Synthesis of (±)-diospongin a: A hetero-diels-alder and c-glycosylation approach

Article abstract of DOI:10.1055/s-0029-1218784

The racemic natural product diospongin A has been prepared using a short and stereoselective sequence. Key steps include a hetero-Diels-Alder reaction and an anchimeric assistance-controlled C-glycosylation. Georg Thieme Verlag Stuttgart · New York.

Full text of DOI:10.1055/s-0029-1218784

PAPER
2419
Synthesis of ( )-Diospongin A: A Hetero-Diels–Alder and C-Glycosylation
Approach
Synthesis
e
of
D
iospon
s
gin A se D. More*
Department of Chemistry, Loyola University Maryland, 4501 North Charles Street, Baltimore, MD 21210, USA
Fax +1(410)6172803; E-mail: jdmore@loyola.edu
Received 21 January 2010; revised 30 March 2010
dropyran framework. The key reaction in our synthesis is
a novel C-glycosylation that installs the phenacyl side-
chain, using neighboring-group participation4 to control
the C3 configuration.
Abstract: The racemic natural product diospongin A has been pre-
pared using a short and stereoselective sequence. Key steps include
a hetero-Diels–Alder reaction and an anchimeric assistance-con-
trolled C-glycosylation
Key words: Diels–Alder reaction, glycosylation, neighboring- Our retrosynthesis is shown in Scheme 1. Compound 1 is
group effects, natural products
derived from a protected oxocarbenium ion precursor
such as 3, which itself is derived from known ketone 4 via
reduction and formal hydration of the glycal olefin. Com-
pound 4 is the product of an HDA–elimination sequence
originating from Danishefsky-type5 diene 5 and benzalde-
hyde 6. Besides facilitating the concise preparation of the
tetrahydropyran ring of 1, the HDA approach offers the
additional advantage of enabling an enantioselective route
using a chiral catalyst.6
Diospongins A (1) and B (2) (Figure 1) are diaryl tetrahy-
dropyran-based natural products isolated from the etha-
nol–water fractions of the rhizome Dioscorea spongiosa,
which is abundant in southern parts of China and is used
in traditional Chinese medicine to treat a number of ail-
ments.1 Compounds 1 and 2 have been found to inhibit re-
lease of calcium from a bone culture system, which
suggests that the diospongins (or some derived analogue)
may have potential as anti-osteoporetic agents. Although
milligram quantities of 1 and 2 were obtained from the
original source, de novo synthesis offers far greater op-
portunities for the preparation of diverse analogues. The
intriguing bioactivity inherent in the relatively simple
structures of 1 and 2 has inspired a number of innovative
synthetic approaches.2,3
OMe
Ph
O
X
O
O
1
+
Ph
H
6
TBSO
OR
3
O
5
4
Scheme 1 Retrosynthetic analysis of 1
The key to our synthesis of 1 was to control the stereo-
chemical course of C-glycosylation – and thereby estab-
lish the C3 configuration – using anchimeric assistance.
This approach complements the work of Sawant and
Jennings, who synthesized 2 via a stereoselective C-gly-
cosylation reaction.2i,7 In this pioneering work, oxocarbe-
nium ion A (Scheme 2) is generated from 3 (R =
triethylsilyl, X = OAc). It is known from the work of
Woerpel and co-workers8 that six-membered oxocarbeni-
um ions such as A favor a half-chair conformation in
which the C5 (diospongin numbering) heteroatom substit-
uent adopts a pseudoaxial position. Nucleophilic attack by
an enol silane nucleophile from an axial trajectory (and
thus via a chair-like transition state) yields compound B,
which is deprotected in situ to give 2.2i
O
O
3
1
7
O
O
5
OH
OH
diospongin B (2)
diospongin A (1)
Figure 1 Structures of the diospongins
Most of the published syntheses of 1 and 2 have involved
buildup of a linear precursor using acyclic stereocontrol,
followed by formation of the tetrahydropyran ring and,
concomitantly, the C3 stereogenic center using an oxy-
Michael,2a–c Prins,2d,e SN2¢,2f or Wacker-type2g reactions.
