First chemical synthesis, aggregation behavior and cholesterol solubilization properties of pythocholic acid and 16α-hydroxycholic acid

Article abstract of DOI:10.1002/ejoc.200700211

The first chemical synthesis of pythocholic acid, a major component of python's bile, and 3α,7α,12α,16α-tetrahydroxy-5β- cholan-24-oic acid (16α-hydroxycholic acid), a minor component in Shoebill stork's bile, was achieved starting from readily available cholic acid in overall yields of 5% and 5.5%, respectively, through a common intermediate. A biomimetic template-directed remote functionalization strategy was utilized to selectively functionalize C-16 of the steroid skeleton. This synthesis involves a series of regio and chemoselective transformations. Pythocholic acid showed unusually low critical micellar concentration (CMC) with high cholesterol solubilization ability. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200700211

FULL PAPER
DOI: 10.1002/ejoc.200700211
First Chemical Synthesis, Aggregation Behavior and Cholesterol Solubilization
Properties of Pythocholic Acid and 16α-Hydroxycholic Acid
Nonappa[a] and Uday Maitra*[a,b]
Keywords: Pythocholic acid / Oxidation / Critical micellar concentration / Regio and chemoselective / Remote functio-
nalization
The first chemical synthesis of pythocholic acid, a major com-
ponent of python’s bile, and 3α,7α,12α,16α-tetrahydroxy-5β-
lized to selectively functionalize C-16 of the steroid skeleton.
This synthesis involves a series of regio and chemoselective
cholan-24-oic acid (16α-hydroxycholic acid), a minor compo- transformations. Pythocholic acid showed unusually low criti-
nent in Shoebill stork’s bile, was achieved starting from read-
ily available cholic acid in overall yields of 5% and 5.5%,
respectively, through a common intermediate. A biomimetic
template-directed remote functionalization strategy was uti-
cal micellar concentration (CMC) with high cholesterol solu-
bilization ability.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
lipoprotein (HDL) and low density lipoprotein (LDL) me-
tabolism.[4a–4c] Bile acids secreted by male sea lamphrey act
as sex pheromones.[4b]
Introduction
Bile acid science with a history of more than a century
and continuing importance in biology and medicine,[1] has
gained considerable attention in supramolecular and mate-
rials chemistry in recent years. Bile is produced by all ver-
tebrates, and bile salts are one of the most important physi-
ological constituents. Structurally modified bile acids have
pharmacological potential to act as carriers of liver-specific
drugs, absorption enhancers and as cholesterol lowering
agents.[1] Bile salts are natural biosurfactants which act as
solubilizer and emulsifier for cholesterol, lipids and proteins
in the intestine. They form mixed micelles with cholesterol/
lipids/fats in the intestine to enable fat digestion and ab-
sorption through the intestinal wall.[1,2]
Bile acids in vertebrates differ mainly in three respects: (i)
the side-chain structure, (ii) the distribution of the number,
position and stereochemistry of the hydroxyl groups on the
steroid nucleus and (iii) the fusion of A/B rings (cis or
trans).[3] In contrast to most of the small molecules found
in vertebrates, bile acid structures are strikingly diverse.[1,3]
Recent literature reports show that bile acids act as po-
tential ligands for nuclear hormone receptors like farnesoid
X receptor (FXR). There is a growing body of evidence
that bile salts are involved in the control of high-density
The most abundant bile salts in humans are cholate,
chenodeoxycholate and deoxycholate, and they are nor-
mally conjugated with glycine (75%) and taurine (25%).
The 15-and 16-hydroxy derivatives of chenodeoxycholic
acid and deoxycholic acid are of sustained interest in syn-
thetic, biological, physiological and metabolic studies be-
cause they are major bile acids in some vertebrates.[3] An
unusual 16α-hydroxy bile acid was reported and structurally
identified many decades ago from the family of snakes
Biodae.[5] It was named Pythocholic acid 1 (Figure 1) to sig-
nify that this class was a major component in python’s bile.
Certain primitive snakes (pythons and boas) including
Cylindrophis all have pythocholic acid (3α,12α,16α-trihy-
droxy-5β-cholan-24-oic acid) in their bile, with the amount
varying from 30% to 90% of the total bile acid. The biosyn-
thesis of pythocholic acid in pythons was reported in 1960
by Bergström et al.[6] Using extensive labelling studies this
group concluded that deoxycholic acid undergoes 16α-hy-
droxylation by the action of a 16α-hydroxylase enzyme.
Over the last five decades there has been no work on a
chemical synthesis of pythocholic acid 1. An attempt to ob-
tain pythocholic acid using ferro-ascorbate system resulted
in the formation of 15-hydroxy bile acid.[7] Thus the synthe-
sis of pythocholic acid and the study of its cholanological
properties remained unexplored. Hagey et al. recently re-
ported the existence of 3α,7α,12α,16α-tetrahydroxy-5β-cho-
lan-24-oic acid (16α-hydroxycholic acid, 2) which is a minor
component (1.6%) in Shoebill stork’s bile.[8] No chemical
synthesis has been reported in the literature for the tetrahy-
droxy bile acid 2 as well.
[a] Department of Organic Chemistry, Indian Institute of Science,
Bangalore 12, India
Fax: +91-80-23600529
E-mail: maitra@orgchem.iisc.ernet.in
[b] Chemical Biology Unit, Jawaharlal Nehru Centre for Advanced
Scientific Research,
Bangalore, India
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Eur. J. Org. Chem. 2007, 3331–3336
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3331

Nonappa, U. Maitra
FULL PAPER
Figure 1. Pythocholic acid (1) and 16α-hydroxycholic acid (2).
