Preparative method for the quantitative separation of methylenic sterols from unsaturated sterol mixtures by chemoselective hydroxymercuration- deoxymercuration

Article abstract of DOI:10.1055/s-2006-942495

An unprecedented selective and quantitative method allowing the separation of the methylenic sterols from unsaturated sterol mixtures is reported. We have extended the hydroxymercuration reaction under modified Brown's conditions to a systematic study on thirty unsaturated sterols as various sterol mixtures. A high-yielding directed chemoselective hydroxymercuration was achieved on methylenic side chain of sterols. The resulting organomercurial intermediates were easily purified by flash chromatography on silica gel column and the reactive alkenes were regenerated by deoxymercuration in a biphasic system (ethyl acetate/l M HCl, 2:1 ratio). These conditions leave the ring double bonds of steroids intact. No isomerization of methylene double bonds was observed. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2006-942495

PAPER
2753
Preparative Method for the Quantitative Separation of Methylenic Sterols
from Unsaturated Sterol Mixtures by Chemoselective Hydroxymercuration–
Deoxymercuration
Q
uantitative Sep
a
aration of
ï
S
terol
M
l
ixtur
e
a
s
El Kihel,*a Georges Charlesb
a
Centre d’Etudes et de Recherche sur le Médicament de Normandie, Université de Caen-Basse Normandie, 5 rue Vaubénard,
14032 Caen cedex, France
Fax +33(2)31931188; E-mail: laila.elkihel@unicaen.fr
Les Goumoines, Route de Bouffard, 17390 La Tremblade, France
b
Received 12 January 2006; revised 18 April 2006
Unfortunately, this method is long, and must be repeated
several times. In addition, it is expensive and imperfect.
Abstract: An unprecedented selective and quantitative method al-
lowing the separation of the methylenic sterols from unsaturated
sterol mixtures is reported. We have extended the hydroxymercura-
tion reaction under modified Brown’s conditions to a systematic
study on thirty unsaturated sterols as various sterol mixtures. A
high-yielding directed chemoselective hydroxymercuration was
achieved on methylenic side chain of sterols. The resulting organo-
mercurial intermediates were easily purified by flash chromatogra-
phy on silica gel column and the reactive alkenes were regenerated
by deoxymercuration in a biphasic system (ethyl acetate/1 M HCl,
2:1 ratio). These conditions leave the ring double bonds of steroids
intact. No isomerization of methylene double bonds was observed.
The aim of this work is to provide an unprecedented, se-
lective and preparative method for the separation of meth-
ylenesterols from unsaturated side chain sterol mixtures
by hydroxymercuration-deoxymercuration.
In previous papers the broad usefulness of the oxymercu-
ration–demercuration process has been demonstrated.17–19
The oxymercuration reaction of alkene double bonds by
mercuric acetate in aqueous tetrahydrofuran combined
with fast, in situ, reduction of the oxymercurial intermedi-
ate by sodium borohydride, provides a convenient, mild
and essentially quantitative, method to achieve the
Markonikov regioselective hydration of carbon–carbon
double bonds without rearrangement, even in presence of
other functional groups. This method is known for a long
time,20 and well documented.21 Various studies have been
accomplished on a wide variety of simple and sterically
hindered olefins. The selectivity of mercuric reactions
was improved using mercuric pivalate. The selective mer-
curation of diolefins by mercuric salt under sonochemical
activation was then studied.22 The oxymercuration–hy-
drodemercuration was explored to synthesize bioactive
compounds such as a bicyclic lactam, a versatile synthon
for the synthesis of carbocyclic nucleosides,23 and in the
industrial preparation of 1a,25-dihydroxy-22-oxavitamin
D3 (Maxacalcitol).24
Key words: alkenes, chemoselectivity, steroids, hydroxymercura-
tion, deoxymercuration
The structural complexity of sterols from plants and ani-
mals, combined with their scarcity and their interesting
biological
– inter alia antibiotic, phytohormonal,
antifeedant – properties, has induced a thriving synthetic
activity in this field.1,2 Introduction of functional groups
on steroidal side chains is crucial for the synthesis of bio-
logically active steroids, such as 25-hydroxycholesterol,3
22-hydroxylanosterol,4 22- and 24-hydroxycholesterol,5
24-aminolanostenol,6 and squalamine.7 Very few unsatur-
ated side chain sterols, are commercially available and are
used as starting material for the synthesis of biologically
active steroids.8–10 Some unsaturated side chain sterols are
commercially available as a mixture such as lanosterol
(Sigma product: 62%, admixed with lanostenol) and stig-
masterol (Sigma product: 95%, admixed with sitosterol).
The preparative separation of the sterol mixture was hard-
ly achieved by classical chromatography methods. In pre-
vious papers, the separation of olefins by chromatography
on silver nitrate impregnated silica gel columns has been
indicated.11–13 The examination of literature only indicat-
ed the existence of a wide analytic application of this
method.14,15 However, the adsorption of olefins on Ag+-
silica gel columns could lead to the formation of epoxides,
and eventually to a misinterpretation by the investigator.16
Nevertheless, it is only a preparative multigram method.
