The truth about false unicorn (Chamaelirium luteum): Total synthesis of 23R,24S-chiograsterol B defines the structure and stereochemistry of the major saponins from this medicinal herb

Article abstract of DOI:10.1002/chem.201100503

Chamaelirium luteum is used in traditional medicine systems and commercial botanical dietary supplements for the treatment of female reproductive health problems. Despite the wide use of this herb, only very limited phytochemical characterisation is available. Our investigation of C. luteum roots led to the isolation of two new steroidal saponins 1 and 2 that contain an unusual aglycone 3. The absolute configurations of these molecules were unable to be determined spectroscopically and thus the total synthesis of 3 was undertaken and achieved in 16 steps and 1.6 % overall yield from pregnenolone. The key step in the synthesis was the stereoselective installation of the side chain at C-17 and C-20, which employed anion-accelerated oxy-Cope methodology. The relative configuration of aglycone 3 was determined by X-ray crystallography of an advanced synthetic intermediate. The absolute configuration was based upon that of the pregnenolone-derived steroidal skeleton and determined to be 23R,24S. Copyright

Full text of DOI:10.1002/chem.201100503

DOI: 10.1002/chem.201100503
The Truth about False Unicorn (Chamaelirium luteum): Total Synthesis of
23R,24S-Chiograsterol B Defines the Structure and Stereochemistry of the
Major Saponins from this Medicinal Herb
Nicholas J. Matovic,[a] Julia M. U. Stuthe,[a] Victoria L. Challinor,[a] Paul V. Bernhardt,[a]
Reginald P. Lehmann,[b] William Kitching,[a] and James J. De Voss*[a]
Abstract: Chamaelirium luteum is used
in traditional medicine systems and
commercial botanical dietary supple-
ments for the treatment of female re-
productive health problems. Despite
the wide use of this herb, only very
limited phytochemical characterisation
is available. Our investigation of C.
luteum roots led to the isolation of two
new steroidal saponins 1 and 2 that
contain an unusual aglycone 3. The ab-
solute configurations of these mole-
cules were unable to be determined
spectroscopically and thus the total
synthesis of 3 was undertaken and ach-
ieved in 16 steps and 1.6% overall
yield from pregnenolone. The key step
in the synthesis was the stereoselective
installation of the side chain at C-17
and C-20, which employed anion-accel-
erated oxy-Cope methodology. The rel-
ative configuration of aglycone 3 was
determined by X-ray crystallography of
an advanced synthetic intermediate.
The absolute configuration was based
upon that of the pregnenolone-derived
steroidal skeleton and determined to
be 23R,24S.
Keywords: natural products · rear-
rangement
· steroids · structure
elucidation · synthesis design
Introduction
Canada.[4] The underground parts (roots and rhizome) of C.
luteum have traditionally been used as a herbal medicine for
the treatment of a wide range of female reproductive health
problems.[5] C. luteum is now a component of many commer-
cial botanical dietary supplements formulated for use during
pregnancy and for the treatment of painful or irregular men-
struation. Over the period 1997–2005, the trade volume of
C. luteum in the USA alone was approximately 9800–
14300 kg dry weight of plant material.[6] Significantly, its use
is rapidly increasing, with 2001 consumption being 37%
greater than that in 2000 and 250% of that in 1997.[7] C.
luteum is currently listed by the United States Department
of Agriculture (USDA) as endangered or threatened in a
number of states,[4] and wild harvesting of C. luteum plants
for medicinal use (wildcrafting) is thought to further threat-
en populations of the species.[8]
The worldwide herbal medicine industry has been estimated
to be worth around $US16–20 billion annually, with $US7
billion spent in Europe (including approximately $US3 bil-
lion in France and Germany alone) and $US4 billion spent
in the USA.[1,2] A 2007 survey by the NIH National Centre
for Complementary and Alternative Medicine indicated that
approximately 18% of adults in the USA had used herbal
products.[3] Despite this wide use, there remains much that is
unknown about the chemical constituents of commonly used
medicinal plants. Chamaelirium luteum (L.) A. Gray, com-
monly known as False Unicorn or Devilꢀs Bit, is a medicinal
herb endemic to the eastern states of the USA and Ontario,
Despite the wide use of C. luteum, its phytochemical pro-
file has received very limited attention, with even authorita-
tive secondary literature, such as Hagerꢀs handbook, relying
on nineteenth century work that reported the presence of
saponins.[9] In 1878, Greene described the isolation of the
“bitter principle” of C. luteum, chamaelirin,[10] and subse-
quently reported the decomposition products of chamaelirin
as glucose and a substance he named chamaeliretin.[11] In
1942, Cataline and Francke “confirmed” the presence of
saponins in C. luteum, with the isolation from the alcoholic
extract of a substance that foamed upon shaking with
water.[12] In the same year, Marker et al. reported the isola-
tion of small amounts of the steroidal aglycone diosgenin
from C. luteum.[13] Despite the almost complete lack of
knowledge of its chemical constituents, and the consequent
[a] Dr. N. J. Matovic, Dr. J. M. U. Stuthe, V. L. Challinor,
Prof. P. V. Bernhardt, Prof. W. Kitching, Prof. J. J. De Voss
School of Chemistry and Molecular Biosciences
The University of Queensland, Brisbane, Qld 4072 (Australia)
Fax : (+61)7-3365-4299
E-mail: j.devoss@uq.edu.au
[b] Dr. R. P. Lehmann
Integria Healthcare, 35 Miles Platting Road
Eight Mile Plans, Qld 4113 (Australia)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100503. It contains general
experimental procedures; crystallography; extraction and isolation;
detailed procedure for the determination of sugar absolute configura-
tion; synthesis and characterisation of 13, 14, 18, 19, 21–23 and 25;
1
crystal and refinement data for 4a; complete H and 13C NMR spec-
1
troscopic assignment of (23R,24S)-3; H and 13C NMR spectra of 1, 2,
23, 29, 35, 40, 41, 47, 4b and (23S,24R)-3.
7578
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

FULL PAPER
impossibility of standardising extracts, supplements contain-
ing C. luteum have been incorporated into a recent clinical
trial[14] and an animal model study, in which the supplement
exhibited growth suppression of prostate tumours.[15]
the additional ring(s) characteristic of a furo- or spirostanol
steroidal saponin. Consistent with this, the 1H NMR spec-
trum revealed three secondary methyl groups at d=1.14 (d,
J=6.9 Hz; H3-26/27), 1.19 (d, J=6.5 Hz; H3-21) and
1.26 ppm (d, J=6.8 Hz; H3-26/27), in addition to two angu-
lar methyl groups, typical of a steroid (d=1.07 (H3-18) and
1.27 ppm (H3-19)). Correlations observed in the HSQC and
HMBC spectra allowed assignment of the five methyl
groups as d=13.6 (C-18), 16.1 (C-19), 20.7 (C-21) and 17.2
and 20.5 ppm (C-26/27). The HMBC correlations of these
characteristic methyl signals were a useful reference point
for structure elucidation of the aglycone. For example, cor-
relations from the methyl signals at d=1.14 and 1.26 ppm
(H-26/27) revealed C-25 (d=30.0 ppm) and C-24 (d=
79.6 ppm), clearly indicating the oxygenation of C-24. Ex-
amination of COSY, 1D TOCSY, HSQC and HMBC spectra
allowed complete assignment of the aglycone in an iterative
process (Table 1). Five secondary hydroxyl groups were
found at C-3 (d=78.0 ppm), C-6 (d=71.0 ppm), C-16 (d=
71.9 ppm), C-23 (d=71.6 ppm) and C-24 (d=79.6 ppm).
The stereochemistry of the different ring junctions and the
relative configurations at C-3, C-6, C-16 and C-17 were de-
termined by using correlations in a 2D ROESY experiment
(Scheme 2). For example, correlations observed between H-
Steroidal saponins are common constituents of a variety
of commercially important herbal medicines.[16,17] They con-
sist of a steroidal nucleus, commonly an oxidised cholesterol
derivative, with a number of sugar moieties at varying posi-
tions. The presence of a hydroxyl moiety at C-16 and a car-
bonyl at C-22 yields the hemiacetal found in furostanol sap-
onins, and an additional free hydroxyl at C-26 results in the
formation of a bicyclic acetal characteristic of spirostanol
saponins; these two motifs are the most common structures
found in steroidal saponins. This family of molecules exhib-
its a large range of biological activities, including anti-in-
flammatory, haemolytic, cytotoxic, anti-fungal and anti-bac-
terial properties, and is thought to be responsible for the
therapeutic effect of a number of medicinal plants.[18]
We thus sought to detail the presence and structure of
any steroidal saponins in the endangered C. luteum and
report herein the structure and absolute configuration of
two new steroidal saponins 1 and 2, elucidated through a
combination of NMR spectroscopy, MS, chemical degrada-
tion, total synthesis and X-ray crystallography. The synthetic
route to the aglycone of 1 and 2 permits a facile, practical
acquisition of this natural sapogenin as well as a range of
polyhydroxylated analogues for future biological evaluation.
Results and Discussion
Isolation and structural elucidation: Semi-preparative re-
verse-phase (RP) HPLC of the crude methanolic extract of
C. luteum underground parts afforded two major steroidal
saponins (1 and 2, Scheme 1) as 0.4 and 0.08% dry weight
of plant material, respectively.
Positive-ion high-resolution ESIMS of 1 provided an ion
at m/z 799.4444 [M+Na]+, corresponding to the molecular
formula C39H68O15. Fragmentation in tandem MS was consis-
tent with the presence of two hexose monosaccharides and a
Scheme 2. 2D ROESY correlations seen in saponins 1 and 2.
C
27 steroidal skeleton with five oxygen atoms. The molecular
1b/H-2b/H3-19, H-1a/H-3, H-3/H-5 and H-5/H-6 indicated a
trans A/B ring junction and b orientation of the substituents
at C-3 and C-6. Similar analysis of observed ROE correla-
tions for 1 indicated the trans fusion of rings B/C and C/D,
and b orientation of substituents at C-16 and C-17. The rela-
tive configuration at C-20 was indicated by a correlation be-
tween H3-18 and H-20.
formula of 1 suggested a saturated steroidal nucleus lacking
Two signals, each characteristic of the anomeric proton of
1
a glycoside, were also present in the H NMR spectrum of 1
at d=5.04 (d, J=7.6 Hz; H-1’) and 5.18 ppm (d, J=7.8 Hz;
H-1’’), and were correlated in the HSQC spectrum with sig-
nals at d=102.2 and 105.4 ppm, respectively. Examination
of the vicinal coupling constants determined by using 1D
TOCSY experiments identified the sugars as two b-glucopyr-
anose residues.[19] In the HMBC spectrum of 1, cross-peaks
were observed between d=5.04 (H-1’, 3-O-b-d-glucose) and
78.0 ppm (C-3, aglycone), and d=5.18 (H-1’’, 6’-O-b-d-glu-
Scheme 1. Steroidal saponins 1 and 2 isolated from C. luteum and their
aglycone 3. Glc=glucose, Fuc=fucose.
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7579

J. J. De Voss et al.
Table 1. 1H (500 MHz) and 13C (125 MHz) NMR spectroscopic data for the aglycone and sugar moieties of saponins 1 and 2 (d [ppm], C5D5N).
