Anti-selective vinylogous aldol reaction by silylated alkyldioxinone dienolate

Article abstract of DOI:10.1246/cl.2010.1042

Diastereoselective vinylogous aldol reaction is reported by exploiting a silylated dioxinone to give anti-adducts in high selectivity.

Full text of DOI:10.1246/cl.2010.1042

doi:10.1246/cl.2010.1042
1042
Published on the web August 21, 2010
anti-Selective Vinylogous Aldol Reaction by Silylated Alkyldioxinone Dienolate
Tomohiro Yoshinari, Ken Ohmori, and Keisuke Suzuki*
Department of Chemistry, Tokyo Institute of Technology, SORST-JST Agency,
O-okayama, Meguro-ku, Tokyo 152-8551
(Received July 8, 2010; CL-100615; E-mail: ksuzuki@chem.titech.ac.jp)
Me
*
Diastereoselective vinylogous aldol reaction is reported by
exploiting a silylated dioxinone to give anti-adducts in high
selectivity.
5
O
−O
O
R
*
RCHO
base
O
O
O
O
O
O
OH
5
VII
6
moderate selectivity
Vinylogous aldol reactions of acetoacetate derivatives allow
the synthesis of ¤-hydroxy-¢-keto acids and derivatives, which
are useful building blocks for biologically active natural
products.1 Cyclic dienolate I derived from dioxinone 1 is
particularly useful2 to give adduct 2, which serves as versatile
intermediate to allow various elaborations (eq 1).
SiMe3
SiMe3
Me3Si
Me
*
5
O
−O
O
R
*
base
RCHO
O
O
O
O
O
O
OH
Z
7
8
VIII
anti-selectivity
O
−O
O
R
*
RCHO
base
ð1Þ
Scheme 1. Vinylogous aldol reaction.
O
O
O
O
O
O
OH
juncture, an idea occurred to us: if a bulky substituent, such as
a trimethylsilyl group, were introduced to dioxinone as in 7,
allylic strain7 would allow stereocontrolled enolization.
This communication describes an affirmative answer to this
scenario, and introduction of a silyl group to the 5-position of
dioxinone indeed realizes the vinylogous aldol reaction in highly
anti-selective manner.
For the required installation of a silyl group, several sets of
conditions were applied to model dioxinone 1, which turned out
to be not straightforward. Direct silylation (Me3SiOTf and
NEt3)8 only gave the wrong regioisomer 9, while vinylmagne-
sium species,9 generated from iodide 11, failed to react with
silylating agent (Figure 2).
1
I
2
Another feature of dioxinone is its ability to serve as the
precursor to ketene species.3 We exploited this feature in our
recent total synthesis of macrocidin A (3), a cyclic tetramic acid
natural product: the dioxinone in III was used for generating the
key ketene species II for macrocyclization (Figure 1).4 As for
control of the C(12)-stereogenic center next to the dioxinone
moiety, the Pfaltz asymmetric hydrogenation5 worked in
excellent selectivity for trisubstituted olefin V.
In addressing the synthesis of the congener with an extra
hydroxy group at C(13), macrocidin B (4: R = OH), we
envisioned that a unified strategy would be realized, given that
the vinylogous aldol reaction proceeded in anti-selective manner
to establish C(12) and C(13) stereogenic centers (VI ¼ IV).
However, we noted an uneasy situation, because in contrast
to the impressive advances in enantioselective reactions for
simple dienolate I (vide supra),2 no enantio- or even diaster-
eoselective reactions have been developed for £-methyl dien-
olate VII (Scheme 1): the syn/anti stereoselectivity is low,2b,6
suggesting the difficulty in controlling its E/Z geometry. At this
SiMe3
5
Me3SiOTf
NEt3
O
O
O
SiMe3
O
O
O
O
O
O
O
O
Et2O
rt
59%
1
9
10a
not obtained
I
SiMe3
5
O
O
Me3SiCl
i-PrMgCl
O
O
THF
O
–30 °C
R
OH
11
10a
O
R
R
HN
O
O
R'HN
.
12
O
O
R'HN
Figure 2. Attempts for the introduction of silyl group.
