Synthesis and olfactory activity of unnatural, sulfated 5β-bile acid derivatives in the sea lamprey (Petromyzon marinus)

Article abstract of DOI:10.1016/j.steroids.2010.11.010

A variety of unnatural bile acid derivatives (9a-9f) was synthesized and used to examine the specificity with which the sea lamprey (Petromyzon marinus) olfactory system detects these compounds. These compounds are analogs of petromyzonol sulfate (PS, 1), a component of the sea lamprey migratory pheromone. Both the stereochemical configuration at C5 (i.e., 5α vs. 5β) and the extent and sites of oxygenation (hydroxylation or ketonization) of the bile acid derived steroid skeleton were evaluated by screening the compounds for olfactory activity using electro-olfactogram recording. 5β-Petromyzonol sulfate (9a) elicited a considerable olfactory response at sub-nanomolar concentration. In addition, less oxygenated systems (i.e., 9b-9e) elicited olfactory responses, albeit with less potency. The sea lamprey sex pheromone mimic 9f (5β-3-ketopetromyzonol sulfate) was also examined and found to produce a much lower olfactory response. Mixture studies conducted with 9a and PS (1) suggest that stimulation is occurring via similar modes of activation, demonstrating a relative lack of specificity for recognition of the allo-configuration (i.e., 5α) in sea lamprey olfaction. This attribute could facilitate design of pheromone analogs to control this invasive species.

Full text of DOI:10.1016/j.steroids.2010.11.010

Steroids 76 (2011) 291–300
Contents lists available at ScienceDirect
Steroids
journal homepage: www.elsevier.com/locate/steroids
Synthesis and olfactory activity of unnatural, sulfated 5␤-bile acid derivatives in
the sea lamprey (Petromyzon marinus)
Aaron C. Burnsa, Peter W. Sorensenb, Thomas R. Hoyea,∗
a Department of Chemistry, College of Science and Engineering, University of Minnesota, 207 Pleasant St. SE, Minneapolis, MN 55455, United States
b Department of Fisheries, Wildlife and Conservation Biology, College of Food, Agriculture, and Natural Science, University of Minnesota, St. Paul, MN 55108, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2010
Received in revised form
28 November 2010
Accepted 30 November 2010
Available online 8 December 2010
A variety of unnatural bile acid derivatives (9a–9f) was synthesized and used to examine the specificity
with which the sea lamprey (Petromyzon marinus) olfactory system detects these compounds. These
compounds are analogs of petromyzonol sulfate (PS, 1), a component of the sea lamprey migratory
pheromone. Both the stereochemical configuration at C5 (i.e., 5␣ vs. 5␤) and the extent and sites of
oxygenation (hydroxylation or ketonization) of the bile acid derived steroid skeleton were evaluated by
screening the compounds for olfactory activity using electro-olfactogram recording. 5␤-Petromyzonol
sulfate (9a) elicited a considerable olfactory response at sub-nanomolar concentration. In addition, less
oxygenated systems (i.e., 9b–9e) elicited olfactory responses, albeit with less potency. The sea lamprey
sex pheromone mimic 9f (5␤-3-ketopetromyzonol sulfate) was also examined and found to produce a
much lower olfactory response. Mixture studies conducted with 9a and PS (1) suggest that stimulation is
occurring via similar modes of activation, demonstrating a relative lack of specificity for recognition of the
allo-configuration (i.e., 5␣) in sea lamprey olfaction. This attribute could facilitate design of pheromone
analogs to control this invasive species.
Keywords:
Steroid synthesis
Olfactory SAR
Pheromone
Bile acid sulfates
© 2010 Elsevier Inc. All rights reserved.
1. Introduction
the cost of production is high, there is interest in producing analogs
that exhibit similar olfactory potency.
It is now well established that naturally occurring bile acids
and closely related compounds function as potent olfactory stimu-
lants for many species of fish. They also have been postulated to
function as pheromones [1–3]. Amongst these species, bile acid
function is especially well understood in the ancient sea lamprey
(Petromyzon marinus) [4–7]. Four behaviorally active bile acid-
like compounds have been shown to be produced, released, and
detected in this species. In particular, a three-component suite of
bile acids [petromyzonol sulfate (PS, 1), petromyzonamine disul-
fate (PADS, 2), and petromyzosterol disulfate (PSDS, 3); Fig. 1] is
known to be produced by larval sea lamprey and to function as
a migratory pheromone for adults. 3-Ketopetromyzonol sulfate
(4), an oxidized analog of 1, is produced and released by sexu-
ally mature male sea lamprey and functions as a sex pheromone
that attracts ovulated females [8,9]. Each of these compounds is
detected at picomolar thresholds by the lamprey olfactory system
with PADS (2) being detected at 10−13 M. Because the sea lamprey
is an invasive species, these novel products are being explored on a
commercial basis for application in trapping programs [3]. Because
Several studies have explored the specificity with which
lamprey olfactory epithelium discerns bile acids using electro-
physiological recording. These have provided evidence of multiple
receptor types, but the ligand structural features associated with
specificity are not yet well understood [4,5,10]. Of special inter-
est have been bile acids with the ‘allo’ (5␣) configuration (cf. 5,
Fig. 2); both lampreys (Family Petromyzontidae) and the minnows
and carps (Family Cyprinidae) produce and release several allo-
steroids and exhibit remarkable sensitivity to them [2,11]. It is
important to emphasize the fairly substantial structural difference,
largely topological in nature, associated with the allocholate-like
(5) vs. cholate-like (6) topologies (cf., Fig. 2).
Because allo steroids are rare amongst mammals, relatively few
have been synthesized. The sea lamprey and related lampreys pro-
duce the biologically active allo steroids, allocholic acid (7, Fig. 2),
PS (1), and PADS (2). Only for the first of these steroids has the anal-
ogous 5␤-isomer, namely cholic acid (10a, Scheme 1), been tested
for olfactory activity in lamprey and other fish [5]. Otherwise, the
importance of the more rare allo-configuration to the underlying
structure–activity relationship (SAR) has not been studied. Simi-
larly, the carps and other cyprinids, the second largest family of
fishes within over 1000 species, produce and detect cyprinol sul-
fate (8, Fig. 2) [12,13]. Other fishes detect 8 with high sensitivity
[11]. However, neither the behavioral function nor the importance
∗
Corresponding author. Tel.: +1 612 625 1891.
E-mail addresses: soren003@umn.edu (P.W. Sorensen), hoye@umn.edu
(T.R. Hoye).
0039-128X/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.steroids.2010.11.010

292
A.C. Burns et al. / Steroids 76 (2011) 291–300
Fig. 1. The three principal components of the sea lamprey migratory pheromone.
Fig. 2. Topological difference between the allocholate (5, 5␣-H) and cholate (6, 5␤-H) structural families. Examples (7 and 8) of the allocholate family.
of the allo-configuration in olfactory detection have been explicitly
tested. Indeed, the overall importance to the fish olfactory system
of most structural features associated with this class of (several
hundred Dalton) sulfated allo steroids has not been systemati-
cally explored. This is notable because bile acids include the largest
olfactory ligands described for a fish (PADS is the largest), so it is
reasonable to ask whether the entire structure is discerned.
The study reported here had two related objectives; to synthe-
size a series of 5␤-bile acids and to determine both the sensitivity
and specificity with which they are discerned by the sea lamprey
Scheme 1. Synthesis of sulfated bile acid derivatives 9a–9c.

A.C. Burns et al. / Steroids 76 (2011) 291–300
293
Fig. 3. 5␤-Bile acid derivatives tested by the electro-olfactogram method in this study.
obtained using an INOVA 500 Varian instrument (1H data recorded
at 500 MHz, 13C data recorded at 125 MHz). 1H chemical shifts
are reported in delta (␦) units, in parts per million (ppm) rela-
tive to tetramethylsilane (0.00 ppm) when CDCl3 was employed or
to CD2HOD (3.31 ppm) when d4-methanol was used. 13C chemical
shifts are reported in delta (␦) units, in parts per million (ppm)
relative to CDCl3 (77.23 ppm) or to CD3OD (49.15 ppm). High-
resolution mass spectrometric data were obtained using a Bruker
BioTOF II (ESI) mass spectrometer. The activity of these products
was studied using electro-olfactogram (EOG) recording as detailed
in Section 2.2.
olfactory system. Toward that end, we prepared the analogs 9a
[the 5␤-epimer of PS (1)], 9b–e (less highly oxygenated analogs
of 9a), and 9f (the C3-keto analog of 9a) (Fig. 3) and then tested
their olfactory activity using electro-olfactogam recording (EOG).
2. Experimental
2.1. General
Unless noted otherwise, all oxygen and moisture-sensitive
reactions were executed in oven-dried glassware sealed under
a positive pressure of dry argon or nitrogen. All commercial
reagents were used as received. Anhydrous THF and methylene
chloride were tapped immediately prior to use after being passed
through a column of activated alumina. Flash chromatography was
2.1.1. 3˛,7˛,12˛-Triformyloxy-5ˇ-cholan-24-oic acid
(11a)[14,15]
A 10 mL flask equipped with a magnetic stirring bar was charged
with cholic acid (10a, 1.0 g, 2.45 mmol) and 4 mL of 88% formic acid.
