.0,
4.3.1. (3R)-3-Benzyl-1-[(trimethylsilyl)methyl]-3-nicotinamido-aze-
tidin-2-one (11b). A solution of the -lactam-nicotinamide 11a
MS3 (267.0): 238.9, 221.0, 205.9, 188.9, 147.9, 120.0. 1H NMR (
d,
b
ppm, CDCl3): 8.56 (d, 1H, J¼4.2 Hz), 7.54 (d, 1H, J¼7.4 Hz), 7.32–7.25
(0.25 mmol, 0.110 g) in CH3CN (5 mL) under nitrogen atmosphere
was heated at 70 ꢀC, and then cesium fluoride (3 equiv, 0.75 mmol,
0.114 g) was added. The reaction mixture was heated at 70 ꢀC for
1 h (consumption of starting material was checked by TLC, eluent
EtOAc). The solvent was evaporated under reduced pressure and
the crude reaction mixture was dissolved in dichloromethane, and
washed with water (2ꢃ5 mL). The organic phase was dried over
MgSO4 and the solvents were evaporated to afford the desired
(m, 5H), 7.16 (m, 1H), 7.06 (br s, 1H), 3.76 (d, 1H, J¼6.3 Hz), 3.63 (d,
1H, J¼6.3 Hz), 3.37 (s, 2H), 2.65 (s, 3H), 2.61 (s, 1H). 13C NMR (
d,
ppm, CDCl3): 170.5, 168.4, 167.0, 156.6, 150.2, 135.5, 131.3, 130.7,
129.0, 127.6, 127.0, 125.0, 121.2, 119.2, 110.2, 68.7, 54.8, 39.0, 38.2,
22.8, ꢁ0.05, ꢁ0.14. Anal. Calcd for C24H35N3O2Si2 (453.72): C, 63.53;
H, 7.78; N, 9.26. Found: C, 63.56; H, 7.18; N, 9.10.
4.5. General procedure for the preparation of pyridinium
salts 14
product, which was used without further purification. Yield:
25
0.088 g (95%); oil; [
a
]
ꢁ11.0 (c¼1.0, CH2Cl2); IR (cmꢁ1, KBr): 3448;
D
2934; 1732 (C]O); 1662 (C]O); 854 (C–Si). HPLC-MS, MeOH/
HCOOH m/z (ion source type: ESI, positive polarity): MSþ1: 368.1,
MS2 (468.2): 253.0 MS3 (253.0): 225.0, 191.9, 174.9, 134.0, 106.1. 1H
To a solution of the corresponding nicotinamide (1.00 mmol) in
CH3CN (10 mL) under N2 atmosphere, methyl iodide (6 mL) was
added. The reaction mixture was stirred at 50 ꢀC for 2–24 h. After
that time, the solvents were evaporated and the residue was dried
under vacuum to afford the pyridinium iodide as a hydroscopic
solid, which was used without further purification.
NMR (d, ppm, CDCl3): 8.97 (br s, 1H), 8.71 (br s, 1H), 8.08 (d, 1H,
J¼8.0 Hz), 7.33–7.25 (m, 5H), 3.63 (d, 1H, J¼6.0 Hz), 3.46 (d, 1H,
J¼6.0 Hz), 3.32 (d, 1H, J¼13.7 Hz), 3.27(d, 1H, J¼13.7 Hz), 2.69 (d,
1H, J¼15.4 Hz), 2.50 (d, 1H, J¼15.4 Hz), ꢁ0.04 (s, 9H). 13C NMR (
d,
ppm, CDCl3): 167.3, 165.7, 152.7, 148.76, 135.2, 130.5, 128.9, 127.5,
123.6, 68.8, 53.7, 39.0, 34.0, 29.9, ꢁ1.8. Anal. Calcd for
C203H25N3O2Si (367.52): C, 65.36; H, 6.86; N, 11.43. Found: C, 65.78;
H, 6.68; N, 11.86.
