Clemmensen reduction of diosgenin and kryptogenin: Synthesis of [16,16,22,22,23,23-2H6]-(25R)-26-hydroxycholesterol

Article abstract of DOI:10.1016/j.steroids.2004.09.009

A new experimental protocol has been established for the Clemmensen reduction of diosgenin and kryptogenin with the aim to prepare deuterated isotopomers of (25R)-26-hydroxycholesterol. Uncontrolled deuteration has been achieved from diosgenin, whereas [16,16,22,22,23,23-²H ₆]-(25R)-26-hydroxycholesterol (1) can be synthesized from kryptogenin.

Full text of DOI:10.1016/j.steroids.2004.09.009

Steroids 69 (2004) 789–794
Clemmensen reduction of diosgenin and kryptogenin: synthesis of
[16,16,22,22,23,23-2H6]-(25R)-26-hydroxycholesterol
Laura Alessandrini, Pierangela Ciuffreda∗, Enzo Santaniello, Giancarlo Terraneo
Dipartimento di Scienze Precliniche LITA Vialba, University of Milan, Via G.B. Grassi, 74-20157 Milano, Italy
Received 29 July 2004; received in revised form 9 September 2004; accepted 17 September 2004
Abstract
AnewexperimentalprotocolhasbeenestablishedfortheClemmensenreductionofdiosgeninandkryptogeninwiththeaimtopreparedeuter-
ated isotopomers of (25R)-26-hydroxycholesterol. Uncontrolled deuteration has been achieved from diosgenin, whereas [16,16,22,22,23,23-
2H6]-(25R)-26-hydroxycholesterol (1) can be synthesized from kryptogenin.
© 2004 Elsevier Inc. All rights reserved.
Keywords: Deuterated isotopomers; 26-Hydroxycholesterol; Clemmensen reduction; Kryptogenin; Diosgenin; Cholesterol metabolism
1. Introduction
using the classical Clemmensen reduction [10-13] (Fig. 2),
although the non-reproducibility of the Clemmensen reac-
tion may constitute a limitation of the approach. Further-
more, kryptogenin (3) is no longer commercially available
and the reported harsh conditions required for the Zn/HCl re-
duction cause extensive enolization of the carbonyl groups.
As a result, five to nine deuterium atoms were incorporated
into the deuterated 26-hydroxycholesterol (1) starting from 3
[14].
A recently described modification of the Clemmensen re-
duction for the synthesis of (25R)-26-hydroxycholesterol (1)
from diosgenin (2) [15] renewed our interest for the classi-
cal procedure that, if adequately controlled, could still rep-
resent an useful method for the introduction of deuterium
label into the side chain and at position 16 as well. We
present here our results on the study carried out to estab-
lish experimental condition for a controlled Clemmensen
reduction of diosgenin (2) that could be applied to the prepa-
ration of deuterated (25R)-cholest-5-en-3␤,16␤,26-triol (4)
(Scheme 1). This compound can be transformed into a deuter-
ated isotopomer of 25(R)-26-hydroxycholesterol (1) through
the procedure described by Williams et al. [15]. Applica-
tion of our protocol to kryptogenin diacetate (5) has then
led directly to (25R)-26-hydroxycholesterol (1) and related
isotopomers.
The biological importance of (25R)-26-hydroxycho-
lesterol (1), a major intermediate in the metabolic pathway
from cholesterol to bile acids [1,2], is well established and
is related to its role in the regulation of cholesterol home-
ostasis [3,4] or inhibition of HMG-CoaA reductase [5]. In
connection with studies concerning the evaluation of serum
levels of 26-hydroxycholesterol (1) in patients with pri-
mary biliary cirrhosis [6], it became necessary to prepare
deuterated 1 for intravenous infusions aimed to evaluate the
26-hydroxylation pathway in bile acid production [7]. In
a previous paper, we have prepared a tetradeuterated 26-
hydroxycholesterol, [7,7,21,21-2H4]-26-hydroxycholesterol
(1-d4), from the deuterated steroid intermediate, (20S)-
[7,7,21,21-2H4]-20-methyl-pregn-5-en-3␤,21-diol [8] cou-
pled to a suitable C-5 synthon [9] (Fig. 1).