12
13
O
COMe
O
COMe 12
R
13
13
As for the latter process, we expected that the corresponding
vinyllithium species may be more reactive, which indeed proved
to react with silylating agents (Table 1). A high-yielding
protocol was established, including the addition of n-butyllithi-
um to the premixed solution of iodide 11 and silyl triflates in
THF at ¹90 °C, by which several trialkylsilyl groups proved to
be installed (Runs 1-3).10
O
R
O
O
II
III
R=H: macrocidin A (3)
R=OH: macrocidin B (4)
13
O
H
R"
O
O
−
12
O
O
O
O
O
R
O
O
This protocol allowed the silylation of homologous dioxi-
none 5 [note that ethyl group is at C(6)], and iodination-
silylation by the above-mentioned conditions gave 7 as colorless
oil (Scheme 2).
anti-selective
vinylogous aldol
Pfaltz
hydrogenation
VI
IV
V
Figure 1. Retrosynthesis of the macrocidins.
Chem. Lett. 2010, 39, 1042-1044
© 2010 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1043
Table 1. Synthesis of 5-silylated dioxinones
CH
3 CH3
SiMe3
SiMe3
O
NOE
I
SiR3
CH3
O
Me3SiO
R3SiX
n-BuLi
Si
H
O
O
a
Me3SiO
O
O
O
O
O
O
O
THF
O
O
–90 °C
15
7
11
10a–c
Run
Electrophile
Product: Yield/%
O
O
O
O
1
2
3
Me3SiOTf
t-BuMe2SiOTf
i-Pr3SiOTf
10a: 94
10b: 94
10c: 92
Me3Si
Me3Si
O
O
<
H
Me
H
Me
H
H
base
favored
Base
I
SiMe3
O
6
O
O
O
Scheme 4. (a) LiN(i-Pr)2, THF, ¹78 ¼ ¹40 °C, 2 h; Me3SiCl,
¹78 °C, 1 h, 63%.
a
b
O
O
O
O
O
5
12
7
Table 2. Mukaiyama conditions
SiMe3
Me3Si
Me
Scheme 2. (a) N-Iodosuccinimide, AcOH, room temp., 18 h,
87%; (b) Me3SiOTf, n-BuLi, THF, ¹90 °C, 10 min, 89%.
Me3SiO
O
Ph
conditions
H
Ph
O
O
O
O
OH
CH2Cl2
−78 °C
O
Upon attempts at the vinylogous aldol reaction of 7, via its
lithium dienolate, with benzaldehyde, we were pleased to find
the desired product 13 was obtained in high anti-selectivity
(syn/anti = 1:6) (Scheme 3).11 By contrast, the corresponding
reaction of the nonsilylated dioxinone 5 gave syn-adduct 14
in moderate stereoselectivity, clearly showing the role of the
trimethylsilyl group for stereocontrol.12
15
13
1.2−1.5 equiv
Run Lewis acida
Time/h Yield/% syn:anti
1
2
Me3SiOTf
BF3¢OEt2
3
4
82b
75
24
(59)b
52
1:1.2
1:3
1:2.4
(1:1.2)
1:15
1:10
1:22
3
TiCl4
3
c
4
5
6
SiCl4
3
5
3d
Li+
–O
SiMe3
SiMe3
O
Me3Si
Me
5
TiCl2(Oi-Pr)2
TiCl2(Oi-Pr)2, MS4A
22
72
O
O
Ph
LDA
THF
–40 °C
PhCHO
O
O
O
O
O
OH
–78 °C
a0.1 equiv. bYield of trimethylsilyl ether of 13. cEtN(i-Pr)2 was
added as base. dThe reaction temperature was gradually raised
to ¹50 °C.
–78
45%
7
2 h
VIII
13
syn:anti = 1:6
cf)
O
Li+
–O
Me
5
By employing MS4A as an additive, fair improvement of the
yield and stereoselectivity was observed (Run 6).14
O
Ph
LDA
THF
–40 °C
2 h
PhCHO
O
O
O
O
O
O
OH
–78 °C
The stereochemical outcome, i.e., anti-selectivity, could be
explained by an open transition state (TS) model. The decisive
factors are severe steric interactions between the phenyl and the
trimethylsilyl groups as well as the methyl group and the
coordinated Lewis acid in the disfavored TS A,15 which are not
present in the favored TS B (Figure 3).
–78
42%
5
14
VII
E:Z = 2:1
syn:anti = 3:1
Scheme 3. Vinylogous aldol reaction with Li+-dienolate.