Our interest in the synthesis of uncommon bile acids,[9]
led us to explore a simple synthesis of both 1 and 2. We
now report the first chemical synthesis of both these bile
acids via a common intermediate involving a series of regio
and chemoselective transformations. Some cholanological
(the term “cholanology” refers to the science of bile ac-
ids[1b]) properties of these two bile acids are also reported,
and compared with other more common bile acids.
Results and Discussion
We adopted a template-directed biomimetic remote func-
tionalization strategy to control the chemical selectivity.[10]
The methyl ester of cholic acid 4 was selectively acetylated
at the 3-and 12-positions using acetic anhydride and anhy-
drous potassium acetate in refluxing toluene.[11] Esterifica-
tion of 5 with 3-iodobenzoyl chloride resulted in 6.[12] It
was crystallized using ethyl acetate in petroleum ether (3:1
v/v) and characterized by spectroscopy and single-crystal
X-ray technique (Figure 2, a).
It is interesting to note that compound 6 is a hydrophobic
bile acid derivative. Supramolecular assemblies of bile acids
are generally directed by hydrophobic interactions and hy-
drogen bonding.[13] A detailed analysis of the packing of
the molecules in the crystal revealed the existence of C–
iodine···Ar interactions (Figure 2, b). This type of C–
halogen···Ar interactions is well documented in the litera-
ture.[14] A variety of proteins and organic compounds have
been reported with these types of interactions. In a majority
of cases the halogen is either fluorine or chlorine, but there
are a few examples with iodine also.[14] C–halogen···Ar in-
teractions are favorable when the distance between halogen
and the interacting carbon/centroid of a π cloud is less than
or equal to the sum of van der Waals radii (ca. 3.68 Å for
C–I). A distance of 3.65 Å was observed between the iodine
and the centroid of aromatic ring with an angle (ЄC–I···
centroid of the aromatic ring) of 139° (Figure 2, b). An
interesting zig-zag arrangement of 3-iodoaryl group can be
clearly observed when the steroidal skeleton is removed
(Figure 2, c) (Scheme 1).
Photochemical reaction of template 6 in dichlorometh-
ane using iodobenzene dichloride (3.0 equiv.) in the pres-
ence of tBuOH (0.3 ) selectively chlorinated C-17. Chlori-
nation was not very facile (50% yield) presumably because
of the deactivating effect of the 12-acetoxy group.[10b] Dehy-
drochlorination in refluxing pyridine followed by hydrobor-
Figure 2. (a) ORTEP diagram of single-crystal X-ray structure; (b)
GS view of C–iodine···Ar interactions (c) Packing pattern of 3-
iodoaryl moiety of compound 6 using mercury 1.4.1.
ation–oxidation (BH3·Me2S in THF) of the resulting olefin phasic reaction using TEMPO-mediated N-chlorosuc-
8 furnished the pentahydroxy steroid 9. Our next task was cinimide oxidation with BnEt3N+Cl– as a phase-transfer
to selectively oxidize the side chain primary hydroxyl group catalyst was unsuccessful.[15] Eventually, TEMPO-mediated
in the presence of four secondary hydroxyl groups. A bi- oxidation using CuCl in DMF under an oxygen atmo-
3332
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3331–3336

Pythocholic Acid
FULL PAPER
more deshielded (δ = 76 ppm) compared to the other C–O
carbon atoms.
Since the methyl ester peracetate of compound 2 has
been reported in the literature,[8] synthetic 2 was converted
into the same derivative 13. Although the observed chemi-
cal shifts of the angular methyl groups of the synthetic sam-
ple are consistent with the predicted values, they differ from
the data reported in the literature for the biological sam-
ple.[8] We can now confirm that our data are correct (see
electronic supporting information) since compound 13 pre-
pared from the biological sample was present only as a
minor product with other impurities in that sample (Fig-
ure 3).[19]
Figure 3. Pythocholic lactone and methyl 3α,7α,12α,16α-tetra-
cetoxy-5β-cholan-24-oate.
The aggregation properties of bile salts have been studied
extensively by various methods and this property of pytho-
cholate and 16α-hydroxycholate in aqueous medium were
studied using pyrene as a fluorescent probe. The ratio of
the two vibronic bands (I3/I1) in the fluorescence spectrum
is indicative of the polarity experienced by the probe solubi-
lized in micellar aggregates.[20] Using this technique we have
measured the CMC values of cholate, deoxycholate, cheno-
deoxycholate (dihydroxy bile salts) and pythocholate. The
CMC of pythocholate was 3 m where as those of deoxy-
cholate, chenodeoxycholate, cholate and 16α-hydroxychol-
ate were 5, 5, 13.5 and 14 m, respectively, under identical
experimental conditions (Figure 4). It is interesting to note
that among the trihydroxy bile acids the aggregation behav-
ior of pythocholate is similar to deoxycholate and differs
considerably from other trihydroxy bile salts (cholate and
avicholate). It thus appears that the 16-hydroxy group does
not significantly affect the aggregation/cholesterol solubili-
zation in the absence of the 7-OH group.
Scheme 1. Synthesis of 1 and 2 from cholic acid.
sphere[16] led to the selective oxidation of the primary
alcohol and resulted in the formation of ε-lactone 10.