In our studies related to the reaction of mercuric acetate
with unsaturated sterols, the oxymercuration–hydrode-
mercuration was accomplished on various mixtures of un-
saturated sterols (e.g. Scheme 1). A high selective
hydroxylation by a Markovnikov addition on 24-methyl-
ene or D24(25) double bond of the side chain sterols was ob-
tained.25 Thus, our approach consists of exploring this
chemoselectivity for separation of reactive sterols under
modified Brown’s conditions: Hg(OAc)2/THF–H2O fol-
lowed by acid-induced deoxymercuration with hydro-
chloric acid using a biphasic system (ethyl acetate/1 M
HCl, 2:1).
To widen the range of substrates studied, the hydroxymer-
curation–deoxymercuration was explored on natural,
commercial and semisynthetic sterol mixtures. The natu-
ral mixtures were obtained from a plant source [Funtumia
SYNTHESIS 2006, No. 16, pp 2753–2759
Advanced online publication: 13.07.2006
DOI: 10.1055/s-2006-942495; Art ID: T00406SS
x
x.
x
x
.
2
0
0
6
© Georg Thieme Verlag Stuttgart · New York

2754
L. El Kihel, G. Charles
PAPER
+
HO
HO
H
H
lanostenol
lanosterol
1) Hg(OAc)2, THF–H2O
2) NaBH4, NaOH
chromatography
HO
H
lanostenol
OH
HO
H
25-hydroxylanostenol
Scheme 1 Hydroxylation of 5a-lanosta-8,24-en-3b-ol (lanosterol) [Sigma product: 62%, admixed with 24,25-dihydro-5a-lanost-8-en-3b-ol
(lanostenol)] under Brown conditions
elastica sp tree (Apocynaceae) leaves], and from a marine On the other hand, we have isolated the lipid extract from
invertebrate [Palythoa variabilis (genus Cnidaria)].
a marine invertebrate, Palythoa variabilis collected from
senegalese coast. Analysis of this extract by gas chroma-
tography showed a mixture of demethylsterols [cholester-
ol (7i: 4.5%) and 24-methylenecholesterol (7f: 94.5%)].
The commercial products studied were lanosterol (2b)
[Sigma product: 62%, admixed with lanostenol (2l)] and
then stigmasterol (7k) [Sigma product: 95%, admixed
with sitosterol (7n)].
The three sterol families (4,4-dimethylsterols, 4-methyl-
sterols and demethylsterols) (Figure 1) were separated as
usual from non-saponifiable lipid extract by silica gel col-
umn chromatography. 4,4-Dimethylsterols were first elut-
ed with cyclohexane–ethyl acetate (9.5:0.5), followed by
methylsterols [cyclohexane–ethyl acetate (7.5:2.5)] and
finally demethylsterols. Demethylsterols extract from
Funtumia elastica leaves contained at least seven differ- The semisynthetic sterol mixtures 7c, 7d, 7e-(E), 7e-(Z)
ent sterols (7b, 7f, 7j, 7k, 7l, 7m, 7n). The ratio of the re- and 7f, 7g, 7h-(E) and 7h-(Z), were obtained by dehydra-
sulting olefinic mixture was established by gas tion with SOCl2 requiring kinetic control, from 20-
chromatography (Table 1).
hydroxycholesteryl 3b-acetate (I),26 and 24-hydroxy-
Table 1 Gas Chromatographic Separation of Funtumia elastica Sterolsa
4,4-Dimethylsterols
4-Methylsterols
Demethylsterols
cycloartenol (1b)
2.25
29-norlanosterol (5b) 39%
cycloeucalenol (4f) 36%
cyclofuntumienol [4j-(Z)] 25%
2.00
2.36
2.66
cholesterol (7l) 8%
1.00
1.19
1.33
1.42
1.67
2.16
2.17
desmosterol (7b) 27%
campesterol (7m) 9%
24-methylenecholesterol (7f) 18%
stigmasterol (7k) 30%
sitosterol (7n) 5%
isofucosterol [7j-(Z)] 3%
a Relative retention time (min)/cholesterol (RRT). GC was carried out with a fused silica capillary column (BPX-35).