However, for studies on the metabolism of cholesterol in-
volving the position 7, sterols labeled in the side chain or in
ring D are much more suitable. For this purpose, synthesis of
deuterated isotopomers of 1 could be realized starting from
two natural isoprenoids, diosgenin (2) and kryptogenin (3)
∗
Corresponding author. Tel.: +39 02 50319695; fax: +39 02 50319723.
E-mail address: Pierangela.Ciuffreda@unimi.it (P. Ciuffreda).
0039-128X/$ – see front matter © 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2004.09.009

790
L. Alessandrini et al. / Steroids 69 (2004) 789–794
Fig. 1. Structure of unlabeled and labeled (25R)-26-hydroxycholesterol (1) and (1-d4).
Fig. 2. Structure of diosgenin (2) and kryptogenin (3).
2. Experimental
flow rate. Column temperature was programmed from 180
to 300 ◦C at 10 ◦C/min. Mass spectral data are given as m/z
2.1. General
(relative abundance).
Melting points were recorded on Stuart Scientific SMP3
instrument and are uncorrected. All reagents were purchased
from Sigma-Aldrich. All reactions were monitored by TLC
on silica gel 60 F254 precoated plates with a fluorescent indi-
cator (Merck). Flash chromatography [16] was performed us-
ing silica gel 60 (230–400 mesh, Merck). Mass spectra were
carried out using a Hewlett Packard 5988A instrument set at
70 eV ion energy, 0.1 mA emission current and 280 ◦C trans-
fer line temperature. All synthetic compounds were analyzed
after derivatisation with trimethylsilylimidazole:piperidine
(1:1) for 10 min at room temperature to obtain the trimethylsi-
lyl (TMS) ethers. GC–MS analyses were performed on an
SPB5 (Supelco Inc.) capillary column 0.32-mm i.d., 0.25 ␮m
film thickness, 30-m long, operating at 20 ml/min helium
2.2. NMR experiments
All NMR spectra were recorded on a Bruker AM-500
spectrometer operating at 500.13 and 125.76 MHz for 1H and
13C, respectively, in CDCl3 solutions. The sample tempera-
ture was 303 K. Chemical shifts are reported as δ (ppm) rel-
ative to CHCl3 fixed at 7.24 ppm for 1H NMR spectra and to
CDCl3 fixed at 77.00 ppm for 13C NMR; coupling constants
(J) are given in hertz. NMR signals were assigned by a com-
bination of 1D 1H, 13C NMR and standard COSY and HSQC
experiments. For overlapped signals of hydrogen atoms im-
portant for the identification of deuterium position, the 1D
HOHAHA technique [17] was used to obtain the chemical
shifts and coupling constants.
Scheme 1. Synthesis of (25R)-cholest-5-en-3␤,16␤,26-triol (4) from diosgenin (2).

L. Alessandrini et al. / Steroids 69 (2004) 789–794
791
Table 1
2.3. (25R)-Cholest-5-en-3␤,16 ␤,26-triol (4)
EI spectra of undeuterated and deuterated (25R)-cholest-5-en-3␤,16 ␤,26-
triol (5)a
A mixture of diosgenin (2) (0.10 g; 0.24 mmol), Zn dust
(2.25 g, 34.4 mmol) and ethanol (15.0 ml) was refluxed for
15 min and then 37% HCl (7.0 ml) was added during a pe-
riod of 15 min. At the end of the addition, the reaction, mon-
itored by TCL (CH2Cl2/MeOH, 97:3, v/v), was cooled to
room temperature and filtered to remove Zn dust. CH2Cl2
(20.0 ml) was then added to the solution and the organic
layer was washed with water, aqueous saturated NaHCO3
and water. The organic phase was dried with Na2SO4 and the
solvent evaporated at reduced pressure. The residue was pu-
rified by flash chromatography (CH2Cl2/MeOH, 99:1, v/v)
to give the product 4 as a white solid (0.050 g, 50%). Mp
175–177 ◦C (lit. [15], Mp 172–174 ◦C); 1H NMR (CDCl3),
(:0.86(3H, s, 18-CH3), 0.89(3H, d, J = 7.0 Hz, 27-CH3), 0.96
(3H, d, J = 7.0 Hz, 21-CH3), 0.99 (3H, s, 19-CH3), 3.42 (1H,
dd, J = 5.6, 10.5 Hz, 26-Ha), 3.46 (1H, dd, J = 6.3, 10.5 Hz,
26-Hb), 3.50 (1H, tt, J = 4.9, 11.2 Hz, 3␣-H), 4.33 (1H, dt,
J = 4.2, 7.3, 16␣-H), 5.33 (1H, dd, J = 1.4, 4.9 Hz, 6-H). MS:
see Table 1.
