Development of silicon-tethered anionic reaction and its application to the synthesis of chiral A-ring moieties of Taxol

Article abstract of DOI:10.1016/j.tetlet.2004.09.143

Silicon-tethered intramolecular nucleophilic additions of a hydroxymethyl unit to ketones in β-hydroxyketones have been developed. The product obtained by this protocol was successfully converted to chiral A-ring moieties of Taxol. Also developed was a pr

Full text of DOI:10.1016/j.tetlet.2004.09.143

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 8647–8651
Development of silicon-tethered anionic reaction and its
application to the synthesis of chiral A-ring moieties of Taxole
Mitsuhiro Iwamoto, Masayuki Miyano, Masayuki Utsugi,
Hatsuo Kawada and Masahisa Nakada*
Department of Chemistry, School of Science and Engineering, Waseda University, 3-4-1 Ohkubo, Shinjuku-ku, Tokyo 169-8555, Japan
Received 27 August 2004; revised 21 September 2004; accepted 22 September 2004
Available online 8 October 2004
Abstract—Silicon-tethered intramolecular nucleophilic additions of a hydroxymethyl unit to ketones in b-hydroxyketones have been
developed. The product obtained by this protocol was successfully converted to chiral A-ring moieties of TaxolTM. Also developed
was a promising silicon-tethered intramolecular a-alkylation reaction of the cyclic hydroxyketone.
Ó 2004 Elsevier Ltd. All rights reserved.
Construction of a stereogenic quaternary carbon is a
challenging problem in natural product synthesis.1
Among various stereogenic quaternary carbons, a chiral
tertiary alcohol can be prepared by enantioselective
addition of nucleophiles to ketones, and when the
ketone is chiral, asymmetric induction is an efficient
method. However, since the diastereoselective addition
to chiral ketones predominantly occurs from the less
hindered side, preferential addition from the more hin-
dered side has rarely been reported.2
of TaxolTM was envisioned to form between the C10 and
C11 bonds by the intramolecular Nozaki–Hiyama reac-
tion (Scheme 2) or the intramolecular Suzuki–Miyaura
coupling reaction (Scheme 3). Hence, 3 and 6 must be
prepared to examine these ring-closing reactions. Addi-
tion of Fragment B as the allylic organometallic reagent
to aldehydes 1a and 1b would stereoselectively generate
the C2 stereogenic center in a chelation-controlled man-
ner; hence aldehydes 1a and 1b were needed for the syn-
thesis of 3 and 6.
We have been studying the asymmetric total synthesis of
TaxolTM via a convergent route,3 and disconnection of
TaxolTM in our plan results in the coupling of two frag-
ments, Fragment A and B (Scheme 1). Since compounds
2 and 5 were surmised to convert to TaxolTM, the B-ring
In the synthesis of 1a and 1b, construction of the chiral
tertiary alcohol was a problem. Although (S)-3-hydr-
oxy-2,2-dimethylcyclohexanone 7 was readily prepared
by bakerꢀs yeast,4 a new method for the diastereoselec-
tive addition of a hydroxymethyl unit to the ketone in
7 had to be developed. We expected the silicon-tethered
AcO
10
O
OH
C
BzHN
Ph
O
9
OH
10
O
OBn
OBn
O
A
B
H
X
3
OH
11
H
1
HO
2
OAc
3
TaxolTM
OBz
H
H
2
MOMO
MOMO
+
Fragment B (4)
Fragment A (1a, 1b)
OBn
OBn
OBn
2
3
OMPM
Scheme 1.
Keywords: Silicon tethered; Intramolecular reaction; TaxolTM; Carban-
ion; Nucleophilic addition; Alkylation.
O
MOMO
1a
4
X
*
Corresponding author. Tel./fax: +813 5286 3240; e-mail: mnakada@
waseda.jp
Scheme 2.
0040-4039/$ - see front matter Ó 2004Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.09.143

8648
M. Iwamoto et al. / Tetrahedron Letters 45 (2004) 8647–8651
Table 1. Preparation of 10 by the silicon-tethered nucleophilic
OBn
10
OBn
I
addition
11
conditions
8
10
3
H
H
2
MOMO
MOMO
Entry
8
Reagents
(equiv)
Solvent
Temp Time Yield
(min) (%)a
OBn
OBn
(°C)
5
6
I
1
2
3
8c n-BuLi (1.0)
8c n-BuLi (1.0)
8c n-BuLi (1.0)
THF
Et2O
À78
À78
5
30
15
61
43
55
+
4
THF/Et2O À78
O
MOMO
(4/1)
THF
1b
4
8c n-BuLi (1.0)
HMPA (2.0)
À78
15
33
Scheme 3.