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York

PAPER
Quantitative Separation of Sterol Mixtures
2755
campesteryl 3b-acetate (II), respectively (Scheme 2 and
Scheme 3). 20-Hydroxycholesteryl 3b-acetate (I) was
prepared by a Grignard reaction between 4-methylpentyl-
magnesium bromide and pregnenolone 3b-acetate as de-
Example of cycles
4,4-Dimethylsterols
R
R
R
17
H
scribed
by
Petrow
and
Stuart-Weeb.27
24-
Hydroxycampesteryl 3b-acetate (II) was prepared ac-
cording to our procedure described in literature.25 The
proportion of different isomer alkenes resulting from de-
hydration of 20-hydroxycholesteryl 3b-acetate (I) was de-
termined by 1H NMR, according to our spectroscopic data
and those of the literature.26
R'
R'
R'
H
H
H
R' = OH: 1
R' = AcO: 1
R' = OH: 2
R' = AcO: 2
R' = OH: 3
R' = AcO: 3
R
R
R
4-Methylsterols
H
OH
H
H
R'
R'
R'
H
H
H
R' = OH: 5
R' = AcO: 5
R' = OH: 4
R' = AcO: 4
R' = OH: 6
R' = AcO: 6
R
AcO
R
I
Desmethylsterols
SOCl2, pyridine
15 min
H
H
H
H
H
C17
R'
R'
R' = OH: 7
R' = AcO: 7
+
+
R' = OH: 8
R' = AcO: 8
C17
C17
7c (Z)
7d
7c (E)
Side chain R
H
Hg(OAc)2, THF–H2O
10 h
+
H
7e
C17
C17
C17
C17
OH
a
b
AcOHg
c (Z,E)
7c (Z) + 7c (E) + 7e
+
C17
H
7d'
Chromatography
C17
C17
C17
7c (Z) + 7c (E) + 7e
d
e (Z,E)
f
unreactive sterols
H
H
H
OH
AcOHg
C17
C17
g
C17
i
h (Z,E)
(Z,E)
C17
7d'
H
EtOAc–1 M HCl, 2:1
2 h
H
H
C17
C17
C17 j (Z,E)
k
l
H
H
m
n
AcO
C17
C17
7d
Figure 1 Sterol structures studied: 4,4-Dimethylsterols: cycloar-
tenol (1b), 24-methylenecycloartanol (1f), cycloartanol (1l), lanoste-
rol (2b), 24-methylenelanostenol (2f), lanostenol (2l), agnosterol (3b),
24,25-dihydroagnosterol (3l); 4-Methylsterols: cycloeucalenol (4f),
cyclofuntumienol [4j-(Z)], 29-norlanosterol (5b), 24-dehydrolophe-
nol (6b); Demethylsterols: 25-dehydrocholesterol (7a), desmosterol
(7b), 20(22)-dehydrocholesterol (7c), 20(21)-dehydrocholesterol
(7d), 17(20)-dehydrocholesterol (7e), ergosta-5,24-dienol (7g), ergo-
sta-5,23-dienol (7h), cholesterol (7l), campesterol (7m), brassicasterol
(7i), fucosterol [7j-(E)], 24-methylenecholesterol (7f), stigmasterol
(7k), sitosterol (7n), isofucosterol [7j-(Z)], ergosterol (8i), 7-dehydro-
cholesterol (8l)
Scheme 2 Dehydration of 20-hydroxycholesteryl 3b-acetate (I) and
separation of 20(21)-dehydrocholesteryl 3b-acetate (7d)
The dehydration of the 24-hydroxycampesteryl 3b-ace-
tate (II) (see Scheme 3) gave a mixture of four unsaturat-
ed side chain sterols all described in literature.28 The
structure of the resulting olefins and their ratio in the ob-
1
tained mixture were established by H NMR spectra.
D
24(28) olefin 7f (33%) depict two characteristic vinyl hy-
drogens with d = 4.71 and 4.65. D23(24)-olefins of both Z-
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York

2756
L. El Kihel, G. Charles
PAPER
OH
The analysis was based on comparison by gas chromatog-
raphy of the initial sterols mixture with unreacted sterols.
The structure of the resulting compounds was established
by spectral comparison (NMR, mass spectra and IR) to re-
lated work on structure elucidation or to our own earlier
research.
AcO
II
The cyclic trisubstituted (D5, D7), tetrasubstituted (D8) and
dienic (D5,7, D7,9) olefins do not react under Brown’s con-
ditions. This is also illustrated in disubstituted D22E (7k),
trisubstituted D20(22) (7c), D24(28) [isofucosterol 7j-(Z)] and
tetrasubstituted, D17(20) (7e) and D24(25) (7g). Likewise all
methylenic olefins studied D20(21) (7d), D24(28) (1f, 2f, 4f,
5f, 7f) and D25(27) (7a) show a good reactivity (Scheme 4).
SOCl2, pyridine
15 min
+
+
C17
C17
7h (E)
C17
7f
7g
+
Hg(OAc)2, THF–H2O
10h
It can be noticed that 20(21)-dehydrocholesterol, which
has been isolated by reverse HPLC from Mandevilla pent-
landiana roots (Apocynaceae) in small amounts (6.8% of
non-saponifiable lipid sterols fraction) is a rare natural
product.30 In our study, this compound was obtained in
important amounts by a semisynthetic method and sepa-
rated by quantitative hydroxymercuration–deoxymercu-
ration. This compound was characterized by NMR and
mass spectra and its spectrum was compared with pub-
lished data.26
7h (Z)
C17
HO
HgOAc
+
7g + 7h (E) + 7h (Z)
C17
7f'
chromatography
7g + 7h (E) + 7h (Z)
Strikingly a high reactivity was too observed on the D24-
prenylic terminal side chain of sterols (compounds 1b, 2b,
5b, 6b, 7b). However, the separation of methylenic and
D24-prenylic terminal side chain sterols mixture failed
with this method. From the work of Tsukayama et al.,31 it
is known that, in the reaction of the alkene mixture (pre-
nylphenol and its methylene isomer) with mercuric ni-
trate, the terminal alkylmercurium ion (Scheme 4) was
quantitatively obtained and the prenylphenol was separat-
ed by silica gel column chromatography. However, it has
been reported that this separation procedure was only used
to separate the prenylphenol from the mixture of the inter-
nal and terminal alkenes. In the same pathway, we have
developed a similar approach using mercuric nitrate for
the quantitative separation of methylenic and D24-prenylic
terminal side chain sterol mixtures.