The resultant solution was heated for 24 h at 55 ◦C, at which time
TLC analysis showed consumption of 10a. The solution was cooled
to room temperature and the formic acid was removed in vacuo.
The crude material was recrystallized from ethyl acetate/hexane to
provide 11a (785 mg, 65%). White crystalline solid; mp 208–209 ◦C;
Rf = 0.48 (1:1 hexane/ethyl acetate); [␣]D (23 ◦C): +90.3◦ (c = 0.050,
˚
performed using Agela Technologies silica gel 40–60 ␮m (40 A);
analytical TLC was performed using 0.25 mm EM silica gel 60
F254 plates that were visualized under UV light (254 nm) or by
staining with anisaldehyde reagent (450 mL of 95% ethanol, 25 mL
concd H2SO4, 15 mL acetic acid, and 25 mL anisaldehyde) and
heating. Optical rotations were obtained using a Perkin-Elmer-
241 polarimeter. IR spectra were recorded using a Prospect 4000
FT-IR spectrophotometer (Midac Corporation). NMR spectra were
CHCl3); IR (neat): 3400 (broad), 2943, 1717 (sharp), 1183 cm−1 1H
;
Fig. 4. EOG response magnitudes of 5␤-petromyzonol sulfate (9a) and analogs 9b–f tested at 10−9 M concentrations. Response magnitudes are presented as a percentage of
10−5 M l-arginine. Vertical bars represent one standard error (n = 4 animals). Each of the above responses was deemed to be significantly different from the others (ANOVA,
P < 0.001).

294
A.C. Burns et al. / Steroids 76 (2011) 291–300
NMR (500 MHz, CDCl3): ␦ 8.17 (s, 1H), 8.11 (s, 1H), 8.03 (s, 1H), 5.27
(dd, J = 2.0, 2.0 Hz, 1H, H12), 5.07 (ddd, J = 2.5, 2.5, 2.5 Hz, 1H, H7),
4.72 (dddd, J = 11.5, 11.5, 4.0, 4.0 Hz, 1H, H3), 2.38 (ddd, J = 15.5,
10.0, 5.5 Hz, 1H, H23a), 2.25 (ddd, J = 16.0, 9.5, 6.5 Hz, 1H, H23b),
2.17–2.10 (m, 2H), 2.02 (ddd, J = 16.0, 5.0, 5.0 Hz, 1H), 1.97–1.85
(m, 2H), 1.82–1.64 (m, 9H), 1.60–1.52 (m, 2H), 1.49–1.39 (m, 2H),
1.37–1.28 (m, 3H), 1.16–1.06 (m, 2H), 0.95 (s, 3H, CH319), 0.86
(d, J = 6.5 Hz, 3H, CH321), 0.76 (s, 3H, CH318); 13C NMR (125 MHz,
CDCl3) ␦ 180.0, 160.8 (2C), 160.7, 75.5, 73.9, 70.9, 47.4, 45.2, 43.2,
41.0, 37.9, 34.9, 34.7, 34.6, 34.5, 31.5, 31.0, 30.6, 28.8, 27.3, 26.8,
25.8, 23.0, 22.6, 17.7, 12.4; HRMS (ESI+) calcd for C27H40O8Na+
(M + Na+) 515.2615, found 515.2613.
2.1.4. Sodium 5ˇ-petromyzonol-24-sulfate (9a, 5␤-PS)[17]
A 10 mL flask equipped with a magnetic stir bar was charged
with 13a (150 mg, 0.227 mmol, 1.0 equiv.) and 8.4 mL of a mixture of
methanol/water (4:1). To this solution was added sodium hydrox-
ide (90 mg, 2.25 mmol, 10.0 equiv.) and the mixture was heated to
50 ◦C for 12 h. The solution was cooled to room temperature and
the volatiles were removed in vacuo to give the crude material,
which was loaded directly onto a silica gel column and puri-
fied by flash chromatography (80:26:2 chloroform/methanol/satd
aq NH3) to provide 9a (104 mg, 92%). White amorphous solid;
mp 180.0–182.0 ◦C; Rf = 0.24 (80:26:2 chloroform/methanol/satd
aq NH3); [␣]D (23 ◦C): +21.1◦ (c = 0.031, MeOH); IR (neat): 3400
(broad), 2939, 1209 cm−1 1H NMR (500 MHz, d4-methanol): ␦
;
3.99–3.94 (m, 3H), 3.79 (ddd, J = 3.0, 3.0, 3.0 Hz, 1H, H7), 3.37 (dddd,
J = 11.0, 11.0, 4.5, 4.5 Hz, 1H, H3), 2.32–2.22 (m, 2H), 2.01–1.71
(m, 8H), 1.67–1.51 (m, 8H), 1.47–1.36 (m, 3H), 1.33–1.25 (m, 1H),
1.19–1.08 (m, 2H), 1.02 (d, J = 6.5 Hz, 3H, CH321), 0.98 (ddd, J = 14.0,
14.0, 3.5 Hz, 1H), 0.92 (s, 3H, CH319), 0.72 (s, 3H, CH318); 13C
NMR (125 MHz, d4-methanol) ␦ 74.2, 73.0, 69.8, 69.2, 48.4, 47.6,
43.3, 43.1, 41.2, 40.6, 37.0, 36.6, 36.0, 35.9, 33.3, 31.3, 29.7, 28.9,
2.1.2. 3˛,7˛,12˛-Triformyloxy-5ˇ-cholan-24-ol (12a)[15,16]
A 25 mL flask equipped with a magnetic stir bar was charged
with 11a (500 mg, 1.02 mmol, 1.0 equiv.) and 10 mL of tetrahy-
drofuran and cooled to −78 ◦C. To this solution was added 10 M
borane–dimethyl sulfide complex (153 ␮L, 1.53 mmol, 1.5 equiv.)
and the resultant solution was warmed to room temperature over
1 h and subsequently allowed to stir for an additional hour. The
excess borane was quenched with 200 ␮L of acetic acid at 0 ◦C and
allowed to stir for 12 h at room temperature. The volatiles were
removed in vacuo to give the crude material, which was purified by
flash chromatography to provide 12a (253 mg, 52%). White amor-
phous solid; mp 86–92 ◦C; Rf = 0.44 (1:1 hexane/ethyl acetate); [␣]D
(23 ◦C): +60.7◦ (c = 0.018, CHCl3); IR (neat): 3400 (broad), 2941,
−
28.0, 27.4, 24.4, 23.3, 18.1, 13.1; HRMS (ESI−) calcd for C24H41SO7
(M − Na+) 473.2578, found 473.2585.
Compounds 9b–13b and 9c–13c were synthesized by analogous
procedures to those detailed above for 9a–13a. Full characteriza-
tion data for each compound are presented below.
2.1.5. 3˛,12˛-Diformyloxy-5ˇ-cholan-24-oic acid (11b)[15,18]
White crystalline solid; mp 198.0–198.5 ◦C; Rf = 0.40 (3:1 hex-
ane/ethyl acetate); [␣]D (23 ◦C): +61.6◦ (c = 0.500, CHCl3); IR (neat):
1717 (sharp), 1183 cm−1 1H NMR (500 MHz, CDCl3): ␦ 8.16 (s, 1H),
;
8.11 (s, 1H), 8.03 (s, 1H), 5.28 (dd, J = 3.0, 3.0 Hz, 1H, H12), 5.07 (ddd,
J = 2.5, 2.5, 2.5 Hz, 1H, H7), 4.72 (dddd, J = 11.5, 11.5, 4.5, 4.5 Hz, 1H,
H3), 3.64–3.56 (m, 2H), 2.15–2.09 (m, 2H), 2.02 (ddd, J = 15.0, 5.0,
3.0 Hz, 1H), 1.97–1.84 (m, 2H), 1.82–1.52 (m, 10H), 1.48–1.25 (m,
6H), 1.15–1.02 (m, 3H), 0.95 (s, 3H, CH319), 0.85 (d, J = 6.5 Hz, 3H,
CH321), 0.76 (s, 3H, CH318); 13C NMR (125 MHz, CDCl3) ␦ 160.8
(2C), 160.7, 75.5, 73.9, 70.9, 63.5, 47.5, 45.1, 43.1, 41.0, 37.9, 35.1,
34.7, 34.6, 34.4, 31.7, 31.5, 29.4, 28.7, 27.4, 26.7, 25.7, 23.0, 22.5,
18.0, 12.3; HRMS (ESI+) calcd for C27H42O7Na+ (M + Na+) 501.2823,
found 501.2820.