It is known in the literature that 16α-hydroxy bile acids
undergo facile latconization.[5] This transformation was ob-
served in several instances – the high resolution mass spec-
tra of both bile acids showed m/z peaks of lactones with
significant intensities.
The selective oxidation of the 7α-OH group of lactone
10 without affecting the other two secondary hydroxyl
groups was readily achieved using N-bromosuccinimide in
25:1 (v/v) acetone/water.[17] 7-ketolactone 11 was success-
fully reduced using the Huang–Minlon modification of the
Wolf–Kishner reduction[18] to furnish pythocholic acid 1 in
an overall yield of 5% from cholic acid. Along with pytho-
cholic acid, pythocholic lactone 12 was also isolated as a
Cholesterol solubilization is another significant cholanol-
minor product and characterized by spectral techniques. On ogical property of bile salts,[21] and this property of pytho-
the other hand, the treatment of lactone 10 with 5% KOH cholate and tetrahydroxy bile salts was also evaluated in a
in MeOH and acidic work up furnished tetrahydroxy bile preliminary experiment using only bile salts. For this study,
acid 2 in overall yield of 5.5% from cholic acid. In order to anhydrous cholesterol was stirred with bile salt solution in
assign the C-16 signal in 13C NMR we carried out 2D- carbonate–hydrogen carbonate buffer (pH10.0) for 24 h at
NMR studies on compound 2 using [D6]DMSO/CDCl3 37 °C. After filtration, the solubilized cholesterol was quan-
(1:9, v/v) (see electronic supporting information). Interest- tified using a commercially available enzymatic assay. The
ingly, in many steroidal and other molecules with a signifi- maximum aqueous solubility of cholesterol was found to be
cant amount of hydrocarbon units one often finds fewer much higher with pythocholate (1.8 m, bile salt:choles-
signals than the number of carbons due to “accidental de- terol ≈ 28:1) than cholate (0.8 m, bile salt: cholesterol ca.
generacy of 13C signals” which is common for steroids and 63:1) and 16α-hydroxycholate (0.7 m, bile salt:cholesterol
triterpenes. Heteronuclear single quantum correlation ca. 70:1). This solubilization ability of pythocholate and the
(HSQC) experiments suggest that the signal from C-16 is 16α-hydroxycholate are consistent with their CMC values.
Eur. J. Org. Chem. 2007, 3331–3336
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3333

Nonappa, U. Maitra
FULL PAPER
Methyl 3α,7α,12α-Trihydroxy-5β-cholan-24-oate (4): A water-free
solution of HCl in methanol was generated by adding CH3COCl
(1 mL, 14.5 mmol) to methanol (50 mL) at 0 °C. Cholic acid
(10.0 g, 24.5 mmol) was added and the mixture was stirred at room
temperature for 10 h. After removal of all volatile components the
residue was dissolved in EtOAc (100 mL). It was washed with satd.
aq. NaHCO3 (2ϫ30 mL) and water (2ϫ25 mL). The organic layer
was separated, dried with anhyd. Na2SO4, filtered and volatile sub-
stances were removed under reduced pressure. The crude product
was recrystallized from hot methanol (25 mL) to afford 9.9 g (95%)
of compound 4; m.p. 153–155 °C (ref.[23] m.p. 155–156 °C). 1H
NMR (300 MHz, CDCl3): δ = 3.97 (br. s, 1 H, 12β-H), 3.85 (br. d,
J = 2.4 Hz, 7β-H), 3.66 (s, 3 H, CO2Me), 3.49–3.39 (br. m, 1 H,
3β-H), 2.43–1.08 (m, steroidal CH2), 0.98 (d, J = 6 Hz, 3 H, 21-
H), 0.89 (s, 3 H, 19-H), 0.68 (s, 3 H, 18-H) ppm. 13C NMR
(75 MHz, CDCl3): δ = 174.8, 73.0, 71.9, 68.5, 51.5, 47.0, 46.5, 41.7,
41.5, 39.6, 39.5, 35.2, 34.7, 31.1, 30.9, 30.4, 28.2, 26.4, 23.2, 22.5,
17.3, 12.5 ppm. IR (CHCl ): ν = 3396, 1737 cm–1.
˜
3
Methyl 3α,12α-Diacetoxy-7α-hydroxy-5β-cholan-24-oate (5): Acetic
anhydride (3.75 mL, 39.7 mmol) and anhyd. potassium acetate
(5.0 g, 50.9 mmol) were added to a stirred solution of compound 4
(5.0 g, 11.8 mmol) in dry toluene (40 mL). The mixture was re-
fluxed for 8 h and quenched with water (20 mL). After washing the
organic layer with water (20 mL) and drying over anhyd. Na2SO4,
the volatile components were removed under reduced pressure. The
crude product was purified on a silica column (3ϫ20 cm) using 3–
5% ethyl acetate in chloroform to afford 3.0 g (50%) of compound
5, which was recrystallized using 10% ethyl acetate in petroleum
ether; m.p. 142–144 °C (ref.[24] m.p. 143–145 °C). 1H NMR
(300 MHz, CDCl3): δ = 5.09 (br. s, 1 H, 12β-H) ppm. 4.56 (br. m,
1 H, 3β-H), 3.88 (br. s, 1 H, 7β-H), 3.66 (s, 3 H, CO2Me), 2.11 (s,
3 H, 12-OAc), 2.06 (s, 3 H, 3-OAc), 1.10–2.33 (m, steroidal H),
0.89 (s, 3 H, 19-H), 0.81 (d, J = 6.6 Hz, 3 H, 21-H), 0.74 (s, 3 H,
18-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 174.5, 170.6, 170.7,
75.5, 74.2, 68.1, 51.5, 47.5, 45.1, 43.5, 41.1, 39.2, 35.3, 34.7, 34.6,
34.4, 30.9, 30.8, 27.7, 27.3, 25.5, 22.6, 21.5, 21.4, 17.5, 12.3 ppm.