1
2
1
2
1H NMR
1.02 (m)
13C NMR 1H NMR
13C NMR
38.9
1H NMR
13C NMR 1H NMR
13C NMR
1a
1b
38.9
1.02 (m[a]
)
)
3-O-b-d-glu-
cose
1’
1.63 (brd,
J=13.1 Hz)
2.20 (m)
1.62 (m[a]
5.04 (d, J=7.6 Hz) 102.2
5.04 (d, J=7.8 Hz) 102.2
2a
2b
3
4a
4b
30.3
2.20 (m)
1.80 (m)
4.19 (m[a]
2.11 (m[a]
2.28 (m[a]
30.3
2’
3’
4’
5’
4.00 (t, J=8.3 Hz)
4.23 (t, J=8.9 Hz)
4.17 (t, J=9.2 Hz)
4.16 (m)
4.39 (dd, J=5.6,
10.7 Hz)
75.2
78.6
71.7
77.4
70.1
4.00 (t, J=8.3 Hz)
4.23 (t, J=9.0 Hz)
4.17 (t, J=9.0 Hz)
4.16 (m)
4.39 (dd, J=4.9,
11.4 Hz)
75.2
78.6
71.8[b]
77.4
70.1
1.80 (m)
4.19 (m[a]
2.10 (m[a]
2.28 (m[a]
)
)
)
78.0
32.9
)
)
)
78.0
32.9
6a’
5
1.09 (m[a]
)
47.9
1.09 (m[a]
)
47.9
6b’
4.87 (brd,
4.86 (brd,
J=10.7 Hz)
J=11.4 Hz)
6
7a
3.95 (m[a]
1.18 (m[a]
)
)
71.0
40.8
3.96 (m[a]
1.17 (m[a]
)
)
71.0
40.7
6’-O-b-d-glu-
cose
7b
2.02 (brd,
2.02 (brd,
1’’
5.18 (d, J=7.8 Hz) 105.4
5.18 (d, J=7.8 Hz) 105.4
J=13.3 Hz)
J=13.2 Hz)
8
9
2.12 (m[a]
)
30.8
54.7
2.12 (m[a]
)
30.8
54.7
2’’
3’’
4.05 (t, J=8.4 Hz)
75.3
78.6
4.05 (t, J=8.4 Hz)
75.3
78.6
0.66 (dt, J=4.2,
0.66 (dt, J=4.4,
4.24 (m[a]
)
4.23 (m[a]
)
11.4 Hz)
11.2 Hz)
10
36.1
21.1
40.4
36.1
21.1
40.4
4’’
5’’
6a’’
4.24 (m[a]
3.94 (m)
4.37 (dd, J=5.6,
11.7 Hz)
4.51 (dd, J=2.8,
11.7 Hz)
)
71.7
78.6
62.8
4.23 (m[a]
3.93 (m)
4.37 (dd, J=5.6,
11.8 Hz)
4.51 (brd,
J=11.8 Hz)
)
71.7[b]
78.6
62.8
11a,b 1.42 (m[a]
)
)
1.43 (m[a]
1.07 (m[a]
)
)
12a
12b
1.06 (m[a]
1.95 (brd,
J=12.4 Hz)
1.96 (brd,
J=12.4 Hz)
6b’’
13
14
42.8
54.2
42.8
54.3
0.89 (m)
0.88 (m)
24-O-b-d-
fucose
1’’’
15a
15b
16
2.29 (td, J=7.5,
12.9 Hz)
1.47 (dt, J=4.5,
13.0 Hz)
36.6
2.26 (m[a]
)
36.6
4.99 (d, J=7.8 Hz) 106.1
1.45 (dt, J=4.2,
13.0 Hz)
4.61 (m)
2’’’
3’’’
4’’’
5’’’
4.37 (brt,
73.4
75.6
72.8
71.5
17.3
J=8.5 Hz)
4.04 (dd, J=3.5,
9.2 Hz)
3.99 (brd,
J=3.5 Hz)
3.76 (brq,
4.63 (m)
71.9
62.8
13.6
71.7
62.7
13.6
17
1.13 (dd, J=6.9,
11.4 Hz)
1.07 s
1.08 (dd, J=6.8,
11.3 Hz)
1.06 (s)
18
J=6.4 Hz)
1.53 (d, J=6.4 Hz)
19
20
21
22a
1.27 (s)
2.54 (m)
1.19 (d, J=6.5 Hz)
1.88 (dd, J=8.7,
14.1 Hz)
16.1
27.8
20.7
40.4
1.27 (s)
2.52 (m)
1.14 (d, J=6.6 Hz) 20.1
1.65 (dd, J=8.7,
14.1 Hz)
16.1
27.5
6a’’’
6b’’’
38.5
22b
2.11 (dd, J=10.7,
14.1 Hz)
4.08 (m)
3.67 (m)
2.37 (m)
2.28 (brt,
J=12.7 Hz)
23
24
25
71.6
79.6
30.0
20.5
17.2
4.17 (m[a]
3.77 (m[a]
2.11 (m[a]
)
)
)
71.2
89.7
30.2
26/27 1.14 (d, J=6.9 Hz)
26/27 1.26 (d J=6.8 Hz)
1.08 (d, J=6.8 Hz) 20.0
1.28 (d, J=6.8 Hz) 19.7
[a] Indicates overlapping signals. [b] Assignments are interchangeable within the column.
cose) and 70.1 ppm (C-6’, 3-O-b-d-glucose), revealing a 1!
6-linked diglucose moiety attached at position C-3 of the
aglycone. The absolute configuration of the sugars was de-
termined by enantioselective GC following acid-catalysed
methanolysis of 1 and per-trifluoroacetylation of the resul-
tant monosaccharides.[20] The structure of 1 was thus eluci-
be determined spectroscopically due to the flexibility of C-
20 to C-27.
Compound 2 was isolated as an off-white solid and posi-
tive-ion high-resolution ESI-MS provided an ion at m/z
945.5037 [M+Na]+, corresponding to the molecular formula
C45H78O19. Fragmentation in tandem MS suggested the pres-
ence of one deoxyhexose (neutral loss of 146 Da) as well as
two hexose units (neutral losses of 162 and 164 Da) and a
C27 steroidal skeleton with five oxygen atoms. The NMR
dated as 3-O-b-d-glucopyranosyl[b-d-glucopyranosyl
6)](3b,6b,16b,23,24)pentahydroxy-5a-cholestane (chamaelir-
G
oside A), with the configuration of C-23 and C-24 unable to
7580
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
data for 2 were very similar to those obtained for 1
(Table 1). Examination of COSY, 1D TOCSY, HSQC and
HMBC spectra revealed the same hydroxylation pattern ob-
served in 1 and correlations observed in a 2D ROESY spec-
trum revealed that 2 also possessed the all-trans ring junc-
tions and b-oriented substituents at C-3, C-6, C16 and C-17.
The major difference between the data for 1 and 2 was that
the NMR spectra of 2 displayed signals consistent with the
presence of an additional deoxyhexose (d=4.99 (d, J=
7.8 Hz; H-1’’’); 106.1 ppm (C-1’’’)). An extra methyl doublet
signal was also observed at d=1.53 ppm (d, J=6.4 Hz; H-
6’’’), consistent with the presence of a 6-deoxyhexose. Ex-
amination of vicinal coupling constants obtained through
1D TOCSY experiments identified the deoxyhexose as b-
fucose.[19] The site of attachment of the fucose residue was
evident from the downfield shift of C-24 from d=79.6 ppm
in 1 to d=89.7 ppm in 2, and
were derived from the threo- and erythro-diols 6, which in
turn are generated from isomerically pure Z- and E-7 by
osmium tetroxide dihydroxylation. Comparison of the
1H NMR spectra of the two synthetic models 5 and the deri-
vatised aglycone 4 clearly indicated that the C-23/C-24 diol
possessed an erythro configuration (Table 2). Both shifts and
coupling constants for the model erythro-5 much more close-
ly matched those of the steroid derivative (Table 2). We thus
set out to synthesise the two C-23/C-24 erythro diastereo-
mers of the aglycone 3.
an HMBC correlation between
d=89.7 ppm (C-24, aglycone)
and the anomeric signal at d=
Table 2. Comparison of selected chemical shifts of natural acetonide 4 with erythro- and threo-5 (500 MHz, d
[ppm], C5D5N).
Assignment
4
erythro-5
threo-5
4.99 ppm
(H-1’,
24-O-b-d-
H-23 or equivalent
4.33 (ddd, J=1.9, 5.5,
11.7 Hz)
3.80 (dd, J=5.5, 9.6 Hz)
4.06 (ddd, J=2.5, 5.3,
11.1 Hz)
3.71 (dd, J=5.3, 9.3 Hz)
3.83 (m)
fucose). Enantioselective GC
analysis identified the sugar
moieties of 2 as one unit of d-
fucose and two units of d-glu-
cose. Both d- and l-enantio-
mers of fucose occur natural-
ly,[21] although the d-enantiomer
is more commonly reported in
steroidal and triterpenoid sapo-
H-24 or equivalent
3.46 (dd, J=6.1,
7.7 Hz)
0.93 (d, J=6.9 Hz)
H-26/27 or equiva-
lent
H-26/27 or equiva-
lent
0.86 (d, J=6.6 Hz)
1.12 (d, J=6.5 Hz)
0.82 (d, J=6.6 Hz)
1.10 (d, J=6.5 Hz)
1.02 (d, J=6.9 Hz)
nins, for example, in surculosides A–C from Dracaena surcu-
Synthesis of aglycone of chamaelirosides A and B: A survey
of commercially available steroids revealed pregnenolone 8
was suitably functionalised to serve as a starting material.
Our initial retrosynthetic analysis (Scheme 3) envisaged the
preparation of the erythro acetonide 4 through dihydroxyla-
tion of cis-alkene 9. Hydroxylation at C-16 would be ach-
ieved from 10 through selective oxidation of the electron-
rich tri-substituted C-16/C-17 olefin to give the C-16 alcohol.
The carbon backbone of the steroid side chain would arise
by displacement at C-22 with, for example, a vinyl cuprate.
Preparation of the requisite C-22-functionalised steroid 11
was expected to utilise procedures developed by Batcho
et al.,[24] which employed hetero-ene methodology for the
stereoselective introduction of side chains at C-20. Thus,
hetero-ene reaction of the Z-ethylidene 12 with paraformal-
dehyde was predicted to install C-22 as the primary alcohol
and generate the correct S configuration at C-20. Thus Z-
alkene 12 became the initial synthetic target, with its synthe-
sis from 8 requiring b-hydroxylation of the C-5/C-6 double
bond, and conversion of the ketone at C-20 to the C-17/C-
20 tri-substituted Z olefin.
losa.[22] The structure of 2 was therefore established as 3-O-
b-d-glucopyranosyl[b-d-glucopyranosyl
copyranosyl)(3b,6b,16b,23,24)pentahydroxy-5a-cholestane
(chamaeliroside B).
G
The aglycone of 1 and 2 is unusual among steroidal sapo-
genins because it lacks any additional rings derived from the
side chain at C-17, as found in the furo- and spirostanols,
and it may be responsible for the unique profile of biologi-
cal activity reported for C. luteum. It shares the same planar
structure as chiograsterol B (3); a very minor metabolite re-
ported from Chionographis japonica in 1968.[23] However,
1
very limited H NMR spectroscopic data are available only
for a derivative of this compound and the absolute configu-
rations at C-23 and C-24 are unknown. We thus undertook
the development of a practical route for the total synthesis
of the aglycone of 1 and 2 to confirm its structure, deter-
mine its absolute stereochemistry and provide analogues for
later biological evaluation.
Relative stereochemistry at C-23 and C-24: Compound 3
was reported to have an erythro configuration at C-23/C-24,
based upon the H-23/H-24 J value of the corresponding ace-
tonide 4.[23] However, the literature NMR spectroscopic data
available for comparison were insufficient to draw any con-
clusion as to the relative stereochemistry of 1 and 2 and
therefore simple models were synthesised. The acetonides 5
Surprisingly, simple hydroboration of 8, as well as a varie-
ty of protected derivatives, gave unsatisfactory yields of the
desired C-6 oxygenated product. However, b-selective epox-
idation of D5-unsaturated steroids with
a mixture of
KMnO4/CuSO4·5H2O in a biphasic solvent mixture had pre-
viously been reported[25] and with pregnenolone acetate 13
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7581

J. J. De Voss et al.
with POCl3 in pyridine at 0–
258C,[28] giving the Z-ethyli-
dene compound 16, which
upon treatment with sodium
cyanoborohydride in the pres-
ence
of
BF3·Et2O[29]
(Scheme 4) afforded the 6b-al-
cohol 17 regioselectively and
in very good yield. At this
stage, desired b-alcohol 17 was
readily separated by column
chromatography from any iso-
meric 6a-alcohol arising from
the a-epoxide impurity.
Acetylation of 17 was very
slow, presumably due to 1,3-di-
axial interactions of the C-6 al-
Scheme 3. Proposed initial synthetic strategy toward the synthesis of 4. Ts=tosyl.
cohol with C-19, but after four
days at room temperature in
provided a mixture of the a and b epoxides in 93% yield
with 94% selectivity for the b-epoxide 14 (Scheme 4).[26]
Separation of the a- and b-epoxide isomers of 14 was unsuc-
cessful by column chromatography and thus the mixture was
carried through, with isomer separation planned for a later
stage.
Reduction of the C-20 ketone of 14 with sodium borohy-
dride provided the two epimeric alcohols 15b and 15a in
the ratio (R/S) 87:13, which were readily separated by
column chromatography. The assignment of the C-20 stereo-
chemistry in these alcohols was based on the report that re-
duction of steroidal C-20 ketones with typical nucleophilic
reducing agents provides the C-20R product as the major
diastereomer.[27] In our hands, the solvent system methanol/
dichloromethane promoted greater selectivity (R/S ratio) in
this reduction than other solvents tried, such as ethanol,
methanol/THF or methanol alone. The R isomer 15b then
experienced formal E2 elimination of water when treated
the presence of a large excess of acetic anhydride in pyri-
dine, a very good yield of the initial target compound 18
was obtained. The b geometry of the C-6 substituent was
confirmed by analysis of the signal ascribable to H-6 in the
1H NMR spectrum of 18 (d=4.93 ppm (q, J=2.9 Hz)),
which displayed coupling values indicative of an equatorial
hydrogen. Treatment of 18 with paraformaldehyde and cata-
lytic BF3·Et2O in dichloromethane at room temperature
smoothly generated the homoallylic alcohol 19, in a highly
stereoselective manner, with the C-18 angular methyl group
steering electrophile approach from the least-hindered a
face of the molecule.[24]
With 19 in hand we explored side-chain attachment. Un-
fortunately, although activation of the primary alcohol of 19
as the mesylate 20 was achieved, reaction with simple[30] or
higher order[31] cuprates failed to produce any detectable
amount of substitution product. The synthetic strategy was
thus revised to explore whether C-16 oxygenation of 18
would deliver
a
steroidal
enone 21 that would undergo
addition to provide the basis
for side-chain construction.