The (Z) geometry of the enolate VIII was confirmed by
trapping experiment by treatment of 7 with LDA followed by
Me3SiCl followed by NOE study (CDCl3, 400 MHz). This
geometric selectivity could be explained by an allylic interaction
between the trimethylsilyl group and the methyl group. This
trapping was also possible in a preparative scale to give dienol
silyl ether 15 as colorless liquid; 63% yield (70-71 °C/
0.3 mmHg) (Scheme 4).
LA
LA
O
O
H
H
H
Me
Me
Ph
H
O
Ph
O
Me3Si
Me3Si
Me3SiO
Me3SiO
O
O
A
B
syn product
anti product
We turned our attention to the Mukaiyama conditions by
using dienol silyl ether 15.13 Several Lewis acids were screened
in the reactions by adding them to a preformed mixture of
benzaldehyde and dienol silyl ether 15 (Table 2). While
Me3SiOTf or BF3¢OEt2 gave low stereoselectivity (Runs 1
and 2), high anti-selectivities were observed by bulky Lewis
acids, such as SiCl4 and TiCl2(Oi-Pr)2 (Runs 4 and 5). An
optimal set of conditions was established: the suitable reaction
temperature was starting with ¹78 °C and was raised to ¹50 °C.
favored
Figure 3. Transition-state model.
The high anti-selectivity was also valid for the reaction with
aliphatic aldehyde 16, giving smoothly 17 (syn/anti = 1:8)
under the conditions stated above (Scheme 5).16
Upon attempted removal of the silyl group in adduct 13 with
n-Bu4NF, sizable epimerization was observed (Run 1, Table 3),
Chem. Lett. 2010, 39, 1042-1044
© 2010 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1044
SiMe3
O
Me3Si
Me
TiCl2(Oi-Pr)2
1689. b) M. Sato, N. Kanuma, T. Kato, Chem. Pharm. Bull.
1982, 30, 1315. c) M. Sato, H. Ogasawara, E. Yoshizumi, T.
Kato, Chem. Pharm. Bull. 1983, 31, 1902.
T. Yoshinari, K. Ohmori, M. G. Schrems, A. Pfaltz, K.
Suzuki, Angew. Chem., Int. Ed. 2010, 49, 881.
a) S. Kaiser, S. P. Smidt, A. Pfaltz, Angew. Chem., Int. Ed.
2006, 45, 5194. b) S. Bell, B. Wüstenberg, S. Kaiser, F.
Menges, T. Netscher, A. Pfaltz, Science 2006, 311, 642.
a) M. De Rosa, M. R. Acocella, M. F. Rega, A. Scettri,
Tetrahedron: Asymmetry 2004, 15, 3029. b) M. R. Acocella,
M. De Rosa, A. Massa, L. Palombi, R. Villano, A. Scettri,
Tetrahedron 2005, 61, 4091. c) M. R. Acocella, A. Massa, L.
Palombi, R. Villano, A. Scettri, Tetrahedron Lett. 2005, 46,
6141. d) B. M. Trost, J. P. N. Papillon, T. Nussbaumer,
J. Am. Chem. Soc. 2005, 127, 17921.
Me3SiO
O
Ph
(0.1 equiv)
H
Ph
CH2Cl2, MS4A
O
O
O
OH
O
−78
−50 °C
59%
syn:anti = 1:8
4
5
10 h
15
1.5 equiv
16
17
Scheme 5. Reaction with aliphatic aldehyde.
6
as rationalized by generation of the dienolate IX under the basic
conditions followed by nonselective protonation. Fortunately,
the stereochemistry was completely retained in the product 18
when the desilylation as attempted under acidic conditions
(CF3CO2H). This anti-product 18 is stereoisomeric to the main
product in 14 (Scheme 3), which was obtained from 5 without a
silyl group at the C-5 position.
7
8
9
R. W. Hoffmann, Chem. Rev. 1989, 89, 1841.
W. Uhlig, A. Tzschach, Z. Chem. 1988, 28, 409.
V. A. Vu, L. Berillon, P. Knochel, Tetrahedron Lett. 2001,
42, 6847.
