Stereoselective α-alkylation of methyl 6-deoxy-3,4-di-O-(tert-butyldimethylsilyl)-2-O-(2-methyl-3-oxobutanoyl)-α-d-glucopyranoside

Article abstract of DOI:10.1016/j.tetlet.2008.12.091

Allylation and benzylation at the α-carbon of α-methylated acetoacetyl (2-methyl-3-oxobutanoyl) group incorporated into the 2-OH of methyl 6-deoxy-3,4-O-(tert-butyldimethylsilyl)-α-d-glucopyranoside provided the respective α,α-differentially alkylated ace

Full text of DOI:10.1016/j.tetlet.2008.12.091

Tetrahedron Letters 50 (2009) 1139–1142
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Stereoselective
a-alkylation of methyl
6-deoxy-3,4-di-O-(tert-butyldimethylsilyl)-
2-O-(2-methyl-3-oxobutanoyl)-
a-D-glucopyranoside
*
Yoko Akashi, Ken-ichi Takao, Kin-ichi Tadano
Department of Applied Chemistry, Keio University, Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Allylation and benzylation at the
incorporated into the 2-OH of methyl 6-deoxy-3,4-O-(tert-butyldimethylsilyl)-
vided the respective
ity. Thus-obtained doubly alkylated products possess an all-carbon quaternary stereogenic center with an
absolute stereochemistry opposite to that introduced by using the 4-O-acetoacetyl regioisomer as the
alkylation substrate.
a
-carbon of
a
-methylated acetoacetyl (2-methyl-3-oxobutanoyl) group
-glucopyranoside pro-
a,a-differentially alkylated acetoacetyl derivatives, both with high diastereoselectiv-
Received 28 October 2008
Revised 16 December 2008
Accepted 18 December 2008
Available online 25 December 2008
a-D
Keywords:
Ó 2008 Elsevier Ltd. All rights reserved.
a-Alkylation
Asymmetric synthesis
Carbohydrates
Chiral auxiliary
Stereoselective synthesis
Stereoselective access to enantiomerically enriched building
blocks with an all-carbon stereogenic center is one of actively
investigated subjects in the field of current synthetic organic
chemistry. For this subject, a number of sophisticated asymmetric
approaches have been realized by using transition-metal-based
chiral catalysts,1 chiral auxiliaries,2 or organocatalysts.3 Especially,
the building blocks with a stereochemically defined all-carbon
quaternary center can be used as structural elements in the syn-
thesis of biologically intriguing compounds.
known stereochemically defined pyrazoline derivative 5 or 7,
respectively.
To further demonstrate the utility of our sugar-based chiral
template approach for access to enantiomerically enriched build-
ing blocks, we have sought other sugar-based templates, which
can realize highly stereoselective carbon–carbon bond-forming
reactions. The results revealed another
9 (Scheme 2). Herein, we describe the synthetic utility of 9, exem-
plified by the -alkylation of the 2-methyl-3-oxobutanoyl ester de-
D-glucose-based template
a
During the past several years, we had demonstrated the syn-
thetic utility of methyl 6-deoxy-2,3-di-O-(tert-butyldimethylsi-
rived from 9. The template 9 was prepared efficiently via stepwise
migration of the two tert-butyldimethylsilyl (TBS) groups attached
on O-3 and O-2 in 1 to the O-4 and O-3 positions by sequential
treatment of two kinds of bases, namely, the TBS group on O-3
migrated to the 4-OH group by the treatment of 1 with NaOMe,
lyl)-
methyl
tive carbon–carbon bond-forming reactions.4 In a recent paper,5
we reported stereoselective sequential alkylation at the -carbon
a-D
-glucopyranoside 1 (Scheme 1), readily prepared from
a
-D
-glucopyranoside, as a chiral template for stereoselec-
a
of the 4-O-acetoacetyl (3-oxobutanoyl) derivative 2, which was
prepared by the acylation of 1. The base-mediated sequential
Xc
O
O
O
HO
a-alkylation of 2 with different carbon electrophiles such as MeI,
TBSO
Xc
then benzyl bromide or allyl bromide eventually provided 4 or 6
via the initially formed mono-C-methylated diastereomeric mix-
ture 3. The second alkylation proceeded with remarkable diastere-
oselectivity in both cases.6 The stereochemical assignment of the
thus-introduced all-carbon quaternary center in the alkylated
product 4 or 6 was ascertained after converting 4 or 6 into the
O
TBSO
OMe
2
1
N
NH
O
O
O
R
O
Xc
Xc
O
1
O
+
O
R
3
5 R: -CH2Ph
7 R: -CH2-CH=CH2
4 R: -CH2Ph
6 R: -CH2-CH=CH2
* Corresponding author.