Other approaches have first built the six-membered ring
and then set the C3 configuration using hydrogenation2h
or nucleophilic addition to a cyclic oxocarbenium ion.2i
This report describes a synthesis of 1 in which a hetero-
Diels–Alder3 (HDA) reaction is used to build the tetrahy-
We reasoned that changing the O-protecting group at C5
to a participating acyl group (e.g., benzoyl) would gener-
ate bicyclic acyloxonium ion C (Scheme 2). The topology
of C would then direct nucleophilic addition to the convex
face, giving the equatorial phenacyl chain as shown in D,
which corresponds to 1. 1,3-Stereocontrol in O-glycosyla-
tion is quite rare (compared to 1,2-stereocontrol), but has
some limited precedent in the synthesis of 2-deoxy-O-gly-
SYNTHESIS 2010, No. 14, pp 2419–2423
Advanced online publication: 10.05.2010
x
x.
x
x
.
2
0
1
0
DOI: 10.1055/s-0029-1218784; Art ID: M00610SS
© Georg Thieme Verlag Stuttgart · New York

2420
J. D. More
PAPER
cosides.9 Nonetheless, we were confident that anchimeric ment,16 we decided to first install an acetoxy group at C3
assistance via a structure corresponding to C could pro- of 8 using Falck’s conditions.17 Unfortunately, this tactic
vide access the target molecule 1. The experimental vali- only resulted in Ferrier rearrangement (not shown) via a
dation of our hypothesis is described below.
putative allyloxocarbenium ion, so the C5 hydroxyl group
was protected as its tert-butyldimethylsilyl (TBS) ether
(9). Acetoxylation of 9 smoothly produced 10 as an incon-
sequential 13:1 (assessed by NMR analysis) mixture of
anomers (Scheme 4). Although no effort was made to de-
termine the anomeric configuration of the major product
(because this stereocenter would be destroyed in a subse-
quent step), it was assumed to be a-(axial) based on liter-
ature precedent.17 Compound 10 was deprotected under
standard conditions to give the somewhat labile alcohol
11, which was converted into the epimeric benzoate 12 in
hot toluene.18
The [4+2] union of commercially available diene 5 and
benzaldehyde (6) was catalyzed by achiral chromium
complex 710 (4 mol%) to give 4 in 79% yield (Scheme 3).
We examined the application of other simple Lewis acid
catalysts [MgBr2, ZnCl2, BF3·OEt2, and Yb(OTf)3] in this
reaction, but found that 7 gave the highest yield and clean-
est reaction. We then needed to set the C5 configuration
using a 1,2-reduction of ketone 4. It is known that reduc-
tion of dihydropyrones such as 4 with achiral reducing
agents gives the C5–C7 (diospongin numbering) syn-ste-
reochemistry, regardless of which reducing agent is
used.11 We briefly examined reduction of 4 with the Co-
rey–Bakshi–Shibata (CBS) reagent12 (not shown) to no
avail. We decided to opt for a diastereoselective reduction
to give the epimeric C5 configuration, followed by inver-
sion later in the synthesis.
O
O
NaBH4
TBSCl, imidazole
4-DMAP, 90%
4
CeCl3⋅7H2O
–80 °C, 43–91%
OTBS
OH
8
9
Ph
O
Ph
Ph
O
R = TES
(ref. 2i)
Nu
3
O
O
OAc
O
OAc
OTES
Ph3P⋅HBr, AcOH
TBAF
82%
OR
B
CH2Cl2, 63%
A
OH
OTBS
10
11
Ph
Ph
O
Ph
O
R = OBz
Nu
3
O
O
OAc
O
O
PhCO2H, Ph3P, DEAD
PhMe, 90 °C, 71%
(this work)
Ph
OR
D
OBz
C
12
Scheme 2 Stereochemical course of C-glycosylations leading to 1
and 2
Scheme 4 Synthesis of 12
Luche reduction13 of 414 gave alcohol 8 as a single isomer
(Scheme 4). In our hands, the yield of this transformation
was somewhat variable, but generally improved as the
scale was increased.
In the key step of our synthesis (Scheme 5), a cold (–80 °C)
dichloromethane solution of acetate 12 was treated se-
quentially with enol silane 13 and BF3·OEt2. Thin layer
chromatography (TLC) after 4.5 hours showed clean con-
version into a new compound, along with some decompo-
sition products. After aqueous workup and filtration
through silica gel, the derived product was deacylated in
methanol/tetrahydrofuran to give 1 in 47% yield over two
steps. No other isomers were detected in the crude reac-
tion mixture. Data for the synthetic material matched
those published1,2i for 1 in all respects (1H, 13C NMR, IR,
HRMS, and TLC). This stereochemical result is consis-
tent with the generation of a bridged bicyclic acyloxonium
ion analogous to C (Scheme 2).
