Synthetic study of aquayamycin. Part 3: First total synthesis

Article abstract of DOI:10.1016/S0040-4039(00)01480-5

The first total synthesis of aquayamycin (1) has been accomplished. The crucial steps include (1) the Hauser reaction between 3-(phenylsulfonyl)phthalide 3 and cyclohexenone 4 to make up the linear BCD tricycle, and (2) the intramolecular pinacol coupling

Full text of DOI:10.1016/S0040-4039(00)01480-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 8393–8396
Synthetic study of aquayamycin. Part 3: First total synthesis
Takashi Matsumoto, Hiroki Yamaguchi, Mitsujiro Tanabe, Yoshizumi Yasui and
Keisuke Suzuki*
Department of Chemistry, Tokyo Institute of Technology, and CREST,
Japan Science and Technology Corporation (JST), O-okayama, Tokyo 152-8551, Japan
Received 3 August 2000; accepted 28 August 2000
Abstract
The first total synthesis of aquayamycin (1) has been accomplished. The crucial steps include (1) the
Hauser reaction between 3-(phenylsulfonyl)phthalide 3 and cyclohexenone 4 to make up the linear BCD
tricycle, and (2) the intramolecular pinacol coupling of keto aldehyde 7 to the full tetracyclic framework.
© 2000 Elsevier Science Ltd. All rights reserved.
We report here the first total synthesis of aquayamycin (1),1 the first angucycline antibiotic
with a C-glycoside structure.2–4 The main challenge of the synthesis stemmed from the
characteristic, highly oxygenated AB ring system, which is prone to rearrange or aromatize
under acidic, basic and also photo-irradiation conditions.1
Our synthetic plan relied on constructing the unstable A ring in the last stage by the pinacol
cyclization of the tricyclic keto aldehyde 2 followed by unsaturation at C(5)ꢀC(6). Compound 2,
in turn, would be accessible via Hauser reaction5 with the C-olivosylated phthalide 36 and the
cyclohexenone 4 (Scheme 1).7
O
1
cyclization
O
OH
MeO
O
HO
A
12
12a
CHO
OH
D
C
B
O
O
O
O
9
OBn
HO
HO
BnO
BnO
5
6a
7
HO
O
BnO
OMe
elimination
O
aquayamycin (1)
2
O
OTBDPS
+
O
O
O
OBn
BnO
BnO
O
SO2Ph
BnO
3
4
Scheme 1.
* Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01480-5

8394
The enone 4 was added to a solution of the lithio anion of the phthalide 3 at −78°C. Upon
warm up to 45°C, a clean reaction occurred to give the unstable hydroquinone 5, which was
immediately protected with (MeO)2SO2 and K2CO3 to give the dimethyl ether 6 in 73% yield
(Scheme 2).
O
RO
O
O
OTBDPS
OTBDPS
a)
+
O
O
O
O
O
O
OBn
BnO
BnO
OBn
Sugar
SO2Ph
BnO
OR
BnO
3
4
5: R = H
6: R = Me
(73%, 2 steps)
b) K2CO3,
(MeO)2SO2
HO
MeO
MeO
O
OBn
HO
e) VCl3(thf)3
Zn, DMF
c) HF•(pyr)n (91%)
CHO
O
O
O
O
OBn
d) Swern oxidation
(90%)
Sugar
Sugar
89%
BnO
OMe
BnO
OMe
MeO
7
8
O
O
MeO
HO
OBn
OH
HO
g) 5% H2SO4 aq.
f) Swern Oxidation
90%
(96%)
O
O
OH
O
HO
HO
h) MsCl, DMAP
(quant.)
i) H2, 10% Pd/C
Sugar
OMs
BnO
OMe
HO
OMe
9
10
O
O
OH
MeO
OH
O
HO
HO
j) BnBr,
Cs2CO3
k) (NH4)2Ce(NO3)6
OH
OMs
OH
OMs
O
O
HO
HO
HO
HO
79%, 2 steps
BnO
OMe
RO
O
12: R = Bn
13: R = H
11
l) H2, 5% Pd/C
then air
O
O
OH
HO
m) (i-Pr)2NEt
58%, 3 steps
O
OH
O
Sugar =
BnO
BnO
HO
HO
HO
O
aquayamycin (1)
Scheme 2. (a) 3, t-BuOLi, THF, −78°C, 30 min; then 4 at −78°C; warm up to 45°C, for 1 h. (b) Acetone, reflux, 5
h. (c) THF, rt, 7 h. (d) (COCl)2, DMSO, CH2Cl2, −78°C; then Et3N, warm up to 0°C. (e) CH2Cl2, rt, 30 min. (f)
(COCl)2, DMSO, CH2Cl2, −78°C; then Et3N, warm up to 0°C. (g) 1,4-Dioxane, 80°C, 2 h. (h) Pyridine, CH2Cl2, rt,
1 h. (i) EtOAc, MeOH, rt, 9.5 h. (j) DMF, 0°C, 1.5 h. (k) Aq. MeCN, 0°C, 3 min. (l) See text. (m) 1,4-Dioxane, 45°C,
1.5 h
Removal of the TBDPS group in 6 by HF·(pyr)n followed by Swern oxidation gave the keto
aldehyde 7, which was then subjected to an intramolecular pinacol coupling for the A ring
cyclization. Among various methods attempted, the Pedersen procedure by utilizing VCl3·(thf)3
and Zn8 effected clean cyclization to give the pinacol 8 as a mixture of the C(1)-epimers
(8a:8b=13:1), which converged to a single ketone 9 by Swern oxidation in high yield.