5
8b t-BuLi (1.95) THF
8c t-BuLi (1.95) THF
À78
À78
0
5
5
84
86
40
—
38
50
6
7
8c Li (10.0)
8c LiNpb (3.0)
THF
THF
60
30
10
15
c
8
À78
À78
À78
8c LiDBBd (3.0) THF
Si
O
X
9
10
OH
8b SmI2 (2.2)
HMPA (4.0)
THF
THF
THF
O
11
12
8c SmI2 (3.0)
HMPA (12.0)
8c SmBr2 (3.0)
À78
5
18
50
O
7
8a (X = Cl); 8b (X = Br);
8c (X = I)
0
15
a Isolated yield.
O
O
Si
Si
b Lithium naphthalenide.
c Decomposed.
d Lithiunm 4,40-di-t-butylbiphenyl.
O
OH
9
10
With 8b and 8c in hand, we first examined the halogen–
lithium exchange reaction of 8c with n-BuLi. As shown
in Table 1, the reaction in THF proceeded in good yield,
and the yield decreased in Et2O (entries 1–3). Use of
HMPA as the additive also decreased the yield (entry 4).
c
a
b
8b
7
8c
8a
Scheme 4. Reagents and conditions: (a) ClMe2SiCH2Cl, imidazole,
CH2Cl2, 1.5h, 96%; (b) ClMe2SiCH2Br, imidazole, CH2Cl2, 1.5h, 85%;
(c) NaI, acetone, reflux, 10h, 98%.
In these reactions (entries 1–4), a certain amount of
unidentified side products formed. These by-products
were surmised to arise from the side reactions, that is,
addition reaction of n-BuLi to 8c and reactions of the
lithium enolate of 8c generated by n-BuLi.10 Next,
t-BuLi was employed for this reaction because not only
is its halogen–lithium exchange reaction faster than that
of n-BuLi, but also no halogenated compounds remain,
and the reaction of bulky t-BuLi with the ketone would
be slow. As expected, the yield with t-BuLi increased to
86% (entry 6), and even the bromide 8b afforded the
desired product 10 in a comparable yield (84%, entry 5).
intramolecular nucleophilic additions to chiral ketones
would be a good solution to the stereoselective genera-
tion of the chiral tertiary alcohol. That is, as shown in
Scheme 4, the anion 9 generated from 8a to c by the halo-
gen–lithium exchange reaction could react intramolecu-
larly with the ketone in 9 to afford the tertiary alcohol
10.
Since the silicon-tethered C–C bond-forming reaction is
a useful method in terms of diastereo- and regioselecti-
vity, a variety of the tethered reactions have been
studied;5,6 however, only a limited number of C–C
bond-forming reactions of carbonyl compounds have
been reported.7 Although the allylation reactions7a,c
and the SmI2-mediated Barbier-type reactions7b have
been reported as the reactions of carbonyl groups, the
silicon-tethered nucleophilic addition has never been
reported as far as we know. Nevertheless, we surmised
the tethered reaction shown in Scheme 4 would proceed
in practical yield because the halogen–lithium exchange
reaction proceeds rapidly enough to permit the intra-
molecular C–C bond-forming reactions;8 furthermore,
the a-silyl carbanion is known to be stabilized by sili-
con.9 Hence, we prepared the silyl ethers 8b and 8c from
7, and investigated the silicon-tethered nucleophilic
addition. The silyl ethers 8a and 8b were prepared by
the reactions of 7 with ClSiMe2CH2Cl and ClS-
iMe2CH2Br, respectively, and the iodide 8c was easily
prepared from 8a by the halogen exchange reaction
(Scheme 4, 98%).
The halogen–lithium exchange reaction using Li powder
also proceeded at 0°C to afford 10, but the yield was
rather low (40%, entry 7). Lithium naphthalenide11
afforded no desired product (entry 8), but LiDBB (lithi-
unm 4,40-di-t-butylbiphenyl)11 gave 10 in 38% yield
(entry 9). In entries 7–9, a certain amount of unidentified
by-products formed, and the yields did not exceed that
of entry 6.
