Regio- and stereocontrolled total synthesis of benanomicin B

Article abstract of DOI:10.1002/anie.200501210

(Chemical Equation Presented) Fully controlled total synthesis of benanomicin B was achieved by exploiting two key steps : a stereocontrolled ring-opening of a lactone (see scheme, step A), and a semipinacol cyclization of an acetal-aldehyde derivative discriminating the two hydroxy groups of the pseudo-C₂-symmetric 1,2-diol moiety (step B).

Full text of DOI:10.1002/anie.200501210

Angewandte
Chemie
Total Synthesis
Two additional issues must be addressed en route to the
fully controlled total synthesis (Figure 1): a) introduction of
the sugar moiety into the aglycon in a regioselective manner,
and b) stereoselective access to the chiral, nonracemic biaryl
Regio- and Stereocontrolled Total Synthesis of
Benanomicin B**
Ken Ohmori, Minoru Tamiya, Mitsuru Kitamura,
Hirohisa Kato, Mami Oorui, and Keisuke Suzuki*
In memory of Tsuguo Mizuochi
Benanomicin/pradimicin antibiotics (BPAs) constitute a class
of natural products with a unique structure consisting of a
benzo[a]naphthacene core, an amino acid, and a disacchar-
ide.[1] Their potent antifungal and anti-HIV activity is
attributed to the specific, Ca2+-mediated binding of BPAs to
the mannose-rich oligosaccharides presented on the fungi or
virus surfaces.[2] Stimulated by these intriguing biological
properties and structural novelty, we initiated a synthesis
study of BPAs.
Figure 1. Tactical issues for the total synthesis of BPAs. a) regioselec-
tive introduction of the sugar moiety and b) stereocontrol of the axial
chirality.
dicarbaldehyde (M)-3, as the key cyclization precursor.
Conjugation of the disaccharide was an intriguing challenge,
because of the local pseudo-C2 symmetry of the 1,2-diol on
the B ring. Although a subsidiary issue, the enantioselective
synthesis of (M)-3 was a serious bottleneck in the sequence of
our synthesis. Herein, we describe clean solutions to these
problems, which have culminated in the first total synthesis of
benanomicin B (1).
In a previous synthesis study we reported the stereo-
selective synthesis of the common aglycon benanomicinone
(pradimicinone, 2)[3] by exploiting the complete transfer of
Scheme 1 features three key points in our strategy;
1) construction of sterically encumbered biaryl structure II
the axial chirality into the central chiralities in the pinacol
cyclization of biaryl dicarbaldehyde (M)-3 into the trans-1,2-
diol (S,S)-4 [Eq. (1)].
Scheme 1. Overview of our synthesis strategy.
ꢀ
by formation of an internal C C bond in the tethered
ꢀ
derivative I; 2) The asymmetric cleavage of the C O tether in
II by employing a chiral agent to give the axially chiral biaryl
III; and 3) Reconnection at two carbonyl groups to give the
tricyclic product IV. A special requirement for the third stage
was the capability of an axial-to-central chirality transfer with
concomitant discrimination of the two oxy functions.
[*] Dr. K. Ohmori, M. Tamiya, Dr. M. Kitamura, Dr. H. Kato, M. Oorui,
Prof. Dr. K. Suzuki
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
The first task was to prepare optically pure 8 as a key
precursor for the pinacol cyclization (Scheme 2). We applied
an asymmetric ring opening of biaryl lactones pioneered by
Bringmann et al. for introducing the biaryl chirality.[4] The
substrate for this purpose, the biaryl lactone 7, was prepared
[**] We thank Meiji Seika Kaisha Ltd. for providing an authentic sample
of benanomicins A and B. This work was partially supported by the
21st Century COE program (Tokyo Institute of Technology).
Angew. Chem. Int. Ed. 2005, 44, 3871 –3874
DOI: 10.1002/anie.200501210
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3871