The requisite 21 was acquired
by selenium dioxide mediated
allylic oxidation of alkene 18
to give the alcohol 22 (79%)
with complete regio- and ste-
reocontrol (Scheme 5). The
stereochemistry of the C-16 al-
cohol 22 was assigned based on
literature precedent[32] and 1D
NOESY NMR spectroscopy,
which showed a spatial correla-
tion between H-16 (d=
4.40 ppm (brd, J=5.5 Hz)) and
the C-18 methyl hydrogen
Scheme 4. Reagents and conditions: a) acetyl chloride, pyridine, CH2Cl2, 08C, 96%; b) KMnO4, CuSO4,
tBuOH, H2O, CH2Cl2, 258C, 93%; c) NaBH4, MeOH/CH2Cl2, 08C; d) POCl3, pyridine, 0–258C, 71%;
e) NaCNBH3, BF3·Et2O, THF, 608C, 87%; f) AcO2, pyridine, 258C, 4 days, 92%; g) (CH2O)n, BF3·Et2O,
CH2Cl2, 258C, 60%; h) mesyl chloride (MsCl), Et3N, CH2Cl2, 08C, 82%.
atoms
(d=0.88 ppm
(s)).
Dess–Martin periodinane oxi-
7582
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
steroidal side chain. This in-
volved 1,2-addition of an allyl
moiety to 21 and subsequent
anionic oxy-Cope rearrange-
ment. Again, there was encour-
aging precedent for this strat-
egy, in pioneering work by Ko-
reeda et al.[39] Thus, when
enone 21 was treated with
3 equivalents of allylmagnesi-
um bromide and 3 equivalents
of AlCl3 in THF at ꢀ788C, a
single diastereomer of the ter-
tiary allylic alcohol 25 was
formed
in
85%
yield
Scheme 5. Reagents and conditions: a) SeO2, tBuOOH, CH2Cl2, 08C, 79%; b) DMSO, (COCl)2, CH2Cl2, Et3N,
ꢀ788C, 80%; c) Dess–Martin periodinane, CH2Cl2, 258C, 82%; d) allyl MgBr, THF, AlCl3, ꢀ308C, 85%;
e) NaOH, MeOH, 258C, 80%; f) KH, dioxane, reflux, 2 h, 5–10%.
(Scheme 5). The stereochemis-
try at C-16 was attributed to
attack of the nucleophile from
the least-hindered a face of
dation of 22 then smoothly afforded enone 21 in 82% yield
after column chromatography. Unexpectedly, attempted oxi-
dation of allyl alcohol 22 under Swern conditions resulted in
complete conversion to the allyl chloride 23, despite reports
of Swern oxidation used for successful conversion of similar
steroidal allyl alcohols to enones.[33]
the molecule, and interestingly, the acetate groups were un-
affected when even 5 equivalents of Grignard reagent were
used. No precedent for the chemoselectivity of this
Grignard/AlCl3 combination was found, and thus, this par-
ticular reagent may be useful for the easy and selective addi-
tion of Grignard reagents to aldehydes and ketones in the
presence of esters.
Initially, conjugate addition to 21 was investigated, as pre-
vious reports indicated creation of the correct stereochemis-
try at both C-17 and C-20.[34] It was hoped that with the hin-
dered steroidal enone 1,4-addition might predominate with
a range of different nucleophiles. Thus, enone 21 was treated
with Grignard reagent 24a in the presence of AlCl3 (to sup-
press g-adduct formation),[35] but yielded mixtures of 1,2-
and 1,4-addition products. Allyl barium reagents had been
reported to add exclusively in a 1,4 fashion to unsaturated
ketones with almost complete suppression of the g adduct
and double-bond isomerisation.[36] However, attempted ad-
dition of allylic barium reagent 24b, even in large excess, to
enone 21 proved fruitless. Finally, 1,4-addition of a simple
allyl system to enone 21 was investigated. Allyl cuprates
failed to give successful 1,4-addition and produced, at best,
only trace quantities of 1,2-addition product. Allylation of
enone 21 was also attempted by exposure to Sakurai condi-
tions,[37] with successful 1,4-allylation occurring when using
the modified conditions (allyl-TMS, I2 cat., CH2Cl2, 258C)
Given the likelihood of undesired transesterification,
compound 25 was converted to 26 prior to anionic oxy-Cope
rearrangement (Scheme 5). A wide variety of conditions
(KH, THF or dioxane at 258C to reflux),[40] yielded only
small amounts of the desired product 27. Reasoning that the
tri-potassium anion of 26 might be modestly soluble, we re-
assessed the protection-group strategy. Thus, conversion of
17 to diol 28 was readily achieved in 85% yield by hydroly-
sis with NaOH in methanol (Scheme 6). tert-Butyldimethyl-
silyl (TBS) protection was investigated, but even vigorous
conditions (TBSCl, KH and [18]crown-6)[41] returned materi-
al only mono-protected at C-3. The sterically hindered, axial
C-6 alcohol also resisted protection as the benzyl ether
under traditional Williamson conditions[42] and gave only a
moderate yield (36%) of 29 when using silver oxide assisted
benzyl bromide alkylation (Scheme 6).[43] The best yields
(72%) were obtained by treating a concentrated solution of
28 with excess sodium hydride followed by 4 equivalents of
freshly prepared benzyl iodide and stirring in the absence of
light for 2 days at room temperature.[44]
Conversion of 29 to enone 30 was then carried out by the
same protocol used for conversion of 18 to 21 (Scheme 6).
Thus, selenium dioxide mediated allylic oxidation of 29 gave
allyl alcohol 31 in 66% yield in diastereomerically pure
form,[32] and further oxidation to enone 30 was then ach-
ieved in 78% yield with Dess–Martin periodinane. Because
benzyl-protected alcohols are generally stable to Grignard
reagents, we decided that allylmagnesium bromide alone
would be sufficient for the conversion of 30 to 32. Interest-
ingly, however, addition of AlCl3 apparently improved both
the reaction rate and the stereoselectivity of the allyl addi-
1
reported by Yadav et al.[38] Disappointingly, H NMR spec-
troscopy analysis showed that two diastereomers had
formed in almost equal amounts. An attempt to epimerise
the C-17 centre failed to alter the ratio of each diastereo-
mer, suggesting the allylated product to be a mixture of C-
20 epimers.
With the failure of conjugate addition to 21, we turned
our attention to a two-step strategy to install the required
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7583

J. J. De Voss et al.
led to epimerisation at C-17
and generation of a 5:1 mix-
ture of 33 and 34, from which
70% of the desired diastereo-
mer was isolated after column
chromatography (Scheme 7).
Completion of the side-chain
backbone was achieved in two
steps from 33, by ozonolysis of
the olefin to give keto-alde-
hyde 35 and a chemoselective
Wittig condensation at ꢀ788C
with the ylide derived from
phosphonium salt 36.[45] The
resulting olefin 37 was ob-
tained in 42% yield over two
Scheme 6. Reagents and conditions: a) NaOH, MeOH, 258C, 85%; b) BnI, NaH, THF/DMF, 258C, 2 days,
72%; c) SeO2, tBuOOH, CH2Cl2, 08C, 66%; d) Dess–Martin periodinane, CH2Cl2, 258C, 78%; e) AlCl3
(3 equiv), THF, ꢀ788C, 89%. Bn=benzyl.
tion compared with allylmagnesium bromide alone. Thus,
treatment of enone 30 with 3 equivalents of allylmagnesium
bromide and 3 equivalents of AlCl3 gave the desired com-
pound 32 in 89% yield and as a single diastereomer.
steps as a 12:1 mixture of the cis and trans isomers by
1H NMR spectroscopy.[46]
At this stage, we anticipated that stereoselective dihydrox-
ylation of the C-23, C-24 olefin could be achieved by utilis-
ing the well-known Sharpless asymmetric dihydroxylation
(SAD) methodology,[47] despite the moderate enantioselec-
tivities reported with cis alkenes.[48–50] However, when
alkene 37 was subjected to typical dihydroxylation condi-
tions, there was no detectable diol formation even after one
week.
It was thus decided to proceed by using non-enantioselec-
tive osmium-catalysed dihydroxylation. The expected eryth-
ro diols 38 and 39, following their conversion to the rigid bi-
cyclic acetals 40 and 41, would reveal their absolute configu-
rations following spectroscopic assignment of their charac-
teristic H-23 coupling constants and comparisons with the
known compound 42.[51]
Molecular modelling (Chem3D Ultra, version 8) estimat-
ed the dihedral angles shown in Scheme 8. The H-22b/H-23
coupling constant was expected to be larger in 40 than that
in 41 (16.5 and 59.38, respectively), whereas the H-22a/H-23
coupling constant was expected to be small for both diaste-
reomers. The estimated dihedral angles between H-23 and
H-24 for 40 and 41 (35.2 and 43.08, respectively) were con-
sidered likely to result in simi-
The anionic oxy-Cope rearrangement of 32 under stan-
dard conditions once again provided only moderate yields,
but heating in 1,2-dimethoxy ethane at reflux for 80 min, fol-
lowed by quenching with water at ꢀ108C gave the 17R,20R-
keto olefin 33 as a single diastereomer in 91% yield
(Scheme 7) in line with literature precedent.[39] Generation
of the C-20R stereochemistry was directed by the C-17/C-20
double-bond geometry and the chair-like transition state op-
erating in the potassium hydride assisted oxy-Cope rear-
rangement,[39] establishing the new C-20/C-21 bond from the
a face of the molecule (Scheme 7). Stereospecific formation
of the 17R stereochemistry results from protonation of the
resulting potassium enolate from the least-hindered a face
upon quenching with water. Cooling the reaction to ꢀ108C
or less prior to quenching was found to be critical in terms
of the stereoselectivity at C-17 and quenching at tempera-
tures above 08C led to significant levels (10–30%) of anoth-
er product, tentatively identified as the 17S epimer 34. This
could be separated from the 17R 33 by column chromatog-
raphy. Treatment of 34 with K2CO3 in methanol overnight
lar vicinal coupling constants
between these hydrogen atoms
for both diastereomers.
Dihydroxylation of 37 was
achieved by using catalytic (2–
3 mol%) osmium tetroxide
with NMO as the primary oxi-
dant.[52] After consumption of
alkene 37 was complete, cycli-
sation to the ketals 40 and 41
was effected by acidification of
the crude reaction mixture
with dilute aqueous HCl.
Semi-preparative normal-phase
(NP) HPLC of the crude reac-
tion mixture afforded pure 40
Scheme 7. Reagents and conditions: a) KH, (MeOCH2)2, reflux, 1 h; b) H2O, ꢀ108C, 91%; c) H2O, 0–258C,
20–40%; d) K2CO3, MeOH, 258C, 70%; e) O3, CH2Cl2, ꢀ788C, PPh3, 55%; f) BuLi, diethyl ether, ꢀ788C,
42%.
7584
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
C-16, C-23 and C-24 have been
reported,[51,53–59] and the route
developed
herein
permits
simple access to this unusual
structural motif. It is interest-
ing that none of the reported
natural compounds have been
found to possess the 23S con-
figuration.
Unfortunately, all attempts
to further transform the acetal
mixture 40 and 41 under a
wide variety of hydrolytic con-
ditions were unsuccessful. It
was decided that reduction of
ketone 37 would be performed
prior to dihydroxylation of the
C-23/C-24 olefin, and that sep-
aration of the diastereomers
might then be achieved at a
later stage. Treatment of 37
with diisobutylaluminium hy-
dride (DIBAL; Scheme 9) at
ꢀ788C resulted in the forma-
Scheme 8. Reagents and conditions: a) OsO4, N-methylmorpholine N-oxide (NMO), THF/tBuOH/H2O, 25 8C;
b) dilute aqueous HCl.
and 41 in an approximately 1:1 ratio. Examination of the
13C NMR spectra for 40 and 41 confirmed the successful for-
mation of diastereomeric acetals, with the absence of any
carbonyl signals and the appearance of a new signal in each
spectrum at d=113.8 ppm for 40 and d=115.0 ppm for 41.
The chemical shifts of H-23 for each diastereomer displayed
significant differences in their coupling constants as expect-
ed, occurring at d=4.42 ppm (ddd, J=2.7, 3.9, 9.6 Hz) in
the major isomer 40, and dH 4.35 ppm (ddd, J=1.4, 3.6,
3.6 Hz) for the minor isomer 41. The largest coupling dis-
played by H-23 for the major isomer 40 (9.6 Hz) was signifi-
cantly larger than that displayed by the minor isomer 41
(3.6 Hz) and thus major acetonide diastereomer was as-
signed as 40 (23R,24S). In addition to this, when the H-23
coupling constants were compared with the known 42
(Table 3),[51] there was a better correlation with H-23 and H-
24 of the major acetonide diastereomer, but not with the
minor acetonide. A number of natural triterpenoid com-
pounds with a similar bicyclic acetal moiety encompassing
tion of the 16b-alcohol 43 with a diastereomeric excess of
>95% (1H NMR spectroscopy). Chromatographic purifica-
tion gave 72% of 43 as a single diastereomer, with the as-
signed C-16 stereochemistry in harmony with hydride deliv-
ery from the least-hindered a face.[28,60] Osmium-catalysed
dihydroxylation of 43 resulted in a mixture of the two dia-
stereomeric erythro diols 44 and 45, which were converted
directly to the acetonides 46 and 47 by stirring in acetone
1
with catalytic TsOH. H NMR spectroscopic analysis of the
crude product showed the diastereomeric ratio of the two
acetonides was approximately 2:1, demonstrating the same
selectivity observed during dihydroxylation of 37. Pleasingly,
compounds 46 and 47 were separable by careful silica gel
column chromatography and were obtained in 53 and 25%
yields, respectively. Removal of the benzyl protection
groups (H2, Pd/C, THF) then gave both acetonide triols 4a
(78%) and 4b (72%).