Figure 4. The ratio of vibronic bands (I3/I1) of pyrene fluorescence
as a function of bile salt concentration. Concentration of pyrene
was ca.1 µ in tris buffer at pH9. Pythocholate (black circles), de-
oxycholate (black triangles), chenodeoxycholate (black diamonds),
cholate (open triangles), 16α-hydroxycholate (grey triangles), avich-
olate (black squares).
Conclusions
We have developed a synthetic route for chemical synthe-
sis of pythocholic acid and 16α-hydroxycholic acid. To the
best of our knowledge this is the first chemical synthesis of
both the bile acids. Unusual aggregation behavior and high
cholesterol solubilization ability of pythocholic acid suggest
that the number, position and distribution of hydroxyl
groups have remarkable effect on the physicochemical prop-
erties of bile acids. It will be interesting to physicochemi-
cally characterize other rare bile acids in order to under-
stand the structure–property landscape.
IR (CHCl ): ν = 3126, 1734 cm–1. [α]24 = +88 (c = 2, EtOH).
˜
3
D
HRMS: m/z calcd. for C29H46O7 + K+: 545.2881; found: 545.2888.
C29H46O7 (506.32): calcd. C 68.73, H 9.15; found C 68.85, H 9.16.
Methyl
3α,12α-Diacetoxy-7α-(3Ј-iodobenzoyl)oxy-5β-cholan-24-
oate (6): 3-Iodobenzoyl chloride was prepared by adding oxalyl
chloride (1.6 mL, 18.3 mmol) and dry DMF (50 µL) to a stirred
solution of 3-iodobenzoic acid (2.0 g, 8.0 mmol) in dichlorometh-
ane (15 mL). After stirring at room temperature for 30 min the vol-
atile components were removed under reduced pressure, the crude
3-iodobenzoyl chloride was dissolved in toluene (8 mL), and added
to a stirred solution of compound 5 in toluene (8 mL) containing
CaH2 (1.2 g, 28.5 mmol) and BnEt3N+Cl– (200 mg, 0.9 mmol). The
reaction mixture was refluxed for 16 h and the volatile components
were removed under reduced pressure. The residue was dissolved
in chloroform (15 mL), filtered through celite, washed with satd.
aq. NaHCO3 (2ϫ10 mL), water (10 mL) and dried with anhyd.
Na2SO4. After removing the volatile components under reduced
pressure the crude product was purified on a silica column using
15–20% ethyl acetate in petroleum ether to afford 1.5 g (51%) of
compound 6 as a white solid, which was recrystallized using 20%
ethyl acetate in petroleum ether; m.p. 150–152 °C. 1H NMR
(300 MHz, CDCl3): δ = 8.38 (t, J = 1.5 Hz, 1.8 Hz, 1 H, 2Ј-H),
8.05 (d, J = 7.8 Hz, 1 H, 6Ј-H), 7.92 (d, J = 8.1, Hz, 1 H, 4Ј-H),
7.24 (d, J = 7.8 Hz, 1 H, 5Ј-H), 5.2 (br. s, 1 H), 5.14 (br. s, 1 H, 12
β-H), 4.50 (br. m, 1 H, 3β-H), 3.63 (s, 3 H, CO2Me), 2.19 (s, 3 H,
12-OAc), 2.03, (s, 3 H, 3-OAc), 0.97 (s, 3 H, 19-H), 0.82 (d, J =
6.6 Hz, 3 H, 21-H), 0.75 (s, 3 H, 18-H) ppm. 13C NMR (75 MHz,
Experimental Section
General Methods: All reactions were carried out in oven-dried
glassware. TLC was checked on pre-coated plates (0.25 mm silica
gel with fluorescent UV254) obtained from Aldrich. After elution
the plates were developed using Liberman–Buchard reagent. Com-
mercial grade solvents were distilled prior to use in column
chromatography. Silica gel (100–200 mesh) columns were run under
gravity. Cholic acid (Fluka, Ͼ99%), 3-iodobenzoic acid (Fluka,
Ͼ98%), TEMPO (Aldrich, 98%), BnEt3N+Cl– (Aldrich, 99%), N-
bromosuccinimide (Aldrich, 99%), BH3·Me2S (Fluka, ca. 95% in
dimethyl sulfide), CuCl (S. D. fine, 96%) and CaH2 (Aldrich, 95%)
were obtained from commercial sources. Iodobenzene dichloride
was prepared from iodobenzene (Lancaster, 98%) using a known
procedure.[22] 1H and 13C NMR was recorded on a JEOL 300 MHz
spectrometer. Infrared spectra were recorded on a JASCO spectro-
photometer, either by making a film of the compounds on a NaCl
plate from a chloroform solution or using KBr pellets. ESI-QTOF
MS was recorded on a Micromass Q-TOF micro Mass Spectrome-
ter. CCDC-617607 (for 6) contains the supplementary crystallo-
graphic 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.
3334
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3331–3336

Pythocholic Acid
FULL PAPER
CDCl3): δ = 174.5, 170.7, 170.2, 163.7, 141.6, 138.6, 132.7, 130.2,
433.2930; found: 433.2930. C24H42O5·H2O (428.31): calcd. C 67.24,
128.9, 93.8, 75.2, 73.5, 71.9, 51.5, 47.3, 44.9, 43.1, 40.5, 38.2, 34.9, H 10.35; found C 67.28, H 10.07.