E-mail address: tadano@applc.keio.ac.jp (K. Tadano).
Scheme 1. Stereoselective a,a
-dialkylation of 4-O-acetoacetate 2 derived from 1.5
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.12.091

1140
Y. Akashi et al. / Tetrahedron Letters 50 (2009) 1139–1142
O
O
O
O
O
HO
TBSO
HO
HO
TBSO
HO
O
O
TBSO
TBSO
a
b
O
HO
HO
a
b
TBSO
TBSO
HO
OEt
OMe
OMe
OMe
OMe
12
9
HO
+
8
1
OMe
O
16
15
O
14
Scheme 2. Synthesis of 9. Reagents and conditions: (a) NaOMe, MeOH, rt, 70% (91%
after one recycle); (b) LiHMDS, THF, rt, 85%.
Scheme 4. Removal of the sugar template from 12. Reagents and conditions: (a)
6 M aq HCl/THF = 1:1, rt; (b) NaOEt, EtOH, rt, 90% for 15 and 97% for 16 for 2 steps.
providing 87 in 70% yield and 30% of 1 was recovered. The recov-
ered 1 was subjected to the same NaOMe treatment to obtain addi-
tional 8. Finally, the 2,4-di-O-TBS derivative 8 was obtained in an
overall yield of 91% after one recycle. Then the treatment of 8 with
lithium hexamethyldisilazide (LiHMDS) solely provided the 3,4-di-
O-TBS derivative 9 in a good yield of 85%.8 None of the 2,3-di-O-TBS
derivative 1 was found in the reaction mixture.9 Therefore, in the
case of 8, the silyl migration from O-2 to O-3 occurs exclusively un-
der the basic conditions employed. Although we have no firm evi-
dence for these facile silyl-group migrations from O-3 to O-4, then
from O-2 to O-3, we presume that the 3,4-di-O-TBS derivative 9 is
the least spatially congested regioisomer, compared with 1 or 8.
We explored the potential of 9 as a chiral template, exemplified
O
O
TBSO
TBSO
N
NH
b
a
O
OMe
+
9
12
O
O
c
17
18
N
NH
Ph
O
+
13
9
19
Scheme 5. Conversion of 12 or 13 into 18 or 19, respectively, for determination of
the enantiomeric excess of the quaternary stereogenic center. Reagents and
conditions: (a) H2, Pd/C, MeOH; (b) NH2NH2ꢂH2O, EtOH, 140 °C in a sealed tube,
82% of 18 and 83% of 9 for 2 steps; (c) NH2NH2ꢂH2O, EtOH, 140 °C in a sealed tube,
quantitative yield.