3400 (broad), 2939, 1698 (sharp), 1179 cm−1 1H NMR (500 MHz,
;
CDCl3): ␦ 8.14 (s, 1H), 8.03 (d, J = 0.5 Hz, 1H), 5.25 (dd, J = 3.0,
3.0 Hz, 1H, H12), 4.84 (dddd, J = 11.0, 11.0, 4.5, 4.5 Hz, 1H, H3), 2.39
(ddd, J = 15.5, 10.0, 5.5 Hz, 1H, H23), 2.25 (ddd, J = 16.0, 9.5, 7.0 Hz,
1H, H23), 2.17–2.10 (m, 2H), 2.02 (ddd, J = 15.5, 5.5, 3.5 Hz, 1H),
1.93–1.77 (m, 4H), 1.74–1.57 (m, 8H), 1.51–1.40 (m, 4H), 1.17–1.01
(m, 3H), 0.93 (s, 3H, CH319), 0.84 (d, J = 6.5 Hz, 3H, CH321), 0.75 (s,
3H, CH318); 13C NMR (125 MHz, CDCl3) ␦ 180.2, 160.9, 160.8, 76.2,
74.3, 49.5, 47.6, 45.2, 41.9, 35.8, 35.0, 34.9, 34.4, 34.5, 34.2, 32.3,
31.1, 30.7, 27.6, 27.0, 26.7, 26.1, 26.0, 23.6, 23.1, 17.6, 12.6; HRMS
(ESI+) calcd for C26H40O6Na+ (M + Na+) 471.2717, found 471.2721.
2.1.3. Triethylammonium
3˛,7˛,12˛-triformyloxy-5ˇ-cholan-24-ol sulfate (13a)
2.1.6. 3˛,12˛-Diformyloxy-5ˇ-cholan-24-ol (12b)[15]
White amorphous solid; mp 57.0–62.5 ◦C; Rf = 0.78 (1:1 hex-
ane/ethyl acetate); [␣]D (23 ◦C): +85.5◦ (c = 0.010, CHCl3); IR (neat):
A
10 mL flask equipped with a magnetic stir bar was
3400 (broad), 2938, 1721 (sharp), 1178 cm−1 1H NMR (500 MHz,
;
charged with 12a (200 mg, 0.418 mmol, 1.0 equiv.) and 4.2 mL
of methylene chloride. To this solution was added triethylamine
(233 ␮L, 1.67 mmol, 4.0 equiv.) and sulfur trioxide–pyridine com-
plex (200 mg, 1.25 mmol, 3.0 equiv.). The reaction mixture was
allowed to stir at room temperature for 24 h. The volatiles were
removed in vacuo to give the crude material, which was loaded
directly onto a silica gel column and purified by flash chromatog-
raphy (90:9:1 chloroform/methanol/satd aq NH3) to provide 13a
(183 mg, 66%). White amorphous solid; mp 95.5–99.0 ◦C; Rf = 0.41
(80:26:2 chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +51.8◦
CDCl3): ␦ 8.14 (s, 1H), 8.03 (d, J = 0.5 Hz, 1H), 5.26 (dd, J = 3.0, 3.0 Hz,
1H, H12), 4.84 (dddd, J = 11.5, 11.5, 4.5, 4.5 Hz, 1H, H3), 3.65–3.57
(m, 2H), 1.92–1.01 (m, 27H), 0.93 (s, 3H, CH319), 0.84 (d, J = 6.5 Hz,
3H, CH321), 0.75 (s, 3H, CH318); 13C NMR (125 MHz, CDCl3) ␦ 160.9,
160.8, 76.3, 74.3, 63.6, 49.4, 47.7, 45.1, 41.9, 35.8, 35.2, 34.8, 34.4,
34.2, 32.3, 31.8, 29.5, 27.6, 27.0, 26.6, 26.1, 25.9, 23.6, 23.1, 18.0,
12.5; HRMS (ESI+) calcd for C26H42O5Na+ (M + Na+) 457.2924, found
457.2932.
(c = 0.026, MeOH); IR (neat): 2944, 1717 (sharp), 1180 cm−1 1H
;
2.1.7. Triethylammonium 3˛,12˛-diformyloxy-5ˇ-cholan-24-ol
sulfate (13b)
NMR (500 MHz, d4-methanol): ␦ 8.22 (s, 1H), 8.15 (s, 1H), 8.05 (d,
J = 1.0 Hz, 1H), 5.28 (dd, J = 5.0, 5.0 Hz, 1H, H12), 5.05 (ddd, J = 3.0,
3.0, 3.0 Hz, 1H, H7), 4.68 (dddd, J = 11.5, 11.5, 4.5, 4.5 Hz, 1H, H3),
3.98–3.92 (m, 2H), 3.22 (q, J = 7.0 Hz, 6H, NCH2CH3), 2.20–2.10 (m,
3H), 2.02 (dd, J = 12.5, 7.5 Hz, 1H), 1.93–1.61 (m, 10H), 1.57–1.41 (m,
5H), 1.40–1.28 (m, 10H), 1.22–1.08 (m, 3H), 1.00 (s, 3H, CH319), 0.88
(d, J = 6.5 Hz, 3H, CH321), 0.82 (s, 3H, CH318); 13C NMR (125 MHz,
d4-methanol) ␦ 162.8 (2C), 162.7, 77.0, 75.2, 72.4, 69.6, 48.1, 46.4
(3C), 44.5, 42.4, 39.1, 36.3, 36.0, 35.7, 35.5, 33.0, 32.6, 29.9, 28.4,
27.9, 27.1, 26.8, 23.9, 22.8, 18.4, 16.4, 12.6, 9.4 (3C); HRMS (ESI−)
White amorphous solid; mp 153.2–154.0 ◦C; Rf = 0.25 (80:18:2
chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +65.5◦ (c = 0.025,
MeOH); IR (neat): 2943, 1718 (sharp), 1176 cm−1
;
1H NMR
(500 MHz, d4-methanol): ␦ 8.20 (s, 1H), 8.05 (d, J = 0.5 Hz, 1H), 5.24
(dd, J = 3.0, 3.0 Hz, 1H, H12), 4.80 (dddd, J = 11.0, 11.0, 5.0, 5.0 Hz,
1H, H3), 3.99–3.92 (m, 2H), 3.21 (q, J = 10.0 Hz, 6H, NCH2CH3),
1.97–1.85 (m, 2H), 1.80–1.30 (m, 27H), 1.20–1.05 (m, 4H), 0.98 (s,
3H, CH319), 0.87 (d, J = 6.5 Hz, 3H, CH321), 0.80 (s, 3H, CH318); 13
C
NMR (125 MHz, d4-methanol) ␦ 162.8, 162.7, 77.5, 75.4, 69.6, 50.8,
−
calcd for C27H41SO10 (M − HNEt3) 557.2426, found 557.2428.
48.1, 46.4 (3C), 43.3, 37.1, 36.4, 35.9, 35.6, 35.3, 33.5, 33.1, 28.6,

A.C. Burns et al. / Steroids 76 (2011) 291–300
295
−
28.1, 27.7, 27.3, 27.2, 26.9, 24.7, 23.5, 18.5, 17.9, 12.9, 9.4 (3C);
23.2, 21.9, 19.2, 12.3, 9.4 (3C); HRMS (ESI−) calcd for C26H41SO8
−
HRMS (ESI−) calcd for C26H41SO8 (M − HNEt3) 513.2528, found
(M − HNEt3) 513.2528, found 513.2530.
513.2533.
2.1.12. Sodium 3˛,7˛-dihydroxy-5ˇ-cholan-24-ol sulfate (9c)
White amorphous solid; mp 172.7–176.2 ◦C; Rf = 0.38 (80:26:2
chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +35.0◦ (c = 0.020,
2.1.8. Sodium 3˛,12˛-dihydroxy-5ˇ-cholan-24-ol sulfate (9b)
White amorphous solid; mp 172.7–176.2 ◦C; Rf = 0.40 (80:26:2
chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +35.0◦ (c = 0.020,
MeOH); IR (neat): 3400 (broad), 2934, 1219 cm−1
;
1H NMR
(500 MHz, d4-methanol): ␦ 3.98–3.93 (m, 3H), 3.52 (dddd, J = 11.0,
11.0, 4.5, 4.5 Hz, 1H, H3), 1.92–1.72 (m, 7H), 1.64–1.38 (m, 13H),
1.29 (dddd, J = 4.5, 4.5, 2.0, 2.0 Hz, 1H), 1.26 (ddd, J = 4.0, 2.0, 2.0 Hz,
1H), 1.21–1.05 (m, 3H), 1.01 (d, J = 6.5 Hz, 3H, CH321), 0.98 (ddd,
J = 14.0, 14.0, 3.0 Hz, 1H), 0.93 (s, 3H, CH319), 0.71 (s, 3H, CH318);
13C NMR (125 MHz, d4-methanol) ␦ 73.0, 69.8, 69.2, 57.6, 51.6,
43.8, 43.3, 41.2, 40.9, 40.6, 39.3, 37.0, 36.7, 36.4, 36.0, 34.2, 33.3,
MeOH); IR (neat): 3400 (broad), 2934, 1219 cm−1
;
1H NMR
(500 MHz, d4-methanol): ␦ 3.98–3.93 (m, 3H), 3.52 (dddd, J = 11.0,
11.0, 4.5, 4.5 Hz, 1H, H3), 1.92–1.72 (m, 7H), 1.64–1.38 (m,
13H), 1.32–1.23 (m, 2H), 1.21–1.05 (m, 3H), 1.01 (d, J = 6.5 Hz,
3H, CH321), 0.98 (ddd, J = 14.0, 14.0, 3.0 Hz, 1H), 0.93 (s, 3H,
CH319), 0.71 (s, 3H, CH321); 13C NMR (125 MHz, d4-methanol)
␦ 74.2, 72.7, 69.8, 49.4, 48.5, 47.7, 43.8, 37.6, 37.4, 37.0, 36.6,
35.5, 35.0, 33.2, 31.2, 30.0, 28.9, 28.6, 27.6, 27.3, 25.0, 23.9, 18.0,
31.5, 29.4, 27.3, 24.8, 23.5, 21.9, 19.3, 12.3; HRMS (ESI−) calcd for
−
C
24H41SO6 (M − Na+) 457.2629, found 457.2627.
