First stereoselective total synthesis of FD-594 aglycon

Full text of DOI:10.1002/anie.200806338

Communications
DOI: 10.1002/anie.200806338
Natural Product Synthesis
First Stereoselective Total Synthesis of FD-594 Aglycon**
Ritsuki Masuo, Ken Ohmori, Lukas Hintermann, Saki Yoshida, and Keisuke Suzuki*
In memory of Katsumi Kakinuma
FD-594 (1) is an antitumor antibiotic isolated from Strepto-
myces sp. TA-0256.[1] The unique structure characterized by
the densely functionalized, curved hexacyclic core having a
trisaccharide unit was elucidated by Kakinuma and co-
workers, wherein an interesting stereochemical behavior,
solvent-dependent atropisomerism, was identified to possibly
have biological relevance.[2] Intrigued by the significant
bioactivities as well as the challenging structural motifs, we
embarked on the synthetic study. Herein, we report the first
total synthesis of the FD-594 aglycon (2).
Scheme 1 outlines our retrosynthetic analysis based on the
chirality-transfer strategy;[3] the chiral centers in diol 2 could
be derived from the pinacol cyclization of axially chiral biaryl
dialdehyde I,[4] which could be related to biaryl lactone III,
given that the axial chirality was established at the stage of
biaryl II by the Bringmann-type asymmetric cleavage with a
chiral nucleophile (Nu*; III!II).[3,5] Furthermore, disconnec-
tion of lactone III, derived by using a Pd-catalyzed cycliza-
tion,[6] suggested ester IV as the precursor, which could then
be dissected into xanthone V and iodophenol VI.
Scheme 2 shows the synthesis of the AB-ring fragment 9,
which began with a three-step conversion of vanillin (3) into
bromide 4.[7] Halogen–lithium exchange of 4 (nBuLi, ꢀ788C)
and subsequent reaction with (R)-propyloxirane (10)[8] at
ꢀ788C in the presence of BF3·OEt2 afforded alcohol 5.[9] The
selective removal of the MOM group in 5 was achieved by
heating the mixture in 1,3-propanediol (1408C, 7 min),[10] and
then carbonylation via triflate 6, prepared by the careful
monotriflation, cleanly afforded lactone 7 in excellent yield.
Demethylation of 7 and subsequent hydrolysis of the dioxane
acetal and reduction of the resulting aldehyde gave alcohol 8.
Regioselective iodination of 8 and protection of the primary
alcohol as a TIPS ether gave the AB-ring fragment 9.
Scheme 1. Synthetic plan.
Scheme 3 illustrates synthesis of the DEF-ring fragment
17, starting from diester 11.[11] Upon treatment with NCS
(1 equiv) in AcOH, diester 11 was cleanly oxidized to the
[*] R. Masuo, Dr. K. Ohmori, Dr. L. Hintermann,[+] S. Yoshida,
Prof. Dr. K. Suzuki
corresponding hydroquinone diester. One of the esters was
reduced using NaBH4 in wet THF, and then acetalization gave
phenol 12. The regioselective bromination of phenol 12 using
pyridinium bromide perbromide (PyHBr3) and a subsequent
four-step conversion (methylation, acetal cleavage, oxidation,
and benzylation) afforded aldehyde 13 in high overall yield.
The F-ring fragment 18 was lithiated and reacted with
aldehyde 13 to afford adduct 14 in high yield. Alcohol 14
was converted into the cyclization precursor 15 by IBX
oxidation[12] and then the MOM group was removed.
Department of Chemistry, Tokyo Institute of Technology
SORST-JST Agency
2-12-1, O-okayama, Meguro-ku, Tokyo 152-8551 (Japan)
Fax: (+81)3-5734-2788
E-mail: ksuzuki@chem.titech.ac.jp
[+] Present address: Institut fꢀr Organische Chemie
RWTH Aachen (Germany)
[**] This work was partially supported by Global COE program
(Chemistry) and a Grant-in-Aid for Scientific Research (JSPS). We
are grateful to Daiso Co. Ltd., Osaka, for the generous supply of (R)-
(ꢀ)-epichlorohydrin.
At the stage of the key SNAr cyclization by using Cs2CO3,
the chemoselectivity was highly dependent on the solvent.[13]
In MeOH, a 1:1 mixture of two cyclized products, 16 and 19,
was obtained in quantitative yield,[14] whereas the use of DMF
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200806338.