by the
a-alkylation of its 2-O-(2-methyl-3-oxobutanoyl) derivative
11. The preparation of 11 and its
a-alkylation with two electro-
philes are summarized in Scheme 3.10 Heating 9 with 2,2,5,6-tetra-
methyl-1,3-dioxin-4-one11 10 in refluxing o-xylene provided 11 as
a diastereomeric mixture.12 This mixture 11 was subjected to
of 18, the ee of the quaternary carbon center in 18, and thus that
in 12, was determined to be 95%.21
Analogously, the hydrazinolysis of 13 provided a pyrazolone
derivative 19, accompanied by isolation of 9. The absolute stereo-
chemistry of the quaternary carbon was determined to be (S) by
comparison with the reported [
mined by chiral HPLC analysis to be 89%.23 As in the case of the for-
mation of 12, the -benzylation of 11 provided 13 with an all-
a
-alkylation with allyl bromide using tert-BuOK as the base, pro-
viding the -differentially alkylated acetoacetate 12 in 93% yield
with almost complete diastereoselectivity.13,14 Analogously, the
a,a
benzylation of 11 provided the
a-benzylated product 13 in 79%
yield as an apparently single diastereomer.15
a]
D value.22 The ee of 19 was deter-
Then, we determined the stereochemistry of the newly intro-
duced quaternary stereogenic center incorporated into the allylated
a
product 12 by detachment of the optically active
a
-methyl-
a
-allyl-
carbon quaternary center, the absolute stereochemistry of which
was opposite to that obtained through our previous approach using
3.5
3-oxobutanoyl moiety from the sugar template. For this purpose,
two TBS-protecting groups in 12 were first removed by acid hydro-
lysis, providing 14 (Scheme 4). The sugar template was smoothly
removed by alcoholysis of 14 with NaOEt in EtOH, providing ethyl
We propose the transition state TS-A depicted in Scheme 6 for
the excellent stereoselectivity observed in the
a-alkylation of 11,
(S)-2-allyl-2-methyl-3-oxobutanoate (15) and methyl 6-deoxy-
a
-
together with the transition state TS-B for our previous results.5
In the transition state TS-A, it is most likely that the (Z)-potas-
sium-chelated enolate formed by the tert-BuOK-treatment of 11
was attacked by alkyl halide to provide 12 or 13. Furthermore,
the attack of the electrophile to the enolate predominantly oc-
curred from the less-congested front side as shown (Si-face of
D
-glucopyranoside (16), both in excellent yields.16 The levorotatory
sign for 15 [½a 2D0
ꢀ
ꢁ25.4 (c 1.18, CHCl3)] confirmed its absolute ste-
reochemistry to be (S) by comparison with the reported [a]
D
for
the (S)-enantiomer.17 Consequently, the chirality of the asymmetric
all-carbon quaternary center in 12 was opposite to that introduced
in the allylation executed for the substrate 3 during our previous
study.5
the a-carbon to the carbonyl in the enolate form), because the rear
side was well shielded by the bulky TBSO group attached at C-3.
This phenomenon was the complete reverse of what we observed
using substrate 3 via transition state TS-B.5 At the present time,
however, we do not have any reasonable explanation for the higher
diastereoselectivity observed in the attack of sterically less-con-
gested allyl bromide, compared with that observed in the attack
of benzyl bromide.24
Next, we determined the enantiomeric excess (ee) of the newly
introduced stereogenic center in 12 accurately by chiral HPLC anal-
ysis. The allyl group in 12 was first hydrogenated to a propyl group,
providing 17 (Scheme 5).18 The hydrazinolysis of 17 at 140 °C19
provided a pyrazolone derivative 18,20 and the sugar template 9
was recovered efficiently. Based on the chiral HPLC measurement
RX
O
Na
Si
O
O
O
O
O
O
O
O
TBSO
TBSO
TBSO
TBSO
O
Re -face
attack
O
a
b
O
OMe
K
O
9
O
Si
OMe
OMe
R
O
Si
OMe
Si - face
attack
O
Si
O
O
O
O
O
O
RX
10
O
12 R= CH2CH=CH2
13 R= CH2Ph
11
TS-A for the enolate derived from 11
TS-B for the enolate derived from 3
4 or 6
Scheme 3. Synthesis of 2-O-(2-methyl-3-oxobutanoyl) derivative 11 and its
a-
12 or 13
alkylation with two electrophiles. Reagents and conditions: (a) 10, o-xylene, reflux,
92%; (b) for 12: allyl bromide, tert-BuOK, THF, ꢁ78 to 5 °C, 93%; for 13: benzyl
bromide, tert-BuOK, THF, ꢁ78 to 0 °C, 79%.
Scheme 6. Plausible transition states for the formation of 4 (or 6) and 12 (or 13).

Y. Akashi et al. / Tetrahedron Letters 50 (2009) 1139–1142
1141
(EtOAc/toluene = 1:10); ½a D20
ꢀ
+73.4 (c 2.52, CHCl3); 1H NMR (300 MHz, CDCl3) d
These
a,a-differentially alkylated 3-oxobutanoic acid ester 15
0.09, 0.11 (2s, each 3H), 0.12, 0.13 (2s, each 3H), 0.91, 0.93 (2s, each 9H), 1.28
(d, 3H, J = 6.8 Hz), 2.31 (d, 1H, J = 10.5 Hz), 3.25 (t, 1H, J = 7.1 Hz) 3.43 (s, 3H),
3.44 (ddd, J = 3.2, 7.9, 10.5 Hz), 3.74–3.80 (m, 2H), 4.69 (d, 1H, J = 3.2 Hz); 13C
NMR (75 MHz) d ꢁ3.8, ꢁ3.6, ꢁ3.1, ꢁ2.9, 18.1 (2C), 18.4, 26.1 (2C), 26.3 (3C),
55.5, 70.3, 72.7, 75.7, 77.2, 98.0; HRMS calcd for C18H39O4Si2 (M+ꢁOMe) m/z
375.2387, found 375.2382.