34.5, 34.3, 31.3, 30.8, 30.7, 28.6, 27.1, 26.7, 25.2, 22.9, 22.4, 21.5,
3α,7α,12α-Trihydroxy-5β-cholane O-24,16α-Lactone (10): CuCl
21.5, 17.5, 12.1 ppm. IR (CHCl ): ν = 1734 cm–1. [α]24 = +90 (c =
˜
3
D
(1.6 mg, 0.016 mmol) was added to a solution containing TEMPO
(2.5 mg, 0.016 mmol) and compound 9 (20 mg, 0.04 mmol) in dry
DMF (1 mL) and the mixture was stirred at room temperature for
3 h under an oxygen atmosphere. The reaction mixture was diluted
by adding diethyl ether (8 mL), washed with satd. aq. CuSO4
(4 mL) and water (4 mL). The organic layer was dried with anhyd.
Na2SO4 and the volatile components were removed under reduced
pressure. The crude product was purified on a silica column
(1ϫ12 cm) with 90–100% EtOAc/petroleum ether to afford
16.4 mg (78%) of compound 10 as a solid; m.p. 277–279 °C. 1H
NMR (300 MHz, CDCl3): δ = 4.63 (t, J = 7.5 Hz, 1 H, 16β-H),
3.88 (br. s, 2 H, 12β-H & 7β-H), 3.42 (br., m, 1 H, 3β-H), 2.04–
1.00 (m, steroidal CH2), 1.09 (d, 3 H, 21-H), 0.87 (s, 3 H, 19-H),
0.76 (s, 3 H, 18-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 176.3,
84.1, 72.6, 71.7, 68.0, 55.9, 48.7, 41.3, 39.3, 38.9, 38.5, 35.9, 35.1,
34.5, 34.3, 33.9, 32.9, 31.5, 30.2, 27.3, 26.2, 22.1, 19.3, 13.3 ppm.
0.5, EtOH). HRMS: m/z calcd. for C36H49IO8 + Na+: 759.2370;
found: 759.2350. C36H49IO8 (736.24): calcd. C 58.70, H 6.70; found
C 58.98, H 6.77.
Methyl 3α,12α-Diacetoxy-7α-(3Ј-iodobenzoyl)oxy-5β-chol-16-en-24-
oate (8): Iodobenzene dichloride (1.5 g, 5.4 mmol) was added to
a solution of compound 6 (1.5 g, 2.0 mmol) in dichloromethane
(150 mL) containing tBuOH (5 mL, 52 mmol) and deoxygenated
by bubbling dry N2 for 3 min. The mixture was irradiated with two
tungsten lamps (200 W each, placed 14 cm from the reaction flask)
at 0 °C (ice-bath) for 2 h. A spot more polar than 6 on the TLC
plate indicated the progress of the reaction. After removing the
volatile components under reduced pressure the crude product was
purified quickly on a silica column (3ϫ16 cm) using 20–30% ethyl
acetate in petroleum ether to yield 1.0 g of the 17-chlorosteroid,
which was dissolved in 30 mL of dry pyridine and refluxed for 14 h.
No major side products were observed during photochemical reac-
tion and unreacted starting material was recovered. Pyridine was
removed under reduced pressure and the crude product was puri-
fied on a silica column (3ϫ12 cm) using 15–20% ethyl acetate in
petroleum ether to afford 0.70 g (50% from 6) of compound 8 as
IR (CHCl ): ν = 3430, 1713, 1704 cm–1. [α]24 = +28 (c 2, EtOH).
˜
3
D
HRMS: m/z calcd. for C24H38O5+ Na+: 429.2617; found: 429.2595.
calcd.
for
C24H38O5+K+:
445.2356;
found
445.2365.
C24H38O5·H2O (424.28): calcd. C 67.88, H 9.50; found C 67.89, H
9.15.
1
a white solid; m.p. 140–142 °C. H NMR (300 MHz, CDCl3): δ =
3α,7α,12α,16α-Tetrahydroxy-5β-cholan-24-oic Acid (2): Compound
10 (40.0 mg, 0.098 mmol) was dissolved in 5% KOH in MeOH
(5 mL) and stirred at room temp. for 12 h. After removing the vola-
tile components under reduced pressure, the solid residue was dis-
solved in water (10 mL) and neutralized by adding 1  HCl in an
ice-water bath with stirring and extracted using ethyl acetate
(2ϫ10 mL). The organic layer was dried with anhyd. Na2SO4, vol-
atile components were removed under reduced pressure to afford
39.6 mg (95%) of compound 2 as a white solid; m.p. 205–206 °C.