O
1. 5, 7 (4 mol%)
PhMe, r.t., 36 h
6
N
O
O
2. TFA, THF,
0 °C to r.t., 79%
7
O
Cr
Cl
( )-4
Scheme 3 Hetero-Diels–Alder reaction of 5 and 6
In conclusion, we have developed a concise and stereose-
lective synthesis of diospongin A (1). The step economy
of the current synthesis compares well with previous syn-
theses. A hetero-Diels–Alder reaction was used to build
the six-membered ring and a participating group at C5
was used to control the stereochemical outcome of the key
C-glycosylation reaction. This is, to our knowledge, the
We next needed to install a leaving group at the anomeric
carbon (C3) and invert the stereochemistry at C5 to pre-
pare a C-glycosylation substrate corresponding to generic
compound 3. The ordering of these steps was crucial to
the success of the sequence. Because Mitsunobu
reaction15 of 8 would give only type-I Ferrier rearrange-
Synthesis 2010, No. 14, 2419–2423 © Thieme Stuttgart · New York

PAPER
Synthesis of Diospongin A
2421
to warm to r.t., and diluted with brine (50 mL). The mixture was ex-
tracted with CH2Cl2 (2 × 20 mL), washed with brine (30 mL), dried
(Na2SO4) and concentrated. Flash chromatography (40
mm × 18 cm column; EtOAc–hexanes, 30%) gave 8 (1.06 g, 91%)
as an oil. All data were in agreement with literature values.13
1H NMR (400 MHz, CDCl3): d = 7.29–7.42 (m, 5 H), 6.52 (d,
J = 6.2 Hz, 1 H), 5.00 (dd, J = 1.9, 11.6 Hz, 1 H), 4.86 (m, 1 H),
4.60 (t, J = 7.4 Hz, 1 H), 2.36–2.47 (m, 1 H), 1.96–2.07 (m, 1 H).
O
OAc
1. 13, BF3⋅OEt2,
CH2Cl2, –80 °C
1
OTMS
2. NaOMe, MeOH
THF, 47% (2 steps)
OBz
12
13
Scheme 5 C-glycosylation to give 1
( )-tert-Butyldimethyl-2-phenyl-3,4-dihydro-2H-pyran-4-
yloxy-silane (9)
Alcohol 8 (0.200 g, 1.14 mmol, 1 equiv) was dissolved in CH2Cl2
(10 mL) under Ar, and imidazole (0.232 g, 3.41 mmol, 3.00 equiv)
and DMAP (0.028 g, 0.227 mmol, 0.20 equiv) were added. TBSCl
(0.257 g, 1.70 mmol, 1.50 equiv) was then added in one portion. The
reaction was allowed to stir for 15 h and then quenched with H2O
(10 mL), diluted with CH2Cl2 (10 mL), washed with H2O (20 mL),
brine (20 mL), dried (Na2SO4) and concentrated. Flash chromatog-
raphy (25 mm × 13 cm column; EtOAc–hexanes, 5%) gave 9
(0.299 g, 90%) as a clear oil. TLC: Rf = 0.53 (EtOAc–hexanes, 5%).
first example of a C-glycosylation controlled by 1,3-
neighboring-group participation and this important prece-
dent should find use in other related contexts.
All reactions were performed in oven- or flame-dried round-bottom
flasks under positive pressure of argon unless otherwise noted.
Flash column chromatography was performed with 32–63 mm (Sili-
cycle) silica gel. Thin-layer chromatography (TLC) was performed
using glass-backed plates coated with 0.25 mm F254 silica gel.
TLC plates were visualized using either UV light or charring after
staining with Verghn’s reagent (prepared by dissolving 40 g ammo-
nium molybdate and 1.6 g ceric ammonium nitrate in 800 mL 10%
v/v H2SO4). All commercial reagents were used without further pu-
rification. 1H NMR spectra were recorded with a 400 MHz instru-
ment in CDCl3. Chemical shifts (d) are reported in parts per million
(ppm) relative to residual solvent peak (CHCl3, d = 7.26 ppm). Data
are given as: chemical shift, multiplicity (s = singlet, d = doublet,
t = triplet, dd = doublet of doublets, dt = doublet of triplets,
ddd = doublet of doublet of doublets, m = multiplet), integration,
coupling constant (Hz). Proton-decoupled 13C NMR spectra were
recorded at 100 MHz. 13C NMR chemical shifts are reported rela-
tive to CDCl3 (d = 77.0 ppm). IR stretches are given in cm–1. HRMS
analysis was performed in the electrospray ionization mode at the
Scripps Research Institute.