13.4; HRMS (ESI−) calcd for C24H41SO6 (M − Na+) 457.2629,
−
found 457.2628.
2.1.13. 3˛-Tetrahydropyranyloxy-5ˇ-cholan-24-ol (12d)[19]
A 100 mL flask equipped with a magnetic stir bar was charged
with lithocholic acid (10d, 2.0 g, 5.31 mmol, 1.0 equiv.) and 18 mL
of methylene chloride. To this solution was added dihydropyran
(4.7 mL, 53.1 mmol, 10 equiv.) and pyridinium p-toluenesulfonate
(133 mg, 0.531 mmol, 0.1 equiv.). The resultant solution was
allowed to stir at room temperature for 24 h at which time the
volatiles were removed in vacuo to provide crude 11d. A stir-
bar was added to the oily 11d in this flask, and 53 mL of ethyl
ether was added. The solution was cooled with an ice-water bath
and lithium aluminum hydride (1.0 g, 26.4 mmol, 5.0 equiv.) was
added. The resultant solution was allowed to stir at room tem-
perature for 3 h and quenched at 0 ◦C with 1 mL of water, 1 mL of
3 N sodium hydroxide, and 3 mL of water. The suspension was fil-
tered and the resultant solution was concentrated in vacuo to give
the crude product, which was purified by flash chromatography to
provide 12d (1.43 g, 60%). Colorless oil; Rf = 0.35 (1:1 hexane/ethyl
2.1.9. 3˛,7˛-Diformyloxy-5ˇ-cholan-24-oic acid (11c)[15]
White crystalline solid; mp 100.0–102.5 ◦C; Rf = 0.53 (1:1 hex-
ane/ethyl acetate); [␣]D (23 ◦C): +32.8◦ (c = 0.017, CHCl3); IR (neat):
3400 (broad), 2948, 1722 (sharp), 1186 cm−1 1H NMR (500 MHz,
;
CDCl3): ␦ 8.08 (s, 1H), 8.03 (d, J = 0.5 Hz, 1H), 5.04 (ddd, J = 3.5, 3.5,
3.5 Hz, 1H, H7), 4.73 (dddd, J = 11.0, 11.0, 4.5, 4.5 Hz, 1H, H3), 2.40
(ddd, J = 15.5, 10.0, 5.0 Hz, 1H, H23a), 2.26 (ddd, J = 16.0, 10.0, 6.5 Hz,
1H, H23b), 2.13 (ddd, J = 13.0, 13.0, 13.0 Hz, 1H), 2.00 (dddd, J = 12.0,
12.0, 4.0, 4.0 Hz), 1.92–1.74 (m, 5H), 1.67–1.62 (m, 3H), 1.54–1.24
(m, 9H), 1.22–1.04 (m, 4H), 0.95 (s, 3H, CH319), 0.94 (d, J = 6.5 Hz,
3H, CH321), 0.66 (s, 3H, CH318); 13C NMR (125 MHz, CDCl3) ␦ 180.3,
161.0 (2C), 74.2, 71.6, 49.5, 55.8, 50.3, 42.9, 41.1, 39.6, 38.0, 35.4,
34.9, 34.7, 34.1, 31.6, 31.1, 30.8, 28.1, 26.9, 23.7, 22.8, 20.8, 18.4,
14.4, 11.9; HRMS (ESI+) calcd for C26H40O6Na+ (M + Na+) 471.2717,
found 471.2721.
acetate); IR (neat): 3397 (broad), 2915, 2852, 1449, 1025 cm−1
;
1H NMR (500 MHz, CDCl3): ␦ 4.73–4.71 (m, 1H), 3.94–3.90 (m,
1H), 3.66–3.58 (m, 3H), 3.51–3.46 (m, 1H), 1.95 (ddd, J = 12.5, 2.5,
2.5 Hz, 1H), 1.92–1.76 (m, 4H), 1.75–1.60 (m, 4H), 1.57–1.51 (m,
6H), 1.47–1.17 (m, 13H), 1.15–0.87 (m, 7H), 0.92 (d, J = 7.0 Hz, 3H,
CH321), 0.91 (s, 3H, CH319), 0.64 (s, 3H, CH318); 13C NMR (125 MHz,
CDCl3) ␦ 97.0, 96.7, 76.2, 76.0, 63.7, 63.0, 62.9, 56.6, 56.4, 42.9, 42.5,
42.2, 40.4, 40.3, 36.6, 36.0, 35.8, 35.7, 35.5, 34.9, 34.7, 34.5, 32.9,
32.4, 32.0, 31.5, 31.0, 30.7, 29.6, 28.7, 28.5, 27.5, 27.4, 26.9, 25.7,
24.4, 23.6, 23.5, 21.0, 20.9, 20.2, 20.1, 18.8, 12.2; HRMS (ESI+) calcd
for C29H50O3Na+ (M + Na+) 469.3652, found 469.3660.
2.1.10. 3˛,7˛-Diformyloxy-5ˇ-cholan-24-ol (12c) [15]
White amorphous solid; mp 49.5–52.0; Rf = 0.45 (1:1 hex-
ane/ethyl acetate); [␣]D (23 ◦C): +11.1◦ (c = 0.114, CHCl3); IR (neat):
3400 (broad), 2937, 1710 (sharp), 1181 cm−1 1H NMR (500 MHz,
;
CDCl3): ␦ 8.08 (s, 1H), 8.03 (s, 1H), 5.03 (ddd, J = 3.0, 3.0, 3.0 Hz,
1H, H7), 4.73 (dddd, J = 11.5, 11.5, 4.5, 4.5 Hz, 1H, H3), 3.65–3.57
(m, 2H), 2.14 (ddd, 14.0, 14.0, 14.0 Hz, 1H), 2.04–1.98 (m, 2H), 1.91
(ddd, J = 14.5, 3.0, 3.0 Hz, 1H), 1.88–1.79 (m, 2H), 1.78–1.73 (m, 1H),
1.66–1.60 (m, 4H), 1.54–1.03 (m, 16H), 0.95 (s, 3H, CH319), 0.94
(d, J = 6.5 Hz, 3H, CH321), 0.66 (s, 3H, CH318); 13C NMR (125 MHz,
CDCl3) ␦ 160.9 (2C), 74.2, 71.6, 63.7, 56.1, 50.3, 42.8, 41.1, 39.6,
38.0, 35.7, 35.0, 34.7, 34.2, 32.0, 31.7, 29.5, 28.3, 26.9, 26.1, 23.7,
22.8, 20.8, 18.8, 11.9; HRMS (ESI+) calcd for C26H42O5Na+ (M + Na+)
457.2924, found 457.2926.
2.1.14. Triethylammonium
3˛-tetrahydropyranyloxy-5ˇ-cholan-24-ol sulfate (13d)
Compound 13d was prepared following the procedure 2.1.3
(for 13a). White amorphous solid; mp 121.5–128.0 ◦C; Rf = 0.17
(90:9:1 chloroform/methanol/satd aq NH3); IR (neat): 2927, 2895,
2.1.11. Triethylammonium 3˛,7˛-diformyloxy-5ˇ-cholan-24-ol
sulfate (13c)
1441, 1253, 1196, 1023 cm−1 1H NMR (500 MHz, CDCl3): ␦ 9.77
;
(bs, 1H), 4.73–4.71 (m, 1H), 4.06–3.98 (m, 2H), 3.94–3.90 (m, 1H),
3.63 (ddd, J = 11.0, 11.0, 4.5, 4.5 Hz, 1H, H3), 3.51–3.46 (m, 1H),
3.18 (q, J = 7.0 Hz, 6H, NCH2CH3), 1.95 (ddd, J = 12.0, 2.5, 2.5 Hz,
1H), 1.89–1.69 (m, 8H), 1.60–1.31 (m, 15H), 1.39 (t, J = 7.5 Hz,
9H, NCH2CH3), 1.29–1.17 (m, 3H), 1.12–0.94 (m, 7H), 0.91 (s, 3H,
CH319), 0.90 (d, J = 6.0 Hz, 3H, CH321), 0.63 (s, 3H, CH318); 13C NMR
(125 MHz, CDCl3) ␦ 96.9, 76.2, 68.8, 62.9, 56.6, 56.4, 46.6 (3C), 42.8,
42.2, 40.4, 40.3, 36.0, 35.7, 35.6, 34.9, 32.9, 32.0, 31.5, 28.7, 28.4,
27.5, 26.5, 26.2, 25.6, 24.4, 2−3.5, 20.9, 20.2, 18.7, 12.2, 8.9 (3C); HRMS
(ESI−) calcd for C29H49SO6 (M − Na+) 525.3255, found 525.3258.