1H NMR (300 MHz, [D6]acetone): δ = 3.99 (t, J = 7.5 Hz, 1 H,
16β-H), 3.89 (br. s, 1 H, 12β-H), 3.79 (br. q, 1 H, 7β-H), 3.35–3.06
(br. m, 1 H, 3β-H), 1.02 (d, J = 6.6 Hz, 3 H, 21-H), 0.89 (s, 3 H,
8.37 (br. s, 1 H, 2Ј-H), 8.04 (d, J = 7.8 Hz, 1 H, 6Ј-H), 7.91 (d, J
= 7.5 Hz, 1 H, 4Ј-H), 7.22 (d, J = 8.1 Hz, 1 H, 5Ј-H), 5.32 (s, 1 H,
12β-H), 5.29 (br. d, 1 H, 16-H), 5.10 (s, 1 H, 7β-H), 4.51 (m, 1 H,
3β-H), 3.65 (s, 3 H, CO2Me), 2.17 (s, 3 H, 12-OAc), 2.07 (s, 3 H,
3-OAc), 2.04–1.00 (m, steroidal CH2), 1.01 (s, 3 H, 19-H), 0.93 (d,
J = 6.3 Hz, 3 H, 21-H),87 (s, 3 H, 18-H) ppm. 13C NMR (CDCl3,
75 MHz): δ = 174.1, 170.6, 170.5, 163.8, 156.3, 141.6, 138.6, 132.6,
130.2, 128.9, 123.3, 93.8, 73.4, 72.9, 51.5, 51.1, 43.3, 40.7, 3.8, 35.0,
34.6, 32.6, 31.9, 31.4, 30.7, 29.9, 26.7, 25.6, 22.5, 21.7, 21.5, 20.5,
17.2 ppm. IR (CHCl ): ν = 1735, 1250 cm–1. [α]24 = +92.0 (c = 2,
˜
3
D
EtOH). HRMS: m/z calcd. for C36H47IO8 + Na+: 757.2213; found:
757.2203. C36H47IO8 (734.23): calcd. C 58.83, H 6.45; found C
58.55, H 6.61.
19-H), 0.72 (s, 3 H, 18-H) ppm. 1H NMR (300 MHz, CDCl3
+
10% [D6]DMSO): δ = 3.99 (t, J = 6.3 Hz, 1 H, 16β-H), 3.88 (br. s,
1 H, 12β-H), 3.81 (br. s, 1 H, 7β-H), 3.47–3.37 (br. m, 1 H, 3β-H),
0.98 (d, J = 6 Hz, 3 H, 21-H), 0.87 (s, 3 H, 19-H), 0.68 (s, 3 H, 18-
H) ppm. 13C NMR (75 MHz, [D6]acetone): δ = 206.2, 175.5, 77.1,
72.7, 72.2, 68.3, 57.1, 48.8, 42.9, 40.5, 40.2, 39.8, 36.9, 36.3, 35.5,
3α,7α,12α,16α,24-Pentahydroxy-5β-cholane (9): BH3·Me2S (1 mL,
10.5 mmol) was added to a stirred solution of compound 8 (0.23 g,
0.56 mmol) in dry THF (2 mL) at 0 °C under nitrogen and the
reaction mixture was stirred for 20 h at room temperature. The mix-
ture was oxidized by dropwise addition of a mixture (13 mL) of 1:1
(v/v) 4  aq. NaOH and 30% H2O2 at 0 °C. After stirring at room
temperature for 2 h it was neutralized by adding 1  HCl (25 mL).
The precipitate obtained was filtered, dried under vacuum and dis-
solved in 5% KOH in MeOH (15 mL) and refluxed for 12 h. The
solid residue obtained after removing volatile components under
vacuum was dissolved in water (10 mL), neutralized with 1  HCl
and filtered. The crude product was purified on a silica column
(2ϫ14 cm) using 10% EtOH in ethyl acetate to afford compound
9 (98 mg, 70%) as a white solid; m.p. 201–203 °C. 1H NMR
(300 MHz, [D6]DMSO): δ = 4.32 (d, J = 4.5 Hz, 1 H, 16-OH), 4.49
34.7, 31.5, 31.4, 27.3, 23.1, 17.9, 14.3 ppm. IR (KBr): ν = 3414,
˜
1704, 1668 cm–1. [α]2D4 = +11 (c = 2, EtOH). HRMS: m/z calcd. for
C24H40O6 + Na+: 447.2723; found 447.2731.
3α,12α-Dihydroxy-7-oxo-5β-cholane O-24,16α-Lactone (11): N-Bro-
mosuccinimide (25 mg, 0.140 mmol) was added to a solution of
compound 10 (42 mg, 0.10 mmol) in 25:1 (v/v) acetone/water
(2.8 mL) and stirred at room temperature for 10 min. After adding
water (2 mL) the crude product was extracted with diethyl ether
and washed with satd. aq. NaHCO3 (2ϫ4 mL). The organic layer
was dried with anhyd. Na2SO4 and the crude product was purified
using 45–60% ethyl acetate in petroleum ether to afford 34 mg
(d, J = 4.8 Hz, 1 H, 12-OH), 4.16 (d, J = 3.6 Hz, 1 H, 7-OH), 4.05 (88%) of compound 11 as a white solid; m.p. 268–269 °C. 1H NMR
(d, J = 5.1 Hz, 1 H, 3-OH), 3.90 (d, J = 2.4 Hz, 1 H, 24-OH), 3.72 (300 MHz, CDCl3): δ = 4.60 (t, J = 7.8 Hz, 1 H, 16β-H), 3.97 (br.