FTIR (thin film): 2956, 2930, 2889, 2857, 1654 cm–1.
1H NMR (400 MHz, CDCl3): d = 7.29–7.37 (m, 5 H), 6.47 (dd,
J = 1, 6.3 Hz, 1 H), 4.97 (dd, J = 1.9, 12.3 Hz, 1 H), 4.75 (dt,
J = 1.8, 6.3 Hz, 1 H), 4.66 (m, 1 H), 2.20 (m, 1 H), 2.04 (ddd,
J = 9.5, 12.5, 26 Hz, 1 H), 0.893 (s, 9 H), 0.0986 (s, 3 H), 0.0775 (s,
3 H).
13C NMR (100 MHz, CDCl3): d = 145.1, 141.1, 129.0, 128.5, 126.6,
107.0, 77.1, 64.7, 40.7, 26.3, 18.6, –4.17, –4.23.
HRMS (ESI): m/z [M + Na] calcd for C17H26O2Si: 313.1594; found:
313.1596.
( )-4-(tert-Butyldimethylsilyloxy)-6-phenyltetrahydro-2H-pyr-
an-2-yl Acetate (10)
Glycal 9 (0.295 g, 1.02 mmol, 1 equiv) was dissolved in CH2Cl2 (7
mL) and AcOH (0.122 g, 2.03 mmol, 2.00 equiv) and Ph3P·HBr
(0.017 g, 0.0508 mmol, 0.050 equiv) were added. The reaction was
stirred for 3.5 h and then quenched with sat. aq NaHCO3, diluted
with CH2Cl2 (10 mL), washed with H2O (15 mL), brine (15 mL),
dried (Na2SO4) and concentrated. Flash chromatography
(25 mm × 15 cm column; EtOAc–hexanes, 5%) gave 10 (0.224 g,
63%) as an approximately 13:1 mixture of anomers. TLC: Rf = 0.27
(EtOAc–hexanes, 5%).
( )-2-Phenyl-2H-pyran-4(3H)-one (4)
A round-bottom flask was charged with catalyst 7 (0.134 g, 0.336
mmol, 0.040 equiv) and barium oxide (2 g). Toluene (30 mL) was
added, followed by diene 5 (1.80 g, 8.40 mmol, 1 equiv) via syringe.
The suspension was stirred at r.t. for 1 h, at which point benzalde-
hyde (6; 1.33 g, 12.6 mmol, 1.50 equiv) was added via syringe. Af-
ter 36 h, the reaction was filtered through Celite, washing with
EtOAc, and the filtrate concentrated. The crude adduct was dis-
solved in THF (25 mL), cooled to 0 °C and treated with TFA (1.44
g, 12.6 mmol, 1.50 equiv). The reaction was allowed to warm to r.t.
and quenched after 4.5 h with sat. aq NaHCO3 (20 mL), extracted
with EtOAc (3 × 15 mL), washed with brine (20 mL), dried
(Na2SO4) and concentrated. Flash chromatography (40
mm × 14 cm column; EtOAc–hexanes, 20%) gave 4 (1.15 g, 79%)
as an oil. All data were in agreement with literature values.5
FTIR (thin film): 2956, 2858, 1753, 1254 cm–1.
1H NMR (400 MHz, CDCl3): d = 7.32 (m, 5 H), 6.42 (d, J = 2.5 Hz,
1 H ), 5.82 (dd, J = 2.2, 10.2 Hz, 1 H, minor isomer), 4.85 (dd, 1 H,
J = 2.2, 12 Hz), 4.52 (dd, J = 2, 12 Hz, 1 H, minor isomer), 4.23 (m,
1 H), 2.11 (s, 3 H), 2.09 (m, 2 H), 1.79 (m, 1 H), 1.67 (dd, J = 12,
26 Hz, 1 H), 0.883 (s, 9 H), 0.0939 (s, 3 H), 0.0817 (s, 3 H).
13C NMR (100 MHz, CDCl3): d = 169.9, 141.5, 128.8, 128.2, 126.7,
126.5, 93.8, 73.1, 64.8, 43.8, 38.9, 26.2, 21.6, 18.4, –4.2.