Suitable crystals for X-ray crystallography of the major
acetonide isomer 4a were grown by vapour diffusion of
hexane into an ethyl acetate solution. Examination of the
crystal structure for 4a (Figure 1) confirmed the stereo-
chemistry of all stereogenic centres unambiguously, but
most importantly, revealed the 23R,24S configuration of the
acetonide. A notable feature is an intramolecular hydrogen-
bond donated by the hydroxyl group (O16) to the acetal
O23 atom. The minor acetal 4b was thus assigned the
23S,24R configuration.
Table 3. Selected 1H NMR spectroscopic data for H-23 and H-24 of 40,
41 (750 MHz, d [ppm], C5D5N) and the known bicyclic acetal 42.[51]
1H and 13C NMR spectroscopy data (in CD3OD) for both
4a and 4b were acquired and compared with one another
and those of acetonide 4 derived from saponins 1 and 2. Be-
sides the difference in chemical shift of the C-21 methyl hy-
drogen atoms, which resonated at d=1.04 ppm (d, J=
6.4 Hz) for 4a and d=1.11 ppm (d, J=6.8 Hz) for 4b, the
Assignment 40
41
42
H-23
4.42 (ddd, J=2.7, 4.35 (ddd, J=1.4, 4.62 (ddd, J=2.5,
3.9, 9.6 Hz)
3.39 (dd, J=4.0,
10.0 Hz)
3.6, 3.6 Hz)
3.60 (dd, J=3.6,
10.5 Hz)
4.2, 9.5 Hz)
3.73 (d, J=4.2 Hz)
H-24
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7585

J. J. De Voss et al.
Scheme 9. Reagents and conditions: a) DIBAL, Et2O, ꢀ788C, 72%; b) OsO4, NMO, THF/tBuOH/H2O, 25 8C, 67% (2:1 44:45); c) acetone, p-toluenesul-
fonic acid (TsOH), 258C, 46: 53%, 47: 25%; d) H2, Pd/C, THF, 258C, 4a: 78%, 4b: 72%; e) Dowex 50-X8, MeOH, 258C, 2 days, (23R,24S)-3: 75%,
(23S,24R)-3: 85%.
for chiograsterol B (m.p. 240–243.58C and [a]2D5 +16.1, c=
0.49, CH3OH) than (23S,24R)-3 (m.p. 191–1928C and [a]D25
+3.6, c=0.52, CH3OH). It thus appears that chiograsterol B
is likely to have the 23R,24S configuration, but comparison
of spectroscopic data from an authentic sample of chiogras-
terol B is necessary before a firm conclusion can be drawn
in this regard.
Conclusion
Two novel saponins 1 and 2 were isolated from alcoholic ex-
tracts of the widely used medicinal plant C. luteum. These
contained an unusual aglycone (23R,24S)-3 that was synthes-
ised in 16 steps and 1.6% overall yield from pregnenolone
8. The key step in the synthesis involved the stereoselective
introduction of the steroidal side chain at C-17 and C-20 by
utilising anionic oxy-Cope methodology. Assignment of the
relative stereochemistry was confirmed by X-ray crystallog-
raphy of an advanced synthetic intermediate. The absolute
configuration was based on that of the pregnenolone-de-
rived steroidal skeleton. The stereochemistry of the isolated
agylcone is probably identical to that of chiograsterol B (3)
from C. japonica,[23] but discrepancies between the reported
physical properties of two compounds makes this conclusion
necessarily tentative. The described route to 3 makes it, and
other analogues/stereoisomers, readily available as well as
sapogenins with the intriguing bicyclic acetal moiety present
in 40 and 41.
Figure 1. ORTEP view of the major acetonide diastereomer 4a. Ellip-
soids are drawn at the 30% level of probability.[62]
most notable changes between each synthetic diastereomer
were observed for H-23 and H-24. In the 13C NMR spectra,
the differences in chemical shifts between acetonides 4a and
4b are small, with the greatest difference being that of C-23
(Dd=2.3 ppm). Pleasingly, when the NMR spectra were
compared, isomer 4a was identical to the naturally derived
compound 4 (Figure 2), clearly indicating that the natural
compounds 1, 2 and 3 posses the 23R,24S configuration.
Coupling constants for the hydrogen atoms as well as melt-
ing points and optical rotations for each diastereomeric ace-
tonide are displayed in Table 4.
Acetal hydrolysis of 4a or 4b (methanol, Dowex 50-X8
acidic resin) gave the pentahydroxy sterols (23R,24S)-3 and
(23S,24R)-3 (Scheme 9). The physical properties of
(23R,24S)-3 and (23S,24R)-3 were compared with those re-
ported for chiograsterol B.[23,61] Both the melting point and
optical rotation of (23R,24S)-3 (m.p. 220–2228C and [a]D25
+9.2, c=0.5, CH3OH) were in closer agreement with those
Experimental Section
Plant material: The underground parts of C. luteum were sourced from
Botanical Liaisons (USA) and Blessed Herbs (USA). A verified speci-
men from Blessed Herbs was deposited (accession number PHARM-
7586
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
Figure 2. Segment of the 13C NMR spectra of the region containing the hydroxylated carbon atoms for 4a, 4b and the naturally derived acetonide 4.
1
Table 4. Melting points, optical rotations and selected H NMR shifts (500 MHz, d [ppm], CD3OD) for 4a, 4b and natural 4.
M.p.
[a]2D5
1H NMR
[8C]
G
H-3
H- 6
H-16
H-23
H-24
4a
182–183
189–191
179–180
+18.4
(c=0.12)
ꢀ7.0
(c=1.2)
+16.2
(c=0.44)
3.54 (tt, J=5.2,
10.8 Hz)
3.54 (tt, J=5.2,
10.8 Hz)
3.54 (tt, J=5.3,
10.8 Hz)
3.74 (m)
4.32 (dt, J=4.3,
7.6 Hz)
4.30 (dt, J=4.8,
7.8 Hz)
4.33 (dt, J=4.3,
7.5 Hz)
4.21 (ddd, J=1.5, 5.8,
11.5 Hz)
4.26 (ddd, J=2.3, 5.7,
10.1 Hz)
4.21 (ddd, J=1.5, 5.8,
11.5 Hz)
3.74 (m)
AHCTUNGTRENNUNG
4b
3.74 (m)
3.74 (m)
3.67 (dd, J=5.6,
8.5 Hz)
3.74 (m)
AHCTUNGTRENNUNG
natural 4
AHCTUNGTRENNUNG
06275) at the Medicinal Plant Herbarium at Southern Cross University,
Lismore, Australia, where it was identified by Dr. Hans Wohlmuth.
(Chirasil-l-Val capillary column) were performed according to proce-
dures used previously.[17] The retention times for the trifluoroacetic acid
anhydride (TFAA) derivatised standards were d-glucose (25.75 and
29.40), l-glucose (25.68 and 29.24), d-fucose (19.70 and 23.79) and l-
fucose (19.83 and 24.08). For 1, peaks were observed at 25.75 and
29.40 min, and for 2 peaks were observed at 19.68, 23.80, 25.74 and
29.39 min. Co-injection of the different saponin hydrolysates with the
TFAA derivatives of authentic d-glucose and d-fucose gave co-eluting
peaks.
Chamaeliroside A (1): Off-white solid; m.p. 209–2108C (decomp); [a]D25
ꢀ19.9 (c 0.24 in CH3OH); 1H (500 MHz, C5D5N) and 13C NMR
(125 MHz, C5D5N), see Table 1; MS (ESI+): m/z: 799; MS (ESIꢀ): m/z:
775, 613, 449; HRMS (ESI): m/z calcd for C39H68NaO15
[M+Na]+; found: 799.4444.
: 799.4450
Chamaeliroside B (2): Off-white solid; m.p. 184–1868C (decomp); [a]D25
ꢀ21.8 (c 0.20 in CH3OH); 1H (500 MHz, C5D5N) and 13C NMR
(125 MHz, C5D5N), see Table 1; MS (ESI+): m/z: 945; MS (ESIꢀ): m/z:
AHCTUNGERTG(NNUN 20R)-3b-Acetoxy-5b, 6b-epoxypregnan-20-ol (15b): Compound 14
921, 775, 759, 613, 595; HRMS (ESI): m/z calcd for C45H78NaO19
945.5030 [M+Na]+; found: 945.5037.
:
(3.6 g, 9.6 mmol) was dissolved in a solution of MeOH (90 mL) and
CH2Cl2 (30 mL) and the mixture was cooled to 08C with an external ice
bath. Next, NaBH4 (380 mg, 10 mmol) was added portionwise with stir-
ring over a 5 min period. The mixture was stirred for a further 40 min,
then quenched by the cautious addition of a 0.1m solution of oxalic acid
(100 mL). The mixture was extracted with CH2Cl2 (2ꢂ80 mL) and the
combined organic layers were washed with water (100 mL), then brine
(100 mL) and dried (MgSO4). Evaporation of the solvent left a solid that
was purified by flash chromatography (20% EtOAc in hexanes, silica gel
60). The first compound to elute was 15b (2.45 g, 68%). M.p. 138–1398C;
[a]2D5 ꢀ33.3 (c 1.07 in CHCl3); 1H NMR (500 MHz, CDCl3): d=0.62 (dt,
J=5.4, 11.2 Hz, 1H), 0.70 (s, 3H), 0.89 (m, 1H), 0.98 (s, 3H), 1.10 (d, J=
6.0 Hz, 3H), 1.08–2.12 (m, 18H), 2.00 (s, 3H), 3.05 (d, J=2.3 Hz, 1H),
ACHTUNGTRENNUNG
(23R,24S)-Chiograsterol B (3): White solid; m.p. 219–2208C; [a]2D2 +11.1
(c 0.38 in CH3OH); 1H (500 MHz, C5D5N) and 13C (125 MHz, C5D5N)
NMR spectra were identical to those for synthetic (23R,24S)-3 reported
1
below. Full H and 13C NMR assignments are provided in the Supporting
Information (Table S2); MS (ESI+): m/z: 475, 491; MS (ESIꢀ): m/z:
451; HRMS (ESI): m/z calcd for C27H48NaO5 475.3394: [M+Na]+; found:
475.3400; elemental analysis calcd (%) for C27H48O5·H2O: C 68.90, H
10.71; found: C 68.90, H 10.51.
Determination of sugar absolute configuration: The saponin (1 mg) was
subjected to acid-catalysed methanolysis before per-trifluoroacetylation
of the resultant methyl glycosides and enantioselective GC analysis
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7587

J. J. De Voss et al.
3.68 (m, 1H), 4.74 ppm (m, 1H); 13C NMR (125 MHz, CDCl3): d=12.2,
17.0, 21.3, 21.7, 23.6, 24.4, 25.5, 27.2, 29.5, 32.5, 35.1, 36.6, 38.0, 39.8, 42.2,
51.0, 55.6, 58.4, 62.5, 63.4, 70.4, 71.3, 170.5 ppm; elemental analysis calcd
(%) for C23H36O4: C 73.37, H 9.64; found: C 73.77, H 9.95;
54.2, 56.0, 71.6, 72.0, 113.3, 150.2 ppm; elemental analysis calcd (%) for
C21H34O2: C 79.19, H 10.76; found: C 78.79, H 10.85.
AHCTUNGERTG(NNUN 17Z)-3b, 6b-Dibenzyloxy-5a-pregn-17(20)-ene (29): NaH (1.44 g,
30 mmol, 50% dispersion in mineral oil) was washed carefully with hex-
anes (3ꢂ15 mL). Next, THF (10 mL) and DMF (10 mL) were added, fol-
lowed by 28 (2.4 g, 7.6 mmol). The mixture was stirred for 30 min at
258C before the addition of freshly distilled benzyl iodide (6.6 g,
30.4 mmol). The mixture was stirred for 48 h at 258C while protected
from direct light. The reaction was then carefully quenched by the addi-
tion of water (30 mL) and the mixture extracted into hexane (3ꢂ25 mL).