4.5.1. N-[(3R)-3-Benzyl-1-[(trimethylsilyl)methyl]-2-oxoazetidin-3-
yl-1-methylpyridinium]-3-carboxamide iodide (14b). The general
procedure was followed (reaction time: 4 h) from 11b (0.234 mmol,
25
0.086 g). Yield: 0.111 g (96%); yellow solid; mp: 159–160 ꢀC; [
a]
D
þ3.75 (c¼1.0, CH2Cl2); IR (cmꢁ1, KBr): 3472; 2958; 1746 (C]O);
1681 (C]O); 854 (C–Si). HPLC-MS, MeOH/HCOOH m/z (ion source
type: ESI, negative polarity): MS-1: 508.2, MS2 (508.2): 426.6,
4.3.2. (3R)-3-Benzyl-1-(tert-butyl)-3-nicotinamido-azetidin-2-one
(11d). To a solution of the nicotinoyl chloride hydrochloride 27
(2 mmol, 356 g) in 25 mL of dry CH3CN cooled to ꢁ5 ꢀC under ni-
trogen atmosphere, N,N-diisopropylethylamine (DIPEA) (6.4 mmol,
366.1, 277.0, 126.9. 1H NMR (
d, ppm, CDCl3): 9.91 (s, 1H); 9.00 (d, 1H,
J¼6.1 Hz); 8.92 (d, 1H, J¼8.1 Hz); 8.77 (s, 1H); 8.75 (br s, 1H); 8.08
(m, 1H); 7.35–7.30 (m, 5H); 4.59 (s, 3H); 3.68 (d, 1H, J¼5.4 Hz); 3.50
(d, 1H, J¼13.8 Hz); 3.46 (d, 1H, J¼13.8 Hz); 3.26 (d, 1H, J¼5.4 Hz);
2.54 (d, 1H, J¼15.4 Hz); 2.46 (d, 1H, J¼15.4 Hz); ꢁ0.08 (s, 9H). 13C
1.1 mL), and a solution of the corresponding
a-amino-b-lactam
(0.8 mmol, 0.186 g) in 10 mL CH3CN were added. The reaction
mixture was stirred for 1 h at 0 ꢀC, then at rt (5–18 h, monitored by
1H NMR). After that time, CH2Cl2 (20 mL) was added and the
NMR (d, ppm, CDCl3): 167.6, 161.5, 147.2, 145.3, 145.0, 134.4, 133.4,

J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
3193
130.8, 128.8, 127.5, 69.8, 51.5, 49.9, 38.8, 33.7, ꢁ1.9. Anal. Calcd for
C21H28IN3O2Si (509.46): C, 49.51; H, 5.54; N, 8.25. Found: C, 49.91;
H, 5.50; N, 8.47.
J¼1.3 Hz); 5.69 (dd, 1H, J1¼8.0 Hz, J2¼1.6 Hz); 5.47 (br s, 1H); 4.69
(m, 1H); 3.50 (d, 1H, J¼5.4 Hz); 3.25 (d, 1H, J¼5.4 Hz); 3.16 (d, 1H,
J¼13.5 Hz); 3.10 (d, 1H, J¼13.5 Hz); 3.06–2.95 (m, 2H); 2.91 (s, 3H);
1.09 (s, 9H). 13C NMR (
d, ppm, CD2Cl2): 168.2, 167.0, 140.3, 135.8,
4.5.2. N-[(3R)-3-Benzyl-1-(tert-butyl)-2-oxoazetidin-3-yl-1-methyl-
pyridinium]-3-carboxamide iodide (14d). The general procedure
130.8, 130.1, 128.7, 127.4, 102.9, 98.6, 65.7, 53.1, 47.9, 41.1, 39.3,
27.6, 22.5.
was followed (reaction time: 2 h) from 11d (0.44 mmol, 0.149 g).