m/z
Ion
Relative intensityb
Undeuterated
634
544
[M]+
[M–Me3SiOH]
[544-Me]
15
40
25
529
454
439
[M–(Me3SiOH)2]
[454-Me]
100
20
364
253
[M–(Me3SiOH)3]
[454-silylated side chain]
10
20
Deuterated
639
638
637
636
549
548
547
546
534
533
532
531
459
458
457
456
444
443
442
441
369
368
367
366
253
a
2H5[M]+
9
20
20
18
20
27
30
26
10
18
14
16
50
82
90
100
12
15
14
10
7
2H4[M]+
2H3[M]+
2H2[M]+
2H5[M–Me3SiOH]
2H4[M–Me3SiOH]
2H3[M–Me3SiOH]
2H2[M–Me3SiOH]
2H5[549-Me]
2H4[548-Me]
2H3[547-Me]
2H2[546-Me]
2H5[M–(Me3SiOH)2]
2H4[M–(Me3SiOH)2]
2H3[M–(Me3SiOH)2]
2H2[M–(Me3SiOH)2]
2H5[459-Me]
2.4. Deuterated (25R)-cholest-5-en-3␤,16␤,26-triol
The reaction was performed following the same condi-
tion described in the previous procedure using deuterated
reagents. From diosgenin (2) (0.1 g, 0.24 mmol) reacted with
Zn dust (2.25 g, 34.4 mmol), EtOD (15.0 ml) and 37% DCl
(7.0 ml), the deuterated compound 4 was obtained as a white
solid (0.051 g, 50%). All 1H NMR signals of hydrogen atoms
are identical to those reported for non-deuterated compound
4. Only the integration of the complex signals between 0.90
and 2.30 ppm indicates the presence of deuterium atoms. For
MS fragmentation data see Table 1.
2H4[458-Me]
2H3[457-Me]
2H2[456-Me]
2H5[M–(Me3SiOH)3]
2H4[M–(Me3SiOH)3]
2H3[M–(Me3SiOH)3]
2H2[M–(Me3SiOH)3]
[459-silylated side chain]c
9
25
14
40
Derivatized as trimethylsilyl ether.
Percentage of total ionisation.
b
c
The same fragmentation would be obtained from m/z 456, 457 and 458.
2.5. Cholest-5-en-3␤,26-diol (1)
H), 3.40 (1H, dd, J = 6.3 and 10.5 Hz, 26-Ha), 3.48 (1H, dd,
J = 5.6 and 10.5 Hz, 26-Hb), 3.49 (1H, tt, J = 4.9, 11.2 Hz, 3␣-
H), 5.33(1H, dd, J = 2.8and5.6 Hz, 6-H). 13CNMR(CDCl3),
δ: 12.5 (C 18), 17.2 (C 27), 19.4 (C 21), 20.1 (C 19), 21.8 (C
11), 24.1 (C 23), 25.0 (C 15), 28.9 (C 16), 32.4 (C 2), 32.6 (C
7), 32.6 (C 8), 34.2 (C 24), 36.1 (C 20), 36.5 (C 25), 36.8 (C
22), 37.2 (C 10), 37.9 (C 1), 40.4 (C 12), 42.8 (C 4), 42.9 (C
13), 50.8 (C 9), 56.6 (C 17), 57.5 (C 14), 69.2 (C 26), 72.5
(C 3), 122.4 (C 6), 141.4 (C 5). MS: see Table 2.