8395
Evidence for the cis stereochemistry of the A/B ring junction was provided by separation of
the epimers of 8 followed by forming the bis-acetonides 14a (from the major isomer) and 14b
(from the minor isomer) [2-methoxypropene, cat. p-TsOH, benzene] (Fig. 1)—if trans, one of the
isomers, 8c, would fail to give the corresponding bis-acetonide because of the topological reason.
The NOE experiments on the bis-acetal 14a proved that the orientation of the C(1) hydroxy
group is a in the major epimer of 8, i.e. 8a. The cis selectivity for the AB ring junction can be
attributed to the CꢀC bond formation on the convex face of the cis-fused 5–6 ring system of the
dioxolane and the B ring.
<the acetonides of AB-cis isomers of 8>
<AB-trans isomers of 8>
6.48 ppm
n.O.e
CH3 1.41 ppm
OBn
H
O
O
O
O
HO
HO
HO
OBn
OBn
HO
OBn
1
1
4a
4a
12b
12b
O
O
O
O
O
O
O
in C6D6
O
8c
8d
14a
(from the major isomer 8a)
14b
(from the minor isomer 8b)
Figure 1.
The ketone 9 was converted to the hexol 10 in high yield via three steps: (1) removal of the
acetonide, (2) selective mesylation of the C(5) hydroxy group and (3) hydrogenolysis of all four
benzyl ethers. Attempted oxidation of the hexol 10 with CAN to the corresponding para-
quinone was unsuccessful under various conditions, giving many unidentified products. Fortu-
nately, however, clean oxidation could be effected when the C(8) phenol was selectively
re-protected by a benzyl group by using Cs2CO3.9
O
O
O
OBn
HO
HO
OBn
HO
OH
OH
OMs
O
O
HO
HO
HO
HO
5
6
BnO
O
HO
OH
15
16
The quinone 12 underwent elimination of a methanesulfonic acid during the chromatographic
purification on silica gel to give compound 15, i.e. the benzyl ether of aquayamycin. Unfortu-
nately, however, all attempts to convert 15 into 1 failed so long as the reductive debenzylation
was employed, because the C(5)ꢀC(6) double bond was concomitantly saturated.10 After
considerable experimentation, a solution to this issue was provided by the following sequence of
reactions. The quinone 12, without purification, was subjected to a quick catalytic hydrogenoly-
sis (1 min) over 5% Pd/C in EtOAc, which effected reduction of the C ring to the corresponding
hydroquinone and removal of the benzyl group. The resulting hydroquinone 16 underwent
spontaneous oxidation upon exposure to air to give the quinone 13, which was treated with
i-Pr2NEt in 1,4-dioxane at 45°C (1.5 h) to effect clean elimination of a methanesulfonic acid to
give aquayamycin (1) in high yield. The synthetic material proved to be identical with the
1
natural product in all respects (mp, [h]D, IR, H and 13C NMR).11
In summary, the first total synthesis of aquayamycin (1), the prototypical and synthetically
most challenging member among the angucycline antibiotics, was achieved in 13 steps (21%
overall yield) from the phthalide 3 and the cyclohexenone 4.