7b
The reaction under Kimꢀs conditions using SmI2 pro-
duced 10 in 50% yield with unidentified side products
(entry 10). Use of 4equiv of HMPA to SmI2 (entry
11), or use of SmBr2 (entry 12) did not improve the
yield.12
Since the desired product 10 was thus prepared in high
yield, 10 was transformed to 1a. As shown in Scheme
5, 10 was converted to the corresponding triol under

M. Iwamoto et al. / Tetrahedron Letters 45 (2004) 8647–8651
8649
acetylation (85%), MOM protection of 17 (90%),
deacetylation (96%), and Dess–Martin oxidation gave
1b (quant.).16
O
OH
f, g
d, e
a, b, c
10
O
O
O
O
This silicon-tethered reaction of some simple cyclic hydr-
oxyketones was examined, and it was found that only
b-hydroxyketones gave the desired products. As shown
in Scheme 7, simple b-hydroxyketones having no sub-
stituent at their a-position gave the desired products.10
Compared with the iodides 18b (42%) and 20b (68%),
yields by bromides 18a (34%) and 20a (22%) were low.
These results could be explained by the different rate
of the halogen–lithium exchange reaction between
iodide and bromide. These reactions in Scheme 7 also
suggest that the high yield of 10 (Table 1, entry 6) arose
from the structure of ketone 8c having dimethyl substit-
uents at the a-position of the ketone. The dimethyl sub-
stituents are so bulky that such side reactions as
addition of t-BuLi to the ketone and reactions of the
lithium enolate formed by t-BuLi would be suppressed.
11
12
OMPM
OMPM
h, i
j, k, l
1a
O
O
OAc
OH
13
14
Scheme 5. Reagents and conditions: (a) 30% H2O2, KHCO3, KF,
THF/MeOH, 10h ; (b) cat. CSA, acetone, 1h, 96% (two steps); (c)
Dess–Martin periodinane, CH2Cl2, 2h, 97%; (d) LDA, Eschenmoserꢀs
salt, THF, À78 to 0°C, 1h, then MeI, acetone, 10h, 64%; (e) NaBH4,
CeCl3Æ7H2O, MeOH, 5min, 96%; (f) SOCl2, Py, Et2O, 3h, 84%; (g)
MPMONa, TBAI, THF/DMF, 60°C, 5h, 92%; (h) cat. PTSA, MeOH/
H2O, 50°C, 3h, 98%; (i) Ac2O, Py, DMAP, CH2Cl2, 20min, 91%; (j)
MOMCl, DIPEA, NaI, CH2Cl2, 45h, 98%; (k) K2CO3, MeOH, 3h,
98% ; (l) Dess–Martin periodinane, CH2Cl2, 99%.
Tamaoꢀs conditions,13 and the crude triol was treated
with acetone under acidic condition to form the aceto-
nide selectively (96%, two steps). Subsequent Dess–Mar-
tin oxidation gave ketone 11 (97%), followed by
methylenation with Eschenmoserꢀs salt (64%) and then
the Luche reduction14 to afford 12 in 96% yield. Alcohol
12 was converted to the chloride by SOCl2 (84%), and
then reacted with sodium p-methoxyphenylmethoxide
to form 13 (92%). Deprotection of the acetonide
(98%), acetylation (91%), protection of the tertiary alco-
hol by the MOM group (98%), removal of the acetyl
group by K2CO3 in methanol (98%), and finally Dess–
Martin oxidation afforded 1a (99%).
We applied this silicon-tethered protocol to a-hydroxy-
methylketone 23, too (Scheme 8). Previously reported
221b was converted to 23 by MOM protection of the sec-
ondary alcohol and subsequent hydrogenation (95%,
two steps). The chloromethyldimethylsilyl ether17 of
Si
X
O
Si
O
t-BuLi
THF, -78 o
5 min
C
O
HO
19
y. 34% (from 18a)
y. 42% (from 18b)
18a (X = Br)
18b (X = I)
Another TaxolTM A-ring moiety 1b was also prepared as
shown in Scheme 6. The ketone 11 was methylated with
MeI and NaH to afford an inseparable mixture of 15
and 16. This mixture was reacted with hydrazine in a
sealed tube, followed by the reaction with I2 and DBU
to generate iodide 17 (54%, three steps).15 Treatment
of 17 with a catalytic amount of p-TsOH in methanol re-
moved the acetonide group (93%), and the following
Si
O
X
O
Si
t-BuLi
THF, -78 o
5 min
C
O
HO
21
20a (X = Br)
20b (X = I)
y. 22% (from 20a)
y. 68% (from 20b)
Scheme 7.