HO
HgOAc
7f'
C17
EtOAc–1 M HCl, 2:1
2h
AcO
7f
Scheme 3 Dehydration of 24-hydroxycampesteryl 3b-acetate (II)
and separation of 24-methylenecholesteryl 3b-acetate (7f)
The starting olefinic sterol mixture D25(27) (34%) and
24(25) (66%) was obtained by dehydration of 25-hydrox-
and E-isomers showed the triplets of the vinyl hydrogens
at 5.13 ppm for the 7h-(E) (20%) and 5.03 ppm for the 7h-
(Z) (10%). D24(25)-olefin showed a chemical shift as a sin-
glet at d = 1.63 for the methyl groups (9 H, CH3-25, CH3-
26 and CH3-28) of 24-methyldesmosteryl 3b-acetate (7g)
(37%). These sterols are all known.29 Each mixture of ste-
rols was treated under modified Brown’s conditions: The
hydroxymercuration was carried out with an excess of
mercuric acetate in tetrahydrofuran–water mixture (1:1).
That reaction did only concern the sterols with a reactive
double bond, the other sterols remained unchanged. After
extraction, the organomercurial intermediate was purified
by flash chromatography on silica gel or alumina column.
Unreacted sterols were eluted first, followed by the orga-
nomercurial. The reactive alkene was regenerated by hy-
drochloric acid treatment of the organomercurial in a
biphasic system (EtOAc–1 M HCl, 2:1) (Schemes 2
and 3).
D
ylanostenyl 3b-acetate (III) which was prepared accord-
HgOAc
Hg(OAc)2
HCl
THF–H2O
H3O+
C17
HO
HgCl
C17
HgCl
+
+
H2O
HOAc
+
C17
HO
C17
HCl
+
HgCl2
C17
Scheme 4 Hydroxymercuration–deoxymercuration mechanism
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York

PAPER
Quantitative Separation of Sterol Mixtures
2757
ing to our procedure.25 Dehydration of this alcohol III Selectivity between methylenic and D24-prenylic terminal
with POCl3–pyridine required thermodynamic control, side chain sterols was observed. In comparison of these
using the procedure described by Sheikh and Djerassi.26 mercuric salts, mercuric nitrate reacts more readily than
The mixture was treated with aqueous mercuric nitrate in mercuric acetate. The regeneration of methylene groups
THF–H2O at room temperature for thirty minutes to af- from the acetoxymercurial intermediate was easily ob-
ford the organomercurial compound which was separated tained after two hours, but the nitroxymercurial interme-
by flash chromatography on silica gel column and 25(27)- diate gave the olefins after 48 hours.
A high
dehydrolanost-8-enyl 3b-acetate (2a) was regenerated un- chemoselectivity by mercuric nitrate was observed on
der the same conditions by deoxymercuration (ethyl ace- methylene double bond.
tate/1 M HCl, 2:1) (see Scheme 5). This compound has
In conclusion, our method includes three phases: hy-
been identified as a natural product, and isolated from the
droxymercuration–chromatography–deoxymercuration.
saquaro cactus pollen of Cereus giganteus. GC analysis of
Thirty sterols as various mixtures have been studied. A
high-yielding directed chemoselective hydroxymercura-
tion was achieved on methylenesterols. The organomer-
curial compound was easily purified by silica gel column
chromatography and the reactive alkene was regenerated
by hydrochloric acid that induced deoxymercuration.
the 4,4-dimethylsterols fraction (300 mg from 1 kg of pol-
len grains) contained trace level of cycloartenol (1b), 24-
methylenecycloartanol (1f) and 25(27)-dehydrolanost-8-
en-3b-ol (2a).32 Our method allows to prepare this com-
pound in important amounts.
These conditions leave intact the ring double bonds of ste-
roids. No isomerization of double bond was observed un-
der these experimental conditions.
OH
In this paper, the results obtained by our method shows
that 20(21)-dehydrocholesteryl 3b-acetate (7d) and
25(27)-dehydrolanost-8-enyl 3b-acetate (2a) (rare sterols)
are obtained in important amounts. This new method, es-
tablished for working on sterols and triterpenols mixtures,
constitutes a chemical screening of compounds with a
methylene group even in presence of other functionalities.
In addition, it avoids the drawbacks of silver-based chro-
matography.
AcO
H
IIIIII
POCl3, pyridine
15 min
+
Importantly, our work suggests that this approach allows
the preparation of other methylenesterols such as natural
sterols identified in trace levels. On the basis of these re-
sults, the sterol mixtures can be used as a source of start-
ing material to obtain sterols with functional groups on the
side chain. For example, 24-methylenecholesterol is a key
intermediate of squalamine synthesis. The efficiency of
this approach to other families of unsaturated methylenic
compounds can be reasonably predicted.