25
Yield: 0.184 g (87%); oil; [
a]
þ8.4 (c¼1.0, CH2Cl2); IR (cmꢁ1, KBr):
4.6.3. (3R)-3-Benzyl-1-[bis(trimethylsilyl)methyl]-3-[N,2-dimethyl-
dihydronicotinamido]-azetidin-2-one (4e). The general procedure
was followed from 14e (using NaHCO3 instead of Na2CO3)
D
1726.0 (C]O); 1674.2 (C]O). HPLC-MS, MeOH/HCOOH m/z (ion
source type: ESI, negative polarity): MS-1: 478.1, MS2 (478.1):
380.7, 352.3, 316.8, 276.8, 232.1,126.8. 1H NMR (
d, ppm, CDCl3): 9.92
(0.14 mmol, 0.083 g). Yield: 0.027 g, (49%); yellow oil; 1H NMR (
d,
(s, 1H); 9.05 (d, 1H, J¼6.1 Hz); 8.964 (d, 1H, J¼8.1 Hz); 8.70 (s, 1H);
8.13 (dd, 1H, J1¼6.2 Hz, J2¼8.0 Hz); 7.35–7.30 (m, 5H); 4.61 (s, 3H);
3.60 (d, 1H, J¼5.2 Hz); 3.47 (d, 1H, J¼13.5 Hz); 3.44 (d, 1H,
ppm, CD2Cl2): 7.27 (m, 5H); 5.80 (br s, 1H); 5.70 (d, 1H, J¼7.9 Hz);
4.56 (m, 1H); 3.58 (d, 1H, J¼5.9 Hz); 3.47 (d, 1H, J¼5.9 Hz); 3.28 (d,
1H, J¼13.7 Hz); 3.19 (d, 1H, J¼13.7 Hz); 2.96–2.92 (m, 5H); 2.57 (s,
J¼13.5 Hz); 3.19 (d, 1H, J¼5.3 Hz); 0.99 (s, 9H). 13C NMR (
d, ppm,
1H); 0.05 (s, 9H); 0.02 (s, 9H). 13C NMR (
d, ppm, CD2Cl2): 169.4,
CDCl3): 165.7, 161.4, 147.3, 145.3, 145.0, 134.1, 133.2, 130.9, 128.4,
128.2, 127.4, 66.9, 52.8, 49.7, 45.7, 38.9, 27.2. Anal. Calcd for
C21H26IN3O2 (479.35): C, 52.62; H, 5.47; N, 8.77. Found: C, 52.82; H,
5.50; N, 8.21.
167.6, 140.1, 136.3, 130.3, 129.06, 127.86, 102.7, 98.8, 68.5, 41.1, 40.0,
28.5, 22.5.
4.7. General procedure for the biomimetic reduction of alkyl
aroylformates with NADH peptidomimetics 4
4.5.3. N-[(3R)-3-Benzyl-1-(bis(trimethylsilyl)methyl)-2-oxoazetidin-
3-yl-1,2-dimethylpyridinium]-3-carboxamide iodide (14e). The gen-
eral procedure was followed (reaction time: 24 h) from 11e
(0.5 mmol, 0.2272 g). Yield: 0.282 g (95%); oil; IR (cmꢁ1, KBr):
1737.0 (C]O); 1674.4 (C]O); 841.7 (C–Si). HPLC-MS, MeOH/
In a typical run, a solution of the NADH model 4 (0.103 mmol,
1.1 equiv), Mg(ClO4)2 (0.103 mmol, 1.1 equiv, 23.0 mg), and carbonyl
compound (0.093 mmol) in 0.7 mL of deoxygenated CD3CN was
prepared in a dried NMR tube in the dark and the reaction was
followed by 1H NMR. After complete consumption of the NADH
model, H2O was added and the aqueous layer was extracted with
CH2Cl2 (4ꢃ2 mL). The combined organic layers were dried over
MgSO4 and the solvent was evaporated under reduced pressure.
The crude reaction mixture was submitted to the HPLC analysis
although in some cases the resulting mandelate was isolated by
preparative thin layer chromatography (eluent: hexanes/EtOAc).
Enantiomeric excesses were determined by HPLC using chiral sta-
tionary phases and hexanes/isopropanol as mobile phase. Reaction
conditions, yields, and HPLC conditions are given in Table 2.
