A practical synthesis of (-)-kazusamycin A (revised)

Article abstract of DOI:10.1016/j.tetlet.2005.07.029

We describe herein a stereo-controlled and practical synthesis of three key building blocks, namely Segment AB′, Segment D, and Segment E′ needed for the total synthesis of (-)-kazusamycin A.

Full text of DOI:10.1016/j.tetlet.2005.07.029

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 6341–6344
A practical synthesis of (ꢀ)-kazusamycin A
Shengfeng Zhou,a Huaxiang Chen,a Wensheng Liao,a,* Shu-Hui Chen,a Ge Li,a
Ryoichi Andob and Isao Kuwajimac
aWuXi PharmaTech Co., Ltd, No. 1 Building, 288 FuTe ZhongLu, WaiGaoQiao Free Trade Zone, Shanghai 200131, PR China
bMitsubishi Pharma Corporation, 1000 Kamoshida-cho, Aoba-ku, Yokohama 227-0033, Japan
cThe Kitasato Institute, 1-15-1 Kitasato, Sagamihara 228-8555, Japan
Received 19 May 2005; revised 6 July 2005; accepted 8 July 2005
Available online 25 July 2005
Abstract—We describe herein a stereo-controlled and practical synthesis of three key building blocks, namely Segment AB0, Segment
D, and Segment E0 needed for the total synthesis of (ꢀ)-kazusamycin A.
Ó 2005 Elsevier Ltd. All rights reserved.
1. Introduction
2. Synthesis of Segment A (building block needed for AB0)
Kazusamycin A, a natural product isolated from the
culture broth of an actinomycete strain 81-484,1 has
demonstrated potent anti-tumor activity against a num-
ber of tumor cell lines such as P388 leukemia, Hela cells,
sarcoma 180, as well as antimicrobial activity.1,2
Kazusamycin A has also exhibited potent inhibitory
activity against Rev protein translocation from the
nucleus to the cytoplasm, suggesting its potential use as
anti-HIV agent.3 The absolute structure of kazusamycin
A was deduced initially from NMR spectroscopic anal-
ysis, and later was confirmed by KuwajimaÕs first elegant
total synthesis.4 Prompted by the promising biological
activity, structural complexity exhibited by kazusamycin
A, and its limited availability from natural source, we
became interested in designing a practical synthetic
route for this natural product.5
The preparation of Segment A was begun with O-benzyl-
ation of inexpensive geraniol. The resulting product 2
was subjected to epoxidation with MCPBA to give 3,
which was transformed into the acid intermediate 5 via
HIO4 mediated epoxide opening followed by the subse-
quent diol cleavage6 and further oxidation with Jones
reagent in an overall yield of 62% from 1. Compound
57 was converted to amide 7 via its mix-anhydride inter-
mediate in 81% yield. EvanÕs chiral auxiliary induced
asymmetric methylation on 7 yielded 8 (72%), which
was converted to Segment A via LiBH4 mediated amide
reduction and final Swern oxidation of the resulting
alcohol 94 with an overall yield of 91%. Compared with
KuwajimaÕs initial route for Segment A, the newly
devised synthetic sequence is more economical (using
geraniol as the starting material), safer (avoiding the
use of dangerous free radical reaction on large scale),
and more amenable for scale-up (75 g of Segment A
made) (Scheme 2).
In this letter, we describe the large-scale preparation of
three key building blocks, namely Segment AB0, Seg-
ment D, and Segment E0. According to the retrosynthetic
analysis outlined in Scheme 1, the availability of these
intermediates should make gram-scale preparation of
kazusamycin A and subsequent extensive biological
and toxicological evaluation of this complex molecule
possible.
3. Synthesis of Segment B (building block needed
for AB0)8
As shown in Scheme 3, saturation of enol ether 10 with
ethanol led to acetal ester 11, which was reduced with
LAH to give diol 12, thereafter the bis-benzylether 13
with 58% overall yield from 10. Treatment of acetal 13
with TFA afforded aldehyde 14, which was reacted with
ethyl Grignard reagent to provide the desired adduct 15
*
Corresponding author. Tel.: +86 21 50461111; fax: +86 21
50463718; e-mail: liao_wensheng@pharmatechs.com
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.07.029