and pyrazolone derivatives 18 and 19 are expected to serve as ver-
satile building blocks for the synthesis of enantiomerically
enriched carbocyclic and heterocyclic compounds with a stereo-
chemically defined all-carbon quaternary stereogenic center. We
had demonstrated the synthetic utility of similar building blocks
in our previous papers.4k,5
In summary, we have further developed our sugar-based chiral
template approach for the stereoselective carbon–carbon bond-
forming reaction, exemplified by the design of a new and effective
template, that is, methyl 6-deoxy-3,4-O-(tert-butyldimethylsilyl)-
9. We examined this silyl-group migration using other bases. For example, 2,4-di-
O-TBS derivative 8 (44%), 3,4-di-O-TBS derivative 9 (16%), and 1 (32% recovery)
were obtained when LiHMDS (1.0 mol equiv) was used in THF at ꢁ78 to ꢁ20 °C.
For convenient purification and high-yield of 9, we adopted the stepwise
migration via the isolation of 8.
10. We also explored the
a-methylation of the 2-O-(3-oxo-butanoyl) derivative,
prepared from 9 by the acylation with 2,2,6-trimethyl-1,3-dioxin-4-one, with
methyl iodide using K2CO3 as base. As a result, 2-O-(2-methyl-3-oxobutanoyl)
derivative 11 was obtained in less satisfactory yield with concomitant
production of the 2-O-(2,2-dimethyl-3-oxobutanoyl) derivative. For this
reason, we prepared 11 by direct introduction of the 2-methyl-3-
oxobutanoyl group at O-2 of 9 under the neutral retro-Diels–Alder strategy
shown in Scheme 3.
a
-
D
-glucopyranoside. The
2-O-(2-methyl-3-oxo-butanoyl) derivative of the sugar template
provided the respective -differentially dialkylated products in
a-allylation and a-benzylation of the
a,a
remarkably high diastereoselectivity. The direction of the electro-
phile attack is highly controlled by the bulky silyl ether located
at C-3 of the sugar template. These facts complement our previous
11. Sato, M.; Ogasawara, M.; Oi, K.; Kato, T. Chem. Pharm. Bull. 1983, 31, 1896–
1901.
12. Based on the 13C NMR analysis, the diastereomeric ratio of 11 was estimated to
be ca. 2:1 to 3:1. As separation of the diastereomers was fruitless, we did not
determine the stereochemistry of the respective diastereomers.
13. Synthesis of 12. The following reaction was carried out under Ar. To a cooled
(0 °C) stirred solution of 11 (94 mg, 0.185 mmol) in THF (1.9 mL) was added
tert-BuOK (27 mg, 0.204 mmol). The mixture was stirred at ꢁ78 °C for 5 min,
results on the stereoselective introduction of the a,a-differentially
alkylated quaternary center achieved using the 4-O-acetoacetate
regioisomer.25
References and notes
and allyl bromide (32
lL, 0.37 mmol) was added. After being stirred at ꢁ78 °C
for 30 min and 5 °C for 18.5 h, the mixture was quenched with satd aq NH4Cl
(5 mL), diluted with EtOAc (10 mL), and washed with satd aq NH4Cl (5 mL ꢃ 3).
The organic layer was dried and concentrated in vacuo. The residue was
purified by column chromatography on silica gel (EtOAc/hexane = 1:80) to
1. Some recent prominent papers on this subject: (a) Trost, B. M.; Pissot-
Soldermann, C.; Chen, I.; Schroeder, G. M. J. Am. Chem. Soc. 2004, 126, 4480–
4481; (b) Trost, B. M.; Xu, J. J. Am. Chem. Soc. 2005, 127, 2846–2847; (c) Mohr, J.