The organic layers were washed with water (50 mL) then brine (50 mL),
dried (MgSO4) and evaporated. Flash chromatographic purification (5%
EtOAc in hexanes, silica gel 60) yielded 29 as a colourless oil (2.72 g,
72%). [a]2D5 ꢀ31.0 (c 1.87 in CHCl3); 1H NMR (500 MHz, CDCl3): d=
0.74 (dt, J=3.7, 11.2 Hz, 1H), 0.92–1.02 (m, 2H), 0.93 (s, 3H), 1.12 (s,
3H), 1.16–1.81 (m, 11H), 1.70 (td, J=1.9, 7.2 Hz, 3H), 1.94–2.04 (m,
2H), 2.09 (dt, J=3.2, 14.0 Hz, 1H), 2.22 (m, 1H), 2.30 (m, 1H), 2.42 (m,
1H), 3.40 (tt, J=5.2, 10.8 Hz, 1H), 3.49 (m, 1H), 4.38 and 4.63 (AB q,
J=12.4 Hz, 2H), 4.59 and 4.61 (AB q, J=12.0 Hz, 2H), 5.16 (qt, J=2.0,
7.1 Hz, 1H), 7.25–7.45 ppm (m, 10H); 13C NMR (125 MHz, CDCl3): d=
13.1, 15.6, 16.9, 21.2, 24.4, 28.3, 30.4, 31.4, 32.4, 34.9, 36.0, 37.2, 38.4, 44.4,
47.8, 54.5, 55.9, 69.7, 71.2, 78.4, 79.4, 113.2, 126.8 (2C), 126.9, 127.3, 127.5
(2C), 128.1 (2C), 128.2 (2C), 139.2, 139.8, 150.3 ppm.
The second compound to elute was 15a (360 mg, 10%). M.p. 128–1298C;
[a]2D5 ꢀ27.0 (c 0.62 in CHCl3); 1H NMR (400 MHz, CDCl3): d=0.61 (s,
3H), 0.60–2.16 (m, 18H), 0.98 (s, 3H), 1.18 (d, J=6.2 Hz, 3H), 2.00 (s,
3H), 3.05 (d, J=2.3 Hz, 1H), 3.67 (m, 1H), 4.74 ppm (tdd, J=5.0, 10.1,
11.8 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=12.3, 17.0, 21.3, 21.6,
23.5, 24.1, 25.6, 27.2, 29.4, 32.4, 35.0, 36.6, 37.9, 38.8, 41.5, 50.9, 55.9, 58.4,
62.5, 63.5, 70.2, 71.3, 170.5 ppm; elemental analysis calcd (%) for
C23H36O4: C 73.37, H 9.64; found: C 73.73, H 9.79.
ACHTUNGTRENNUNG(17Z)-3b-Acetoxy-5b, 6b-epoxypregn-17(20)-ene (16): A solution of 15b
(1.7 g, 4.52 mmol) in pyridine (50 mL) was cooled to 08C with an exter-
nal ice bath. Then, POCl3 (6.9 g, 45 mmol) was added dropwise with stir-
ring over 30 min and the mixture was left to stir overnight in the melting
ice bath. Next, the mixture was diluted with EtOAc (200 mL) and
poured slowly into a mixture of 4m HCl (100 mL) and crushed ice
(200 g). The organic layer was separated and the aqueous phase was ex-
tracted with further EtOAc (100 mL). The combined organic layers were
washed with brine (2ꢂ100 mL), dried (MgSO4) and evaporated. Flash
chromatographic purification (10% EtOAc in hexanes, silica gel 60)
yielded 16 (1.15 g, 71%) as a white solid. M.p. 83–848C; [a]2D5 ꢀ11.5 (c
1.28 in CHCl3); 1H NMR (400 MHz, CDCl3): d=0.65 (m, 1H), 0.83 (s,
3H), 0.99 (s, 3H), 0.98–2.41 (m, 20H), 1.61 (td, J=2.0, 7.1 Hz, 3H), 2.01
(s, 1H), 3.07 (d, J=2.4 Hz, 1H), 4.75 (tdd, J=4.9, 10.1, 11.9 Hz, 1H),
5.09 ppm (tq, J=2.0, 7.1 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=13.1,
16.5, 17.0, 21.3, 22.0, 24.4, 27.2, 29.3, 31.3, 32.3, 35.1, 36.7, 37.0, 38.0, 44.0,
51.0, 55.9, 62.5, 63.5, 71.3, 113.5, 150.0, 170.5 ppm; elemental analysis
calcd (%) for C23H34O3: C 77.05, H 9.56; found: C 77.17, H 9.83.
AHCTUNGTERG(NNUN 17Z)-3b,6b-Dibenzyloxy-5a-pregn-17(20)-en-16-one (30): Compound 31
(190 mg, 0.37 mmol) was dissolved in CH2Cl2 (15 mL) and Dess–Martin
periodinane (188 mg, 0.44 mmol) was added in one portion. The mixture
was stirred for 2 h by which time a white precipitate had developed. The
mixture was diluted with diethyl ether (20 mL) and filtered through a
plug of Celite. The Celite was washed with diethyl ether (2ꢂ10 mL) and
the combined filtrates were evaporated. Flash chromatographic purifica-
tion (10% EtOAc in hexanes, silica gel 60) yielded 30 as a white solid
(147 mg, 78%). M.p. 113–1158C; [a]2D5 ꢀ145.8 (c 0.36 in CHCl3);
1H NMR (400 MHz, CDCl3): d=0.76–0.98 (m, 3H), 0.99 (s, 3H), 1.09 (s,
3H), 1.15 (m, 1H), 1.37–1.96 (m, 12H), 1.83 (d, J=7.6 Hz, 3H), 2.13 (m,
1H), 2.28 (m, 1H), 3.35 (tt, J=5.2, 10.8 Hz, 1H), 3.46 (m, 1H), 4.38 and
4.54 (AB q, J=12.3 Hz, 2H), 4.53 and 4.56 (AB q, J=12.0 Hz, 2H), 6.47
(q, J=7.5 Hz, 1H), 7.20–7.35 ppm (m, 10H); 13C NMR (100 MHz,
CDCl3): d=13.2, 15.7, 17.7, 20.7, 26.9, 28.2, 29.7, 32.3, 35.1, 36.1, 36.2,
37.9, 38.1, 43.4, 47.8, 49.7, 54.2, 69.8, 71.6, 78.3, 79.2, 126.9 (2C), 127.1,
127.4, 127.6, 128.1 (2C), 128.3 (2C), 129.0, 139.1, 139.7, 147.9, 206.4 ppm;
elemental analysis calcd (%) for C35H44O3: C 81.99, H 8.65; found: C
81.85, H 8.86.
A
(17):
Compound
16
(420 mg, 1.17 mmol) and NaBH3CN (378 mg, 6.0 mmol) were added to
THF (20 mL). A small amount (approx. 5–10 mg) of bromocresol green
indicator was added to the stirred solution, which resulted in a deep
green/blue colour. Next, the solution was heated to 508C and freshly dis-
tilled BF3·Et2O was added dropwise until the reaction changed to a
bright yellow colour. BF3·Et2O was continually added one drop at a time
to maintain the yellow colour; however, it should not be added so rapidly
that hydrogen evolution occurs. After 35 min the yellow colour persisted
without further addition of BF3·Et2O, with approximately 0.6 mL of
BF3·Et2O added in total. The solution was stirred for a further 2 h at
508C, before being cooled to room temperature. Water (100 mL) was
added and the mixture extracted into diethyl ether (3ꢂ20 mL). The di-
ethyl ether layers were washed with water (50 mL), then brine (50 mL),
dried (MgSO4) and evaporated. Flash chromatographic purification
(20% EtOAc in hexanes, silica gel 60) yielded 17 as a white solid
(366 mg, 87%). M.p. 160–1618C; [a]2D5 ꢀ13.0 (c 0.52 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=0.71 (m, 1H), 0.88 (s, 3H), 0.99 (dt, J=3.8,
11.8 Hz, 1H), 1.04 (s, 3H), 1.11–1.25 (m, 3H), 1.36–1.86 (m, 13H), 1.63
(td, J=2.0, 7.2 Hz, 3H), 2.01 (s, 3H), 2.09–2.39 (m, 3H), 3.79 (m, 1H),
4.72 (tt, J=5.4, 11.1 Hz, 1H), 5.12 ppm (tq, J=2.1, 7.2 Hz, 1H);
13C NMR (100 MHz, CDCl3): d=13.1, 15.6, 16.9, 21.1, 21.4, 24.4, 27.4,
29.9, 31.3, 31.4, 35.5, 37.0, 38.2, 39.5, 44.4, 47.1, 54.1, 55.9, 71.8, 73.9,
113.4, 150.2, 170.7 ppm; elemental analysis calcd (%) for C23H36O3: C
76.62, H 10.06; found: C 76.59, H 10.33.
A
(31):
tBuOOH
(0.12 mL, 0.6 mmol, 5m tBuOOH solution in decane) was added to a
stirred suspension of selenium dioxide (28 mg, 0.25 mmol) in CH2Cl2
(8 mL) at 08C. After stirring for 20 min, compound 29 (200 mg,
0.42 mmol) was added in one portion and stirring was continued for 4 h.
Then, a 10% aqueous solution of NaHSO3 (10 mL) was added and the
mixture was stirred vigorously for a further 15 min. The CH2Cl2 layer was
collected and dried (MgSO4), then evaporated to leave an oil. Flash chro-
matographic purification (20% EtOAc in hexanes, silica gel 60) yielded
31 as a colourless oil (143 mg, 66%). [a]2D5 ꢀ29.9 (c 1.99 in CHCl3);
1H NMR (400 MHz, CDCl3): d=0.75 (m, 1H), 0.83–1.18 (m, 4H), 0.86
(s, 3H), 1.07 (s, 3H), 1.15 (td, J=2.5, 12.9 Hz, 1H), 1.38–1.76 (m, 9H),
1.73 (dd, J=2.0, 7.2 Hz, 3H), 1.88–2.04 (m, 3H), 2.25 (m, 1H), 3.36 (tt,
J=5.2, 10.7 Hz, 1H), 3.46 (m, 1H), 4.41 (brd, J=5.7 Hz, 1H), 4.35 and
4.58 (AB q, J=12.3 Hz, 2H), 4.54 and 4.56 (AB q, J=12.0 Hz, 2H), 5.56
(dq, J=1.0, 7.1 Hz, 1H), 7.20–7.37 ppm (m, 10H); 13C NMR (100 MHz,
CDCl3): d=13.2, 15.6, 17.6, 21.0, 28.2, 29.7, 32.4, 34.8, 35.0, 36.0, 37.3,
38.3, 44.5, 47.8, 52.2, 54.5, 69.7, 71.3, 74.3, 78.3, 79.4, 119.4, 126.8 (2C),
126.9, 127.3, 127.5 (2C), 128.1 (2C), 128.3 (2C), 139.1, 139.8, 155.4 ppm;
elemental analysis calcd (%) for C35H46O3: C 81.67, H 9.01; found: C
81.50, H 9.15.
ACHTUNGTRENNUNG(17Z)-5a-Pregn-17(20)-en-3b,6b-diol (28): NaOH (270 mg, 6.7 mmol)
was added in one portion to a solution of 17 (1.21 g, 3.4 mmol) in MeOH
(15 mL) at 258C. The solution was stirred for 4 h, then water (30 mL)
was added followed by brine (30 mL). The mixture was extracted into
CH2Cl2 (3ꢂ15 mL). The combined organic layers were washed with
water (30 mL), then brine (30 mL) and dried (MgSO4). The crude solid
material (28) obtained after removal of the volatile compounds was suffi-
ciently pure for the next reaction (0.91 g, 85%). M.p. 124–1268C; [a]D25
ꢀ0.4 (c 2.4 in CHCl3); 1H NMR (400 MHz, CDCl3): d=0.69 (m, 1H),
0.88 (s, 3H), 1.02 (s, 3H), 0.92–1.89 (m, 16H), 1.63 (td, J=2.0, 7.1 Hz,
3H), 2.08–2.39 (m, 3H), 3.62 (tt, J=5.3, 10.8 Hz, 1H), 3.79 (m, 1H),
5.09 ppm (tq, J=2.1, 7.2 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=13.1,
15.7, 16.9, 21.2, 24.4, 30.0, 31.4, 31.5, 35.3, 35.4, 37.1, 38.4, 39.4, 44.4, 47.3,
A
(32):
AlCl3 (117 mg, 0.88 mmol) was dissolved in THF (6 mL) at room temper-
ature and the stirred solution was cooled to ꢀ308C. Next, a 1m solution
of allyl magnesium bromide in diethyl ether (0.88 mL, 0.88 mmol) was
added dropwise, causing the formation of a white precipitate. The mix-
7588
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
ture was stirred for a further 10 min, then 30 (150 mg, 0.29 mmol) was
added. The mixture was stirred for 20 min at ꢀ308C, before being
quenched with a saturated aqueous solution of NH4Cl (10 mL). The or-
ganic layer was collected and the aqueous phase extracted with diethyl
ether (10 mL). The combined organic layers were washed with water
(10 mL) then brine (10 mL). After drying (MgSO4), the solvent was
evaporated and the residue was purified by flash chromatography (20%
EtOAc in hexanes, silica gel 60) yielding 32 as a white solid (142 mg,
89%). M.p. 51–528C; [a]D25 ꢀ18.5 (c 0.14 in CHCl3); 1H NMR (400 MHz,
CDCl3): d=0.69 (m, 1H), 0.89 (m, 2H), 1.05 (s, 3H), 1.06 (s, 3H), 1.01–
2.05 (m, 14H), 1.72 (d, J=7.2 Hz, 3H), 2.18–2.37 (m, 3H), 3.34 (tt, J=
5.2, 10.7 Hz, 1H), 3.44 (m, 1H), 4.32 and 4.57 (AB q, J=12.3 Hz, 2H),
4.54 and 4.55 (AB q, J=12.0 Hz, 2H), 5.04–5.17 (m, 2H), 5.49 (q, J=
7.2 Hz, 1H), 5.86 (m, 1H), 7.21–7.37 ppm (m, 10H); 13C NMR
(100 MHz, CDCl3): d=13.1, 15.6, 18.1, 20.9, 28.2, 29.6, 32.4, 34.8, 36.0,
37.0, 38.2, 38.3, 44.6, 47.0, 47.8, 51.1, 54.5, 69.8, 71.2, 78.4, 79.2, 79.7,
117.0, 118.1, 126.9 (2C), 127.0, 127.3, 127.5 (2C), 128.1 (2C), 128.3 (2C),
134.8, 139.1, 139.7, 157.5 ppm; elemental analysis calcd (%) for C38H50O3:
C 82.26, H 9.08; found: C 81.97, H 8.77.
for 30 min, then the ice bath was removed and the reaction was allowed
to warm slowly to room temperature. After stirring for 30 min at room
temperature, a saturated aqueous solution of NH4Cl (5 mL) was added.