(br. d, J = 6.3 Hz, 1 H, 16β-H), 3.67 (br. s, 1 H, 12β-H), 3.57 (br.
s, 1 H, 7β-H), 3.19 (m, 1 H, 3β-H), 2.04–1.00 (m, steroidal CH2),
t, 1 H, 12β-H), 3.58–3.50 (br. m, 1 H, 3β-H), 2.04–1.02 (m, steroidal
CH2), 1.17 (s, 3 H, 19-H), 1.08 (d, J = 6.6 Hz, 3 H, 21-H), 0.76 (s,
0.90 (d, J = 6.6 Hz, 3 H, 21-H), 0.78 (s, 3 H, 19-H), 0.57 (s, 3 H, 3 H, 18-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 212.0, 176.0,
18-H) ppm. 13C NMR (75 MHz, [D6]DMSO): δ = 75.4, 70.8, 70.4,
84.2, 77.2, 70.8, 54.7, 49.1, 48.9, 46.1, 45.1, 37.8, 37.7, 35.9, 34.6,
34.0, 33.1, 32.9, 29.9, 29.6, 29.3, 22.8, 19.6, 13.7 ppm. IR (CHCl3):
66.4, 61.5, 55.9, 47.3, 41.5, 35.9, 35.2, 34.7, 34.4, 33.9, 31.3, 30.4,
29.8, 28.2, 26.1, 22.6, 17.9, 13.9 ppm. IR (CHCl ): ν = 3423 cm–1.
ν = 3419, 1712 cm–1. [α]24 = +32 (c = 0.5, EtOH). HRMS: m/z
˜
˜
3
D
[α]2D4 = +11 (c = 2, EtOH). HRMS: m/z calcd. for C24H42O5 + Na+:
calcd. for C24H36O5 + Na+: 427.2460; found. 427.2449.
Eur. J. Org. Chem. 2007, 3331–3336
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3335

Nonappa, U. Maitra
FULL PAPER
New York, 1973, vol. 2, pp. 1–32; b) S. Mukhopadhyay, U.
Maitra, Curr. Sci. 2004, 87, 1666–1683; c) A. Enhsen, W.
Kramer, G. Wess, Drug Discovery Today 1998, 3, 409–418.
[2] a) A. F. Hofmann, Biochem. J. 1963, 89, 57–68; b) M. C. Carey,
Ann. Rev. Physiol. 1983, 45, 651–677.
[3] a) G. A. D. Haselwood, Biol. Rev. 1964, 39, 537–574; b) T. Iida,
M. Hikosaka, G. Kakiyama, K. Shiraishi, C. D. Schteingart,
L. R. Hagey, H. T. Ton-Nu, A. F. Hofmann, N. Mano, J. Goto,
T. Nambara, Chem. Pharm. Bull. 2002, 50, 1327–1334.
[4] a) A. Moschetta, A. L. Bookout, D. J. Mangelsdorf, Nat. Med.
2004, 10, 1352–1358; b) M. Watanabe, M. S. Houten, C. Ma-
taki, M. A. Christoffolete, B. W. Kim, H. Sato, N. Messaddeq,
J. W. Harney, O. Ezaki, T. Kodama, K. Schoonjans, A. C. Bi-
anco, J. Auwerx, Nature 2006, 439, 484–489; c) D. L. Motola,
C. L. Cummins, V. Rottiers, K. K. Sharma, T. Li, Y. Li, K. S.
Powell, H. E. Xu, R. J. Auchus, A. Antebi, D. J. Mangelsdorf,
Cell 2006, 124, 1209–1223.
3α,12α,16α-Trihydroxy-5β-cholan-24-oic Acid (1): To a solution of
compound 11 (50 mg, 0.12 mmol) in digol (1 mL), hydrazine hy-
drate (0.1 mL, 2.0 mmol) and KOH (100 mg, 1.8 mmol) were
added. The reaction mixture was refluxed at 125 °C for 3 h and
then heated to 205 °C for 1 h on a sand bath. After cooling to
room temperature and adding water (2 mL) it was neutralized with
6  HCl. The acid was extracted with ethyl acetate (3ϫ3 mL) and
dried with anhyd. Na2SO4. After removing the volatile components
under reduced pressure, it was dried under vacuum to afford 45 mg
(88%) of compound 1 as a white solid; m.p. 188–189 °C (ref.[2] m.p.
186–187 °C). 1H NMR (300 MHz, [D6]acetone): δ = 3.99 (t, J =
7.8 Hz, 1 H, 16β-H), 3.89 (br. t, 1 H, 12β-H), 3.57–3.46 (m, 1 H,
3β-H), 2.04–1.00 (m, steroidal CH2), 1.01 (d, J = 6.3 Hz, 21-H),
0.91 (s, 3 H, 19-H), 0.723 (s, 3 H, 18-H) ppm. 1H NMR (300 MHz,
CDCl3): δ = 4.03 (t, J = 7.5 Hz, 1 H, 16β-H), 3.88 (br. s, 1 H, 12β-
H), 3.71–3.61 (br. m, 1 H, 3β-H), 1.00 (d, J = 6.6 Hz, 21-H), 0.89
(s, 3 H, 19-H), 0.70 (s, 3 H, 18-H) ppm. 13C NMR (75 MHz, [D6]-
acetone): δ = 175.6, 76.7, 72.7, 71.6, 57.0, 48.8, 45.8, 43.1, 37.7,
37.3, 36.3, 36.1, 34.9, 34.5, 34.2, 31.5, 31.3, 31.0, 28.0, 27.2, 23.6,
[5] a) G. A. D. Haslewood, V. Wootton, Biochem. J. 1951, 49, 67–
71; b) G. A. D. Haslewood, V. Wootton, Biochem. J. 1950, 47,
584–597; c) M. Kuroda, H. Arata, J. Biol. Chem. 1951, 29, 225–
226.
18.0, 14.6 ppm. [α]2D4 = +28 (c = 0.5, EtOH). IR (KBr): ν = 3415,
[6] S. Bergström, H. Danielsson, T. Kazuno, J. Biol. Chem. 1960,
˜
235, 983 –988.