1H NMR (400 MHz, CDCl3): d = 7.39–7.49 (m, 6 H), 5.53 (dd,
J = 1, 6 Hz, 1 H), 5.43 (dd, J = 3.2, 14.4 Hz, 1 H), 2.92 (dd,
J = 14.4, 16.8 Hz, 1 H), 2.67 (ddd, J = 1.2, 3.2, 16.8 Hz, 1 H).
HRMS (ESI): m/z [M + Na] calcd for C19H30O4Si: 373.1805; found:
373.1822.
( )-4-Hydroxy-6-phenyltetrahydro-2H-pyran-2-yl Acetate (11)
Acetate 10 (0.218 g, 0.622 mmol, 1 equiv) was dissolved in THF (6
mL) and TBAF (1 M in THF, 0.809 mL, 0.809 mmol, 1.30 equiv)
was added. The reaction was stirred at r.t. for 1.5 h, then additional
TBAF solution (0.200 mL) was added. After an additional 12 h, the
reaction was diluted with H2O (10 mL) and EtOAc (5 mL), the
aqueous layer was extracted with EtOAc (5 mL), and the combined
organic layer was washed with brine (5 mL), dried (Na2SO4), and
concentrated. Rapid flash chromatography (25 mm × 20 cm;
EtOAc–hexanes, 50%) gave 11 (0.121 g, 82%) as an oil [Rf = 0.28
( )-2-Phenyl-3,4-dihydro-2H-pyran-4-ol (8)
Ketone 4 (1.15 g, 6.60 mmol, 1 equiv) was dissolved in CH2Cl2 (41
mL) and EtOH (21 mL) and CeCl3·7H2O (3.20 g, 8.58 mmol, 1.30
equiv) were added. The suspension was cooled to –80 °C and a so-
lution of NaBH4 (0.300 g, 7.92 mmol, 1.20 equiv) in EtOH (26 mL)
was added dropwise via addition funnel over 30 min. TLC after 3 h
showed that some starting ketone remained, so additional
CeCl3·7H2O (1.60 g, 4.29 mmol, 0.650 equiv) and NaBH4 (0.150 g,
3.97 mmol, 0.600 equiv, in 10 mL EtOH) were added. After an ad-
ditional 5 h the reaction was quenched with brine (20 mL), allowed
Synthesis 2010, No. 14, 2419–2423 © Thieme Stuttgart · New York

2422
J. D. More
PAPER
(EtOAc–hexanes, 50%)]. The material was somewhat unstable and
used immediately in the next step.
1H NMR (400 MHz, CDCl3): d = 7.31 (m, 5 H), 6.44 (d, J = 2 Hz,
1 H), 4.88 (dd, J = 2, 12 Hz, 1 H), 4.29 (m, 1 H), 2.24 (m, 2 H), 2.10
(s, 3 H), 1.75 (m, 1 H), 1.63 (m, 1 H).
Acknowledgment
The author thanks Loyola University Maryland for funding, inclu-
ding a Junior Faculty Sabbatical leave. The author also thanks Pro-
fessors George Greco and Ruquia Ahmed-Schofield for generous
and unrestricted use of the 400 MHz NMR spectrometer in the
Goucher College Department of Chemistry.
( )-2-Acetoxy-6-phenyltetrahydro-2H-pyran-4-yl Benzoate
(12)
Alcohol 11 (1.29 g, 2.60 mmol, 1 equiv) was combined with Ph3P
(0.249 g, 0.948 mmol, 2.00 equiv) and benzoic acid (0.116 g, 0.948
g, 2.00 equiv) and dissolved in toluene (6 mL) under Ar and cooled
to 0 °C. DEAD (40% by weight in toluene, 0.430 mL, 0.948 mmol,
2.00 equiv) was added dropwise via syringe, the reaction was
warmed to r.t. and then placed in an oil bath set to 90 °C. After stir-
ring for 4 h, the solution was diluted with EtOAc (10 mL), washed
with sat. aq NaHCO3 (10 mL), brine (10 mL), dried (Na2SO4), and
concentrated. Flash chromatography (25 mm × 18 cm; EtOAc–hex-
anes, 20%) gave 12 (0.114 g, 71%) as an oil [Rf = 0.29 (EtOAc–hex-
anes, 20%)].