3462
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3462 –3465

Angewandte
Chemie
Scheme 2. Synthesis of the AB-ring fragment 9. Unless otherwise
noted, the reactions were performed at ambient temperature. a) Br2,
MeOH, 1 h (94%). b) 1,3-propanediol, 1 mol% Bu4N+Br3ꢀ, HC(OEt)3,
25 min. c) MOMCl, NaH, DMF, 15 min (2 steps, 94%). d) nBuLi
(1.1 mol equiv), Et2O, ꢀ788C, 35 min. 10 (1.2 mol equiv), BF3·OEt2
(1.3 mol equiv), ꢀ788C, 80 min, then 268C, 30 min (68%). e) 1,3-
propanediol, 1408C, 7 min (91%). f) PhNTf2 (1.2 mol equiv), K2CO3
(1.1 mol equiv), DMF, 15 h (82%). g) CO (1 atm), Pd(OAc)2
(10 mol%), dppp (10 mol%), Et3N (2 mol equiv), DMF, 1008C, 4 h
(91%). h) BCl3, CH2Cl2, 08C, 15 min. i) 0.5m H2SO4, 1,4-dioxane,
1008C, 5.5 h (2 steps, 99%). j) NaBH4, THF/MeOH (10:1), ꢀ788C,
1.5 h (99%). k) BnMe3N+ICl2ꢀ, NaHCO3, CH2Cl2, MeOH, ꢀ108C, 57 h
(85%). l) TIPSCl, imidazole, DMF, 4 h (quant.). MOM=methoxy-
methyl, DMF=N,N-dimethylformamide, Tf=trifluoromethanesulfonyl,
dppp=1,3-bis(diphenylphosphino)propane, Bn=benzyl, TIPS=triiso-
propylsilyl.
Scheme 3. Synthesis of the DEF-ring fragment 17. Unless otherwise
noted, reactions were performed at ambient temperature. a) NCS
(1.0 mol equiv), AcOH, 808C, 1 h (89%). b) NaBH4, THF/H2O (9:1),
1 h. c) 2,2-dimethoxypropane, TsOH, acetone, 1 h. d) PyHBr3, pyridine,
1 h. e) (MeO)2SO2, K2CO3, DMF, 31 h. f) 10% H2SO4 aq., 1,4-dioxane,
508C, 2.5 h. g) MnO2, EtOAc, 30 min (6 steps, 66%). h) BnBr, K2CO3,
DMF, 1.5 h (99%). i) 18 (1.1 mol equiv), nBuLi (1.1 mol equiv), tolu-
ene, 08C, 1.5 h; 13, THF, ꢀ78!ꢀ508C, 30 min (85%). j) IBX, DMSO,
3 h (quant.). k) 0.7m H2SO4 aq., 1,4-dioxane, 908C, 1 h (97%).
l) Cs2CO3, cyclohexane, reflux, 81 h (84%). m) LiOH, H2O, 1,4-dioxane,
2 h (99%). NCS=N-chlorosuccinimide, Ts=p-toluenesulfonyl, IBX=
o-iodoxybenzoic acid, DMSO=dimethyl sulfoxide.
the lactone moieties. This conversion was achieved by N-
methylation and subsequent treatment with NaBH(OMe)3.
The resulting N,O-acetal was hydrolyzed with acid to give
aldehyde 24.[17] The desilylation of 24 and oxidation of the
resulting alcohol gave dialdehyde 25 in high yield, ready for
the key pinacol cyclization.[18]
However, we were disappointed by the poor results for
this key step; SmI2 in THF at 08C gave the desired product
only in low yield (ca. 20%) and poor stereoselectivity (trans/
cis = 3:1). After many unproductive trials, we became con-
vinced that the difficulties mainly originated from the
presence of the xanthone moiety, and decided to convert
the xanthone into the corresponding xanthene temporarily.
Thus, xanthone 24 was reduced by a two-step process [l-
Selectride and NaBH3(CN)] to give xanthene 26, and removal
of the silyl group and subsequent oxidation of the resulting
diol gave dialdehyde 27.