HCOOH m/z (ion source type: ESI, positive polarity): M
loss): 468.2, MS2 (468.2): 281.0, 263.2, 237.0, MS3 (237.0): 220.9,
221.0, 205.9,188.9, 147.9,120.0. 1H NMR (
, ppm, CDCl3): 8.63 (d,1H,
*
ꢁI (iodine
d
J¼6.1 Hz); 8.38 (br s, 1H); 8.14 (d, 1H, J¼8.5 Hz); 7.80 (m, 1H); 7.48
(d, 1H, J¼8.2 Hz); 7.35–7.28 (m, 3H); 4.26 (s, 3H); 3.92 (d, 1H,
J¼5.7 Hz); 3.53 (d, 1H, J¼5.7 Hz); 3.50 (d, 1H, J¼14.4 Hz); 3.42 (d,
1H, J¼14.4 Hz); 2.76 (s, 3H); 2.71 (s,1H); 0.17 (s, 9H); 0.11 (s, 9H). 13C
NMR (d, ppm, CDCl3): 166.5, 164.5, 154.7, 147.0, 143.7, 138.2, 135.3,
130.8, 128.9, 127.5, 125.9, 69.4, 54.0, 47.8, 39.2, 37.9, 19.7, 0.2, 0.1.
4.6. General procedure for the preparation of
1,4-dihydropyridines 4
4.8. Computational methods
To a solution of the corresponding pyridinium salt (0.5 mmol) in
12 mL of deoxygenated CH2Cl2/MeOH (3/1) a deoxygenated solu-
tion of Na2S2O4 (10.0 mmol) in 0.5 M aqueous Na2CO3 (12.5 mL)
was added and the mixture was vigorously stirred under nitrogen
in the dark for 16 h (soon the color of the solution turned from
orange to yellow). The organic layer was separated, the aqueous
phase was extracted with CH2Cl2 (deoxygenated; 10 mLꢃ2), and
the combined organic phase was dried (MgSO4) and concentrated
in vacuo to give the crude corresponding dihydropyridine, which
was used in the following reaction without further purification.
These compounds can be kept for 24–48 h at ꢁ30 ꢀC without oxi-
dation or decomposition.
All structures were optimized using the functional B3LYP24a and
the 6-31G
basis sets as implemented in Gaussian 03.25 All energy
*
minima and transition structures were characterized by frequency
analysis. The energies reported in this work include zero-point
vibrational energy corrections (ZPVE) and are not scaled. The sta-
tionary points were characterized by frequency calculations in or-
der to verify that they have the right number of negative
eigenvalues. The intrinsic reaction coordinates (IRC)28 were fol-
lowed to verify the energy profiles connecting each transition
structure to the correct associated local minima. Single-point cal-
culations with the self-consistent reaction field (SCRF) based on the
IEF-PCM27 solvation model (MeCN,
3
¼36.64) were carried out at
B3LYP/6-311þþG** level on the previously optimized most relevant
4.6.1. (3R)-3-Benzyl-1-[(trimethylsilyl)methyl]-3-(N-methyl-dihydro-
nicotinamido)-azetidin-2-one (4b). The general procedure was fol-
lowed from 14b (0.185 mmol, 0.094 g). Yield: 0.059 g (84%); orange
structures.
Acknowledgements
solid; 1H NMR (
d, ppm, CD2Cl2): 7.32–7.25 (m, 5H); 6.88 (d, 1H,
´
J¼1.4 Hz); 5.69 (dd, 1H, J1¼8.0 Hz, J2¼1.6 Hz); 5.47 (br s, 1H); 4.68
(m, 1H); 3.56 (d, 1H, J¼5.4 Hz); 3.34 (d, 1H, J¼5.4 Hz); 3.19 (d, 1H,
J¼13.7 Hz); 3.15 (d, 1H, J¼13.7 Hz); 3.02–2.95 (dm, 2H); 2.91 (s, 3H);
2.68 (d,1H, J¼15.4 Hz); 2.54 (d,1H, J¼15.4 Hz); 0.01 (s, 9H). 13C NMR
We thank the Ministerio de Educacion y Ciencia (MEC, Spain)
(Project: CTQ2006-13891/BQU), UPV-EHU and Gobierno Vasco
(ETORTEK inanoGUNE IE-08/225) for financial support. Thanks are
also due to SGI/IZO-SGIker for the generous allocation of compu-
tational resources and to SGIker (UPV-EHU) for NMR facilities. A
grant from UPV-EHU to P.F. is acknowledged.