The organic layer was separated, then washed with brine (10 mL), dried
(MgSO4) and evaporated. Flash chromatographic purification (5%
EtOAc in hexanes, silica gel 60) yielded 37 as a colourless oil (69 mg,
42% over two steps). [a]D25 ꢀ100.7 (c 1.3 in CHCl3); 1H NMR (500 MHz,
CDCl3): d=0.79 (s, 3H), 0.80 (m, 1H), 0.88–0.99 (m, 2H), 0.90 (d, J=
6.7 Hz, 3H), 0.91 (d, J=6.8 Hz, 3H), 0.95 (d, J=6.8 Hz, 3H), 1.07 (s,
3H), 1.15 (td, J=2.5, 12.9 Hz, 1H), 1.32–2.18 (m, 17H), 2.41 (td, J=3.7,
13.9 Hz, 1H), 2.59 (m, 1H), 3.35 (tt, J=5.2, 10.8 Hz, 1H), 3.45 (m, 1H),
4.36 and 4.52 (AB q, J=12.3 Hz, 2H), 4.53 and 4.55 (AB q, J=12.0 Hz,
2H), 5.18–5.25 (m, 2H), 7.20–7.36 ppm (m, 10H); 13C NMR (125 MHz,
CDCl3): d=13.6, 15.7, 18.5, 20.5, 23.1, 23.2, 26.5, 28.2, 29.8, 31.6, 32.4,
33.2, 35.2, 36.1, 38.2, 38.8, 39.0, 43.3, 47.8, 50.4, 54.3, 67.3, 69.8, 71.5, 78.3,
79.2, 125.4, 126.9 (2C), 127.1, 127.4, 127.5 (2C), 128.1 (2C), 128.3 (2C),
138.9, 139.1, 139.7, 218.7 ppm; elemental analysis calcd (%) for C41H56O3:
C 82.50, H 9.46; found: C 82.51, H 9.49.
Cyclic ketals (23R,24S)-40 and (23S,24R)-41 of 3b,6b-dibenzyloxy-23,24-
dihydroxy-5a-cholesten-16-one: Compound 37 (20 mg, 0.034 mmol) was
dissolved in THF (110 mL), then water (11 mL) was added and the solu-
tion stirred at 258C for 5 min. Following the addition of NMO (5.8 mg,
0.05 mmol), stirring was continued for a further 5 min. OsO4 (2.5% w/v
in tBuOH, 33 mL, 0.003 mmol) was then added and the solution was
stirred overnight at 258C. The reaction was quenched with sodium sulfite
(approx. 5 mg) and stirred for a further 30 min. The mixture was extract-
ed into CH2Cl2 (3ꢂ2 mL). The combined organic layers were washed
with water (3ꢂ2 mL), dried (MgSO4) and the solvent removed. The resi-
due was dissolved (70% EtOAc in hexanes, 4 mL), filtered and purified
by semi-preparative NP-HPLC (10% ethyl acetate in hexanes,
2 mLminꢀ1). The first diastereomer to elute was 40 (tR 13.1 min) as a col-
ourless oil (3.2 mg, 16%). [a]2D5 ꢀ15.5 (c 0.24 in CHCl3); 1H NMR
(750 MHz, C5D5N): d=0.66–1.39 (m, 11H), 0.70 (d, J=6.7 Hz, 3H), 0.75
(s, 3H), 0.93 (d, J=6.6 Hz, 3H), 1.13 (d, J=6.5 Hz, 3H), 1.14 (s, 3H),
1.47 (m, 1H), 1.54–1.83 (m, 6H), 1.85 (ddd, J=6.7, 9.6, 13.4 Hz, 1H),
1.99 (m, 1H), 2.03 (td, J=3.1, 14.0 Hz, 1H), 2.08 (q, J=12.0 Hz, 1H),
2.21 (dd, J=6.4, 13.1 Hz, 1H), 3.39 (dd, J=4.0, 10.0 Hz, 1H), 3.42 (m,
2H), 4.35 and 4.57 (AB q, J=12.3 Hz, 2H), 4.42 (ddd, J=2.7, 3.9, 9.6 Hz,
1H), 4.64 and 4.66 (AB q, J=12.0 Hz, 2H), 7.22–7.57 ppm (m, 10H);
13C NMR (188 MHz, C5D5N): d=13.6, 16.0, 18.1, 20.2, 21.0, 21.2, 23.1,
28.8, 29.3, 29.8, 30.2, 33.0, 35.5, 36.2, 37.3, 38.5, 40.5, 43.3, 47.8, 51.4, 54.5,
66.9, 69.9, 71.7, 73.5, 78.8, 79.9, 84.4, 113.8, 127.5 (3C), 127.6, 127.9 (2C),
128.6 (2C), 128.7 (2C), 140.4 ppm (2C); HRMS (ESI): m/z calcd for
C41H56NaO4: 635.4071 [M+Na]+; found: 635.4068.
3b,6b-Dibenzyloxy-5a-cholan-23-en-16-one (33): KH (60 mg, 0.5 mmol,
35% dispersion in mineral oil) was washed carefully with hexanes (2ꢂ
2 mL). Next, 1,2-dimethoxy ethane (3 mL) was added followed by 32
(130 mg, 0.23 mmol) and the mixture was heated to a gentle reflux with
stirring for 60 min. The reaction was returned to room temperature, then
cooled to ꢀ10 to ꢀ208C by using an external ice/ethanol bath and care-
fully quenched by the addition of water (7 mL). The mixture was extract-
ed into diethyl ether (3ꢂ10 mL) and the combined organic layers were
washed with water (25 mL) then brine (20 mL) and dried (MgSO4). After
evaporation of the solvent, the residue was purified using flash chroma-
tography (10% EtOAc in hexanes, silica gel 60) to yield 33 as a white
solid (112 mg, 91%). M.p. 82–838C; [a]2D5 ꢀ113.2 (c 1.1 in CHCl3);
1H NMR (400 MHz, CDCl3): d=0.78–1.01 (m, 3H), 0.79 (s, 3H), 0.94 (d,
J=6.9 Hz, 3H), 1.06 (s, 3H), 1.14 (m, 1H), 1.31–2.18 (m, 17H), 2.50 (m,
1H), 3.35 (tt, J=5.2, 10.7 Hz, 1H), 3.46 (m, 1H), 4.35 and 4.52 (AB q,
J=12.3 Hz, 2H), 4.52 and 4.55 (AB q, J=12.0 Hz, 2H), 4.94–5.04 (m,
2H), 5.76 (m, 1H), 7.20–7.36 ppm (m, 10H); 13C NMR (100 MHz,
CDCl3): d=13.7, 15.7, 18.4, 20.5, 28.2, 29.8, 31.0, 32.4, 35.2, 36.1, 38.1,
38.9, 39.0, 39.9, 43.3, 47.8, 50.3, 54.2, 67.1, 69.8, 71.5, 78.3, 79.2, 116.1,
126.9 (2C), 127.1, 127.4, 127.6 (2C), 128.2 (2C), 128.3 (2C), 137.3, 139.1,
139.7, 218.5 ppm; elemental analysis calcd (%) for C38H50O3: C 82.26, H
9.08; found: C 82.23, H 9.17.
ACHTUNGTRENNUNG(23Z)-3b,6b-Dibenzyloxy-5a-cholest-23-en-16-one (37): Compound 33
(150 mg, 0.27 mmol) was dissolved in CH2Cl2 (15 mL) and chilled to
ꢀ788C by using an external dry-ice bath. Ozone was then bubbled
through the solution and as soon as the reaction had changed to a faint
blue colour bubbling was ceased and the mixture purged with nitrogen to
discharge the blue colour. Next, triphenylphosphine (100 mg, 0.4 mmol)
was added and stirring was commenced. The ice bath was removed and
after the mixture had returned to room temperature CH2Cl2 was evapo-
rated. Purification of the residue by flash chromatography (20% EtOAc
in hexanes, silica gel 60) gave 35 (80 mg, 55%). 1H NMR (500 MHz,
CDCl3): d=0.80 (s, 3H), 0.80–1.06 (m, 4H), 1.01 (d, J=6.6 Hz, 3H), 1.07
(s, 3H), 1.14 (m, 1H), 1.33–1.97 (m, 13H), 2.05 (m, 1H), 2.16 (dd, J=7.6,
18.3 Hz, 1H), 2.35 (m, 1H), 3.04 (m, 1H), 3.35 (tt, J=5.2, 10.8 Hz, 1H),
3.45 (m, 1H), 4.36 and 4.51 (AB q, J=12.3 Hz, 2H), 4.53 and 4.55 (AB
q, J=12.0 Hz, 2H), 7.21–7.35 (m, 10H), 9.74 ppm (m, 1H); 13C NMR
(125 MHz, CDCl3): d=13.4, 15.7, 20.2, 20.5, 25.9, 28.2, 29.8, 32.4, 35.2,
36.1, 38.1, 38.8, 39.0, 43.3, 47.8, 50.1 (2C), 54.2, 66.8, 69.8, 71.6, 78.3, 79.2,
126.9 (2C), 127.1, 127.4, 127.6 (2C), 128.2 (2C), 128.3 (2C), 139.1, 139.6,
202.4, 218.5 ppm.
The second diastereomer to elute was 41 (tR 14.9 min) as a colourless oil
(3.1 mg, 15%). [a]2D5 ꢀ51.2 (c 0.21 in CHCl3); 1H NMR (750 MHz,
C5D5N): d=0.67–2.02 (m, 20H), 0.72 (d, J=6.5 Hz, 3H), 0.90 (s, 3H),
0.93 (d, J=6.5 Hz, 3H) 1.12 (d, J=6.5 Hz, 3H), 1.15 (s, 3H), 2.09 (q, J=
12.0 Hz, 1H), 2.21 (m, 1H), 2.26 (dd, J=9.0, 14.2 Hz, 1H), 3.43 (m, 2H),
3.60 (dd, J=3.6, 10.5 Hz, 1H), 4.33 and 4.64 (AB q, J=12.3 Hz, 2H),
4.35 (ddd, J=1.4, 3.6, 3.6 Hz, 1H), 4.65 and 4.66 (AB q, J=12.0 Hz, 2H),
7.24–7.58 ppm (m, 10H); 13C NMR (188 MHz, C5D5N): d=12.6, 16.0,
19.0, 20.2, 20.7, 21.0, 27.6, 28.0, 28.8, 30.3, 33.0, 34.9, 35.2, 36.1, 36.3, 38.4,
39.7, 42.1, 47.9, 53.3, 55.0, 65.8, 69.9, 71.5, 78.5, 78.8, 79.7, 85.8, 115.0,
127.5 (3C), 127.6, 127.9 (2C), 128.7 (4C), 140.4 ppm (2C); HRMS (ESI):
m/z: calcd for C41H56NaO4: 635.4071 [M+Na]+; found: 635.4067.