White amorphous solid; mp 174.9–176.1 ◦C; Rf = 0.25 (80:18:2
chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +14.0◦ (c = 0.026,
MeOH); IR (neat): 2941, 1718 (sharp), 1187 cm−1
;
1H NMR
(500 MHz, d4-methanol): ␦ 8.13 (s, 1H), 8.05 (d, J = 1.0 Hz, 1H), 5.02
(ddd, J = 2.0, 2.0, 2.0 Hz, 1H, H7), 4.69 (dddd, J = 11.5, 11.5, 4.5, 4.5 Hz,
1H, H3), 3.99–3.92 (m, 2H), 3.22 (q, J = 7.0 Hz, 6H, NCH2CH3), 2.18
(ddd, J = 13.5, 13.5, 13.5 Hz, 1H), 2.11 (ddd, J = 15.5, 5.0, 3.5 Hz, 1H),
2.05 (ddd, J = 12.5, 3.5, 3.5 Hz, 1H), 1.96–1.84 (m, 3H), 1.77–1.70
(m, 3H), 1.65–1.61 (m, 2H), 1.57–1.32 (m, 10H), 1.32 (t, J = 7.5 Hz,
9H), 1.26–1.09 (m, 5H), 1.00 (s, 3H, CH319), 0.97 (d, J = 6.5 Hz, 3H,
CH321), 0.72 (s, 3H, CH318); 13C NMR (125 MHz, d4-methanol) ␦
162.9, 162.8, 75.5, 72.9, 69.7, 57.5, 51.6, 48.1 (3C), 43.9, 42.5, 41.0,
39.3, 36.9, 36.1, 36.0, 35.9, 35.5, 33.2, 32.7, 29.2, 28.0, 27.2, 24.6,
2.1.15. Sodium 3˛-hydroxy-5ˇ-cholan-24-ol sulfate (9d)
A 5 mL flask equipped with a magnetic stir bar was charged
with 13d (100 mg, 159 ␮mol, 1.0 equiv.), 1.6 mL of 95% ethanol,

296
A.C. Burns et al. / Steroids 76 (2011) 291–300
and pyridinium p-toluenesulfonate (4 mg, 16 ␮mol, 0.1 equiv.).
The reaction mixture was heated at 50 ◦C for 3 h at which time
400 ␮L of aq sodium hydroxide (15% by weight) solution was
added and the mixture was allowed to stir for 15 min at this tem-
perature. The mixture was cooled to room temperature and the
volatiles were removed in vacuo to give the crude material that
was purified by flash chromatography to provide 9d (46 mg, 62%).
White amorphous solid; mp 210.0–212.0 ◦C; Rf = 0.45 (80:18:2
chloroform/methanol/satd aq NH3); [␣]D (23 ◦C): +24.7◦ (c = 0.016,
MeOH); IR (neat): 2926, 2855, 1441, 1212 cm−1; 1H NMR (500 MHz,
d4-methanol): ␦ 3.99–3.92 (m, 2H, H24), 3.53 (dddd, J = 11.0, 11.0,
5.0, 5.0 Hz, 1H, H3), 2.02 (ddd, J = 12.5, 2.5, 2.5 Hz, 1H), 1.94–1.70
(m, 5H), 1.64–1.40 (m, 11H), 1.35 (ddd, J = 14.5, 3.0, 3.0 Hz, 1H),
1.34–1.24 (m, 3H), 1.19 (dd, J = 12.5, 3.5 Hz, 1H), 1.22–1.04 (m, 5H),
1.02–0.96 (m, 4H), 0.95 (s, 3H, CH319), 0.69 (s, 3H, CH318); 13C NMR
(125 MHz, d4-methanol) ␦ 72.6, 69.8, 58.1, 57.8, 44.0, 43.7, 42.0,
41.7, 37.4, 37.3, 36.9, 36.6, 35.8, 33.2, 31.3, 28.7, 29.4, 28.5, 27.8,
35.6, 35.4, 35.0, 33.0, 31.3, 31.2, 28.4, 27.4, 27.1, 26.5, 24.4, 23.6,
21.0, 18.5, 12.2.
A 5 mL flask equipped with a magnetic stir bar was charged
with the crude sample of 12e and 1 mL of ethyl ether. To this solu-
tion lithium aluminum hydride (9.5 mg, 250 ␮mol, 5.0 equiv.) was
added and the solution was allowed to stir for an additional 3 h.
The reaction was subsequently quenched 10 ␮L of water, 10 ␮L of
3 N sodium hydroxide, and 20 ␮L of water. The suspension was
filtered and the resultant solution was concentrated in vacuo to
give the crude alcohol that was purified by flash chromatogra-
phy to provide cholanol (13e) (11.1 mg, 64%). White amorphous
solid; mp 129.5–130.5 ◦C; Rf = 0.43 (3:1 hexane/ethyl acetate); [␣]D
(23 ◦C): +22.2◦ (c = 0.019, CHCl3); IR (neat): 2937, 2864, 1472, 1375,
1054 cm−1 1H NMR (500 MHz, CDCl3): ␦ 3.65–3.57 (m, 2H), 1.96
;
(ddd, J = 12.0, 3.5, 3.5 Hz, 1H), 1.89–1.53 (m, 7H), 1.45–1.02 (m,
22H), 0.93 (d, J = 7.0 Hz, 3H, CH321), 0.91 (s, 3H, CH319), 0.65 (s, 3H,
CH318); 13C NMR (125 MHz, CDCl3) ␦ 63.8, 56.8, 56.4, 43.9, 42.9,
40.7, 40.5, 37.8, 36.1, 35.8, 35.6, 32.0, 29.6, 28.5, 27.7, 27.4, 27.2,
−
27.3, 25.4, 24.1, 22.1, 19.2, 12.6; HRMS (ESI−) calcd for C24H41SO5
(M − Na+) 441.2680, found 441.2680.
26.8, 24.5, 24.4, 21.5, 21.0, 18.9, 12.3; LRMS (EI) calcd for C24H42
(M) 346.3, found 346.
O
2.1.16. (4R)-Methyl 4-((3R,5R,10S,13R)-3-((1H-imidazole-1-
carbonothioyl)oxy)-10,13-dimethylhexadecahydro-1H-
cyclopenta[a]phenanthren-17-yl)pentanoate
2.1.18. Sodium cholan-24-ol sulfate (9e)
Compound 9e was prepared from 13e by procedure 2.1.3 fol-
lowed by cation exchange and purification similar to that used
in procedure 2.1.4. White amorphous solid; mp 195.5–198.0 ◦C;
Rf = 0.44 (90:9:1 chloroform/methanol/satd aq NH3); [␣]D (23 ◦C):
+18.3◦ (c = 0.026, MeOH); IR (neat): 2930, 2860, 1449, 1402,
(11e)
A 100 mL flask equipped with a magnetic stir bar was charged
with methyl lithocholate (10e, 2 g, 5.12 mmol, 1.0 equiv.),
thiocarbonyldiimidazole (2.7 g, 15.4 mmol, 3.0 equiv.), 4-
dimethylaminopyridine (63 mg, 512 ␮mol, 0.1 equiv.), and 50 mL
of methylene chloride. After 5 h the mixture was quenched with
25 mL of saturated aqueous ammonium chloride. The aqueous layer
was extracted twice with 50 mL of ethyl acetate. The combined
organic layers were combined, washed with 50 mL of saturated
aqueous sodium bicarbonate and 50 mL of saturated brine, and
dried over anhydrous sodium sulfate. The volatiles were removed
in vacuo to give the crude material, which was purified by flash
chromatography to provide 11e (1.62 g, 63%). White amorphous
solid; mp 146.0–147.5 ◦C; Rf = 0.61 (1:1 hexane/ethyl acetate);
[␣]D (23 ◦C): +47.6◦ (c = 0.029, CHCl3); IR (neat): 2949, 2870, 1740
1213 cm−1 1H NMR (500 MHz, d4-methanol): ␦ 3.99–3.92 (m,
;
2H, H24), 2.01 (ddd, J = 12.5, 3.5, 3.5 Hz, 1H), 1.95–1.71 (m, 6H),
1.61–1.09 (m, 23H), 0.96 (d, J = 7.5 Hz, CH321), 0.95 (s, 3H, CH319),
0.69 (s, 3H, CH318); 13C NMR (125 MHz, d4-methanol) ␦ 69.8, 58.1,
57.8, 45.4, 44.0, 42.0, 41.8, 38.9, 37.4, 36.9, 36.6, 33.3, 29.5, 28.8,
28.5, 28.4, 27.9, 27.3, 25.5, 25.0, 22.5, 22.1, 19.3, 12.7; HRMS (ESI−)
calcd for C24H41SO4 (M − Na+) 425.2731, found 425.2736.