(d, ppm, CD2Cl2): 168.7, 167.7, 140.1, 136.3, 130.7, 130.4, 129.0, 127.6,
102.8, 99.0, 68.4, 41.2, 39.9, 33.7, 30.2, 22.6, ꢁ1.8.
4.6.2. (3R)-3-Benzyl-1-(tert-butyl)-3-(N-methyl-dihydro-
nicotinamido)-azetidin-2-one (4d). The general procedure was fol-
lowed from 14d (0.37 mmol, 0.180 g). Yield: 0.096 g (74%); oil; 1H
Supplementary data
NMR spectra of compounds 4a–e, chiral HPLC chromatograms of
2-hydroxy esters 15–19, and Cartesian coordinates of all computed
NMR (d, ppm, CD2Cl2): 7.31–7.27 (m, 5H); 7.22 (br s,1H); 6.89 (d,1H,

3194
J.M. Aizpurua et al. / Tetrahedron 66 (2010) 3187–3194
Baba, N.; Amano, M.; Oda, J.; Inouye, Y. J. Am. Chem. Soc. 1984, 106, 1481–1486;
(f) Makino, T.; Nunozawa, T.; Baba, N.; Oda, J.; Inouye, Y. Tetrahedron Lett. 1979,
20, 1683–1686.
stationary points are included as Supplementary data. These data
can be found in the online version, at doi:10.1016/j.tet.2010.02.085.
13. Saito, R.; Naruse, S.; Takano, K.; Fukuda, K.; Katoh, A.; Inoue, Y. Org. Lett. 2006, 8,
2067–2070.
14. Aizpurua, J. M.; Palomo, C.; Fratila, R. M.; Ferron, P.; Benito, A.; Gomez-Bengoa,
E.; Miranda, J. I.; Santos, J. I. J. Org. Chem. 2009, 74, 6691–6702.
15. (a) Freidinger, R. M.; Veber, D. F.; Perlow, D. S.; Brooks, J. R.; Saperstein, R.
Science 1980, 210, 656–658; (b) Freidinger, R. M.; Schwenk, D.; Veber, D. F. J. Org.
Chem. 1982, 47, 104–109; (c) Freidinger, R. M. J. Org. Chem. 1985, 50, 3631–3633;
(d) Freidinger, R. M. J. Med. Chem. 2003, 46, 5553–5566.
16. (a) Palomo, C.; Aizpurua, J. M.; Benito, A.; Galarza, R.; Khamrai, U. K.; Vazquez,
J.; DePascual-Teresa, B.; Nieto, P. M.; Linden, A. Angew. Chem., Int. Ed. 1999, 38,
3056–3058; (b) Palomo, C.; Aizpurua, J. M.; Benito, A.; Miranda, J. I.; Fratila, R.
M.; Matute, C.; Domercq, C. M.; Gago, F.; Martı´n-Santamaria, S.; Linden, A. J. Am.
Chem. Soc. 2003, 125, 16243–16260.
References and notes
1. (a) Branden, C. I.; Eklund, H. In Dehydrogenases Requiring Nicotinamide
Coenzymes; Jeffery, J., Ed.; Birkhauser: Basel/Boston/Stuttgart, 1980; pp 40–84;
(b) Branden, C. I.; Jornwall, H.; Eklund, H.; Furugren, B. In The Enzymes, 3rd ed;
Boyer, P. D., Ed.; Academic: New York, NY, 1975; Vol. XI, pp 104–190.
2. (a) Loewus, F. A.; Ofner, P.; Fisher, H. F.; Westheimer, F. H. J. Biol. Chem. 1953,
202, 699–704; (b) You, K. Crit. Rev. Biochem. 1984, 17, 313–451.
3. Walsh, C. Enzymatic Reactions Mechanisms; Freeman, W.H.: San Francisco, 1979;
pp 777–827.