AHCTUNGERTG(NNUN 23Z)-3b,6b-Dibenzyloxy-5a-cholest-23-en-16b-ol (43): DIBAL (71 mg,
0.5 mmol) was added dropwise to a stirred solution of 37 (250 mg,
0.42 mmol) in diethyl ether (15 mL) at ꢀ788C. After stirring for 3 h, the
reaction was quenched by the addition of 1m HCl (5 mL). The mixture
was allowed to return to room temperature and stirring was continued
for a further 2 h. The diethyl ether layer was separated, then washed with
brine (15 mL), dried (MgSO4) and the solvent was removed. Purification
of the residue by flash chromatography (10% EtOAc in hexanes, silica
gel 60) yielded 43 as a colourless oil (180 mg, 72%). [a]2D5 ꢀ35.0 (c 0.19 in
CHCl3); 1H NMR (400 MHz, CDCl3): d=0.64 (dt, J=4.2, 11.3 Hz, 1H),
0.82–1.15 (m, 6H), 0.87 (s, 3H), 0.93 (d, J=6.6 Hz, 3H), 0.94 (d, J=
6.6 Hz, 3H), 0.97 (d, J=6.2 Hz, 1H), 1.05 (s, 3H), 1.33–2.04 (m, 15H),
Meanwhile, a solution of the required ylide was prepared as follows: Iso-
butyl phosphonium bromide[45] (36) (250 mg, 0.4 mmol) was suspended in
diethyl ether (5 mL) and then butyllithium (0.23 mL, 1.4m solution in
hexanes, 0.32 mmol) was added dropwise with stirring at 258C. After
complete addition, the deep-red solution was stirred for a further 30 min,
before being cooled to ꢀ788C by using an external ice bath. Next, the al-
dehyde (35) prepared above was dissolved in diethyl ether (5 mL) and
added dropwise to the ylide solution. The reaction was stirred at ꢀ788C
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7589

J. J. De Voss et al.
2.17–2.29 (m, 2H), 2.59 (m, 1H), 3.34 (tt, J=5.2, 10.7 Hz, 1H), 3.43 (m,
1H), 4.32 (m, 1H), 4.32 and 4.57 (AB q, J=12.3 Hz, 2H), 4.54 and 4.55
(AB q, J=12.0 Hz, 2H), 5.23–5.33 (m, 2H), 7.21–7.37 ppm (m, 10H);
13C NMR (100 MHz, CDCl3): d=13.3, 15.7, 18.5, 20.6, 23.1, 23.2, 26.5,
28.2, 30.4 (2C), 32.4, 33.9, 34.9, 36.0, 36.8, 38.4, 39.9, 42.6, 47.8, 53.9, 54.5,
61.3, 69.7, 71.2, 72.4, 78.4, 79.3, 125.6, 126.9 (2C), 127.0, 127.3, 127.5 (2C),
128.1 (2C), 128.3 (2C), 138.7, 139.1, 139.8 ppm; elemental analysis calcd
(%) for C41H58O3: C 82.22, H 9.76; found: C 82.46, H 9.41.
(500 MHz, CDCl3): d=0.62 (dt, J=3.6, 11.4 Hz, 1H), 0.79–2.18 (m,
21H), 0.86 (d, J=6.6 Hz, 3H), 0.89 (s, 3H), 1.00 (d, J=6.5 Hz, 3H), 1.03
(s, 3H), 1.09 (d, J=6.8 Hz, 3H), 1.31 (s, 3H), 1.43 (s, 3H), 2.19 (td, J=
7.6, 12.9 Hz, 1H), 2.70 (d, J=3.6 Hz, 1H), 3.33 (tt, J=5.2, 10.8 Hz, 1H),
3.42 (m, 1H), 3.63 (dd, J=5.3, 9.4 Hz, 1H), 4.24 (td, J=4.9, 9.8 Hz, 1H),
4.31 and 4.56 (AB q, J=12.2 Hz, 2H), 4.33 (m, 1H), 4.52 and 4.54 (AB
q, J=12.0 Hz, 2H), 7.20–7.37 ppm (m, 10H); 13C NMR (125 MHz,
CDCl3): d=13.3, 15.7, 19.2, 20.2, 20.7, 21.2, 26.1, 27.2, 27.4, 28.3, 28.5,
30.4, 32.4, 35.0, 36.0, 36.5, 38.4, 38.6, 40.1, 43.3, 47.8, 53.9, 54.5, 62.6, 69.7,
71.2, 72.9, 77.1, 78.4, 79.3, 84.2, 107.7, 126.9 (2C), 127.0, 127.3, 127.6 (2C),
128.1 (2C), 128.3 (2C), 139.2, 139.8 ppm.
ACHTUNGTRENNUNG(23R,24S)-3b,6b-Dibenzyloxy-5a-cholest-23,24-isoproylidene-16b-ol (46)
and (23S,24R)-3b,6b-dibenzyloxy-5a-cholest-23,24-isoproylidene-16b-ol
(47): A 0.1m solution of OsO4 was prepared as follows: OsO4 (51 mg,
0.2 mmol) was dissolved in tBuOH (2 mL) to which 2 drops of a 5m solu-
tion of tBuOOH was added as a stabiliser; this solution was stored at
48C until needed. Next, compound 43 (30 mg, 0.05 mmol) was dissolved
in a mixture of tBuOH/THF/H2O (10:3:1, 1 mL), then NMO (15 mg,
0.13 mmol) was added to the stirred solution. The above pre-prepared
0.1m solution of OsO4 (50 mL) was added and the mixture was stirred
overnight at 258C. Next, 5% sodium sulfite solution (5 mL) was added
and the mixture was vigorously stirred for 25 min. The mixture was ex-
tracted into CH2Cl2 (2ꢂ4 mL). The organic layers were washed with
water (5 mL), dried (MgSO4) and the solvent was removed. Purification
by flash chromatography (30% EtOAc in hexanes, silica gel 60) yielded a
mixture of the two erythro diols in approximately a 1:2 ratio (22 mg,
67%). 1H NMR (400 MHz, CDCl3) indicated dihydroxylation was suc-
cessful by the disappearance of the alkene resonances at d=5.23–
5.33 ppm and the appearance of two multiplets at d=3.27 (dd, J=3.9,
7.8 Hz) and 3.72 ppm (dd, J=4.0, 10.0 Hz) for the major isomer. For the
purpose of completing the synthesis, the mixture of diols was taken on to
the next step, which allowed separation of each diastereomer. A pure
sample of the major isomer, 23R,24S diol (44), was obtained by removal
of the acetonide from 46 (methanol, acidic resin) to give a white solid.
M.p. 178–1798C; [a]2D5 ꢀ17.5 (c 0.12 in CHCl3); 1H NMR (500 MHz,
CDCl3): d=0.63 (dt, J=3.7, 11.3 Hz, 1H), 0.83–0.92 (m, 3H), 0.86 (s,
3H), 0.87 (d, J=6.9 Hz, 3H), 1.00 (d, J=6.6 Hz, 3H), 1.02 (d, J=6.6 Hz,
3H), 1.04 (s, 3H), 1.00–1.78 (m, 12H), 1.85–2.06 (m, 5H), 2.13 (m, 1H),
2.23 (td, J=7.5, 12.8 Hz, 1H), 3.27 (dd, J=3.9, 7.8 Hz, 1H), 3.33 (tt, J=
5.2, 10.7 Hz, 1H), 3.42 (m, 1H), 3.72 (dd, J=4.0, 10.0 Hz, 1H), 4.31 and
4.56 (AB q, J=12.3 Hz, 2H), 4.39 (dt, J=4.5, 7.5 Hz, 1H), 4.52 and 4.54
(AB q, J=12.0 Hz, 2H), 7.19–7.36 ppm (m, 10H); 13C NMR (125 MHz,
CDCl3): d=13.2, 15.7, 18.7, 19.0, 20.2, 20.6, 27.0, 28.3, 29.7, 30.4, 32.4,
35.0, 36.0, 36.3, 36.6, 38.4, 39.8, 42.5, 47.8, 53.7, 54.4, 61.6, 69.8, 71.2, 71.5,
72.4, 78.4, 79.3, 79.5, 126.9 (2C), 127.0, 127.3, 127.6 (2C), 128.1 (2C),
128.3 (2C), 139.2, 139.8 ppm; elemental analysis calcd (%) for C41H60O5:
C 77.81, H 9.56; found: C 77.74, H 9.49.
A
(4a)
and
(23S,24R)-5a-cholest-23,24-isoproylidene-3b,6b,16b-triol (4b): Com-
pound 46 (46 mg, 0.088 mmol) was dissolved in THF (2 mL) and 10%
Pd/C (4 mg) was added. The suspension was stirred under an H2 atmos-
phere for 2 h, then filtered through a plug of Celite and the filter was
washed with diethyl ether (2 mL). The combined filtrates were evaporat-
ed and purified by flash chromatography (50% EtOAc in hexanes, silica
gel 60) to yield 4a as a white solid (33 mg, 78%). M.p. 182–1838C; [a]D25
+5.6 (c 0.25 in CH3OH); 1H NMR (500 MHz, CD3OD): d=0.69 (dt, J=
3.7, 11.4 Hz, 1H), 0.88 (d, J=6.6 Hz, 3H), 0.91–2.12 (m, 20H), 0.93 (s,
3H), 1.01 (d, J=6.5 Hz, 3H), 1.04 (s, 3H), 1.04 (d, J=6.4 Hz, 3H), 1.32
(s, 3H), 1.45 (s, 3H), 1.64 (td, J=3.5, 13.0 Hz, 1H), 2.21 (td, J=7.6,
12.9 Hz, 1H), 3.54 (tt, J=5.2, 10.8 Hz, 1H), 3.74 (m, 2H), 4.21 (ddd, J=
1.5, 5.8, 11.5 Hz, 1H), 4.32 ppm (dt, J=4.3, 7.6 Hz, 1H); 13C NMR
(125 MHz, CD3OD): d=13.7, 16.2, 19.1, 19.5, 20.5, 21.9, 25.9, 28.3, 28.4,
28.8, 31.4, 32.2, 36.3, 36.6, 37.0, 37.3, 39.8, 40.7, 41.5, 43.8 (2C), 55.3, 55.8,
63.7, 72.5 (2C), 72.8, 77.4, 84.7 108.8 ppm; elemental analysis calcd (%)
for C30H52O5: C 73.13, H 10.64; found: C 73.08, H 10.58.
Compound 4b was also obtained under identical conditions starting from
47 (72%). M.p. 189–1918C. [a]2D5 ꢀ7.0 (c 0.012 in CH3OH); 1H NMR
(500 MHz, CD3OD): d=0.69 (dt, J=3.6, 11.4 Hz, 1H), 0.87–2.13 (m,
22H), 0.89 (d, J=6.6 Hz, 3H), 0.93 (s, 3H), 0.98 (d, J=6.5 Hz, 3H), 1.04
(s, 1H), 1.11 (d, J=6.8 Hz, 3H), 1.30 (s, 3H), 1.43 (s, 3H), 1.64 (td, J=
3.5, 13.0 Hz, 1H), 2.21 (td, J=7.5, 12.7 Hz, 1H), 3.54 (tt, J=5.2, 10.8 Hz,
1H), 3.67 (dd, J=5.6, 8.5 Hz, 1H), 3.74 (m, 1H), 4.26 (ddd, J=2.3, 5.7,
10.1 Hz, 1H), 4.30 ppm (dt, J=4.8, 7.8 Hz, 1H); 13C NMR (125 MHz,
CD3OD): d=13.7, 16.2, 19.9, 20.4, 20.8, 21.9, 26.2, 28.6, 28.9, 30.8, 31.3,
32.2, 36.3, 36.6, 37.0, 38.1, 39.8, 40.7, 41.5, 43.8 (2C), 55.3, 55.8, 63.3, 72.5
(2C), 73.0, 79.7, 85.4, 108.5 ppm.
AHCTUNGTERG(NNUN 23R,24S)- 3b,6b,16b,23,24-Pentahydroxy-5a-cholestane ((23R,24S)-3)
and (23S,24R)-3b,6b,16b,23,24-pentahydroxy-5a-cholestane ((23S,24R)-
3): Compound 4a (8 mg, 0.012 mmol) was dissolved in MeOH (0.5 mL)
and Dowex 50-X8 acidic resin (20 mg) was added. The mixture was
stirred for 2 days at 258C, then filtered from the resin and the solvent
was removed. Purification by column chromatography (10% MeOH in
CH2Cl2, silica gel 60) gave (23R,24S)-3 as a white solid (6 mg, 75%).
M.p. 220–2228C; [a]2D5 +9.2 (c 0.6 in CH3OH); 1H NMR (500 MHz,
C5D5N): d=0.76 (dt, J=3.7, 11.3 Hz, 1H), 0.92–2.65 (m, 22H), 1.10 (s,
3H), 1.15 (d, J=6.9 Hz, 3H), 1.20 (d, J=6.5 Hz, 3H), 1.26 (d, J=6.7 Hz,
3H), 1.38 (s, 3H), 3.66 (brt, J=5.5 Hz, 1H), 3.98 (tt, J=5.2, 10.7 Hz,
1H), 4.04 (m, 1H), 4.07 (dd, J=7.4, 10.0 Hz, 1H), 4.63 ppm (dt, J=4.7,
7.2 Hz, 1H); 13C NMR (125 MHz, C5D5N): d=13.7, 16.4, 17.3, 20.6, 20.7,
21.3, 27.8, 30.1, 30.9, 32.8, 36.2, 36.7, 37.2, 39.2, 40.5, 40.6, 41.0, 42.9, 48.6,
54.4, 55.0, 62.8, 71.3, 71.5, 71.6, 72.0, 79.6 ppm; elemental analysis calcd
(%) for C27H48O5: C 71.64, H 10.69; found: C 71.11, H 10.54.
The mixture of diols obtained above (33 mg, 0.05 mmol) was dissolved in
acetone (1 mL) and a small crystal of TsOH was added. The mixture was
stirred at 258C for 1 h, then a saturated aqueous solution of Na2CO3
(3 mL) was added and the mixture was extracted into CH2Cl2 (2ꢂ3 mL).