C17
2b
C17
2a
Hg(NO3)2, THF–H2O
30 min
HgONO2
OH
C17
C17
chromatography
All solvents were distilled and dried prior to use. Reagents and ma-
terials were obtained from commercial suppliers and were used
without further purification. The reactions were monitored by TLC
on Kieselgel-G (Merck Si 254 F, 0.25 mm thick). The spots were
detected by spraying with H2SO4–EtOH (2:8) and heating. Melting
points were determined on a Kofler block and are uncorrected. IR
spectra were recorded on a PerkinElmer 1600 FT-IR spectrometer.
EI mass spectra were recorded on a Jeol-GCmate (GC-MS system)
spectrometer with ionization energy from 30 eV. 1H NMR spectra
were obtained in CDCl3, DMSO-d6 or CD3OD solutions at 400
MHz (Jeol JNM 400) and 13C NMR spectra at 100 MHz with the
same instrument. Chemical shifts are reported relative to TMS; J
C17
HgONO2
OH
2a'
C17
EtOAc–1 M HCl, 2:1
48h
1
values are given in Hz. 13C NMR spectra are H-decoupled. Gas
chromatography performed on a Varian 3000 gas chromatograph
analyzer. Operating conditions: column temperature, 240 °C; detec-
tor temperature, 320 °C; injector temperature, 300 °C, silica capil-
lary column (BPX-35).
2a
AcO
H
Palythoa variabilis, a marine invertebrate was collected from the
senegalese coast. Funtumia elastica sp tree leaves were collected
from Cameroun.
Scheme 5 Dehydration of 25-hydroxylanost-8-enyl 3b-acetate (III)
and separation of 25(27)-dehydrolanost-8-enyl 3b-acetate (2a)
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York

2758
L. El Kihel, G. Charles
PAPER
Extraction and Isolation Procedures of Sterols
24-Methylenecholesteryl 3b-Acetate (7f)
Plant Lipid Extract: Funtumia elastica sp tree leaves were ground
into powder (500 g) and refluxed with CH2Cl2 for 9 h. The solvent
was renewed every 3 h. The extracted solvents were combined and
concentrated by vacuum evaporation. The crude extract (30.60 g)
was then saponified with NaOH (6.20 g) in EtOH (500 mL) at r.t.
for 12 h. The solution was acidified to neutral pH with 1 M HCl at
0 °C. The solvent was evaporated and the residue was dissolved in
CHCl3 and washed with aq 5% NaHCO3, H2O, and dried (Na2SO4).
The crude product was purified by flash chromatography (eluent:
cyclohexane–EtOAc, 7:3) to give total sterols (10.2 g). The total
sterols were then separated by chromatography on silica gel col-
umn. 4,4-Dimethylsterols were first eluted with cyclohexane–
EtOAc (9.5:0.5), followed by methylsterols, which were eluted with
cyclohexane–EtOAc (7.5:2.5) and finally demethylsterols. Yields
obtained: 4,4-dimethylsterols: 2.10 g, methylsterols: 0.4 g, demeth-
ylsterols: 0.87 g (see Table 1).
Mp 139–137 °C (Lit.33 mp 138–136 °C).
IR (KBr): 3090 (C=CH2), 2950 (C–H alkane), 1730 (C=O ester),
1672 (CH=CH), 1640 (C=CH2) cm–1.
1H NMR (400 MHz, CDCl3): d = 0.68 (s, 3 H, CH3-18), 0.96 (d,
J = 6.5 Hz, 3 H, CH3-21), 1.02 (s, 3 H, CH3-19), 1.03 (d, J = 6.5 Hz,
6 H, CH3-26 and CH3-27), 1.05–1.62 (m, 17 H), 1.82–2.01 (m, 4 H),
2.03 (s, 3 H, COCH3), 2.04–2.32 (m, 5 H), 4.55–4.65 (m, 1 H, H-
3a), 4.65 and 4.71 (d, J = 1.3 Hz, 2 H, =CH2), 5.37 (br d, 1 H, H-6).
13C NMR (100 MHz, CDCl3): d = 12.0 (CH3-18), 18.7 (CH3-21),
19.3 (CH3-19), 21.4 (CH3CO), 21.1, 21.9, 22.0, 24.3, 27.8, 31.0,
31.9 (C-7 and C-8), 32.0, 33.8, 34.8, 35.8, 36.5, 37.1, 38.2, 39.9,
42.4, 50.0, 56.1, 56.9, 73.9 (CH3CO2CH), 106.0 (C-28), 122.6 (C-
6), 139.6 (C-5), 156.7 (C-24), 170.4 (C=O).
MS (EI, 30 eV): m/z (%) = 440.7 (28, [M+]), 380.3 (100, [M+ –
AcOH]), 365.2 (26, [M+ – (AcOH + CH3)]), 296.4 (49).