6342
S. Zhou et al. / Tetrahedron Letters 46 (2005) 6341–6344
OH
O
OTBS O
O
HO
R3O
O
OH
O
Segment AB'
(-)-Kazusamycin A
X
O
+
O
O
H
OH
O
R3O
R3O
H
OR2
OR1
Segment C
O
Segment AB
Segment A
+
O
O
H
O
O
R4O
CO2Me
O
+
OR2
O
OR1
Segment B
X
Segment E'
OR5
Segment D
Segment E
Scheme 1. Retrosynthetic analysis for kazusamycin A.
O
O
BnO
BnO
v
RO
ii
iii
X
O
4
5
X = H
X = OH
iv
Li
1
2
R = H
R = Bn
i
3
N
O
O
O
Bn
6
vii
viii
BnO
BnO
CHO
BnO
N
OH
O
R
7
8
R = H
R = Me
9
vi
Segment A
Bn
Scheme 2. Synthesis of Segment A: (i) NaH, BnBr; (ii) MCPBA; (iii) HIO4; (iv) Jones oxidation, 62% (four steps); (v) t-BuC(O)Cl, 6, 81%; (vi)
NaHMDS, MeI, 72%; (vii) LiBH4, 91%; (viii) Swern oxidation, 92%.
O
OEt
OH
CO2Et
CO2Et
v
iv
ii
R
i
EtO
H
OBn
OBn
EtO
OBn
OBn
R
10
11 R = CO2Et
12 R = CH2OH
13 R = CH2OBn
14
15
iii
O
O
O
vi
x
ix
OPMB
OR
OPMB
OTBDPS
OR'
OH
(R)-18
Segment B
16 R = R' = Bn
17 R = R' = H
vii
viii
18 R = PMB, R' = H
Scheme 3. Synthesis of Segment B: (i) cat. EtONa, EtOH; (ii) LAH; (iii) NaH, BnBr, 58% (three steps); (iv) TFA; (v) EtMgBr, 60–80% (two steps);
(vi) Jones oxidation, 69%; (vii) Pd/C, H2, (viii) CSA, Cl3CC@NH(OPMB), 71% (two steps); (ix) Lipase AK, vinyl acetate, 22%; (x) TBDPSCl,
imidazole, DMF, 95%.
in 80% overall yield from 13. Jones reagent mediated
oxidation of 15 led to the keto-derivative 16 (69%),
which was then subjected to de-benzylation and subse-
quent mono-PMB protection to give the mono alcohol
18 (71% from 16). Treatment of 18 with lipase AK
afforded chiral alcohol (R)-18 (22%),8 which was further
converted to its silyl ester Segment B (95%). It is
worthwhile to mention that we streamlined the synthetic