O
O
OR1
O
OR2
OMOM
a
11
+
O
Si
c, d
O
O
Ph
Ph
O
O
O
I
16
15
22 (R1 = Bn; R2 = H)
I
a, b
e
23 (R1 = H; R2 = MOM)
24
OMOM
d-h
b, c
OMOM
1b
O
O
Ph2Si
AcO
AcO
f, g
O
OH
OH
17
25
26
Scheme 6. Reagents and conditions: (a) MeI, NaH, THF/DMF, 19h;
(b) hydrazine, TEA, EtOH, sealed tube, 130°C, 4d; (c) I2, DBU, Et2O,
rt, 15min, then PhH, reflux, 3h, 54% (three steps); (d) cat. PTSA,
MeOH/H2O, 13h, 93%; (e) Ac2O, Py, DMAP, CH2Cl2, 20min, 85%;
(f) MOMCl, DIPEA, NaI, CH2Cl2, 45h, 90%; (g) K2CO3, MeOH, 3h,
96%; (h) Dess–Martin periodinane, CH2Cl2, quant.
Scheme 8. Reagents and conditions: (a) MOMCl, DIPEA, NaI,
CH2Cl2, reflux, 2d, 95%; (b) H2, Pd/C, MeOH, 10h, quant. (c)
ClPh2SiCH2Cl, imidazole, CH2Cl2, 1h, 74%; (d) NaI, acetone, reflux,
10h, 96%; (e) t-BuLi, À78°C, 5min, 64% ; (f) 30% H2O2, KHCO3, KF,
THF/MeOH, 10h; (g) Ac2O, DMAP, CH2Cl2, 10h, quant. (two steps).

8650
M. Iwamoto et al. / Tetrahedron Letters 45 (2004) 8647–8651
Utsugi, M.; Watanabe, H.; Umino, A.; Matsumura, T.;
Si
I
O
OH
O
Hagihara, T.; Takano, M.; Nakada, M. J. Am. Chem. Soc.
2003, 125, 2860–2861; (b) Iwamoto, M.; Kawada, H.;
Tanaka, T.; Nakada, M. Tetrahedron Lett. 2003, 44, 7239–
7243; Honma, M.; Nakada, M. Tetrahedron Lett. 2003,
44, 9007–9011.
a
b
Si
O
O
O
8c
27
28
OH
2. Addition of the organotitanium reagent from the more
hindered side: Reetz, M. T. In Organotitanium Reagents in
Organic Synthesis; Reactivity and Structure Concepts in
Organic Chemistry; Springer: Berlin Heidelberg, 1986;
Vol. 24, pp 123–124.
Scheme 9. (a) LDA, THF, À78 to 0°C, 1h, 80%; (b) 30% H2O2,
KHCO3, KF, THF/MeOH, rt, 3.5h.
3. Nakada, M.; Kojima, E.; Iwata, Y. Tetrahedron Lett.
1998, 39, 313–316.
4. Mori, K.; Mori, H. Tetrahedron 1985, 41, 5487–
5493.
5. Reviews of silicon-tethered reactions: (a) Bols, M.;
Skrydstrup, T. Chem. Rev. 1995, 95, 1253–1277; (b)
Fensterbank, L.; Malacria, M.; Sieburth, S. McN. Syn-
thesis 1997, 813–854.
6. References in the review of the oxidation of the carbon–
silicon bond, see: Jones, G. R.; Landais, Y. Tetrahedron
1996, 52, 7599–7662.
7. (a) Reetz, M. T. Pure Appl. Chem. 1985, 57, 1781–1788;
(b) Park, H. S.; Lee, I. S.; Kwon, D. W.; Kim, Y. H.
Chem. Commun. 1998, 2745–2746; (c) Robertson, J.; Hall,
M. J.; Green, S. P. Org. Biomol. Chem. 2003, 1, 3635–
3638.
8. (a) Cooke, M. P., Jr.; Houpis, I. N. Tetrahedron Lett.
1985, 26, 4987–4990; (b) Kihara, M.; Kashimoto, M.;
Kobayashi, Y. Tetrahedron 1992, 48, 67–78.
9. Panek, J. S. In Comprehensive Organic Synthesis; Trost, B.
M., Ed.; Pergamon, 1991; Vol. 1, pp 579–627.
10. Dimers derived from the starting material were not found
in the products.
the primary alcohol 23 was found to be too unstable to
purify by silica gel chromatography; hence, a more sta-
ble chloromethyldiphenylsilyl ether was prepared (74%),
and the silyl ether was converted to iodide 24 (96%).