8396
Acknowledgements
We thank Dr. Shinichi Kondo, Institute of Microbial Chemistry, for providing us with the
valuable sample of aquayamycin. Financial support from Nissan Science Foundation is deeply
acknowledged.
References
1. (a) Sezaki, M.; Hara, T.; Ayukawa, S.; Takeuchi, T.; Okami, Y.; Hamada, M.; Nagatsu, T.; Umezawa, H. J.
Antibiot. 1968, 21, 91. Nagatsu, T.; Ayukawa, S.; Umezawa, H. J. Antibiot. 1968, 21, 91. (b) Sezaki, M.; Kondo,
S.; Maeda, K.; Umezawa, H.; Ohno, M. Tetrahedron 1970, 26, 5171. Also, see: (c) Imamura, N.; Kakinuma, K.;
Ikekawa, N.; Tanaka, H.; Omura, S. Chem. Pharm. Bull. 1981, 29, 1788.
2. For reviews on the angucycline antibiotics, see: Rohr, J.; Thiericke, R. Nat. Prod. Rep. 1992, 103. Krohn, K.;
Rohr, J. Top. Curr. Chem. 1997, 188, 127.
3. (a) Total syntheses of angucyclines possessing the C-glycoside (urdamycinone B): Yamaguchi, M.; Okuma, T.;
Horiguchi, A.; Ikeura, C.; Minami, T. J. Org. Chem. 1992, 57, 1647. Sulikowski, G. A.; Boyd, V. A. J. Am.
Chem. Soc. 1995, 117, 8472. Matsuo, G.; Miki, Y.; Nakata, M.; Matsumura, S.; Tashima, K. Chem. Commun.
1996, 224. (C104): Matsumoto, T.; Sohma, T.; Yamaguchi, H.; Kurata, S.; Suzuki, K. Synlett 1995, 263.
Matsumoto, T.; Sohma, T.; Yamaguchi, H.; Kurata, S.; Suzuki, K. Tetrahedron 1995, 51, 7347. (b) Selected
examples of total synthesis of the angucyclines without the C-glycoside (emycin, ochromycinone): Larsen, D. S.;
O’Shea, M. D. Brooker, S. Chem. Commun. 1996, 203. (MM 47755, X-14881) Krohn, K.; Khanbabaee, K.;
Micheel, J. Liebigs Ann. Chem. 1995, 1529. (SF 2315B): Kim, K.; Sulikowski, G. A. Angew. Chem., Int. Ed. Engl.
1995, 34, 2396.
4. For synthetic studies on the aquayamycin-type skeleton, see: Kraus, G. A.; Wan, Z. Tetrahedron Lett. 1997, 38,
6509. Krohn, K.; Micheel, J. Tetrahedron, 1998, 54, 4827.
5. Hauser, F. M.; Rhee, R. P. J. Org. Chem. 1978, 43, 178. Review: Mitchell, A. S.; Russell, R. A. Tetrahedron 1995,
51, 5207. For an example of applying the Hauser reaction to aryl C-glycoside synthesis, see: Tatsuta, K.; Ozeki,
H.; Yamaguchi, M.; Tanaka, M.; Okui, T. Tetrahedron Lett. 1990, 31, 5495.
6. As the protecting group for the phenol we employed benzyl instead of methyl, because the preliminary study
showed that removal of the methyl protecting group was not possible. The phthalide 3 was prepared by a similar
method described in the first paper in this series: Matsumoto, T.; Yamaguchi, H.; Hamura, T.; Tanabe, M.;
Kuriyama, Y.; Suzuki, K. Tetrahedron Lett. 2000, 41, 8383.
7. Yamaguchi, H.; Konegawa, T.; Tanabe, M.; Nakamura, T.; Matsumoto, T.; Suzuki, K. Tetrahedron Lett. 2000,
41, 8389.
8. Takahara, P. M.; Freudenberger, J. H.; Konradi, A. W.; Pedersen, S. F. Tetrahedron Lett. 1989, 30, 7177.
Freudenberger, J. H.; Konradi, A. W.; Pedersen, S. F. J. Am. Chem. Soc. 1989, 111, 8014.
9. The corresponding 8-acetoxy or 8-benzoyloxy derivatives were recovered unchanged under similar conditions.
10. The product, 5,6-dihydroaquayamycin, exhibited spectroscopic data identical to those reported: Henkel, T.;
Ciesiolka, T.; Rohr, J.; Zeeck, A. J. Antibiot. 1989, 42, 299.
11. Mp 197–198°C (dec.) (benzene), [h]2D4 +140 (c 0.11, 1,4-dioxane) [Ref.1c 175–180°C (hexane–benzene), [h]D20 +130
(c 1, 1,4-dioxane)]. Anal. calcd for C25H26O10: C, 61.72; H, 5.39. Found: C, 61.43; H, 5.49.
.