References
(1) Yin, J.; Kouda, K.; Tezuka, Y.; Tran, Q.-L.; Miyahara, T.;
Chen, Y.; Kadota, S. Planta Med. 2004, 70, 54.
(2) (a) Bressy, B.; Allais, F.; Cossy, J. Synlett 2006, 3455.
(b) Bates, R. W.; Song, P. Tetrahedron 2007, 63, 4497.
(c) Sabitha, G.; Padmaja, P.; Yadav, J. S. Helv. Chim. Acta
2008, 91, 2235. (d) Yadav, J. S.; Padmavani, B.; Reddy, B.
V. S.; Venugopal, Ch.; Rao, A. B. Synlett 2007, 2045.
(e) Hiebel, M.-A.; Pelotier, B.; Piva, O. Tetrahedron 2007,
63, 7874. (f) Lee, K.; Kim, H.; Hong, J. Org. Lett. 2009, 11,
5202. (g) Kawai, N.; Hande, S. M.; Uenishi, J. Tetrahedron
2007, 63, 9049. (h) Wang, H.; Shuhler, B. J.; Xian, M.
Synlett 2008, 2651. (i) Sawant, K. B.; Jennings, M. P.
J. Org. Chem. 2006, 71, 7911.
(3) While this manuscript was in preparation, a hetero-Diels–
Alder approach to the diospongins was reported, see:
Kumaraswamy, G.; Ramakrishna, G.; Naresh, P.; Jagadeesh,
B.; Sridhar, B. J. Org. Chem. 2009, 74, 8468.
(4) For general reviews of neighboring-group participation in
glycosylation, see: (a) Demchenko, A. In Handbook of
Chemical Glycosylation; Demchenko, A., Ed.; Wiley-VCH:
Weinheim, 2008, 5–6. (b) Hunig, S. Angew. Chem., Int. Ed.
Engl. 1964, 3, 548.
FTIR (thin film): 3066, 3034, 2964, 2923, 1749, 1720, 1278 cm–1.
1H NMR (400 MHz, CDCl3): d = 8.18 (m, 2 H), 7.62 (m, 1 H), 7.50
(t, J = 8 Hz, 2 H ), 7.34 (m, 5 H), 6.43 (d, J = 4 Hz, 1 H ), 5.51 (t,
J = 3 Hz, 1 H ), 5.35 (dd, J = 12, 2 Hz, 1 H), 2.25 (m, 2 H), 2.06 (m,
2 H), 2.00 (s, 3 H).
13C NMR (100 MHz, CDCl3): d = 169.9, 166.1, 141.3, 133.6, 130.9,
130.1, 128.89, 128.85, 128.3, 126.3, 91.9, 68.5, 66.5, 37.1, 32.0,
21.6.
HRMS (ESI): m/z [M + Na] calcd for C20H20O5: 363.1203; found:
363.1218.
( )-Diospongin A (1)
(5) Danishefsky, S.; Kerwin, J. F.; Kobayashi, S. J. Am. Chem.
Acetate 12 (0.037 g, 0.11 mmol, 1 equiv) was dissolved in CH2Cl2
(1.5 mL) under Ar and cooled to –80 °C. 1-Phenyl-1-trimethylsil-
oxyethylene (13; 0.071 mL, 0.35 mmol, 3.2 equiv) was added via
syringe, followed by BF3·OEt2 (0.044 mL, 0.345 mmol, 3.2 equiv).
After 1 h, the reaction was quenched with sat. aq NaHCO3 (2 mL)
slowly allowed to warm to r.t., extracted with CH2Cl2 (5 mL),
washed with brine (5 mL), and dried over Na2SO4. The crude prod-
uct was filtered through silica gel, washing with acetone–hexane
(15%) and concentrated. The product was then dissolved in MeOH–
THF (1:1, 1.6 mL) and treated with NaOMe (1 M in MeOH, 0.22
mL, 0.17 mmol, 1.6 equiv). TLC after 4.5 h showed complete dis-
appearance of starting material, and the reaction was concentrated.