1704 cm–1. HRMS: m/z calcd. for C24H40O5 + Na+: 431.2773;
found: 431.2757. C24H40O5 (408.28): calcd. C 70.54, H 9.87; found
C 70.65, H 10.00.
[7] M. Kimura, M. Kawata, A. Fujino, K. Yamasaki, Tetrahedron
Lett. 1970, 23, 2021–2024.
[8] L. R. Hagey, C. D. Schteingart, H. T. Ton-Nu, A. F. Hofmann,
J. Lipid Res. 2002, 43, 685–690.
[9] S. Mukhophadhyay, U. Maitra, Org. Lett. 2004, 6, 31–34.
[10] a) R. Breslow, P. White, J. Am. Chem. Soc. 1990, 112, 6842–
6847; b) R. Breslow, Acc. Chem. Res. 1980, 13, 170–177.
[11] V. Schwarz, P. Pihera, J. Halasková, Cesk. Farm. 1984, 33, 327–
330.
[12] R. V. Oppenauer, Monatsh. Chem. 1966, 97, 62–66.
[13] K. Kato, M. Sugahara, N. Tohnai, K. Sada, M. Miyata, Cryst.
Growth Des. 2004, 4, 263–272.
[14] a) I. Saraogi, V. G. Vijay, S. Das, K. Sekar, T. N. Guru Row,
Cryst. Eng. 2003, 6, 69–77; b) M. D. Prasanna, T. N.
Guru Row, Cryst. Eng. 2000, 3, 135–154; c) P. Metrangolo, T.
Pilati, G. Resnati, CrystEngComm 2006, 8, 946–947.
[15] J. Einhorn, C. Einhorn, F. Ratajczak, J. L. Piere, J. Org. Chem.
1996, 61, 7452–7454.
Characterization of Pythocholic Lactone (12): Wolf–Kishner re-
duction followed by acidic work up of compound 11 afforded a
less polar product along with the desired pythocholic acid. Spectral
characterization confirmed that it to be pythocholic lactone 12;
m.p. 260–261 °C. (ref.[4] m.p. 263–264 °C). 1H NMR (300 MHz,
CDCl3): δ = 4.60 (t, J = 7.5 Hz, 1 H, 16β-H), 3.93 (t, J = 3.3 Hz,
1 H, 12β-H), 3.67–3.59 (br. m, 1 H, 3β-H), 2.04–1.16 (steroidal
CH2), 1.07 (d, J = 6.3 Hz, 3 H, 21-H), 0.91 (s, 3 H, 19-H), 0.77 (s,
3 H, 18-H) ppm. 13C NMR (75 MHz, CDCl3): δ = 175.7, 83.6,
72.3, 71.7, 55.9, 48.8, 44.9, 41.9, 36.3, 35.9, 35.4, 35.1, 34.1, 33.6,
32.9, 32.2, 30.3, 28.3, 26.9, 25.9, 23.0, 19.6, 13.7 ppm. IR (CHCl3):
ν = 3430, 1713, 1704 cm–1. [α]24 = +48 (c = 0.5, EtOH). HRMS:
˜
D
m/z calcd. for C24H38O4 + Na+: 413.2668; found: 413.2662.
[16] M. F. Semmelhack, C. R. Schmidt, D. A. Cortes, C. S. Chou,
J. Am. Chem. Soc. 1984, 106, 3374–3376.
Supporting Information (see also the footnote on the first page of
1
this article): H and 13C NMR spectra of compounds 1, 2, 6, 8, 9,
[17] B. Matkovics, B. Samuelsson, Acta Chem. Scand. 1962, 16,
10, 11, 12, 2D-NMR spectrum of compound 2, crystal data of
compound 6.
683–688.
[18] H. Minlon, J. Am. Chem. Soc. 1946, 68, 2487–2488.
[19] C. D. Schteingart, A. F. Hofmann, personal communication.
[20] K. Kalyanasundaran, J. K. Thomas, J. Am. Chem. Soc. 1977,
99, 2039–2044.
Acknowledgments
[21] a) M. C. Carey, J. C. Montent, M. C. Phillips, M. J. Armstrong,
N. A. Mazer, Biochemistry 1981, 20, 3637; b) K. Matsuoka, Y.
Kuranaga, Y. Moroi, Biochim. Biophys. Acta 2002, 1580, 200–
214.
We thank Dr. A. F. Hofmann and Dr. C. D. Schteingart, UCSD
for useful discussions. The NMR Research Centre and X-ray CCD
facility IISc, Bangalore are also thanked for services. We also
acknowledge Department of Science and Technology (DST), New
Delhi for funding (Grant SR/SI/OC/-11/2004) and N. thanks the
Council of Scientific and Industrial Research (CSIR), New Delhi
for a fellowship.
[22] Vogel’s Text Book of Practical Organic Chemistry, 5th ed., Addi-
son Wesley Longman Limited, Harlow, England, 1996, pp. 931.
[23] L. F. Fieser, S. Rajagopalan, J. Am. Chem. Soc. 1949, 71, 3935–
3938.
[24] R. Hirschmann, N. L. Wendler, J. Am. Chem. Soc. 1953, 75,
2361–2364
[1] a) H. Danielsson, in The Bile Acids: Chemistry, Physiology and
Metabolsm (Eds.: P. P. Nair, D. Kritchevsky), Plenum Press,
Received: March 9, 2007
Published Online: May 16, 2007
3336
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3331–3336