Pleasingly, xanthene 27 behaved nicely in the pinacol
cyclization (Scheme 5); upon treatment with SmI2, the
reaction proceeded far more smoothly than the case of
xanthone 25, albeit the trans/cis selectivity remained low
(Table 1, entry 1). Additional screening revealed that addi-
tives could improve this situation.[19] Whereas the addition of
tetraglyme did not affect the stereoselectivity (Table 1,
entry 2), various crown ethers gave much improved stereose-
lectivities (Table 1, entries 3–6). Furthermore, pybox ligands
were effective for improving the yield and the stereoselectiv-
ity (Table 1, entries 7 and 8).
slightly improved the selectivity (16/19 = 3:1). Additional
screening showed that cyclohexane was the solvent of choice,
giving the desired xanthone 16 in high selectively (16/19 =
23:1), which was easily isolated by re-precipitation (CHCl3/
petroleum ether= 1:3). The saponification of methyl ester 16
gave the DEF-fragment 17.
Two fragments, 9 and 17, were combined via the acid
chloride, giving the corresponding ester quantitatively
(Scheme 4). After the removal of the benzyl group to give
20, the palladium-catalyzed cyclization gave hexacycle 21 in
high yield.[15] For inducing the axial chirality, biaryl lactone 21
was subjected to the asymmetric ring-opening using (S)-
valinol as a chiral nucleophile.[5] The stereoselectivity proved
to be highly dependent on the solvent, and THF gave the best
result, affording diastereomeric amides 22 and 22ꢀ in a 93%
combined yield with excellent selectivity (14:1);[16] the
diastereomers were easily separated by using silica gel
column chromatography (n-hexane/EtOAc = 2:1). After sep-
aration, the major isomer (22) was converted into dialdehyde
25: treatment of 22 with PPh3 and I2 afforded a mixture of the
corresponding iodide and oxazoline, which, without separa-
tion, was treated with BnBr and Cs2CO3 in DMF, wherein the
two phenols were benzylated and the cyclization to the
oxazoline was complete, giving oxazoline 23. The next stage
was the selective conversion of the oxazoline into the
corresponding aldehyde without altering the xanthone and
Angew. Chem. Int. Ed. 2009, 48, 3462 –3465
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3463

Communications
Scheme 4. Pinacol cyclization precursors 25 and 27. Unless otherwise noted, reactions performed at ambient temperature. a) 17, (COCl)2, DMF,
CH2Cl2, 0.5 h; 9 (1.1 mol equiv), DMAP, pyridine, 1 h (quant.). b) BBr3, CH2Cl2, ꢀ58C, 15 min (91%). c) [Pd2(dba)3]·CHCl3 (15 mol%), tBuCO2Na
(3 mol equiv), DMA, 608C, 40 min. d) (S)-valinol (3.2 mol equiv), THF, 268C, 20 min (2 steps, 93%, 14:1 d.r.); separation by silica gel column
chromatography (87% yield for isolated 22, 6.4% yield for isolated 22’) e) I2, PPh3, imidazole, CH2Cl2, 30 min. f) BnBr, Cs2CO3, DMF, 408C, 2 h (2
steps, 97%). g) MeOTf, 2,6-di-tert-butylpyridine, CH2Cl2, 2 h. h) NaBH(OMe)3, ꢀ78!ꢀ208C, 30 min. i) sat. citric acid aq. soln., THF, 0.5 h (76%
from 23). j) nBu4NF, THF, 08C, 0.5 h (92%). k) Dess–Martin periodinane, CH2Cl2, 0.5 h (87%). l) l-Selectride, THF, ꢀ78!ꢀ258C, 20 min.
m) NaBH3(CN), AcOH, CH2Cl2, 0.5 h (2 steps, 96%). n) nBu4NF, THF, 15 min (94%). o) (COCl)2, DMSO, CH2Cl2, ꢀ788C, 50 min; Et3N, ꢀ788C,
1 h (90%). DMAP=4-N,N-dimethylaminopyridine, dba=dibenzylideneacetone, DMA=N,N-dimethylacetamide, l-Selectride=lithium tri-sec-
butylborohydride.