T.; Behenna, D. C.; Harned, A. M.; Stoltz, B. M. Angew. Chem., Int. Ed. 2005, 44,
6924–6927.
provide 94 mg (93%) of 12 as
a
colorless oil: TLC Rf 0.45 (EtOAc/
hexane = 1:7); ½a D23
ꢀ
+59.1 (c 1.14, CHCl3); 1H NMR (300 MHz, CDCl3) d 0.11,
2.
A
recent review on this subject: Arya, P.; Qin, H. Tetrahedron 2000, 56,
0.12 (2s, each 6H), 0.87, 0.91 (2s, each 9H), 1.35 (d, 3H, J = 6.4 Hz), 1.54 (s,
3H), 2.20 (s, 3H), 2.58 (dd, 1H, J = 7.6, 14.0 Hz), 2.65 (dd, 1H, J = 6.9, 14.0 Hz),
3.34 (s, 3H), 3.37 (dd, 1H, J = 5.1, 9.0 Hz), 3.72 (m, 1H), 3.91 (dd, 1H, J = 5.1,
6.8 Hz), 4.82 (d, 1H, J = 3.8 Hz), 4.91 (dd, J = 3.8, 6.8 Hz), 5.10–5.16 (m, 2H),
5.60–5.71 (m, 1H); 13C NMR (75 MHz) d ꢁ3.9, ꢁ3.6, ꢁ3.3, ꢁ2.7, 18.0, 18.2,
18.9, 19.0, 26.0 (3C), 26.1(3C), 26.5, 39.6, 54.9, 59.5, 69.0, 73.1, 73.3, 77.9,
95.7, 118.9, 132.8, 172.3, 204.7; HRMS calcd for C27H52O7Si2 (M+) m/z
544.3252, found 544.3258.
917–947.
3. Some recent prominent papers on this subject: (a) Bella, M.; Jørgensen, K. A. J.
Am. Chem. Soc. 2004, 126, 5672–5673; (b) Wilson, R. M.; Jen, W. S.; MacMillan,
D. W. C. J. Am. Chem. Soc. 2005, 127, 11616–11617.
4. An account on our sugar-based chiral template approach for stereoselective
carbon–carbon bond-forming reactions: (a) Totani, K.; Takao, K.; Tadano, K.
Synlett 2004, 2066–2080; Our previous publications on this topic: (b) Totani,
K.; Nagatsuka, T.; Takao, K.; Ohba, S.; Tadano, K. Org. Lett. 1999, 1, 1447–1450;
(c) Munakata, R.; Totani, K.; Takao, K.; Tadano, K. Synlett 2000, 979–982; (d)
Totani, K.; Nagatsuka, T.; Yamaguchi, S.; Takao, K.; Ohba, S.; Tadano, K. J. Org.
Chem. 2001, 66, 5965–5975; (e) Nagatsuka, T.; Yamaguchi, S.; Totani, K.; Takao,
K.; Tadano, K. Synlett 2001, 481–484; (f) Totani, K.; Asano, S.; Takao, K.; Tadano,
K. Synlett 2001, 1772–1776; (g) Nagatsuka, T.; Yamaguchi, S.; Totani, K.; Takao,
K.; Tadano, K. J. Carbohydr. Chem. 2001, 20, 519–535; (h) Tamai, T.; Asano, S.;
Totani, K.; Takao, K.; Tadano, K. Synlett 2003, 1865–1867; (i) Asano, S.; Tamai,
T.; Totani, K.; Takao, K.; Tadano, K. Synlett 2003, 2252–2254; (j) Sasaki, D.;
Sawamoto, D.; Takao, K.; Tadano, K.; Okue, M.; Ajito, K. Heterocycles 2007, 72,
103–110; (k) Kubo, H.; Kozawa, I.; Takao, K.; Tadano, K. Tetrahedron Lett. 2008,
49, 1203–1207.
14. We also explored the sequential
a-alkylation in the order of allylation and
methylation of 2-O-(3-oxobutanoyl) derivative. The initial allylation (allyl
bromide, NaOMe, THF, ꢁ78 °C to rt) provided mono-allylated product (44%).
The second methylation (MeI, NaOMe, THF, rt) of the
a-allylated product
provided the
a,a-dialkylated product in 44% yield. Unfortunately, the
diastereomeric ratio of the products was approximately 1:1 based on 1H
NMR analysis.