Communications
for which we planned the semipinacol cyclization of acetal-
aldehyde V [Eq. (3)]. Activation of the acetal with Lewis acid
followed by a net two-electron reduction would hopefully
give the mono-masked diol VI.[13]
With this strategy in mind, we envisioned the acetal-
aldehyde (M)-15 as the key substrate, which was readily
available from the lactone ring-opening product 11
(Scheme 3). After selective benzylation of two of the phenol
Scheme 2. Preparation and the ring-opening reaction of 7. a) 2,4,6-tri-
chlorobenzoyl chloride, Et3N, DMAP, toluene, 258C, 20 min (99%).
b) CF3CO2H, CH2Cl2, 258C, 4 h (96%). c) Pd(OAc)2 (30 mol%), PPh3
(60 mol%), tBuCO2Na (3 equiv), DMA, 1108C, 20 min (60%). d) l-
valine (3.3 equiv), NaBH4 (3.0 equiv), THF, (8%, 2% ee). DMAP=4-
N,N-dimethylaminopyridine, DMA=N,N-dimethylacetamide.
in high yield by the Yamaguchi esterification[5] of carboxylic
acid 5[3b] with phenol 6[6] followed by the Pd-mediated
cyclization.[3b,7]
Initial attempts were centered at the enantioselective
reduction of 7 with NaBH4 by using stoichiometric amounts of
various chiral modifiers,[8–10] all of which resulted in disap-
pointing enantioselectivities. Unexpectedly, however, we
obtained an important hint when l-valine was employed:
thus, although the reaction itself was not very effective (8%
yield and 2% ee), a sizable amount of a highly polar by-
product was noted, which was proved to be the amide 9—the
direct adduct of the amino acid. An important observation
was that amide 9 was highly enriched in one of the
diastereomers (90:10 d.r.).
Encouraged by this finding, we decided to pursue a
diastereoselective ring-opening reaction, and optimization
experiments showed that d-valinol (10) was an excellent
chiral nucleophile, which afforded amide alcohol 11 in 90%
yield with 91:9 diastereoselectivity [Eq. (2)].[11] Purification of
Scheme 3. a) BnBr, Cs2CO3, DMF, 08C!RT, 70 min (97%). b) I2, PPh3,
imidazole, CH2Cl2, 08C, 20 min (quant.). c) MeOTf, 2,6-di-tert-butylpyri-
dine, CH2Cl2, RT, 1 h. d) l-Selectride, 08C, 20 min; SiO2, RT, 17 h (96%
from 13). e) BnOTMS, TMSOTf, toluene, ꢀ158C, 6 h. f) nBu4NF, THF,
08C, 1 h. g) MnO2, CH2Cl2, RT, 16 h (94% from 14). TIPS=triisopro-
pylsilyl; Bn=benzyl, Tf=trifluoromethanesulfonyl, l-selectride=
lithium tri-sec-butylborohydride, TMS=trimethylsilyl.
groups in 11, the hydroxyamide moiety in the product 12 was
converted into an oxazoline functionality by treatment with
PPh3 and I2 to give 13 in high yield. After N-methylation of 13,
reduction and exposure to silica gel gave the corresponding
aldehyde 14 cleanly in 96% yield.[14] Conversion of 14 into the
corresponding dibenzyl acetal[15] followed by detachment of
the silyl group and oxidation afforded the enantiopure acetal-
aldehyde (M)-15 for the key cyclization.
The initial attempts at the semipinacol cyclization of (M)-
15 by using SmI2 and BF3·OEt2 suffered from not only modest
yields, but also the lack of reproducibility (Table 1).[16] The
yields widely fluctuated in the range 10–50%, even by a slight
change in reaction parameters (temperature, reaction time,
amounts of reducing agent, and Lewis acid). After extensive
effort, we finally found that the presence of a suitable proton
source was the key to secure the reproducibility. A clean
reaction occurred in the presence of H2O to give semipinacol
the major diastereomeric product by column chromatography
(silica gel, hexane/EtOAc = 1/2) gave single crystals that were
suitable for X-ray crystallographic analysis.[12] This analysis
proved the compound had the requisite M configuration
around the biaryl axis.
Now the stage was set for the pivotal stage: construction
of the B-ring diol with discrimination of the two oxy functions,
3872
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 3871 –3874