Kryptogenin diacetate (5) was prepared by an overnight
reaction from kryptogenin (3) (0.2 g, 0.46 mmol), acetic an-
hydride (0.4 ml) and pyridine (0.8 ml) in 82% yield (0.206 g).
To a solution of kryptogenin diacetate (5) (0.1 g, 0.18 mmol)
inEtOH(14.0 ml), Zndust(1.64 g, 25.0 mmol)wasaddedand
the mixture stirred and refluxed for 15 min. Then 37% HCl
(3.0 ml)wasaddeddropwiseduringaperiodof15 minandthe
reaction, monitored by TLC (petroleum ether/AcOEt, 70:30,
v/v), was slowly cooled to room temperature. After filtra-
tion of Zn dust, CH2Cl2 (20.0 ml) was added and the organic
layer was washed with water, aqueous saturated NaHCO3 and
water. The organic solution was dried with Na2SO4 and the
solvent evaporated at reduced pressure. The residue was puri-
fiedbyflashchromatography(petroleumether/AcOEt, 80:20,
v/v) to give the product 1 as a white solid (0.055 g, 76%). Mp
169–170 ◦C(fromAcOEt;lit. [15]Mp168–170 ◦C).1HNMR
(CDCl3), δ: 0.65 (3H, s, 18-CH3), 0.89 (6H, d, J = 7.0 Hz, 27-
CH3 and 21-CH3), 0.98 (3H, s, 19-CH3), 0.99–1.04 (1H, m,
24-Ha), 1.24–1.30 (1H, m, 24-Hb), 1.54–1.60 (1H, m, 25-
2.6. [16,16,22,22,23,23-2H6]-cholest-5-en-3␤,26-diol
(1-d6)
The reaction was performed following the same con-
ditions described in the previous procedure using deuter-
ated reagents. Treatment of kryptogenin diacetate (5) (0.1 g,
0.18 mmol)withZndust (1.64 g, 25.0 mmol), EtOD(15.0 ml)
and 37% DCl (3.0 ml) afforded the compound 1-d6 as white
solid (0.056 g, 76%). 1H NMR (CDCl3), δ: 0.65 (3H, s, 18-

792
L. Alessandrini et al. / Steroids 69 (2004) 789–794
Table 2
der controlled monitoring of the transformation of diosgenin
(2) and it was preferable to stop at 50–60% conversion in
order to avoid undesired products that would complicate the
final purification.
EI spectra of undeuterated (1) and deuterated (25R)-cholest-5-en-3␤,26-diol
(7)a
m/z
Ion
Relative intensityb
Undeuterated
In order to obtain the deuterated derivatives of compound
4, we performed the Clemmensen reduction under our con-
ditions using deuterated reagents. NMR spectra of the prod-
uct could only suggest that labeling was located in the side
chain and the integration of the complex signals between
0.90 and 2.30 ppm indicated involvement of positions going
from C21 to C24. Analysis of mass spectra shows that a vari-
able number of deuterium atoms were randomly incorporated
into the structure. Table 1 reports the fragmentation pattern
of unlabeled and labeled (25R)-cholest-5-en-3␤,16␤,26-triol
(4) derivatized as trimethylsilyl ether (mol. wt. 634). The
most intense peak in the non-deuterated compound is at m/z
454 representing the [M-(Me3SiOH)2] ion that becomes m/z
456 in deuterated 4 corresponding to an enrichment of two
deuterium atoms and is accompanied by close ions at m/z
457, 458, 459 of variable intensities. All other peaks show
a similar pattern and it can be concluded that a mixture of
deuterated isotopomers have been prepared from diosgenin
(2). The fragment m/z 253 in the spectra of unlabeled and
labeled 4 confirms the indications obtained by NMR about
the presence of the label in the side chain, since this ion can
be derived from M+ by loss of three Me3SiOH and of the side
chain.