Marine Lipid Extract: P. variabilis was crushed (800 g) and re-
fluxed with MeOH–CH2Cl2 mixture (1:3 ratio) for 48 h. The mix-
ture was filtered and the solution was washed with H2O, dried
(Na2SO4) and evaporated to give 5.54 g of extract. The crude prod-
uct (5.54 g) was then treated with KOH (1.50 g) in EtOH (50 mL)
at r.t. for 12 h. The solution was acidified (0.1 M HCl) to neutral pH
and concentrated. The crude product was diluted with CH2Cl2 and
washed with 5% NaHCO3, H2O and dried (Na2SO4). The solution
was evaporated to give 1.60 g of saponified lipid mixture. GC anal-
ysis showed the mixture of 24-methylenecholesterol and cholesterol
(94.5:4.5 ratio).
Anal. Calcd for C30H48O2: C, 81.76.; H, 10.98. Found: C, 81.80; H,
10.20.
20(21)-Dehydrocholesteryl 3b-Acetate (7d)
Sterol mixture 7c + 7d + 7e-(Z,E) (1.2 g) was treated as above to
give 0.38 g of 7d.
7d
Mp 122–118 °C.
IR (KBr): 3009 (C=CH2), 2960 (C–H alkane), 1735 (C=O ester),
1675 (C=CH2), 819 (CH=CH) cm–1.
Sterol Mixtures by Dehydration of Hydroxysterols
1H NMR (400 MHz, CDCl3): d = 0.57 (s, 3 H, CH3-18), 086–0.88
(d, J = 6.7 Hz, 6 H, CH3-26 and CH3-27), 1.02 (s, 3 H, CH3-19),
1.1–1.90 (m, 20 H), 2.03 (s, 3 H, COCH3), 2.05–2.38 (m, 7 H),
4.55–4.65 (m, 1 H, H-3a), 4.77 and 4.86 (d, J < 1 Hz, 2 H, =CH2),
5.39 (br d, 1 H, H-6).
13C NMR (100 MHz, CDCl3): d = 12.9, 19.3, 21.0, 21.4, 22.5, 22.6,
22.7, 24.3, 25.9, 27.7, 27.9, 31.8, 32.1, 36.6, 37.0, 38.0, 38.6, 38.9,
39.2, 43.0, 50.2, 55.7, 56.6, 73.9, 122.5, 122.6, 139.7, 149.7, 170.6.
Sterol mixture 7c, 7d, 7e-(E) and 7e-(Z) was obtained by dehydra-
tion with SOCl2 from 20-hydroxycholesteryl 3b-acetate (I) accord-
ing the established literature method.26 Proportion was determined
by 1H NMR spectra: D17(20): 29%, D20(21): 30%, (E)-D20(22): 30%, (Z)-
D
20(22): 11%.
Sterol mixture 7f, 7g, 7h-(E) and 7h-(Z) was obtained by dehydra-
tion with SOCl2 from 24-hydroxycampesteryl 3b-acetate (II) ac-
cording to the published procedure.26 Proportion was determined by
1H NMR spectra: (E)-D23(24): 20%; (Z)-D23(24): 10%, D24(25): 37%,
MS (30 eV, EI): m/z (%) = 426 (12, [M+]), 411.5 (30, [M+ – CH3]),
366.5 (100, [M+
–
CH3CO2H]), 354.4 (22, [M+–
D
24(28): 33%.
(CH2CH2CH(CH3)2)]), 304,3 (11), 299 (27).
Sterol mixture 2a + 2b was obtained by dehydration with SOCl2
from 25-hydroxylanostenyl 3b-acetate (III) according to published
procedure.26 Proportion was determined by 1H NMR spectra:
Anal. Calcd for C29H46O2: C, 81.63; H, 10.87. Found: C, 81.65; H,
10.89.
D
24(25): 66%, D25(27): 34%.
24-Methylenecholesterol (7f)
Palythoa variabilis lipid extract containing 7l + 7f (1.1 g) was sa-
ponified under the above conditions to give 1 g of 7f.
Hydroxymercuration with Mercuric Acetate and Deoxymercu-
ration; General Procedure 1
Hydroxymercuration with Hg(OAc)2 of Unsaturated Side Chains of
Sterol Mixture 7f + 7g + 7h-(E) + 7h-(Z): Hg(OAc)2 (2.30 g, 7.3
mmol) in H2O (15 mL) was added dropwise to a solution of the ste-
rol mixture (0.80 g, 1.8 mmol) in THF (15 mL) and the mixture was
stirred at r.t. for 12 h. The solution was filtered, diluted with H2O
(50 mL) and extracted with CHCl3 (3 × 20 mL). The organic layer
was washed with brine, dried (Na2SO4) and evaporated. The residue
was purified by flash chromatography [cyclohexane–EtOAc (8:2)
and pure EtOAc] afforded organomercurial 7f¢ (0.38 g) and unreact-
ed sterols 7g, 7h-(E) and 7h-(Z) (0.49 g).
7f
Mp 141 °C (Lit.33 mp 142 °C).
IR (KBr): 3435 (O–H alcohol), 2928–2854 (C–H alkane) cm–1.
1H NMR (400 MHz, CDCl3): d = 0.68 (s, 3 H, CH3-18), 0.94 (d,
J = 6.6 Hz, 3 H, CH3-21), 1.01 (s, 3 H, CH3-19), 1.02 (d, J = 6.5 Hz,
6 H, CH3-26 and CH3-27), 1.05–1.62 (m, 15 H), 1.81–2.4 (m, 11 H),
3.52 (m, 1 H, H-3), 4.65 and 4.71 (d, J = 1.3 Hz, 2 H, =CH2), 5.34
(d, J6,7b = 5.1 Hz, 1 H, H-6).