The crude residue was partitioned between H2O (4 mL) and EtOAc
(4 mL), the organic layer was washed with brine and dried
(Na2SO4). Flash column chromatography on silica gel (EtOAc–hex-
anes, 40%) gave diospongin A (1; 0.015 g, 0.051 mmol, 47% over
two steps).
Soc. 1982, 104, 358.
(6) For a review, see: Pellissier, H. Tetrahedron 2009, 65, 2839.
(7) Sawant and Jennings also synthesized 1 by reduction of an
analogous oxocarbenium ion. See ref. 2i.
(8) Ayala, L.; Lucero, C. G.; Romero, J. A. C.; Tabacco, S. A.;
Woerpel, K. A. J. Am. Chem. Soc. 2003, 125, 15521.
(9) For examples of 1,3-stereocontrol in O-glycosylation via
anchimeric assistance, see: (a) Wiesner, K.; Tsai, T. Y. R.;
Jin, H. Helv. Chim. Acta 1985, 68, 300. (b) Binkley, R. W.;
Koholic, D. J. J. Carbohydr. Chem. 1988, 7, 487.
(c) Tanaka, H.; Yoshizawa, A.; Takahashi, T. Angew. Chem.
Int. Ed. 2007, 46, 2505. For reviews of 2-deoxyglycoside
synthesis, see: (d) Hou, D.; Lowary, T. L. Carbohydr. Res.
2009, 344, 1911. (e) Marzabadi, C.; Franck, R. W.
Tetrahedron 2000, 56, 8385.
(10) (a) Joly, G. D.; Jacobsen, E. N. Org. Lett. 2002, 4, 1795.
(b) Joly, G. D. Ph.D. Thesis; Harvard University: Cambridge
USA, 2004.
TLC: Rf = 0.23 (EtOAc–hexanes, 40%).
(11) (a) Moilanen, S. B.; Tan, D. S. Org. Biomol. Chem. 2005, 3,
798. (b) Kaneno, D.; Tomoda, S. Tetrahedron Lett. 2009,
50, 329.
(12) For a review, see: Corey, E. J.; Helal, C. J. Angew. Chem. Int.
Ed. 1998, 37, 1986.
FTIR (thin film): 3424, 2921, 1681, 1450, 1059 cm–1.
1H NMR (400 MHz, CDCl3): d = 7.95–7.99 (m, 2 H), 7.43–7.58 (m,
3 H), 7.21–7.37 (m, 5 H), 4.93 (dd, J = 11.8, 1.6 Hz,1 H), 4.65 (m,
1 H), 4.37 (t, J = 2.7 Hz, 1 H), 3.42 (dd, J = 16.0, 5.8 Hz, 1 H), 3.08
(dd, J = 16.0, 6.9 Hz, 1 H), 1.85–1.97 (m, 2 H), 1.6–1.8 (m, 4 H).
13C NMR (100 MHz, CDCl3): d = 198.9, 143.1, 137.7, 133.5,
128.95, 128.75, 128.67, 127.6, 126.2, 74.2, 69.5, 65.0, 45.6, 40.4,
38.9.
(13) Gemal, A. L.; Luche, J. L. J. Am. Chem. Soc. 1981, 103,
5454.
(14) Berkowitz, D. B.; Danishefsky, S. J.; Schulte, G. K. J. Am.
Chem. Soc. 1992, 114, 4518.
(15) Mitsunobu, O. Synthesis 1980, 1.
HRMS (ESI): m/z [M + H] calcd for C19H20O3: 297.1485; found:
297.1481.
(16) For a review, see: Ferrier, R. J. Top. Curr. Chem. 2001, 215,
153.
(17) Bolitt, V.; Mioskowski, C.; Lee, S.-G.; Falck, J. R. J. Org.
Chem. 1990, 55, 5812.
Synthesis 2010, No. 14, 2419–2423 © Thieme Stuttgart · New York

PAPER
Synthesis of Diospongin A
2423
(18) Reaction at lower temperatures gave little or no conversion,
which is congruous with sluggish SN2 displacement of an
equatorial leaving group. In the case of 11, a ring-flip to give
an axial leaving group would force the phenyl group into a
disfavored axial position. See: Eliel, E. L.; Wilen, S. H.;
Mander, L. N. In Stereochemistry of Organic Compounds;
John Wiley and Sons: New York, 1994, 723.
Synthesis 2010, No. 14, 2419–2423 © Thieme Stuttgart · New York