Table 3. Removal of silyl group
Me3Si
Me
Me
O
Ph
O
Ph
10 For trapping with other electrophiles, see the following.
PhCHO 95%; CH2=C(Me)CHO, 90% (1,2-adduct); cyclo-
C6H11CHO, 82%; PhCOMe, 81%; PhCH=NSO2Ph, 75%;
MeOTf, 59%; PhCON(Me)OMe, 74%; PhSSPh, 72%.
conditions
O
O
OH
O
O
OH
syn:anti = 1:15
13
18
1
Yield/%
(syn:anti)
11 The syn/anti ratio was estimated by H NMR. anti-13 was
Run Reagent
Temp/°C Time/h
isolated by recrystallization (hexane), and the stereochem-
istry assigned by comparison of the NMR data,2b after
removal of the silyl group.
1
2
n-Bu4NF, THF
CF3CO2H, CH2Cl2
0
25
0.5
0.5
51 (1:1)
96 (1:15)
1
12 The syn/anti ratio was estimated by H NMR. Stereochem-
istry of syn-14 was determined by X-ray crystallographic
analysis. The data has been deposited with Cambridge
Crystallographic Data Centre as supplementary publication
No. CCDC-773065.
H
Me
−O
Ph
O
O
O−
IX
13 T. Mukaiyama, K. Banno, K. Narasaka, J. Am. Chem. Soc.
1974, 96, 7503.
In summary, a highly anti-selective vinylogous aldol
14 Similar transition state model was proposed, see: S.
Shirokawa, M. Kamiyama, T. Nakamura, M. Okada, A.
Nakazaki, S. Hosokawa, S. Kobayashi, J. Am. Chem. Soc.
2004, 126, 13604.
reaction was developed.17 Further work is now in progress to
develop an asymmetric version of the reaction.
1
This work was partially supported by the Global COE
Program (Chemistry) and a Grant-in-Aid for Scientific Research
(JSPS). JSPS Research Fellowship for Young Scientists to T.Y.
is also acknowledged. We are also grateful to Mr. Takuya
Hashimoto for early phase of this work.
15 The syn/anti ratio was estimated by H NMR. anti-17 was
isolated by recrystallization (hexane), and the stereochem-
istry assigned by comparison of the NMR data of ref. 6a,
after removal of the silyl group.
16 Supporting Information is available electronically on the
CSJ-Journal Web site, http://www.csj.jp/journals/chem-lett/
index.html.
References and Notes
1
G. Casiraghi, F. Zanardi, G. Appendino, G. Rassu, Chem.
Rev. 2000, 100, 1929.
17 General procedure: To a mixture of benzaldehyde (20.4 ¯L,
0.200 mmol) and 15 (90.1 mg, 0.300 mmol) and activated
MS 4A (100 mg) in CH2Cl2 (2.0 mL) was added TiCl2(Oi-
Pr)2 (1.0 M toluene solution, 20 ¯L, 20 ¯mol) at ¹78 °C.
After stirring for 3 h at ¹50 °C, the reaction mixture was
poured into sat. aq. NaHCO3. The mixture was filtered
through a Celiteµ pad, and the products extracted (EtOAc,
©3). The combined organic extracts were washed with brine,
dried (Na2SO4), and concentrated in vacuo. Purification by
preparative TLC (hexane/EtOAc = 4/1) gave adduct 13
(47.9 mg, 72%, syn/anti = 1:22).
2
a) M. Sato, S. Sunami, Y. Sugita, C. Kaneko, Chem. Pharm.
Bull. 1994, 42, 839. b) M. Sato, S. Sunami, Y. Sugita, C.
Kaneko, Heterocycles 1995, 41, 1435. c) R. A. Singer, E. M.
Carreira, J. Am. Chem. Soc. 1995, 117, 12360. d) D. A.
Evans, J. A. Murry, M. C. Kozlowski, J. Am. Chem. Soc.
1996, 118, 5814. e) V. B. Gondi, M. Gravel, V. H. Rawal,
Org. Lett. 2005, 7, 5657. f) J. Krüger, E. M. Carreira, J. Am.
Chem. Soc. 1998, 120, 837. g) S. E. Denmark, G. L. Beutner,
J. Am. Chem. Soc. 2003, 125, 7800.
3
a) G. Jäger, J. Wenzelburger, Liebigs Ann. Chem. 1976,
Chem. Lett. 2010, 39, 1042-1044
© 2010 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/