13C NMR (100 MHz, CDCl3): d = 11.8, 18.7, 19.4, 21.8, 22.0, 22.7,
24.3, 28.2, 29.7, 31.0, 31.6, 31.9, 33.8, 34.7, 35.7, 36.5, 37.2, 39.8,
42.2, 42.3, 50.1, 56.0, 56.7, 71.8, 105.9, 121.7, 140.7, 156.8.
Deoxymercuration: Aq 1 M HCl (2.5 mL) was added to a solution
of organomercurial 7f¢ (0.30 g, 0.4 mmol) in EtOAc (5 mL) and the
mixture was stirred at r.t. for 2 h. The solution was filtered, diluted
with H2O (20 mL) and extracted with EtOAc (3 × 15 mL). The or-
ganic layer was washed with aq 5% NaHCO3, H2O, dried (Na2SO4)
and evaporated to give the crude product, which was purified by
chromatography on silica gel column (cyclohexane–EtOAc, 8:2) to
give 24-methylenecholesteryl 3b-acetate (7f) as a white solid; yield:
0.18 g (95%).
MS (30 eV, EI): m/z (%) = 398.5 (3, [M+]), 314.4 (28, [M+
CHC(CH2)CH(CH3)2]), 256.3 (48), 129.1 (71), 97.1 (79), 73.9
(100).
–
Anal. Calcd for C28H46O: C, 84.36; H, 11.63. Found: C, 84.39; H,
11.67.
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York

PAPER
Quantitative Separation of Sterol Mixtures
2759
Hydroxymercuration with Mercuric Nitrate and Deoxymercu-
ration; General Procedure 2
(4) Amann, A.; Ourisson, G.; Luu, B. Synthesis 1987, 696.
(5) Leoni, V.; Masterman, T.; Diczfalusy, U.; De Luca, G.;
Hillert, J.; Bjorkhem, I. Neuroscience Lett. 2002, 331, 163.
(6) Chung, S. K.; Ryoo, C. H.; Yang, H. W.; Shim, J. Y.; Kang,
M. G.; Lee, K. W.; Kang, H. Tetrahedron 1998, 54, 15899.
(7) Wehrli, S. L.; Moore, K. S.; Roder, H.; Durell, S.; Zasloff,
M. Steroids 1993, 58, 370.
(8) Jones, S. R.; Selinsky, B. S. J. Org. Chem 1998, 63, 3786.
(9) Okumura, K.; Nakamura, Y.; Takeuchi, S.; Kato, I.;
Fujimoto, Y.; Ikekawa, N. Chem. Pharm. Bull. 2003, 51,
1177.
Hydroxymercuration of Unsaturated Side Chains of Sterols 2a + 2b
with Mercuric Nitrate: A solution of Hg(NO3)2 (2.63 g, 7.7 mmol)
in H2O (25 mL) was added dropwise to a solution of the sterol mix-
ture (1.20 g, 2.6 mmol) in THF (25 mL) and the mixture was stirred
at r.t. for 30 min. The mixture was diluted with H2O (50 mL) and
extracted with CHCl3 (3 × 25 mL). The combined organic layers
were washed with brine, dried (Na2SO4) and evaporated. The resi-
due was purified by flash chromatography [cyclohexane–EtOAc
(8:2) and EtOAc–AcOH (9.6:0.4)] to afford organomercurial 2a¢
(0.78 g, 41%) and unreacted sterols as pure lanosteryl acetate (2b);
yield: 0.65 g (54%).
(10) Shu, Y.; Jones, S. R.; Kinney, W. A.; Selinsky, B. S. Steroids
2002, 67, 291.
(11) Vroman, H. E.; Cohen, C. F. J. Lipid Res. 1967, 8, 150.
(12) Nes, W. D.; Varkey, T. E.; Crump, D. R.; Gut, M. J. Org.
Chem. 1976, 41, 3429.
(13) Aitzetmueller, K.; Guaraldo Goncalves, L. A. J.
Chromatogr. 1990, 519, 349.
(14) Van Beek, T. A.; Subrtova, D. Phytochem. Anal. 1995, 6, 1.
(15) Thompson, R. Q.; Phinney, K. W.; Sander, L. C.; Welch, M.
J. Anal. Bioanal. Chem. 2005, 381, 1432.
(16) Chen, S. L.; Stein, R. A.; Mead, J. F. Chem. Phys. Lipids
1976, 16, 161.
Deoxymercuration: Aq 1 M HCl (3 mL) was added to a solution of
organomercurial 2a¢ (0.50 g, 0.7 mmol) in EtOAc (9 mL) and the
mixture was stirred at r.t. for 48 h. The mixture was filtered, diluted
with H2O (25 mL) and extracted with EtOAc (3 × 10 mL). The or-
ganic layer was washed with aq 5% NaHCO3 and H2O, dried
(Na2SO4) and evaporated to give the crude product, which was pu-
rified by chromatography on silica gel column (hexane–EtOAc,
8:2) to give 25(27)-dehydrolanostenyl 3b-acetate (2a); yield: 0.20 g
(64%); mp 139–137 °C.