The organic layers were separated, washed with water (3 mL), dried
(MgSO4) and evaporated. Careful flash chromatographic purification of
the residue (10% EtOAc in hexanes, silica gel 60) allowed separation of
the two erythro acetonides. The first isomer to elute was the major
23R,24S isomer as a white solid (46; 18 mg, 53%). M.p. 71–728C; [a]D25
1
ꢀ6.2 (c 0.56 in CHCl3); H NMR (500 MHz, CDCl3): d=0.63 (dt, J=4.0,
11.3 Hz, 1H), 0.86 (d, J=6.6 Hz, 3H), 0.88 (s, 3H), 0.81–2.04 (m, 20H),
0.94 (dd, J=6.6, 11.6 Hz, 1H), 1.01 (d, J=6.5 Hz, 3H), 1.01 (d, J=
6.6 Hz, 3H), 1.03 (s, 3H), 1.32 (s, 3H), 1.46 (s, 3H), 2.17 (td, J=7.6,
13.0 Hz, 1H), 3.13 (d, J=1.5 Hz, 1H), 3.33 (tt, J=5.2, 10.7 Hz, 1H), 3.42
(m, 1H), 3.71 (dd, J=6.0, 9.3 Hz, 1H), 4.11 (dd, J=6.2, 10.9 Hz, 1H),
4.29 and 4.56 (AB q, J=12.1 Hz, 2H), 4.34 (m, 1H), 4.53 and 4.55 (AB
q, J=12.0 Hz, 2H), 7.20–7.36 ppm (m, 10H); 13C NMR (125 MHz,
CDCl3): d=13.0, 15.7, 19.2, 19.4, 20.1, 20.6, 25.3, 27.4, 27.8, 27.9, 28.3,
30.4, 32.4, 34.9, 35.1, 36.0, 37.1, 38.4, 39.9, 42.4, 47.8, 53.9, 54.5, 62.1, 69.7,
71.0, 71.8, 76.7, 78.5, 79.1, 82.9, 107.6, 127.0 (3C), 127.3, 127.5 (2C), 128.1
(2C), 128.3 (2C), 139.2, 139.7 ppm; elemental analysis calcd (%) for
C44H64O5: C 78.53, H 9.59; found: C 78.75, H 9.79.
Compound (23S,24R)-3 was also obtained under identical conditions
starting from 4b (85%). M.p. 191–1928C; [a]2D5 +3.6 (c 0.52 in CH3OH);
1H NMR (500 MHz, C5D5N): d=0.75 (dt, J=3.6, 11.2 Hz, 1H), 0.89–2.23
(m, 17H), 1.11 (s, 3H), 1.20 (d, J=6.9 Hz, 3H), 1.23 (d, J=6.7 Hz, 3H),
1.34 (d, J=6.6 Hz, 3H), 1.38 (s, 3H), 2.37–2.63 (m, 4H), 2.32 (td, J=7.5,
12.7 Hz, 1H), 3.81 (dd, J=3.1, 7.2 Hz, 1H), 3.98 (tt, J=5.1, 10.5 Hz, 1H),
4.04 (m, 1H), 4.18 (m, 1H), 4.68 ppm (m, 1H); 13C NMR (125 MHz,
C5D5N): d=13.7, 16.3, 16.4, 20.8, 21.1, 21.3, 28.6, 29.9, 30.8, 32.8, 36.1,
37.2, 37.3, 39.2, 40.6, 40.9, 43.1, 43.5, 48.5, 54.5, 54.9, 63.2, 71.2, 71.4, 71.5,
72.1, 80.7 ppm.
The minor 23S,24R isomer (47) was the second isomer to elute as a col-
ourless oil (8 mg, 25%). [a]2D5 ꢀ37.9 (c 1.1 in CHCl3); 1H NMR
7590
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7578 – 7591

Total Synthesis of 23R,24S-Chiograsterol B
FULL PAPER
[29] R. O. Hutchins, I. M. Taffer, W. Burgoyne, J. Org. Chem. 1981, 46,
5214–5215.
Acknowledgements
[30] G. H. Posner, Org. React. 1975, 22, 253–400.
[31] B. H. Lipshutz, R. S. Wilhelm, D. M. Floyd, J. Am. Chem. Soc. 1981,
103, 7672–7674.
[32] B. B. Snider, B. Shi, Tetrahedron 1999, 55, 14823–14828.
[33] W. S. Yu, Z. D. Jin, J. Am. Chem. Soc. 2002, 124, 6576–6583.
[34] N. R. Schmuff, B. M. Trost, J. Org. Chem. 1983, 48, 1404–1412.
[35] Y. Yamamoto, K. Maruyama, J. Org. Chem. 1983, 48, 1564–1565.
[36] A. Yanagisawa, S. Habaue, K. Yasue, H. Yamamoto, J. Am. Chem.
Soc. 1994, 116, 6130–6141.
[37] A. Hosomi, H. Sakurai, J. Am. Chem. Soc. 1977, 99, 1673–1675.
[38] J. S. Yadav, B. V. S. Reddy, K. Sadasiv, G. Satheesh, Tetrahedron
Lett. 2002, 43, 9695–9697.
[39] M. Koreeda, Y. Tanaka, A. Schwartz, J. Org. Chem. 1980, 45, 1172–
1174.
We thank L. Lambert (Centre for Advanced Imaging, The University of
Queensland), and are grateful for funding from Australian Research
Council Linkage grant LP0453473, Australian Postgraduate Awards to
N.J.M., J.M.U.S. and V.L.C., and the United States Pharmacopeia (USP
Fellowship 2008–2009 to V.L.C.).
[1] J. Barnes, Br. J. Clin. Pharmacol. 2003, 55, 226–233.
[2] G. B. Mahady, J. Nutr. 2001, 131, 1120S-1123S.
[3] P. M. Barnes, B. Bloom, R. L. Nahin, Complementary and Alterna-
tive Medicine Use Among Adults and Children: United States 2007,
National Health Statistics Report no. 12, National Center for Health
Statistics, Hyattsville, MD, 2008.
[4] United States Department of Agriculture Natural Resources Con-
servation Service, The PLANTS Database, National Plant Data
Center, http://plants.usda.gov, 26th July 2010.
[5] A. Chevallier, The Encyclopedia of Medicinal Plants, Dorling Kind-
ersley, London, 1996.
[6] E. P. Burkhart, M. G. Jacobson, Agrofor. Syst. 2009, 76, 437–453.
[7] Analysis of the Economic Viability of Cultivating Selected Botanicals
in North Carolina, Strategic Reports, North Carolina Consortium on
Natural Medicinal Products, Raleigh, NC, 2003.
[8] M. McKenzie, A. Kirakosyan, P. B. Kaufman in Recent Advances in
Plant Biotechnology, Vol. 4 (Eds.: A. Kirakosyan, P. B. Kaufman),
Springer, Heidelberg, 2009, pp. 363–387.
[40] D. A. Evans, A. M. Golob, J. Am. Chem. Soc. 1975, 97, 4765–4766.
[41] T. F. Braish, P. L. Fuchs, Synth. Commun. 1986, 16, 111–115.
[42] K. Kanai, I. Sakamoto, S. Ogawa, T. Suami, Bull. Chem. Soc. Jpn.
1987, 60, 1529–1531.
[43] A. Bernardi, D. Arosio, L. Manzoni, F. Micheli, A. Pasquarello, P.
Seneci, J. Org. Chem. 2001, 66, 6209–6216.
[44] E. E. Van Tamelen, S. R. Zawacky, R. K. Russell, J. G. Carlson, J.
Am. Chem. Soc. 1983, 105, 142–143.
[45] P. M. Gannett, D. L. Nagel, P. J. Reilly, T. Lawson, J. Sharpe, B.
Toth, J. Org. Chem. 1988, 53, 1064–1071.
[46] R. J. Abraham, M. Canton, L. Griffiths, Magn. Reson. Chem. 2001,
39, 421–431.
[9] Hagers Handbuch der Pharmazeutischen Praxis, Vol. 5 (Eds.: F. v.
Bruchhausen, S. Ebel, E. Hackenthal, U. Holzgrabe), Springer,
Berlin, 1999.
[47] K. B. Sharpless, W. Amberg, Y. L. Bennani, G. A. Crispino, J. Har-
tung, K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, J. Org.
Chem. 1992, 57, 2768–2771.
[10] F. V. Greene, Am. J. Pharm. (1835–1936) 1878, 50, 250–253.
[11] F. V. Greene, Am. J. Pharm. (1835–1936) 1878, 50, 465–468.
[12] E. L. Cataline, D. E. Francke, J. Am. Pharm. Assoc.(1912–1977)
1942, 31, 519–521.
[13] R. E. Marker, R. B. Wagner, P. R. Ulshafer, J. Am. Chem. Soc. 1942,
64, 1283–1285.
[14] K. M. Newton, S. D. Reed, L. Grothaus, K. Ehrlich, J. Guiltinan, E.
Ludman, A. Z. LaCroix, Maturitas 2005, 52, 134–146.
[15] S. S. W. Ng, W. D. Figg, Anticancer Res. 2003, 23, 3585–3590.
[16] P. Y. Hayes, A. H. Jahidin, R. Lehmann, K. Penman, W. Kitching,
J. J. De Voss, Phytochemistry 2008, 69, 796–804.
[17] P. Y. Hayes, R. Lehmann, K. Penman, W. Kitching, J. J. De Voss,
Phytochemistry 2009, 70, 105–113.
[18] S. G. Sparg, M. E. Light, J. van Staden, J. Ethnopharmacol. 2004, 94,
219–243.
[19] C. Altona, C. A. G. Haasnoot, Org. Magn. Reson. 1980, 13, 417–
429.
[48] K. S. Jeong, P. Sjo, K. B. Sharpless, Tetrahedron Lett. 1992, 33, 3833–
3836.
[49] L. Wang, K. B. Sharpless, J. Am. Chem. Soc. 1992, 114, 7568–7570.
[50] H. Becker, K. B. Sharpless, Angew. Chem. 1996, 108, 447–449;
Angew. Chem. Int. Ed. Engl. 1996, 35, 448–451.
[51] T. Nanba, S. Kadota, T. Myahara (Kaken Pharma Co., Ltd., Tokyo,
Japan), Extraction of terpenes from Cimicifuga foetida for bone dis-
ease, JP 1995-184117 09030977, 1997.
[52] V. VanRheenen, R. C. Kelly, D. Y. Cha, Tetrahedron Lett. 1976, 17,
1973–1976.
[53] A. Kusano, M. Shibano, G. Kusano, Chem. Pharm. Bull. 1995, 43,
1167–1170.
[54] A. Kusano, M. Takahira, M. Shibano, T. Miyase, G. Kusano, Chem.
Pharm. Bull. 1999, 47, 511–516.
[55] L. Zhou, J. Yang, J. Zou, G. Tu, Chem. Pharm. Bull. 2004, 52, 622–
624.
[56] G. F. Lai, Y. F. Wang, L. M. Fan, J. X. Cao, S. D. Luo, J. Asian Nat.
Prod. Res. 2005, 7, 695–699.
[57] L. Sun, J. Yan, S. Pei, M. Qiu, Yunnan Zhiwu Yanjiu 2005, 27, 331–
336.
[20] W. A. Kçnig, I. Benecke, H. Bretting, Angew. Chem. 1981, 93, 688–
690; Angew. Chem. Int. Ed. Engl. 1981, 20, 693–694.
[21] H. M. Flowers, Adv. Carbohydr. Chem. Biochem. 1981, 39, 279–345.
[22] A. Yokosuka, Y. Mimaki, A. Sashida, J. Nat. Prod. 2000, 63, 1239–
1243.
[58] L. Zhou, J.-S. Yang, G.-Z. Tu, J.-H. Zou, Chem. Pharm. Bull. 2006,
54, 823–826.
[23] K. Takeda, H. Minato, A. Shimaoka, Chem. Commun. (London)
1968, 104–106.
[59] J. C. Gao, F. Huang, J. C. Zhang, G. Y. Zhu, M. S. Yang, P. G. Xiao,
J. Nat. Prod. 2006, 69, 1500–1502.
[24] A. D. Batcho, D. E. Berger, S. G. Davoust, P. M. Wovkulich, M. R.
Uskokovic, Helv. Chim. Acta 1981, 64, 1682–1687.
[25] M. S. Syamala, J. Das, S. Baskaran, S. Chandrasekaran, J. Org.
Chem. 1992, 57, 1928–1930.
[60] I. Izzo, M. Di Filippo, R. Napolitano, F. De Riccardis, Eur. J. Org.
Chem. 1999, 3505–3510.
[61] K. Takeda, A. Shimaoka, M. Iwasaki, H. Minato, Chem. Pharm.
Bull. 1965, 13, 691–694.
[26] J. A. R. Salvador, M. L. Sꢃe Melo, A. S. Campos Neves, Tetrahedron
Lett. 1996, 37, 687–690.
[27] M. M. Midland, Y. C. Kwon, J. Am. Chem. Soc. 1983, 105, 3725–
3727.
[62] CCDC-823483 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[28] L. K. A. Blaehr, F. Bjorkling, M. J. Calverley, E. Binderup, M. Begtr-
up, J. Org. Chem. 2003, 68, 1367–1375.
Received: February 15, 2011
Published online: May 20, 2011
Chem. Eur. J. 2011, 17, 7578 – 7591
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7591