S. Zhou et al. / Tetrahedron Letters 46 (2005) 6341–6344
6343
operation by performing the first three-step in a
5. Synthesis of Segment D
continuous fashion. In order to be cost efficient, we
replaced Dess–Martin periodinane used by Kuwajima
and co-workers8 with inexpensive Jones reagent
(from 15 to 16) and carried out this sequence at 1 kg
scale.
PharmaTechÕs approach toward the preparation of Seg-
ment D began with mono-silylation of 1,4-butanediol 32
to obtain mono-alcohol 33 (87%), which was oxidized
with PDC to provide acid 34 (61%), followed by treat-
ment with ClC(O)C(O)Cl to give acid chloride 35.9
Reaction of 35 with the lithiated Evans chiral auxiliary
36 provided the desired adduct 37 in an overall yield
of 72% from 34. Asymmetric C-methylation on 37
yielded 38 with high level of diastereoselectivity. Com-
pound 38 was converted to alcohol 3910 and then alde-
hyde 4011 via LiBH4 reduction and subsequent Swern
oxidation. This aldehyde was subjected to AndoÕs cis-
olefination12 to afford 4113 (78% for two steps), which
was treated with DIBAL-H to produce the desired Seg-
ment D in 92% yield. Compared with the literature route
for Segment D, the synthetic route employed at Pharma-
Tech was more amenable for scale-up due to relatively
short sequence and readily available starting material.
Furthermore, the more expensive Dess–Martin reagent
was replaced by Swern oxidation with similar yield
(Scheme 5).
4. Synthesis of Segment AB04
Condensation between Segment A and Segment B was
promoted by Sn(OTf)2 and afforded the adduct AB
(80%) in a stereoselective fashion. The keto moiety in
Segment AB was reduced selectively by NaBH4/Et2-
BOMe to provide the syn-diol 20, which was further
desilylated to give the triol 21. Acetonide protection
of 21 yielded the mono-alcohol 22, which was further
transformed into its TBS ether 23. The overall yield
was 60% (from Segment AB to 23). It should be pointed
out that we conducted this four-step sequence at 100 g
scale in a continuous fashion without purifying the
intermediates involved. The PMB moiety in 23 was then
removed by DDQ, and the resulting hydroxy group in
24 (65%) was oxidized under Swern conditions (instead
of expensive Dess–Martin periodinane used previously)4
to provide the corresponding aldehyde 25. Reaction of
25 with the Wittig reagent 26 gave the unsaturated ester
27 (80% from 24). Compound 27 was subjected to DI-
BAL-H reduction to obtain alcohol 28 (94%), thereafter
the bis-silylated derivative 29 (95%). Removal of the
benzyl moiety was accomplished via sodium in liquid
ammonia and provided allylic alcohol 30 in variable
yields. This intermediate was further converted to
Segment AB0 via O-acylation and followed by subse-
quent O-desilylation with an overall yield of 90%
(Scheme 4).
6. Synthesis of Segment E0
As highlighted in Scheme 6, N- and O-acylation of ami-
no alcohol 42 yielded 43 in excellent yield. The following
titanium enolate mediated aldol condensation between
43 and BnOCH2CHO provided the syn-adduct 4414 with
high enantioselectivity. The synthetic approach used
herein for establishing absolute stereochemistry (two
newly formed chiral centers) was different from that
used by Kuwajima and co-workers.4 In that case,
the Evans chiral auxiliary was employed as the
OH OH
OH
O
ii
i
BnO
BnO
CHO
BnO
OPMB
OPMB
O
OR
iii
OTBDPS
OPMB
20 R = TBDPS
21 R = H
Segment A
AB
Segment B
OTBDPS
iv
BnO
vii
viii
TBSO
O
O
TBSO
O
O
O
OR
O
O
BnO
COOEt
BnO
Ph3P
OR'
26
27
25
COOEt
22 R = H, R' = PMB
23 R = TBS, R' = PMB
24 R = TBS, R' = H
v
vi
TBSO
O
O
TBSO
O
O
xii
R'O
ix
R'O
28 R = H, R' = Bn
31 R = TIPS, R' = Alloc
AB' R = H, R' = Alloc
OR
x
OR
xiii
29 R = TIPS, R' = Bn
30 R = TIPS, R' = H
xi
Scheme 4. Synthesis of Segment AB0: (i) Sn(OTf)2, triethylamine, 80%; (ii) Et2BOMe, NaBH4; (iii) TBAF, THF; (iv) Me2C(OMe)2, PPTS; (v)
TBSOTf, 2,6-lutidine, 60% (four steps); (vi) DDQ, 65%; (vii) Swern; (viii) 26, toluene, 80% (two steps); (ix) DIBAL-H, 94%; (x) TIPSCl, imidazole,
DMF, 95%; (xi) Na/NH3, 33–80%; (xii) AllocCl, pyridine; (xiii) TBAF, 90% (two steps).