Angewandte
Chemie
Table 1: Semipinacol cyclization of acetal-aldehyde (M)-15.[a]
was obtained in a fully regiocontrolled manner, thus com-
pleting the full carbon framework. Removal of the four acyl
groups and hydrolysis of the methyl ester moiety in 22 was
achieved by treatment with 1m NaOH to give 23. Final
hydrogenolysis of 23 in the presence of aqueous HCl gave
benanomicin B hydrochloride (1·HCl), which was fully iden-
tical with an authentic sample.[21]
In conclusion, the first total synthesis of benanomicin B
(1) was achieved based on a chiral transmission strategy.
Stereoselective incorporation of valinol into lactone 7 effec-
tively gave the axially chiral product 11. The semipinacol
cyclization of acetal-aldehyde (M)-15 successfully provided
the semiprotected trans-diol (S,S)-16, which allowed the
selective introduction of the sugar moiety, thus enabling
access to 1.
Entry
Additive[b]
Yield [%]
trans/cis
ee [%]
1
2
3
none
H2O
MeOH
0–50
82
95
>99/<1
>99/<1
>99/<1
>99
>99
>99
[a] 2.5 equiv of SmI2, 3.0 equiv of BF3·OEt2. [b] 1.0 equiv.
16 in 82% yield; furthermore, methanol proved to be even
more effective (95% yield). The reaction proceeded with
stereochemical integrity, namely it was trans-selective and
with complete chiral transmission, and (M)-15 (> 99% ee)
was cleanly converted into (S,S)-16 (> 99% ee).[17]
Having finally overcome two major synthetic hurdles, the
stage was set for the introduction of the amino acid and the
sugar moieties (Scheme 4). Thus, tetracycle 16 was saponified,
and condensed with d-alanine methyl ester by using the
benzotriazole derivative BOP to give the corresponding
amide 17 in excellent yield. The metallocene-based activator
[Cp2HfCl2]/2AgOTf,[18] allowed the coupling of glycosyl
fluoride 18[19] with the acceptor 17 to give the b-glycoside 19
in 72% yield along with its a-anomer in 9% yield.
Received: April 6, 2005
Published online: May 18, 2005
Keywords: antibiotics · atropisomerism · biaryls · samarium ·
.
total synthesis
[1] a) T. Oki, M. Konishi, K. Tomatsu, K. Tomita, K. Saitoh, M.
Tsunakawa, M. Nishio, T. Miyaki, H. Kawaguchi, J. Antibiot.
1988, 41, 1701 – 1704; b) T. Takeuchi, T. Hara, H. Naganawa, M.
Hamada, H. Umezawa, S. Gomi, M. Sezaki, S. Kondo, J.
Antibiot. 1988, 41, 807 – 811.
[2] a) M. Watanabe, S. Gomi, H. Tohyama, K. Ohtsuka, S. Shiba-
hara, S. Inouye, H. Kobayashi, S. Suzuki, S. Kondo, T. Takeuchi,
H. Yamaguchi, J. Antibiot. 1996, 49, 366 – 373; b) T. Ueki, K.
Numata, Y. Sawada, T. Nakajima, Y. Fukagawa, T. Oki, J.
Antibiot. 1993, 46, 149 – 161; c) T. Mizuochi, M. Nakata, Jpn. J.
Clin. Med. 1995, 53, 2340 – 2349.
After oxidation of 19 by Ce(NH4)2(NO3)6 (CAN), the
resulting chloroquinone 20 was subjected to a Diels–Alder
reaction with siloxydiene 21,[3b,20] from which pentacycle 22
Scheme 4. The final stage toward 1. a) 5m NaOH, EtOH, in sealed tube, 1008C, 2 h. b) d-Ala-OMe·HCl, BOP, Et3N, DMF, RT, 1 h (84% from 16).
c) 18, [Cp2HfCl2]/2AgOTf, 4-ꢀ MS, CH2Cl2, ꢀ788C, 20 min then ꢀ358C, 11 h (72% for the b anomer 19, 9% for the a anomer). d) Ce(NH4)2-
(NO3)6, CH3CN, H2O, 08C, 5 min. e) 21, THF, RT, 2 h; SiO2, RT, 12 h; K2CO3, CH2Cl2, THF, 3 h (74% from 19). f) 5m NaOH, MeOH, 258C, 9 h.
g) H2, Pd/C, MeOH, 1m HCl, DMF, 2 h (53% from 22). d-Ala-OMe=d-alanine methyl ester, DMAP=4-N,N-dimethylaminopyridine, DMA=N,N-
dimethylacetamide, BOP=benzotriazol-1-yloxytris(dimethylamino)phosphonium hexafluorophosphate, Tf=trifluoromethanesulfonyl, MS=molec-
ular sieves, Bz=benzoyl.
Angew. Chem. Int. Ed. 2005, 44, 3871 –3874
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3873