Reaction of 24 with t-BuLi afforded 25 in 64% yield.10
This rather low yield could arise from the bulky phenyl
groups of the silyl ether; that is, the large phenyl groups
was surmised to slow down not only the halogen–lith-
ium exchange reaction but also the addition reaction,
resulting in the decrease of the yield. Tamao oxidation
of 25 proceeded smoothly, and subsequent acetylation
afforded 26 quantitatively (two steps).18
It should be noted that we have also found another sili-
con-tethered anionic reaction of 8c (Scheme 9). Thus,
treatment of 8c with LDA instead of t-BuLi caused
intramolecular alkylation to afford 27 in 80% yield,10
and the following Tamao oxidation gave 28 (71%). Since
the intramolecular reaction is fast and the C–Si bond is
rather long, this reaction would not suffer from the steric
hinderance around the silicon atom, giving the product
in high yield. This tethered reaction is expected to be a
general method for the regio- and stereoselective instal-
lation of a hydroxymethyl group at the a-position of
ketones in cyclic as well as acyclic hydroxyketones;
hence, its scope and limitations are now under
investigation.
11. Freeman, P. K.; Hutchinson, L. L. J. Org. Chem. 1980, 45,
1924–1930.
12. About inner-sphere versus outer-sphere electron transfer
in reactions of Sm(II) reductants: Miller, R. S.; Sealy, J.
M.; Shabangi, M.; Kuhlman, M. L.; Fuchs, J. R.; Flow-
ers, R. A., II J. Am. Chem. Soc. 2000, 122, 7718–
7722.
13. (a) Tamao, K.; Ishida, N.; Kumada, M. J. Org. Chem.
1983, 48, 2120–2122; (b) Tamao, K.; Ishida, N.; Tanaka,
T.; Kumada, M. Organometallics 1983, 2, 1694–1696; (c)
Tamao, K. J. Synth. Org. Chem. Jpn. 1988, 48, 861–
878.
14. (a) Luche, J. L. J. Am. Chem. Soc. 1978, 100, 2226–2227;
(b) Gemal, A. L.; Luche, J. L. J. Am. Chem. Soc. 1981,
103, 5454–5459.
In summary, we have developed intramolecular silicon-
tethered nucleophilic additions of a hydroxymethyl unit
to ketones in cyclic b-hydroxyketones, and the products
obtained by this protocol were successfully converted to
chiral A-ring moieties of TaxolTM. Also found was the
promising diastereoselective silicon-tethered a-alkyla-
tion of the cyclic hydroxyketone. To our knowledge,
these silicon-tethered anionic reactions have not been
reported; therefore, now our attention is focused on
their further development and applications including
the studies in an acyclic system. The further develop-
ment will be reported in due course.
15. (a) Barton, D. H. R.; Bashiardes, G.; Fourrey, J.-L.
Tetrahedron 1988, 44, 147–162; (b) Grandi, M. J. D.; Jung,
D. K.; Krol, W. J.; Danishefsky, S. J. J. Org. Chem. 1993,
58, 4989–4992.
27
16. Absolute structure of 1b (½aꢀD À76.2 (c 0.98, CHCl3)) was
elucidated by the reaction mechanism and comparison of
its [a]D value with that of the same compound as prepared
below.
Acknowledgements
This work was financially supported in part by Waseda
University Grant for Special Research Projects. We are
also indebted to 21COE ꢁPractical Nano-Chemistryꢀ.
I
I
I
a, b
c, d
O
I
O
OH
I
OEE
References and notes
e, f
g, h
[?]D24 +63.2
(c 0.98, CHCl3)
O
MOMO
1. Our contribution to constructing a stereogenic quaternary
carbon, see: (a) Honma, M.; Sawada, T.; Fujisawa, Y.;
MOMO
83% ee
OEE

M. Iwamoto et al. / Tetrahedron Letters 45 (2004) 8647–8651
8651
(a) Me3SI, NaH, DMSO/THF, 19h, 85%; (b) 2,2,6,6-
tetramethylpiperidine, n-BuLi, Et2AlCl, toluene, 0°C, 6h,
67%; (c)TBHP, L-(+)-DET, Ti(O–i-Pr)4, MS 4 A, CH2Cl2,
DIPEA, NaI, CH2Cl2, 15h, quant.; (g) cat. PPTS, MeOH,
1.5h, 80%; (h) Dess–Martin periodinane, CH2Cl2, 1h, 96%.
17. Allen, J. M.; Aprahamian, S. L.; Sans, E. A.; Shechter, H.
J. Org. Chem. 2002, 67, 3561–3574.
˚
À20°C, 1h, 90%, 83% ee; (d) EVE, cat. PPTS, CH2Cl2,
20min, 96%; (e) LiAlH4, Et2O, 20min, 91%; (f) MOMCl,
18. Structure of 26 was elucidated by the NOE experiments.