−
2.1.19. Sodium 3-keto-5ˇ-petromyzonol sulfate (9f)
A 25 mL flask equipped with a magnetic stir bar was charged
with 9a (300 mg, 604 ␮mol, 1.0 equiv.), 400 ␮L of water, and
1.6 mL of t-butanol. To this solution was added potassium bromide
(144 mg, 1.21 mmol, 2.0 equiv.), potassium bicarbonate (605 mg,
6.04 mmol, 10.0 equiv.), and 2,2,6,6-tetramethylpiperidine-1-oxyl
(104 mg, 664 ␮mol, 1.1 equiv.). The solution was cooled to 0 ◦C
and 5% sodium hypochlorite (7.2 mL, 4.83 mmol, 8.0 equiv.) was
added in 1.8 mL portions at 1-h intervals. After the final addi-
tion of sodium hypochlorite, the mixture was allowed to stir for
one additional hour (5 h total). The resultant reaction mixture was
quenched at 0 ◦C with 1.2 g sodium thiosulfate pentahydrate in
10 mL of water and allowed to come to room temperature. The
aqueous layer was extracted ten times with 10% methanol in
chloroform (10 × 40 mL). The volatiles were removed in vacuo to
give the crude material, which was purified by flash chromatog-
raphy (80:18:2–80:26:2 chloroform/methanol/satd aq ammonia)
to provide 9f (188 mg, 63%). White amorphous solid; mp > 230 ◦C;
Rf = 0.43 (80:26:2 chloroform/methanol/satd aq NH3); [␣]D (23 ◦C):
+23.6◦ (c = 0.035, MeOH); IR (neat): 2946, 2870, 1700 (sharp),
(sharp), 1465, 1387, 1339, 1285, 1233 cm−1 1H NMR (500 MHz,
;
CDCl3): ␦ 8.35 (dd, J = 1.0, 1.0 Hz, 1H, H2), 7.65 (dd, J = 2.0, 1.0 Hz,
1H, H5), 7.02 (dd, J = 2.0, 1.0 Hz, 1H, H4), 5.45 (dddd, J = 11.0, 11.0,
4.5, 4.5 Hz, 1H, H3), 3.67 (s, 3H), 2.35 (ddd, J = 15.5, 10.0, 5.0 Hz, 1H,
H23a), 2.20 (ddd, J = 16.0, 10.0, 6.5 Hz, 1H, H23b), 2.06–1.77 (m, 8H),
1.66–1.53 (m, 3H), 1.45–1.02 (m, 15H), 0.98 (s, 3H, CH319), 0.92
(d, J = 6.5 Hz, 3H, CH321), 0.66 (s, 3H, CH318); 13C NMR (125 MHz,
CDCl3) ␦ 183.7, 174.9, 137.0, 130.8, 118.1, 84.6, 56.6, 56.2, 51.7,
42.9, 42.1, 40.7, 40.2, 36.0, 35.5, 35.0, 34.8, 31.6, 31.2, 31.1, 28.4,
27.2, 26.5, 26.1, 24.4, 23.4, 21.0, 18.5, 12.2; HRMS (ESI+) calcd for
C
29H44N2SO3Na+ (M + Na+) 523.2965, found 523.2963.
2.1.17. Cholanol (13e)[20]
A 10 mL flask equipped with a magnetic stir bar was charged
with tri-n-butyltin hydride (54 ␮L, 200 ␮mol, 4.0 equiv.) and 3.3 mL
of toluene. The flask was equipped with a condenser and heated at
reflux while a solution of 11e (25 mg, 50 ␮mol, 1.0 equiv.) and azo-
bisisobutyronitrile (1.6 mg, 5 ␮mol, 0.2 equiv.) in 1.7 mL of toluene
was added over 1 h. The solution was maintained at reflux for
an additional 3 h and subsequently cooled to room temperature,
at which time the volatiles were removed in vacuo to provide
the crude sample of 12e, which was used without purification in
the subsequent step. 1H NMR (500 MHz, CDCl3): ␦ 3.67 (m, 2H),
2.35 (ddd, J = 15.5, 10.5, 5.0, 1H, H23a), 2.21 (ddd, J = 16.0, 9.5, 6.5,
1H, H23b), 1.95 (ddd, J = 6.5, 2.5, 2.5 Hz, 1H), 1.89–1.69 (m, 6H),
1.59–1.51 (m, 2H), 1.43–0.93 (m, 19H), 0.92 (s, 3H, CH319), 0.90
(d, J = 6.5 Hz, 3H, CH321), 0.64 (s, 3H, CH318); 13C NMR (125 MHz,
CDCl3) ␦ 175.0, 64.5, 56.6, 56.1, 51.7, 42.9, 42.2, 40.5, 40.3, 36.0,
1221 cm−1 1H NMR (500 MHz, d4-methanol): ␦ 4.02 (dd, J = 2.5,
;
2.5 Hz, 1H, H12), 3.99–3.95 (m, 2H, H24), 3.86 (ddd, J = 3.5, 3.5,
3.5 Hz, 1H, H7), 3.54 (dd, J = 15.5, 14.0 Hz, 1H, H4␣), 2.60 (ddd,
J = 14.5, 14.5, 5.5 Hz, 1H, H2␣), 2.44 (ddd, J = 12.5, 12.5, 4.5 Hz, 1H,
H2␤), 2.15–1.09 (m, 21H), 1.04 (d, J = 6.5 Hz, 3H, CH321), 1.03 (s,
3H, CH319), 0.76 (s, 3H, CH318); 13C NMR (125 MHz, d4-methanol)
␦ 216.8, 74.0, 69.8, 69.0, 47.7, 45.1, 45.0, 43.1, 41.2, 38.0, 37.8,
37.1, 36.3, 35.2, 35.1, 33.3, 30.0, 28.9, 28.5, 27.4, 24.3, 22.3, 18.1,
13.2; HRMS (ESI−) calcd for C24H39SO7− (M − Na+) 471.2422, found
471.2412.

A.C. Burns et al. / Steroids 76 (2011) 291–300
297
2.2. Biology: olfactory activity and specificity
same sets of olfactory receptor(s). We have used both in the past
[5]. Briefly, to determine the ICI, EOG responses were measured to
equipotent concentrations of each member of an odorant pair (Ra
and Rb) and then to a mixture of the two stimuli (Rab). The ratio of
the sum of the former to the latter was then calculated (Eq. (1)).
An ICI value of 1.0 describes total independence. To determine the
MDI, EOG responses were measured to twice the initial concentra-
tion of each odorant (R2a or R2b), and these values were summed
and averaged. The ratio of this value to the EOG response elicited to
the simple mixture of the two (Rab) was calculated using Eq. (2). An
MDI value of 1.0 describes a total lack of independence and occurs
when odorants are discerned by exactly the same set of receptor(s).
Each of these indices is a nonlinear measure and the two need to be
interpreted in concert when odorants are discerned by overlapping
yet different sets of receptors.
The olfactory activity of synthesized steroids was assayed
using electro-olfactogram (EOG) recording, which is an extra-
cellular, multiunit electrophysiological recording method that
measures summated generator potentials from olfactory recep-
tors [2]. This technique had previously been successfully employed
to describe the olfactory potency and structure–activity relation-
ships of steroidal odorants in the sea lamprey [4,5,7]. Three types of
experiments were performed in a sequential fashion to first identify
structures that had significant olfactory activity and then to deter-
mine whether and how they bound to the olfactory receptor(s) that
recognize native allo-compounds. The last step involved two types
of mixture experiments that used both natural (allo-) and unnat-
ural (5␤−) compounds to test whether and how different steroids
bound with similar receptors.
For EOG recording small (<1 mL) aliquots of test odorants are
applied to the fish’s nose while electrical responses are measured
from the surface of the olfactory tissue that contains the olfac-
tory receptors using established protocols [4,5]. For this study,
sea lamprey were anaesthetized with an intramuscular injec-
tion of metomidate hydrochloride (Syndel, Vancouver, BC, Canada,
3 mg kg−1 body weight), immobilized with an intramuscular injec-
tion of Flaxedil (gallamine triethiodide; Sigma, 150 mg kg−1 body
weight), and placed on a stand where their gills were perfused
with 11 ◦C well water. Their olfactory epithelium was exposed
by surgically removing the dorsal portion of their cartilaginous
nasal capsules and perfused with flowing well water. Differen-
tial EOG responses were recorded using two Ag/AgCl electrodes
(type EH-1S; WPI, Sarasota, FL, USA) filled with 3 mol L−1 KCl and
bridged to saline (8%) gelatin-filled glass capillaries with tip diame-
ters of about 400 ␮m. The recording electrode was placed between
two lamellae and adjusted to maximize responses to the standard
(10−5 M l-arginine) while minimizing responses to blank water
control. A reference electrode was placed on the skin near the
naris. The preparation was grounded with a surgical needle that
was inserted under the skin and the needle was attached with a
wire to the recording cage. Electrical signals were amplified using
a DC-preamplifier (Model P16, Grass, Quincy, MA, USA), digitized
by a MacLab/4 (Analog Digital Instruments, Castle Hill, Australia),
and displayed on an Apple Macintosh computer. Test stimuli were
applied in a stream of well water as 5-s pulses using a device that
minimized pressure and temperature fluctuations. All stimuli were
tested at least twice with 2 min breaks to permit recovery. Con-
trols were performed at the beginning and end of each series of
compounds being tested or once an hour (whichever came first).