6344
S. Zhou et al. / Tetrahedron Letters 46 (2005) 6341–6344
O
O
O
iv
ii
TBDPSO
RO
TBDPSO
N
O
O
OH
R
34 R = OH
35 R = Cl
Li
37 R = H
38 R = Me
R
32 R = H
33 R =TBDPS
iii
v
i
N
O
Bn
36
Bn
TBDPSO
TBDPSO
TBDPSO
R
ix
vi
viii
Me
Me
CO2Et
OH
Me
vii
Segment D
39 R = CH2OH
40 R = CHO
41
Scheme 5. Synthesis of Segment D: (i) n-BuLi, TBDPSCl, 87%; (ii) PDC, DMF, 61%; (iii) (COCl)2; (iv) 36, 72% (two steps); (v) NaHMDS, MeI,
74%; (vi) LiBH4, 95%; (vii) Swern; (viii) (o-MeOPhO)2P(O)CH(Et)CO2Et, NaH, 78% (two steps); (ix) DIBAL-H, 92%.
O
O
OH
O
R
NHR'
ii
i
iii
iv
O
NHTs
BnO
NHTs
O
O
O
Bn
42 R = R' = H
43 R = C(O)Et, R' = Ts
Me
Bn
Me
Bn
45
44
OHC
O
CO2Me
O
vi
v
O
O
O
O
Br
O
OR
OH
Org. Lett. 2004, 6, 2845
Me
Segment E'
Me
Br
Me
46
47
Segment E
Scheme 6. Synthesis of Segment E0: (i) TsCl, Et3N, then EtC(O)Cl, Et3N, >90%; (ii) TiCl4, DIEA, BnOCH2CHO; (iii) H2, Pd/C, then Me2C(OMe)2;
(iv) LAH, 80% (four steps); (v) Swern, then CBr4, PPh3, 50% (two steps); (vi) n-BuLi, ClCO2Me, 81%.
chirality-inducing source. With compound 44 in hand,
the O-Bn moiety of which was removed by standard
hydrogenation, and the resulting diol was subjected to
acetonide protection to yield 45. LAH mediated ester
reduction on 45 provided alcohol 4615 in 80% overall
yield (four steps from 43). Alcohol 46 was oxidized under
Swern conditions to provide its corresponding aldehyde
intermediate, which was further treated with triphenyl-
phospine and carbon tetrabromide to afford the a,a-di-
bromo-olefin 47 in 50% overall yield. Final treatment
of 47 with n-BuLi and chloromethylformate thus pro-
vided the desired alkyne derivative Segment E0. It should
be pointed out that further conversion of Segment E0 to
References and notes
1. Umezawa, I.; Komiyama, K.; Oka, H.; Okada, K.;
Tomisaka, S.; Miyano, T.; Takano, S. J. Antibiot. 1984,
37, 706–711.
2. Komiyama, K.; Okada, K.; Hirokawa, Y.; Masuda, K.;
Tomisaka, S.; Umezawa, I. J. Antibiot. 1985, 38, 224–229.
3. Wang, Y.; Ponelle, M.; Sanglier, J.-J.; Wolff, B. Helv.
Chim. Acta 1997, 80, 2157–2167.
4. Arai, N.; Chikaraishi, N.; Omura, S.; Kuwajima, I. Org.
Lett. 2004, 6, 2845–2848.
5. Zhou, S-F.; Chen, H-X.; Liao, W.; Chen, S. H.; Li, G.;
Ando, R.; Kuwajima, I. Presented at the Poster Section in
Organic Chemistry Division, 13–18 March 2005, San
Diego, CA. Abst.#ORGN-576.
Segment
E
was reported by Kuwajima and
co-workers.4
6. Chen, K.-M.; Semple, J. E.; Joullie, M. M. J. Org. Chem.
1985, 50, 3997–4005.
7. Ple, P. A.; Hamon, A.; Jones, G. Tetrahedron 1997, 53,
3395–3400.
8. Arai, N.; Chikaraishi, N.; Ikawa, M.; Omura, S.; Kuwa-
jima, I. Tetrahedron: Asymmetry 2004, 15, 733–741.
9. Tori, M.; Toyoda, N.; Sono, M. J. Org. Chem. 1998, 63,
306–313.
In summary, we have accomplished practical, large-scale
preparations of Segment AB0, Segment D, and Segment
E0, needed for total synthesis of kazusamycin A. All of
the three building blocks were prepared via newly
designed synthetic sequences with the intention to avoid
using expensive starting materials, hazardous reagents,
and labor-intensive chromatographic separation.
10. Aiguade, J.; Hao, J.-L.; Forsyth, C. J. Tetrahedron Lett.
2001, 42, 817–820.
11. Lautens, M.; Stammers, T. A. Synthesis 2002, 1993–2012.
12. Ando, K. J. Org. Chem. 1997, 62, 1934–1939.
13. Vicario, J. L.; Job, A.; Wolberg, M.; Muller, M.; Enders,
D. Org. Lett. 2002, 4, 1023–1026.
Acknowledgements
14. Ghosh, A. K.; Kim, J.-H. Tetrahedron Lett. 2002, 43,
5621–5624.
15. Ojika, M.; Kigoshi, H.; Ishigaki, T.; Nisiwaki, M.;
Tsukada, I.; Mizuta, K.; Yamada, K. Tetrahedron Lett.
1993, 34, 8505–8508.
We thank Mr. Jianyi Ma, Chonggang Liu, Shengbin Li,
and Xia Li for their help with synthetic work. We are
also indebted to the chemists from Analytic Department
at WuXi PharmaTech for their support.