IR (KBr): 3010 (C=CH2), 2959 (C–H alkane), 1736 (C=O ester),
1677 (C=CH2), 819 (CH=CH) cm–1.
(17) Brown, H. C.; Geoghehan, P. J. J. Am. Chem. Soc. 1967, 89,
1522.
1H NMR (400 MHz, CDCl3): d = 0.70 (s, 3 H, CH3-18), 0.84 (s, 3
H, CH3-29), 0.88 (s, 3 H, CH3-30), 0.89 (s, 3 H, CH3-28), 0.92 (d,
J = 6.3 Hz, 3 H, CH3-21), 1.03 (s, 3 H, CH3-19), 1.05–1.70 (m, 13
H), 1.71 (s, 3 H, CH3-26), 1.73–1.91 (m, 7 H), 2.06 (s, 3 H, COCH3),
2.08–2.19 (m, 5 H), 4.48–4.52 (m, 1 H, H-3a), 4.67 and 4.69 (d,
J = 7.6. Hz, 2 H, =CH2).
(18) Brown, H. C.; Geoghehan, P. J. J. Org. Chem. 1981, 46,
3810.
(19) Carnevale, G.; Davini, E.; Iavarone, C.; Trogolo, C. J. Org.
Chem. 1988, 53, 5343.
(20) Brown, H. C.; Geoghegan, P. J. J. Org. Chem. 1970, 35,
1844.
(21) Larock, R. C. Solvation/Demercuration Reactions in
Organic Synthesis; Springer: Berlin, 1986.
(22) Einhorn, J.; Einhorn, C.; Luche, J. L. J. Org. Chem. 1989,
54, 4479.
(23) Palmer, C. F.; McCague, R. J. Chem. Soc., Perkin Trans. 1
1998, 2977.
(24) Shimizu, H.; Shimizu, K.; Kubodera, N.; Yakushijin, K.;
Horne, D. A. Tetrahedron Lett. 2004, 45, 1347.
(25) El Kihel, L.; Charles, G.; Letourneux, Y.; Ourisson, G. C. R.
Acad. Sci. 2003, 6, 79.
(26) Sheikh, Y. M.; Djerassi, C. J. Org. Chem. 1973, 38, 3545.
(27) Petrow, V.; Stuart-Webb, I. A. J. Chem. Soc. 1956, 4675.
(28) Guo, D.; VenkaTramesh, M.; Born, W. D. Lipids 1995, 30,
203.
13C NMR (100 MHz, CDCl3): d = 15.5 (CH3-18), 16.2 (CH3-29),
17.9 (CH3-21), 18.0 (C-11), 18.9 (CH3-19), 20.8 (C-6), 20.9 (CH3-
30), 21.0 (C-32), 22.1 (CH3-28), 22.5 (CH3-26), 23.9 (C-2), 26.2 (C-
23), 27.6 (C-7), 28.0 (C-12), 29.1 (C-16), 30.8 (C-15), 35.1 (C-1),
35.7 (C-20), 36.2 (C-22), 36.7 (C-10), 37.6 (C-4), 38.0 (C-24), 44.3
(C-13), 49.5 (C-14), 50.3 (C-5 and 17), 81.3 (C-3), 109,2 (C-27),
134.1 (C-8), 134.1 (C-9), 147.2 (C-25), 171.5 (C=O).
MS (EI, 30 eV): m/z (%) = 468.3 (62.6, [M+]), 453.4 (100, [M+ –
CH3]), 393.4 (70, [M+ – (AcOH + CH3)]), 353.3 (25, [M+–
(CH3CO2H + H2C=C(CH3)CH2)]), 283.3 (13), 109.1 (56).
Anal. Calcd for C32H52O2: C, 81.99; H, 11.18. Found: C, 82.02; H,
11.21.
(29) Goad, L. J.; Akihisa, T. Lipids 1997, 30, 203.
(30) Cabrera, G.; Palermo, J. A.; Seldes, A. M.; Gros, E. G.;
Oberti, J. C. Phytochemistry 1991, 30, 1239.
(31) Tsukayama, M.; Kikuchi, M.; Kawamura, Y. Chem. Lett.
1994, 1203.
(32) Nes, W. D.; Schmidt, J. O. Phytochemistry 1988, 27, 1705.
(33) Goad, L. J.; Akihisa, T. Analysis of Sterols; Blackie
Academic and Professional: London, 1997.
References
(1) Okamura, W. H.; Aurrecoechea, J. M.; Gibbs, R. A.;
Norman, A. W. J. Org. Chem. 1989, 54, 4072; and
references cited therein.
(2) Schütz, B.; Orjala, J.; Sticher, O.; Rali, T. J. Nat. Prod. 1998,
61, 96.
(3) Morisaki, M.; Lightbourn, J. R.; Ikekawa, N. Chem. Pharm.
Bull. 1973, 21, 457.
Synthesis 2006, No. 16, 2753–2759 © Thieme Stuttgart · New York