Communications
[3] a) K. Ohmori, M. Kitamura, K. Suzuki, Angew. Chem. 1999, 111,
iPrOH 8/2, flow rate 0.5 mLminꢀ1
S,S isomer, 23.1 min for the R,R isomer).
[18] a) K. Suzuki, H. Maeta, T. Matsumoto, G. Tsuchihashi, Tetrahe-
dron Lett. 1988, 29, 3571 – 3574; b) K. Suzuki, H. Maeta, T.
Matsumoto, Tetrahedron Lett. 1989, 30, 4853 – 4856.
[19] H. Kato, K. Ohmori, K. Suzuki, Tetrahedron Lett. 2000, 41,
6827 – 6832.
[20] J. Savard, P. Brassard, Tetrahedron 1984, 40, 3455 – 3464.
[21] Although the 1H NMR spectra of free benanomicin B (1) are
highly dependent on the pH value and temperature, the mixed
sample of synthetic/authentic benanomicin B hydrochloride
showed fully coincided 1H NMR spectra. Furthermore, all
physical data (1H and 13C NMR, IR, HRMS (FAB), [a]D, and
m.p.) of the synthetic material were fully identical with those of
1·HCl.[1b]
, tR = 14.2 min for the
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.
[4] For enantioselective reactions, see a) G. Bringmann, T. Hartung,
Angew. Chem. 1992, 104, 782 – 783; Angew. Chem. Int. Ed. Engl.
1992, 31, 761 – 762; b) G. Bringmann, R. Walter, R. Weirich,
Tetrahedron 1993, 49, 7891 – 7902; c) G. Bringmann, M. Breun-
ing, Synlett 1998, 634 – 636; for diastereoselective reactions, see
d) Angew. Chem. 1989, 101, 1725 – 1726; Angew. Chem. Int. Ed.
Engl. 1989, 28, 1672 – 1673; e) G. Bringmann, M. Breuning, R.
Walter, A. Wuzik, K. Peters, E.-M. Peters, Eur. J. Org. Chem.
1999, 3047 – 3055; f) G. Bringmann, M. Breuning, S. Tasler, H.
Endress, C. L. J. Ewers, L. Gꢀbel, K. Peters, E.-M. Peters, Chem.
Eur. J. 1999, 5, 3029 – 3038; for a review, see g) G. Bringmann, M.
Breuning, S. Tasler, Synthesis 1990, 102, 1006 – 1019; Angew.
Chem. Int. Ed. Engl. 1990, 29, 977 – 991, and references therein.
[5] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989 – 1993.
[6] Prepared from the known orsellinic acid derivative, namely,
methyl 4-hydroxymethyl-2-methoxymethoxy-6-methylbenzoa-
te,[3b] by a three-step sequence [1) iPr3SiCl, imidazole, DMF;
2) TFA, CH2Cl2; 3) BnMe3NCl·ICl, NaHCO3, MeOH, CH2Cl2
(83% over 3 steps)].
[7] a) P. P. Deshpande, O. R. Martin, Tetrahedron Lett. 1990, 31,
6313 – 6316; b) T. Hosoya, E. Takashiro, T. Matsumoto, K.
Suzuki, J. Am. Chem. Soc. 1994, 116, 1004 – 1015.
[8] Initial attempts at the enantioselective ring opening of 7 were
centered on the borane activated by chiral oxazaborolidines,[9]
which uniformly led to poor enantioselectivity. Attempts were
also made to use NaBH4 with stoichiometric amounts of various
chiral compounds, such as l-(+)-tartaric acid diethyl ester,[10a]
N,N-dibenzoylcycteine,[10b] and (S)-(ꢀ)-1,1’-binaphthol.[10c]
[9] a) E. J. Corey, C. J. Helal, Angew. Chem. 1998, 110, 2092 – 2118;
Angew. Chem. Int. Ed. 1998, 37, 1986 – 2012, and references
therein.
[10] a) M. Yatagai, T. Ohnuki, J. Chem. Soc. Perkin Trans. 1 1990,
1826 – 1828; b) K. Soai, T. Yamanoi, H. Oyamada, Chem. Lett.
1984, 251 – 254; c) R. Noyori, I. Tomino, Y. Tanimoto, M.
Nishizawa, J. Am. Chem. Soc. 1984, 106, 6709 – 6716.
1
[11] The diastereomer ratio was assessed by H NMR spectroscopic
analysis (CDCl3, 400 MHz).
[12] The major diastereomer of 11 gave a single crystal (EtOAc/
hexane) suitable for X-ray analysis, which confirmed the stereo-
structure of 11. CCDC-262629 contains the supplementary
crystallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or
from the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB21EZ, UK; fax: (+ 44)1223-336-033; or
deposit@ccdc.cam.ac.uk).
[13] K. Ohmori, M. Kitamura, Y. Ishikawa, H. Kato, M. Oorui, K.
Suzuki, Tetrahedron Lett. 2002, 43, 7023 – 7026.
[14] 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.
[15] T. Tsunoda, M. Suzuki, R. Noyori, Tetrahedron Lett. 1980, 21,
1357 – 1358.
[16] For preparation of SmI2/THF, see P. Girard, J. L. Namy, H. B.
Kagan, J. Am. Chem. Soc. 1980, 102, 2693 – 2698.
[17] The enantiomeric purity of 15 was assessed by HPLC analysis
(Daicel Chemical Industries, Ltd, CHIRALPAK AD-H (25 cm,
0.46 cm diameter), hexane/iPrOH 9/1, flow rate 0.5 mLminꢀ1
,
tR = 11.4 min for the M isomer, 15.2 min for the S isomer). Also,
16 was analyzed (Daicel Chemical Industries Ltd,
CHIRALPAK AD-H (25 cm, 0.46 cm diameter), hexane/
3874
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 3871 –3874