If control responses differed by more than 20% (a rare occurrence),
the entire series of measurements was re-run. For analysis, peak
responses were measured, blank response subtracted (if present),
and the net values expressed relative to the most recent standard.
All odorants were tested on at least four lampreys. In the first exper-
iment the sensitivity of the sea lamprey olfactory epithelium was
determined to nanomolar concentrations of six analogs of PS (1)
to gain initial insight into structure–activity relationships and to
identity the most stimulatory compounds. Next, dose–response
relationships were determined for the most stimulatory agents by
testing them at concentrations ranging from 10−12 to 10−7 M. Last,
two types of binary mixture studies were conducted using the test
compounds and PADS (2) to address whether these compounds
stimulate the same (or different) sets of olfactory receptors and,
thus, might be discerned from each other by sea lampreys.
Rab
ICI =
(1)
(2)
Ra + Rb
Rab
MDI =
0.5(R2a + R2b
)
3. Results and discussion
3.1. Chemical synthesis
The synthesis of 5␤-petromyzonol sulfate (9a) [17] was
achieved by protection of cholic acid (10a) as the triformate
ester 11a (formic acid, 55 ◦C) [14] and subsequent chemoselective
reduction [15] of the carboxylic acid with borane–dimethylsulfide
complex to provide the primary alcohol 12a (Scheme 1). Sulfation
of 12a was effected by a modified procedure in which a suspen-
sion of sulfur trioxide–pyridine complex in methylene chloride was
solubilized by the addition of triethylamine (4 equiv.); this permits
sulfation to occur at ambient temperature rather than the higher
temperatures typically required for sulfation using sulfur–trioxide
in pyridine. The triformate 13a was saponified (aq NaOH, MeOH)
with concomitant triethylammonium–sodium exchange to provide
9a. The analogous compounds 9b and 9c, derived from deoxycholic
acid (10b) and chenodeoxycholic acid (10c) respectively, were syn-
thesized using a very similar protocol.
Lithocholic acid (10d) was converted to the primary alcohol 12d
using a one-flask/two-reaction sequence (Scheme 2), presumably
via the THP-ester 11d [19]. The primary alcohol 12d was sulfated
(SO3·Py, Et3N) to give the THP-ether 13d, which was converted to
the secondary alcohol 9d (pPTS, EtOH).
For the synthesis of 9e (Scheme 3), methyl lithocholate (10e)
was converted to the thiocarmabate 11e (TCDI, DMAP; Scheme 3)
followed by a modified Barton–McCombie deoxygenation [21,22].
The deoxygenation was sensitive to the reaction conditions;
formation of a by-product [23] arising from the reduction of
the thiocarbamate moiety was observed under the standard
Barton–McCombie protocol. This could be overcome by the slow
addition (i.e., over 1 h) of the substrate and azobisisobutyronitrile
(AIBN) to a dilute solution of an excess of tri-n-butyltin hydride
(0.06 M, 4.0 equiv.) at toluene reflux [22].
Selective oxidation of 5␤-petromyzonol sulfate (9a) with
TEMPO under aqueous conditions gave 3-keto-5␤-petromyzonol
sulfate (9f, Scheme 4). Under these conditions the regioselective
oxidation of the C3 carbinol in the presence of the unprotected C7
and C12 alcohols is noteworthy; the use of a stoichiometric amount
of inexpensive TEMPO and the slow addition of the stoichiometric
oxidant (5% NaOCl) was important to minimize unselective back-
ground oxidation effected by sodium hypochlorite.
The Independent Component Index (ICI) was determined
because it provides insight into the extent to which odorants stim-
ulate different (independent) sets of olfactory receptors while the
Mixture Discrimination Index (MDI) was determined because it
provides insight into the extent to which odorants stimulate the

298
A.C. Burns et al. / Steroids 76 (2011) 291–300
Scheme 2. Synthesis of lithocholic acid derivative 9d.
Scheme 3. Synthesis of cholan-24-ol sulfate (9e).
3.2. Olfactory activity
response. Ketone 9f, a mimic of the natural sex pheromone, 3-
ketopetromyzonol sulfate [8–10] produced an olfactory response,
but it was small. This may imply that 3-ketopetromyzonol sul-
fate receptors are less promiscuous than the PS receptors; indeed,
earlier cross-adaptation studies [7] demonstrate that receptors for
3-ketopetromyzonol sulfate and PS (1) are not completely indepen-
dent of each other. Finally, the des-oxygenated compound 9e did
not elicit an olfactory response, suggesting that some oxygenation
is required for binding (although micellization of this detergent like
molecule cannot be ruled out as a contributing factor). The entire
steroid framework appears to contribute to some extent to binding
by the PS receptor.
All EOG waveforms measured in this study were typical of those
we have measured before [5]. They showed a characteristic sharply
negative phasic component preceding a peak (typically between 0.2
and 3 mV in magnitude) that decayed back to baseline over a 5–10 s
period. The averaged results from EOG tests of 5␤-petromyzonol
sulfate (9a) and five additional analogs (9b–f) are summarized in
Fig. 4. Although the tris-oxygenation present in 9a, the 5␤-analog
of natural PS (1), was not required to elicit an olfactory response
at a concentration of 10−9 M, this attribute certainly contributes to
olfactory activity because 9a is the most stimulatory compound in
this series. Nevertheless, bis-oxygenation at both C3, C7 and C3, C12
(9c and 9b, respectively) elicited a modest olfactory response. The
C3 mono-oxygenated compound 9d generated only a small EOG
As shown in Fig. 5, 5␤-petromyzonol sulfate (9a) elicited
relatively large EOG responses from the sea lamprey olfactory
epithelium. The dose–response curve for 9a was shifted to the right
Scheme 4. Selective oxidation of 5␤-petromyzonol sulfate.

A.C. Burns et al. / Steroids 76 (2011) 291–300
299
than 1.0 (MDI = 1.32, P = 0.024)]. These data are in agreement with
previous cross-adaptation results [7] that indicate that PS (1) and
PADS (2) are discerned from one another and have different recep-
tor activation structural requirements.
4. Conclusions
This study has resulted in the synthesis of several new bile
acid derivatives. These were used to demonstrate that sea lamprey
olfactory receptors for PS (1) are relatively insensitive to the allo-
configuration at C5 but that structural features elsewhere in the
molecules are discerned. 5␤-Petromyzonol sulfate (9a) elicits EOG
responses that are only slightly smaller than those of natural (allo)
PS (1). Further, mixture studies demonstrated that 9a and 1 inter-
act with the same receptor (ICI = ca. 1). This result is analogous to
previous findings for allocholic (7) vs. cholic (10a) acid [5]. In the
only other example of a steroidal odorant where binding properties
have been explored in detail – the goldfish sex pheromone 17,20␤-
dihydroxy-4-pregenen-3-one – single changes in the structural
features decreased binding affinity by 1–2 orders of magnitude
[24]. The present work also sheds some insight on possible differ-
ences between olfactory receptors for modern teleost (bony) fish
and lampreys. In particular, single unit recording from the catfish
olfactory bulb suggests that this teleost’s olfactory receptors are
more sensitive to hydroxylation at the C7 and C12 positions than
those of the sea lamprey [25]. In other ways both groups of fishes
are similar. They both exhibit specific and remarkable sensitivities
for steroids with allo-configurations as well as taurine, glycine, and
sulfated conjugates, although the details of these relationships have
yet to be fully explored.
The structural influence on olfactory activity in these unnat-
ural 5␤-steroidal sulfates was further examined as a function of
oxygenation at the C3-, C7-, and C12-positions of the steroid skele-
ton. Interestingly, the C7-hydroxy substituent appears to be more
important to olfactory activity than that of the C12-substituent
(cf. 9b vs. 9c, Fig. 5). Previous results have shown that the C24-O-
sulfate is also critical [5]. It appears that the PS (1) binding pocket
is discerning a large portion, but not the entirety, of each of these
structures. Because allo-configured steroids are more difficult and
expensive to synthesize than their 5␤-counterparts, this informa-
tion should be useful in the further development of pheromone
analogs (e.g., 5␤-PADS) for use in controlling the sea lamprey,
which is invasive in the Great lakes of North America.
Fig. 5. Semi-logarithmic plots of the dose–response relationship of PS (1) and 9a.
Response magnitudes are presented as a percentage of the standard, 10−5 M l-
arginine. Vertical bars represent one standard error (n = 5 animals).