15. Compound 13 as a colorless oil: TLC Rf 0.42 (EtOAc/hexane = 1:7); ½a D21
ꢀ
+51.9 (c
0.91, CHCl3); 1H NMR (300 MHz, CDCl3) d 0.10, 0.11 (2s, each 3H), 0.13 (s, 6H),
0.88, 0.89 (2s, each 9H), 1.26 (d, 3H, J = 6.4 Hz), 1.29 (s, 3H), 2.24 (s, 3H), 3.10
(d, 1H, J = 13.8 Hz), 3.21 (s, 3H), 3.31 (d, 1H, J = 13.8 Hz), 3.37 (dd, 1H, J = 5.3,
8.4 Hz), 3.70 (m, 1H), 3.96 (dd, 1H, J = 5.3, 7.1 Hz), 4.73 (d, 1H, J = 3.8 Hz), 4.93
(dd, 1H, J = 3.8, 7.1 Hz), 7.11–7.14 (m, 2H), 7.22–7.28 (m, 3H); 13C NMR
(75 MHz) d ꢁ3.9, ꢁ3.7, ꢁ3.2, ꢁ2.6, 18.0, 18.1, 18.8 (2C), 26.0 (3C), 26.1(3C),
26.7, 40.8, 54.9, 60.9, 69.1, 73.0, 73.3, 77.8, 95.8, 126.8, 128.1 (2C), 130.4 (2C),
136.3, 172.1, 204.8; HRMS calcd for C30H51O6Si2 (M+ꢁOMe) m/z 563.3224,
found 563.3242.
5. Kozawa, I.; Akashi, Y.; Takiguchi, K.; Sasaki, D.; Sawamoto, D.; Takao, K.;
Tadano, K. Synlett 2007, 399–402.
6. When the two alkylations were carried out in the order of benzylation and
methylation, the diastereomeric ratio for
4 significantly changed with a
decrease in the formation of 4 (4.5:1 to 4:1 in favor of the formation of 4).4k,5
7. All new compounds were fully characterized by spectral means [1H and 13C
NMR, IR, and HRMS]. Yields refer to isolated products after purification by
column chromatography on silica gel.
16. The detachment of the sugar template from 12 did not proceed under the
ethanolytic conditions. We had encountered the same difficulty in the
detachment of the sugar template from other
derivatives; see Ref. 4k.
a,a-dialkylated acetoacetate-
8. The preparation of 9 from 1 via 8. To a cooled (0 °C) stirred solution of 1
(500 mg, 1.23 mmol) in MeOH (10 mL) was added NaOMe (1.0 M solution in
MeOH, 1.8 mL, 1.8 mmol). The mixture was stirred at rt for 15 h, quenched with
satd aq NH4Cl (5.0 mL), diluted with EtOAc (40 mL), and then washed with satd
aq NH4Cl (20 mL ꢃ 3). The organic layer was dried and concentrated in vacuo.
The residue was purified by column chromatography on silica gel (EtOAc/
hexane = 1:25 then 1:10) to provide 349 mg (70%) of 8 and 153 mg (30%) of 1.
The same procedure was repeated for the recovered 1 (153 mg, 0.376 mmol) to
provide 105 mg of additional 8 (46 mg of 1 was recovered). Thus, 454 mg (91%)
17. For the reported value of 15: ½a D22
ꢁ27.1 (c 1.17, CHCl3), see: Fráter, G. Helv.
ꢀ
Chim. Acta 1979, 62, 2825–2828.
18. The hydrazinolysis of 12 was accompanied by the hydrogenation of the allylic
olefin to some extent. Thus, 12 was subjected to hydrogen addition prior to
hydrazinolysis.
19. For the hydrazinolysis conditions, see: Moreno-Mañas, M.; Trepat, E.;
Sebastián, R. M.; Vallribera, A. Tetrahedron: Asymmetry 1999, 10, 4211–4224.
20. Compound 18 as colorless crystals: mp 61–62 °C; ½a D19
ꢀ
+49.1 (c 0.265, CHCl3);
of 8 was obtained as a colorless oil: TLC Rf 0.52 (EtOAc/hexane = 1:5); ½a D20
ꢀ
1H NMR (300 MHz, CDCl3) d 0.88 (t, 3H, J = 7.1 Hz), 0.92–1.31 (m, 2H), 1.22
(s, 3H) 1.52 (ddd, 1H, J = 4.5, 4.7, 12.2 Hz), 1.75 (ddd, J = 4.5, 4.7, 12.2 Hz),
1.98 (s, 3H), 8.47 (br s, 1H); 13C NMR (75 MHz) d 13.1, 14.0, 17.9, 20.3, 37.4,
51.9 164.6, 180.2; HRMS calcd for C8H14N2O (M+) m/z 154.1106, found
154.1092.