In order to find a more reliable deuteration procedure, we
havealsoinvestigatedthereductionofasampleofcrudekryp-
togenin (3) available to us as a gift from Prof. Javitt and puri-
fied as the diacetate 5. This compound was efficiently trans-
formed (76% of purified product) into 26-hydroxycholesterol
(1) under controlled conditions that use a lower amount of
concentrated HCl (16 ml/mmol instead of 30 ml/mmol). Fur-
thermore, using deuterated ethanol and HCl, a deuterated
isotopomer of 1 was cleanly obtained. Comparing the mass
spectra of the above deuterated with unlabeled compound 1,
it was possible to establish that six deuterium atoms were
present in the structure (Table 2). In fact the most abundant
ions present in the spectrum of unlabeled 1 [546(43), 531(10),
456(100), 441(27), 417(30)] became [552 (45), 537(10),
462(100), 447(25), 423 (30)] with nearly the same relative
intensities. Moreover, the fragment at m/z 255 (loss of one
Me3SiOH and the silylated side chain) in non-deuterated 1
becames 257 in the deuterated isotopomer. These data high-
light that four deuterium atoms were inserted in the side chain
and two in the D ring. In order to assign the position of deu-
terium atoms, a detailed NMR study was performed. In the
1H NMR spectrum of 1 the 16-H atoms resonate within a
complex range between 1.76–1.86 and 1.45–1.60 ppm and
the presence of deuterium atoms in position 16 could be
clearly evidenced in the 13C NMR spectrum that showed the
absence of the resonance at 28.9 ppm corresponding to the
C16. Assignment of the position of deuterium atoms in the
side chain was achieved by the following experiments. De-
coupling at 0.89 ppm (21-CH3) simplified the double quartet
546
531
456
441
417
255
[M]+
[M–Me]
[M–Me3SiOH]
[456-Me]
[M-129]
43
10
100
27
30
20
[456-silylated side chain]
Deuterated
552
537
462
447
2H6[M]+
45
10
100
25
30
20
2H6[M–Me]
2H6[M–Me3SiOH]
2H6[462-Me]
423
257
2H6[M-129]
2H6[462-silylated side chain]
a
Derivatized as trimethylsilyl ether.
Percentage of total ionisation.
b
CH3), 0.89 (6H, d, J = 7.0 Hz, 27-CH3 and 21-CH3), 0.99
(3H, s, 19-CH3), 1.02 (1H, t, J = 13.3 Hz, 24-Ha), 1.35 (1H,
dq, J = 7.0 and 9.8 Hz, 20-H), 1.27 (1H, dd, J = 4.9 and
13.3 Hz, 24-Hb), 1.54–1.60 (1H, m, 25-H), 3.40 (1H, dd,
J = 6.3 and 10.5 Hz, 26-Ha), 3.48 (1H, dd, J = 5.6 and 10.5 Hz,
26-Hb), 3.49 (1H, dddd, J = 4.9, 4.9, 11.2, 11.2 Hz, 3␣-H),
5.33 (1H, dd, J = 2.8 and 5.6 Hz, 6-H). 13C NMR (CDCl3),
δ: 12.5 (C 18), 17.2 (C 27), 19.4 (C 21), 20.1 (C 19), 21.8 (C
11), 25.0 (C 15), 32.4 (C 2), 32.6 (C 7), 32.6 (C 8), 34.2 (C
24), 36.1 (C 20), 36.5 (C 25), 37.2 (C 10), 37.9 (C 1), 40.4
(C 12), 42.8 (C 4), 42.9 (C 13), 50.8 (C 9), 56.6 (C 17), 57.5
(C 14), 69.2 (C 26), 72.5 (C 3), 122.4 (C 6), 141.4 (C 5). MS:
see Table 2.