(less responsive) of that of PS (1) by only 1–2 orders of magnitude.
The latter had a detection threshold of ∼10−11 M and the former
∼10−10 M. This result is analogous to earlier findings with allocholic
(7) vs. cholic acids (10a) [5] where the allo-configured ligand also
elicited a stronger response. Moreover, the dose–response curves
for 7 vs. 10a were parallel with the latter shifted to the right of the
former (allo isomer) by 1–2 orders of magnitude. Mixture exper-
iments using 7 and 10a showed an ICI suggesting that they also
interact with the same receptor [5]. Together, these results sug-
gest moderate tolerance of 5␤-steroids by sea lamprey PS olfactory
receptor(s).
Binary mixture experiments using either PS (1) or PADS (2) with
9a were performed. The results are shown in Fig. 6 and suggest
that 1 and 9a are detected by the same set of olfactory recep-
tor(s) while 2 and 9a are not. In the first instance, while the ICI
for 1/9a was less than 1.0 (ICI = 0.67; p = 0.005 when compared
with 1.0 by t-test), the MDI was not (MDI = 1.05; p = 0.789 when
compared with 1.0). This can be taken as additional evidence for
recognition by the same binding sites (the PS-receptor), and indi-
cates that the PS-receptor sites are fairly promiscuous with respect
to the topology of the steroid nucleus (i.e., cis-decalin-like vs. trans-
decalin-like). In the second instance, while the ICI for the 2/9a
mixture was very close to 1.0 (ICI = 0.82), the MDI was far greater
Acknowledgments
This work is the result of research sponsored by: (a) The
Minnesota Sea Grant College Program supported by the NOAA
office of Sea Grant, United States Department of Commerce, under
grant no. NA07OAR4170009. The U.S. Government is authorized
to reproduce and distribute reprints for government purposes, not
withstanding any copyright notation that may appear hereon. This
paper is journal reprint no. JR 578 of the Minnesota Sea Grant Col-
lege Program. (b) The United States National Institutes of Health
(CA-76497). Neither agency was involved in study design, data
collection or interpretation, or writing or decision to submit this
manuscript.
References
[1] Døving KB, Selset R, Thommesen G. Olfactory sensitivity to bile acids in
salmonid fishes. Acta Physiol Scand 1980;108:123–31.
[2] Sorensen PW, Caprio J. Chemoreception. In: Evans DH, editor. The physiology
of fishes. Boca Raton, USA: CRC; 1998. p. 251–61.
[3] Sorensen PW, Hoye TR. A critical review of the discovery and application of a
migratory pheromone in an invasive fish, the sea lamprey, Petromyzon marinus
L. J Fish Biol 2007;71(Suppl. D):100–14.
Fig. 6. The independent component index (ICI) and mixture discrimination index
(MDI) of binary mixtures of 9a with PS (1) (n = 5 animals) or with petromyzonamine
disulfate (PADS, 2) (n = 3 animals). Vertical bars represent one standard error.

300
A.C. Burns et al. / Steroids 76 (2011) 291–300
[4] Li W, Sorensen PW, Gallaher DD. The olfactory system of migratory adult sea
[13] Hwang D, Yeh YH, Lai YS, Deng JF. Identification of cyprinol and cyprinol sulfate
from grass carp bile and their toxic effects in rats. Toxicon 2001;39:411–4.
[14] Cortese F, Bauman L. A synthesis of conjugated bile acids. I. Gycolic acid. J Am
Chem Soc 1935;57:1393–5.
[15] Babu P, Maitra U. Synthesis and in vitro cholesterol dissolution by 23- and
24-phosphonobile acids. Steroids 2005;70:681–9.
lamprey (Petromyzon marinus) is specifically and acutely sensitive to unique
bile acids released by conspecific larvae. J Gen Physiol 1995;105:569–87.
[5] (a) Li W, Sorensen PW. Highly independent olfactory receptor sites for naturally
occurring bile acids in the sea lamprey, Petromyzon marinus. J Comp Physiol A
1997;180:429–38;
(b) Fine JM, Vrieze LA, Sorensen PW. Evidence that petromyzontid lampreys
employ a common migratory pheromone that is partially comprised of bile
acids. J Chem Ecol 2004;30:2091–110.
[16] Starchenkov I, Trapencieris P, Kauss V, Jas G, Kalvinsh I. A convenient syn-
thesis of 5␤-cholestan-26-oic and 5␤-cholestan-26,27-dioic acids. Steroids
2000;65:143–7.
[6] Sorensen PW, Fine JM, Dvornikovs V, Jeffrey CS, Shao F, Wang J, et al. Mixture of
new sulfated steroids functions as a migratory pheromone in the sea lamprey.
Nat Chem Biol 2005;1:324–8.
[17] Kuramoto T, Kihira K, Matsumoto N, Hoshita T. Determination of the sulfated
position in 5␤-bufol sulfate by a carbon-13 nuclear magnetic resonance study.
Chem Pharm Bull 1981;29:1136–9.
[7] Fine JM, Sorensen PW. Isolation and biological activity of the multi-component
sea lamprey migratory pheromone. J Chem Ecol 2008;34:1259–67.
[8] Li W, Scott AP, Siefkes MJ, Yan H, Liu Q, Yun S, et al. Bile acid secreted
by male sea lamprey that acts as a sex pheromone. Science 2002;296:138–
41.
[18] Hoehn WM, Moffett RB. Preparation of phenyl ketones from bile acids. J Am
Chem Soc 1945;67:740–3.
[19] Banerjee S, Trivedi GK. Synthesis of potential spin probes for
biomembranes–TEMPO and proxyl nitroxides of lithocholic acid. Tetrahedron
Lett 1992;48:9939–50.
[9] Johnson NS, Yun S, Thompson HT, Brant CO, Li W. A synthesized pheromone
induces upstream movement in female sea lamprey and summons them into
traps. Proc Natl Acad Sci 2009;106:1021–6.
[20] Cushman M, Golebiewski WM, Pommier Y, Mazumder A, Reymen D, Clercq
ED, et al. Cosalane analogues with enhanced potencies as inhibitors of HIV-1
protease and integrase. J Med Chem 1995;38:443–52.
[10] Siefkes MJ, Li W. Electrophysiological evidence for detection and discrimination
of pheromonal bile acids by the olfactory epithelium of female sea lampreys
(Petromyzon marinus). J Comp Physiol A 2004;190:193–9.
[11] (a) Sorensen PW, Stacey NE. Brief review of fish pheromones and discussion
of their possible uses in the control of non-indigenous teleost fishes. N Z J Mar
Freshwater Res 2004;38:399–417;
[21] Barton DHR, McCombie SW. A new method for the deoxygenation of secondary
alcohols. J Chem Soc, Perkin Trans 1975;1:1574–82.
[22] Chochrek P, Wicha J. An expedited approach to the vitamin D trans-hydrindane
building block from the Hajos dione. Org Lett 2006;8:2551–3.
[23] The by-product observed is assigned the thiohemiacetal (HSCH2O-steroid)
based on its 1H NMR spectrum (500 MHz, CDCl3): (4.80 (d, J = 9.0 Hz, HSCH2OR)
and 2.02 (t, J = 9.0 Hz, HSCH2OR) See, e.g., Kulkarni YS, Niwa M, Ron E,
Snider BB. Synthesis of terpenes containing the bicyclo[3.1.1]heptane ring
system by the intramolecular [2 + 2] cycloaddition reaction of vinylketenes
with alkenes. Preparation of chrysanthenone, ␤-pinene, ␤-cis-bergamotene,
(b) Goto T, Holzinger F, Hagey LR, Cerre C, Ton-Nu HT, Schteingart CD, et al.
Physiochemical and physiological properties of 5␣-cyprinol sulfate, the toxic
bile salt of cyprinid fish. J Lipid Res 2003;44:1643–51;
(c) Baker CF, Carton AG, Fine JM, Sorensen PW. Can bile acids function as migra-
tory pheromones in banded kokopu Galaxius fasciatus (Gray)? Ecol Freshwater
Fishes 2006;15:275–83;
␤-trans-bergamotene, ␤-copaene, and ␤-ylangene and lemnalol
J Org
Chem.1987;52 1568–76.
(d) Tammar AR. Bile salts in fishes. In: Florkin M, Scheer BT, editors. Chemical
zoology, vol. VIII. San Francisco: Academic Press; 1974. p. 595–612.
[12] Asakawa M, Noguchi T, Seta H, Furihata K, Fujikura K, Hashimoto K. Struc-
ture of the toxin isolated from carp (Cyprinus carpio) bile. Toxicon 1990;28:
1063–9.
[24] Sorensen PW, Hara TJ, Stacey NE, Dulka JG. Extreme olfactory specificity of
male goldfish to the preovulatory steroidal pheromone 17␣,20␤-dihydroxy-4-
pregnen-3-one. J Comp Physiol A 1990;166:373–83.
[25] Rolen SH, Caprio J. Processing of bile acid salt odor information by single olfac-
tory bulb neurons in the channel catfish. J Neurophysiol 2007;97:4058–68.