+76.7 (c 1.19, CHCl3); 1H NMR (300 MHz, CDCl3) d 0.09, 0.10 (2s, each 3H), 0.11,
0.14 (2s, each 3H), 0.89, 0.91 (2s, each 9H), 1.23 (d, 3H, J = 6.2 Hz), 2.08 (d, 1H,
J = 2.1 Hz), 3.18 (t, 1H, J = 9.0 Hz), 3.38 (s, 3H), 3.52 (dd, 1H, J = 3.6, 9.0 Hz),
3.61–3.70 (m, 2H), 4.55 (d, 1H, J = 3.6 Hz); 13C NMR (75 MHz) d ꢁ4.6, ꢁ4.5, ꢁ3.8
(2C), 18.2 (2C), 18.3, 25.8 (2C), 25.9 (3C), 55.1, 67.7, 74.1, 76.5, 77.5, 99.7; HRMS
calcd for C18H39O4Si2 (M+ꢁOMe) m/z 375.2387, found 375.2389. The following
reaction was carried out under Ar. To a cooled (0 °C) stirred solution of 8
(544 mg, 1.34 mmol) in THF (10 mL) was added LiHMDS (1.0 M solution in THF,
2.0 mL, 2.0 mmol). The mixture was stirred at rt for 1 h, quenched with satd aq
NH4Cl (5.0 mL), diluted with EtOAc (40 mL), and then washed with satd aq
NH4Cl (20 mL ꢃ 3). The organic layer was dried and concentrated in vacuo. The
residue was purified by column chromatography on silica gel (EtOAc/
toluene = 1:80) to provide 462 mg (85%) of 9 as a colorless oil: TLC Rf 0.38
21. The chiral HPLC conditions: CHIRALPAK AD-H column, hexane/2-
propanol = 20:1. Racemic 18 was prepared from ethyl 2-methylacetoacetate
as follows: (1) allyl bromide, tert-BuOK, THF, 0 °C; (2) H2, Pd/C, MeOH; (3)
NH2NH2ꢂH2O, EtOH, 140 °C in a sealed tube.
22. Compound 19, (4S)-4-benzyl-3,4-dimethyl-2-pyrazolin-5-one as colorless
crystals: mp 90–91 °C; ½ ꢀ
a 2D2 +162 (c 0.73, CHCl3); 1H NMR (300 MHz,
CDCl3) d 1.34 (s, 3H), 2.04 (s, 3H), 2.83 (d, 1H, J = 13.6 Hz), 3.12 (d, 1H,
J = 13.6 Hz), 7.07–7.11 (m, 2H), 7.12–7.23 (m, 3H), 8.28 (br s, 1H); 13C NMR
(75 MHz) d 14.3, 20.6, 41.2, 53.5, 127.2 (2C), 128.4 (2C), 128.9, 135.2, 163.4,

1142
Y. Akashi et al. / Tetrahedron Letters 50 (2009) 1139–1142
179.5; HRMS calcd for C12H14N2O (M+) m/z 202.1106, found 202.1106. For
Scheme 6 by the following reason. If the potassium-chelate forms between the
OMe and enolate, the sterically less-congested space might turn out to be Re-
the reported [
Ref. 19.
a
]
D
for 5, the enantiomer of 19: ½a D15
ꢁ186 (c 1.24, CHCl3), see
ꢀ
face (not the Si-face) of the a-methylated enolate.
23. The conditions: CHIRALPAK OD column, hexane/2-propanol = 20:1.
24. A referee suggested the participation of anomeric OMe for the potassium-
chelate formation. Although we have no evidence which rules out this
possibility, we insist on the presence of the chelate structure depicted in
25. As described previously,4k,5 the highly stereoselective introduction of the (S)-
quaternary carbon center into the acetoacetate ester 2 could not be attained by
changing the order of the addition of the electrophiles, that is, allylation (or
benzylation) and then methylation.