Table 1: Pinacol cyclization of xanthene dialdehyde 27.[a]
Entry
Additive
trans [%]
cis [%]
1
2
3
4
5
6
7
8
none
53
51[b]
63
55
61
66
72
71
15
16[b]
10
17
7
8
10
10
tetraglyme
[12]crown-4
[15]crown-5
[18]crown-6
[24]crown-8
(S,S)-iPr-pybox
(R,R)-iPr-pybox
[a] Three molar equivalents of SmI2 and six molar equivalents of the
additive were used. [b] The yield was assessed after acetylation. (S,S)-iPr-
pybox=2,6-bis[(4S)-(ꢀ)-isopropyl-2-oxazolin-2-yl]pyridine.
With diol 28 in hand, the final stages included two
synthetic hurdles; 1) the regeneration of the xanthone and
2) the removal of the methylene protecting group for the
catechol without touching the methyl and the benzyl groups.
We were pleased to find that the first problem was solved by
using DDQ. Thus, after masking the diol in 28 by acetylation,
treatment with DDQ afforded xanthone 29 in quantitative
yield.
As for the second issue, the methylene acetal in 29 was
oxidatively removed by treatment of 29 with Pb(OAc)4
allowing the clean formation of acetoxy acetal 30, which
was smoothly hydrolyzed in acidic methanol to provide the
Scheme 5. Final stages for the synthesis of 2. Unless otherwise noted,
reactions were performed at ambient temperature. a) Ac2O, DMAP,
pyridine, 20 min (98%). b) DDQ, CH2Cl2, 1,4-dioxane, H2O, 11 h
(quant.). c) Pb(OAc)4, benzene, reflux, 1 h (94%). d) TsOH, MeOH,
H2O, 10 h (99%). e) K2CO3, MeOH, 0.5 h (87%). f) Pd(OH)2/C, H2
(1 atm), MeOH, 0.5 h (quant.). DDQ=2,3-dichloro-5,6-dicyano-1,4-
benzoquinone.
3464 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 3462 –3465

Angewandte
Chemie
6313 – 6316; c) T. Hosoya, E. Takashiro, T. Matsumoto, K.
desired catechol 31. Finally, the removal of the two acetyl
groups in 31 and the subsequent catalytic hydrogenolysis of
the two benzyl groups furnished the FD-594 aglycon (2) as a
yellow powder, which was identical with the authentic sample
Suzuki, J. Am. Chem. Soc. 1994, 116, 1004 – 1015.
[7] R. Gopinath, S. J. Haque, B. K. Patel, J. Org. Chem. 2002, 67,
5842 – 5845.
[8] Prepared from (R)-(ꢀ)-epichlorohydrin by a two-step sequence
[1) EtMgBr, CuI, THF; 2) NaOH (55% over 2 steps)]; a) C.
Crause, F. R. van Heerden, S. Afr. J. Chem. 1998, 51, 35;
b) C. A. G. M. Weijers, A. L. Botes, M. S. van Dyk, J. A. M.
de Bont, Tetrahedron: Asymmetry 1998, 9, 467 – 473.
[9] M. Yamaguchi, I. Hirao, Tetrahedron Lett. 1983, 24, 391 – 394.
[10] H. Miyake, T. Tsumura, M. Sasaki, Tetrahedron Lett. 2004, 45,
7213 – 7215.
by direct comparison; the rotation was ½aꢁ2D6 = + 5.4 ꢀ 102 (c =
26
0.43, MeOH), and that of the authentic sample was ½aꢁD
=
+ 5.4 ꢀ 102 (c = 0.35, MeOH), and the melting point of the
synthetic sample was 209–2128C, compared to 207–2108C for
the authentic sample.[1b,20]
In conclusion, the first total synthesis of FD-594 aglycon
(2) was achieved. Currently, we are studying the glycosyla-
tion, which is directed at the total synthesis of the natural
product.
[11] a) A. B. Padias, H. K. Hall, Jr., J. Org. Chem. 1985, 50, 5417 –
5419; b) F. V. Bagrov, D. F. Bagrov, Russ. J. Org. Chem. 1994, 30,
637 – 639.
[12] J. D. More, N. S. Finney, Org. Lett. 2002, 4, 3001 – 3003.
[13] L. Hintermann, R. Masuo, K. Suzuki, Org. Lett. 2008, 10, 4859 –
4862.
Received: December 27, 2008
Published online: April 3, 2009
[14] The product ratio (16/19) was assessed by 1H NMR analaysis
(CDCl3, 300 MHz).
Keywords: antibiotics · atropisomerism ·
.