4. Ohnishi, Y.; Kagami, M.; Ohno, A. J. Am. Chem. Soc. 1975, 97, 4766–4768.
5. For reviews on NADH models, see: (a) Inouye, Y.; Oda; Baba, N. Asymmetric
Synthesis; Academic: New York, 1983; Vol. 2, pp 91–124; (b) Yasui, S.; Ohno, A.
Bioorg. Chem. 1986, 14, 70–96; (c) Burgess, V. A.; Davies, S. G.; Skerlj, R. T. Tet-
rahedron: Asymmetry 1991, 2, 299–328; (d) Dupas, G.; Levacher, V.; Bourgui-
gnon, J.; Que´guiner, G. Heterocycles 1994, 39, 405–429; (e) Wang, N. X.; Zhao, J.
Synlett 2007, 2785–2791.
6. (a) Ohno, A.; Ikeguchi, M.; Kimura, T.; Oka, S. J. Am. Chem. Soc. 1979, 101, 7036–
7040; (b) Mikata, 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; (d) de Kok, P. M. T.;
Bastiaansen, L. A. M.; van Lier, P. M.; Vekemans, J. A.; Buck, H. M. J. Org. Chem.
1989, 54, 1313–1320; (e) Meyers, A. I.; Oppenlaender, T. J. J. Am. Chem. Soc. 1986,
108, 1989–1996; (f) Meyers, A. I.; Brown, J. D. J. Am. Chem. Soc. 1987, 109, 3155–
3156.
17. Ohno, A.; Ikeguchi, M.; Kimura, T.; Oka, S. J. Chem. Soc., Chem. Commun. 1978,
328–329.
18. (a) Palomo, C.; Aizpurua, J. M.; Legido, M.; Galarza, R.; Deya, P. M.; Dunogues, J.;
Picard, J. P.; Ricci, A.; Seconi, G. Angew. Chem., Int. Ed. 1996, 35, 1239–1241; (b)
Palomo, C.; Aizpurua, J. M.; Legido, M.; Mielgo, A.; Galarza, R. Chem.dEur.
J. 1997, 3, 1432–1441; (c) Palomo, C.; Aizpurua, J. M.; Ganboa, I.; Benito, A.;
Cuerdo, L.; Fratila, R. M.; Jimenez, A.; Loinaz, I.; Miranda, J. I.; Pytlewska,
K. R.; Micle, A.; Linden, A. Org. Lett. 2004, 6, 4443–4446.
19. Palomo, C.; Aizpurua, J. M.; Balentova, E.; Jimenez, A.; Oyarbide, J.; Fratila, R. M.;
Miranda, J. I. Org. Lett. 2007, 9, 101–104.
20. (a) Seebach, D.; Aebi, J. D. Tetrahedron Lett. 1984, 25, 2545–2548; (b) Seebach,
D.; Aebi, J. D.; Gander-Coquoz, M.; Naef, R. Helv. Chim. Acta 1987, 70, 1194–1216.
21. Vicario, J. L.; Badı´a, D.; Carrillo, L. J. Org. Chem. 2001, 66, 9030–9032.
22. Kanomata, N.; Nakata, T. J. Am. Chem. Soc. 2000, 122, 4563–4568.
23. Li, J.; Liu, Y.; Deng, J. G.; Li, X.; Cui, X.; Li, Z. Tetrahedron: Asymmetry 2000, 11,
2677–2682.
24. (a) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789; (b) Becke, A. D.
J. Chem. Phys. 1993, 98, 5648–5652; (c) Kohn, W.; Becke, A. D.; Parr, R. G.
J. Phys. Chem. 1996, 100, 12974–12980.
25. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.;
Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.;
Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G.
A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;
Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, Revision D.03; Gaussian: Wallingford, CT, 2004.