3. Results and discussion
The recent report on a new experimental protocol for the
Clemmensen reduction of diosgenin (2) [15] prompted us to
repeat the procedure prior to application to the preparation
of deuterated isotopomers of 26-hydroxycholesterol (1). The
ratio of ethanol and concentrated HCl versus diosgenin (2)
(ca. 100 ml and 75 ml per mmol) seemed too high especially
in view of a deuteration process, but we repeated the reac-
tion as described [15]. We obtained a mixture of products
that was too complicated to be purified by crystallization and
isolation of low yield of pure 5 required a careful purification
by silica gel chromatography. Therefore, we monitored the
formation of the product by TLC during the dropwise addi-
tion of HCl as described in the original procedure and found
that, after addition of 30 ml/mmol, diosgenin (2) was cleanly
converted to 4 in 50% yield. Additional HCl transformed the
remaining diosgenin (2), but yield of 4 were only marginally
increased and several side products were formed, including
chlorinated sterols. In our hands, therefore, the Clemmensen
reduction worked well with addition of concentrated HCl un-

L. Alessandrini et al. / Steroids 69 (2004) 789–794
793
Scheme 2. Synthesis of [16,16,22,22,23,23-2H6]-cholest-5-en-3␤,26-diol (1-d6) from kryptogenin diacetate (5).
(dq) at 1.35 ppm that corresponds to the 20-H to a doublet
Acknowledgement
(J20,17 = 9.8 Hz), showing that no protons are present in po-
sition 22.
This work was supported by the Ministero dell’Istruzione
dell’Universita` e della Ricerca (COFIN 2002064429).
By decoupling the 26-H (3.40 and 3.48 ppm), it was
possible to assign the 25-H resonance between 1.54 and
1.60 ppm that was used to identify one of position 24-H at
1.27 ppm. The splitting pattern of this hydrogen is a doublet
(Jgem = 13.3 Hz), which suggests that no proton in position 23
was present. Furthermore, by 1D HOHAHA experiment [17]
on 25-H the other 24-H was detected for which the splitting
pattern is a t (1.02 ppm, Jgem = 13.3 and J24,25 = 13.3 Hz) con-
firmingtheabsenceofprotonatomsinposition23. Additional
evidence of label at positions 22 and 23 was furnished by the
13C NMR spectrum of deuterated compound 1-d6 where the
lack of peaks at 24.1 ppm (C23) and 36.8 ppm (C22) was
observed. It can be therefore concluded that Clemmensen
reduction of kryptogenin (3) allows a clean preparation of
[16,16,22,22,23,23-2H6] isotopomer 1-d6 (Scheme 2). We
have, therefore, observed no introduction of label into posi-
tions C 15, C 17 and C 20 during the strongly acidic Clem-
mensen reduction conditions. This is probably due to the fact
that enolization into these positions is slow compared to the
rates of enolization at C 23 and the reduction process.
Results so far obtained may constitute a general approach
to the synthesis of deuterium-labeled 26-hydroxycholesterol
(1) required for biomedical applications. One drawback may
be related to the reproducibility of the Clemmensen reduc-
tion that depends, among other factors, on the form of zinc
to be used in the reaction. In fact, when zinc was activated
by washing with 37% HCl, water and acetone, a heptadeuter-
ated isotopomer of 1-d6 was obtained. Additionally, it should
be pointed out that, although samples of kryptogenin (3) are
no longer available, it is still possible to rely on the method
described by Barton for the conversion of diosgenin (2) into
kryptogenin (3) [18]. We have repeated this procedure and
confirmed the reported yield of 35–40% that could not be im-
proved by attempted modifications of the experimental pro-
tocol. Nonetheless, the overall 30% yield for the conversion
of diosgenin (2) to kryptogenin (3) and the following con-
trolled Clemmensen reduction of kryptogenin diacetate (5) to
hexadeuterated 1-d6 make our procedure an easy three-step
preparation of labeled 26-hydroxycholesterol. Application of
controlled conditions for the Clemmensen reaction to other
steroid ketones is under current investigation, in order to ver-
ify whether the observed selectivity for kryptogenin (3) is a
more general feature of the reaction.
References
[1] Bjoerkhem I. Mechanism of degradation of the steroid side
chain in the formation of bile acids. J Lipid Res 1992;33:455–
71.