[15] We also attempted the cyclization before removal of the benzyl
group. However, the reaction produced only many unidentified
by-products.
[16] The same reaction with (R)-valinol gave two diastereomers in
high selectivity (18:1). After separation, the major isomer of the
(R)-valinol adducts was converted by the same synthetic
sequence described in the text. The final product was the 6,7-
bis(epimer) of 2, implying that the stereochemical course of the
lactone cleavage is mainly decided by the reagent control (by the
chirality of the nucleophile), but not by the substrate control (by
natural product synthesis · samarium · total synthesis
[1] a) Y. Qiao, T. Okazaki, T. Ando, K. Mizoue, K. Kondo, T.
Eguchi, K. Kakinuma, J. Antibiot. 1998, 51, 282 – 287; b) K.
Kondo, T. Eguchi, K. Kakinuma, K. Mizoue, Y. Qiao, J. Antibiot.
1998, 51, 288 – 295.
[2] T. Eguchi, K. Kondo, K. Kakinuma, H. Uekusa, Y. Ohashi, K.
Mizoue, Y. Qiao, J. Org. Chem. 1999, 64, 5371 – 5376. The
absolute and relative stereochemistries of 1 were determined by
X-ray analysis.
[3] a) K. Ohmori, M. Tamiya, M. Kitamura, H. Kato, M. Oorui, K.
Suzuki, Angew. Chem. 2005, 117, 3939 – 3942; Angew. Chem. Int.
Ed. 2005, 44, 3871 – 3874; b) M. Tamiya, K. Ohmori, M.
Kitamura, H. Kato, T. Arai, M. Oorui, K. Suzuki, Chem. Eur.
J. 2007, 13, 9791 – 9823.
[4] a) K. Ohmori, M. Kitamura, K. Suzuki, Angew. Chem. 1999, 111,
1304 – 1307; Angew. Chem. Int. Ed. 1999, 38, 1226 – 1229; b) M.
Kitamura, K. Ohmori, T. Kawase, K. Suzuki, Angew. Chem.
1999, 111, 1308 – 1311; Angew. Chem. Int. Ed. 1999, 38, 1229 –
1232.
[5] a) G. Bringmann, M. Breuning, S. Tasler, Synthesis 1999, 525 –
558; b) G. Bringmann, A. J. P. Mortimer, P. A. Keller, M. J.
Gresser, J. Garner, M. Breuning, Angew. Chem. 2005, 117, 5518 –
5563; Angew. Chem. Int. Ed. 2005, 44, 5384 – 5427, and
references therein.
the chirality in lactone 21). For
a related paper, see S.
Masamune, S. A. Ali, D. L. Snitman, D. S. Garvey, Angew.
Chem. 1980, 92, 573 – 575; Angew. Chem. Int. Ed. Engl. 1980, 19
557 – 558. For the details see the Supporting Information.
[17] a) A. I. Meyers, M. Shipman, J. Org. Chem. 1991, 56, 7098 – 7102;
b) S. Boisnard, L. Neuville, M. Bois-Choussy, J. Zhu, Org. Lett.
2000, 2, 2459 – 2462.
[18] For preparation of SmI2 in THF, see: P. Girard, J. L. Namy, H. B.
Kagan, J. Am. Soc. Chem. 1980, 102, 2693 – 2698.
[19] Recently, Greeves et al. reported notable ligand effects on dl/
meso selectivity of the SmI2-mediated pinacol coupling of
benzaldehyde. See, H. C. Aspinall, N. Greeves, C. Valla, Org.
Lett. 2005, 7, 1919 – 1922.
[20] All physical data (1H and 13C NMR , IR, elemental analysis, [a]D,
and m.p.) of the synthetic material were to be consistent with an
authentic sample of the natural product (see, Ref. [1b]), kindly
provided by Prof. Dr. Tadashi Eguchi, Tokyo Institute of
Technology.
[6] a) G. Bringmann, J. R. Jansen, H.-P. Rink, Angew. Chem. 1986,
98, 917 – 919; Angew. Chem. Int. Ed. Engl. 1986, 25, 913 – 915;
b) P. P. Deshpande, O. R. Martin, Tetrahedron Lett. 1990, 31,
Angew. Chem. Int. Ed. 2009, 48, 3462 –3465
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3465