7. (a) Burgess, V. A.; Davies, S. G.; Skerlj, R. T. J. Chem. Soc., Chem. Commun. 1990,
1759–1762; (b) Burgess, V. A.; Davies, S. G.; Skerlj, R. T.; Whittaker, M. Tetra-
hedron: Asymmetry 1992, 3, 871–901; (c) Davies, S. G.; Skerlj, R. T.; Whittaker,
M. Tetrahedron: Asymmetry 1990, 1, 725–728.
8. (a) Seki, M.; Baba, N.; Oda, J.; Inouye, Y. J. Am. Chem. Soc. 1981, 103, 4613–4615;
(b) Hoshide, F.; Ohi, S.; Baba, N.; Oda, J.; Inouye, Y. Agric. Biol. Chem. 1982, 46,
2173–2175; (c) Seki, M.; Baba, N.; Oda, J.; Inouye, Y. J. Org. Chem. 1983, 48, 1370–
1373; (d) Skog, K.; Wennerstro¨m, O. Tetrahedron Lett. 1992, 33, 1751–1754.
9. (a) de Vries, J. G.; Kellogg, R. M. J. Am. Chem. Soc. 1979, 101, 2759–2761; (b) Jouin,
P.; Troostwijk, C. B.; Kellogg, R. M. J. Am. Chem. Soc. 1981, 103, 2091–2093; (c)
Talma, A. G.; Jouin, P.; De Vries, J. G.; Troostwijk, C. B.; Werumeus Buning, G. H.;
Waninge, J.; Visscher, J.; Kellogg, R. M. J. Am. Chem. Soc. 1985, 107, 3981–3997.
10. (a) Imanishi, T.; Hamano, Y.; Yoshikawa, H.; Iwata, C. J. Chem. Soc., Chem.
Commun. 1988, 473–475; (b) Obika, S.; Nishiyama, T.; Tatematsu, S.; Miyashita,
K.; Iwata, C.; Imanishi, T. Tetrahedron 1997, 53, 593–602; (c) Obika, S.; Nish-
iyama, T.; Tatematsu, S.; Miyashita, K.; Imanishi, T. Chem. Lett. 1996, 25, 853–
854; (d) Obika, S.; Nishiyama, T.; Tatematsu, S.; Miyashita, K.; Imanishi, T.
Tetrahedron 1997, 53, 3073–3082; (e) Obika, S.; Nishiyama, T.; Tatematsu, S.;
Nishimoto, M.; Miyashita, K.; Imanishi, T. Heterocycles 1998, 49, 261–267.
11. (a) Ohno, A.; Ushida, S. Mechanistic Models of Asymmetric Reductions; Springer:
Heidelberg, 1986; (b) Gase, R. A.; Pandit, U. K. J. Am. Chem. Soc. 1979, 101, 7059–
7064; (c) Ohno, A.; Kimura, T.; Yamamoto, H.; Kim, S. G.; Oka, S.; Ohnishi, Y. Bull.
Chem. Soc. Jpn 1977, 50, 1535–1538; (d) Ohno, A.; Yamamoto, H.; Okamoto, T.;
Oka, S.; Ohnishi, Y. Bull. Chem. Soc. Jpn 1977, 50, 2385–2386.
26. (a) Kluge, S.; Weston, J. Biochemistry 2005, 44, 4877–4885; (b) Cha, J.-N.;
Cheong, B.-S.; Cho, H.-G. J. Phys. Chem. A 2001, 105, 1789–1796.
27. (a) Cance`s, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107, 3032–3047; (b)
Cossi, M.; Barone, V.; Mennuci, B.; Tomasi, J. J. Chem. Phys. Lett. 1998, 286, 253–
12. (a) Endo, T.; Hayashi, Y.; Okawara, M. Chem. Lett. 1977, 6, 391–394; (b) Endo, T.;
Kawasaki, H.; Okawara, M. Tetrahedron Lett. 1979, 20, 23–26; (c) Katoh, A.;
Naruse, S.; Ohkanda, J.; Yamamoto, H. Heterocycles 1997, 45, 1441–1446; (d)
Baba, N.; Oda, J.; Inouye, Y. J. Chem. Soc., Chem. Commun. 1980, 815–817; (e)
`
260; (c) Tomasi, J.; Mennuci, B.; Cances, E. J. Mol. Struct. (Theochem) 1999, 466,
211–226.
28. Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523–5527.