[2] Javitt NB. Bile acid synthesis from cholesterol: regulatory and aux-
iliary pathways. FASEB J 1994;8:1308–11.
[3] Lund E, Bjorkhem I. Role of oxysteroids in the regulation of
cholesterol homeostasis:
1995;28:241–9.
[4] Wolf G. The role of oxysteroids in cholesterol homeostasis. Nutr
Rev 1999;57:196–8.
a critical evaluation. Acc Chem Res
[5] Defray R, Astruc ME, Roussillion S, Descomps B, Crastes de Paulet
A. DNA synthesis and 3-hydroxy-3-methylglutaryl CoA reductase
activity in PHA-stimulated human lymphocytes: a comparative study
of some oxysterols with special reference to side chain hydroxy-
lated derivatives. Biochem Biophys Res Commun 1982;106:362–
72.
[6] Del Puppo M, Galli Kienle M, Crosignani A, Petroni ML, Amati
B, Zuin M, Podda M. Cholesterol metabolism in primary biliary
cirrhosis during simvastatin and UDCA administration. J Lipid Res
2001;42:437–41.
[7] Duane WC, Javitt NB. 27-hydroxycholesterol: production rates in
normal human subjects. J Lipid Res 1999;40:1194–9.
[8] Ciuffreda P, Casati S, Bollini D, Santaniello E. Synthe-
sis of (20S)-[7,7,21,21-2H4]-3␤-(tert-butyldimethylsilanyloxy)-20-
methyl-pregn-5-en-21-ol, a useful intermediate for the preparation
of deuterated isotopomers of sterols. Steroids 2003;68:193–8.
[9] Ciuffreda P, Casati S, Alessandrini L, Terraneo G, Santaniello
E. Synthesis of deuterated isotopomers of 7␣ and (25 R,S)-26-
hydroxycholesterol, internal standards for in vivo determination of
the two biosynthetic pathways of bile acids. Steroids 2003;68:733–
8.
[10] Varma RK, Koreeda M, Yagen B, Nakanishi K, Caspi E, Syn-
thesis. C-25 chirality of 26-hydroxycholesterols.
1975;40:3680–6.
[11] Shoda J, Axelson M, Sjoevall J. Synthesis of potential C27-
intermediates in bile acid biosynthesis and their deuterium labeled
analogs. Steroids 1993;58:119–25.
[12] Arunachalam T, MacKoul PJ, Green NM, Caspi E. Synthesis of 26-
halo-26-(phenylseleno)- and 26-indolylcholesterol analogues. J Org
Chem 1981;46:2966–8.
[13] Ni Y, Kim H-S, Wilson WK, Kisic A, Schroepfer Jr GJ. A re-
visitation of the Clemmensen reduction of diosgenin. Characteriza-
tion of by-products and their use in the preparation of (25R)-26-
hydroxysterols. Tetrahedron Lett 1993;34:3687–90.
J Org Chem

794
L. Alessandrini et al. / Steroids 69 (2004) 789–794
[14] Javitt NB, Kok E, Lloyd J, Benscath A, Field FH. Cholest-5-ene-
3␤, 26-diol: synthesis and biomedical use of a deuterated compound.
Biom Mass Spectrom 1982;9:61–3.
[17] Davis DGD, Bax A. Simplification of 1H NMR spectra by se-
lective excitation of experimental subspectra. J Am Chem Soc
1985;107:7197–8.
[15] Williams JR, Chai DP, Wright D. Synthesis of (25R)-26-
hydroxycholesterol. Steroids 2002;67:1041–4.
[16] Still WC, Kahn M, Mitra A. Rapid chromatographic technique
for preparative separations with moderate resolution. J Org Chem
1978;43:2923–5.
[18] Barton DHR, Magnus PD, Smith G, Streckert G, Zurr D. Exper-
iments on the synthesis of tetracycline. Part XI. Oxidation of ke-
tone acetals and ethers by hydride transfer. JCS Perkin I 1971:542–
52.