Synthesis and Evaluation of Osteogenic Oxysterols as Hedgehog Pathway Activators

Article abstract of DOI:10.1002/cmdc.201500550

Oxysterols (OHCs) are metabolic byproducts of cholesterol that are known to function as agonists of the Hedgehog (Hh) signaling pathway. Previously, we reported 23(S)-hydroxycholesterol [23(S)-OHC, 4] as a potent activator of Hh signaling with the ability to functionally differentiate mouse embryonic fibroblasts to an osteogenic fate. To obtain 23(S)-OHC in quantities suitable for invivo evaluation, we developed a revised synthetic route that decreases the number of steps and chromatographic purifications, and which also enhances the stereoselective nature of the synthesis. This new route also allows access to the C21 methyl group of the OHC scaffold, and several new analogues with varying stereochemistry at this location were evaluated for their ability to up-regulate the Hh pathway.

Full text of DOI:10.1002/cmdc.201500550

DOI: 10.1002/cmdc.201500550
Full Papers
Synthesis and Evaluation of Osteogenic Oxysterols as
Hedgehog Pathway Activators
Chad A. Maschinot, Audrey R. Corman, Albert M. DeBerardinis, and M. Kyle Hadden*[a]
Oxysterols (OHCs) are metabolic byproducts of cholesterol that
are known to function as agonists of the Hedgehog (Hh) sig-
naling pathway. Previously, we reported 23(S)-hydroxycholes-
terol [23(S)-OHC, 4] as a potent activator of Hh signaling with
the ability to functionally differentiate mouse embryonic fibro-
blasts to an osteogenic fate. To obtain 23(S)-OHC in quantities
suitable for in vivo evaluation, we developed a revised synthet-
ic route that decreases the number of steps and chromato-
graphic purifications, and which also enhances the stereoselec-
tive nature of the synthesis. This new route also allows access
to the C21 methyl group of the OHC scaffold, and several new
analogues with varying stereochemistry at this location were
evaluated for their ability to up-regulate the Hh pathway.
Introduction
The Hedgehog (Hh) signaling pathway is a developmental sig-
naling pathway that is essential for proper differentiation of
a variety of tissues during embryonic development.[1] Aberrant
activation of Hh signaling has been linked to several human
malignancies, and multiple small-molecule inhibitors of Hh
signal transmission are in preclinical or clinical trials as anti-
cancer chemotherapeutics. In contrast, the development of
pathway agonists as therapeutic agents has received less at-
tention. The therapeutic potential of Hh agonists exists primar-
ily with neurological disorders[2] such as Parkinson’s disease
(PD) and amyotrophic lateral sclerosis (ALS), as well as osteode-
generative disorders[3] in which de novo bone formation is of
interest, such as bone fracture repair and osteoporosis. Among
the Hh agonists that have been studied, oxysterols (OHCs)
have shown promise with the activation of Hh signaling, result-
ing in osteo-inductive effects in vitro and bone formation
in vivo.[4–7]
Figure 1. Natural and synthetic OHC agonists of Hh signaling.
tance of incorporating a hydroxy group in the alkyl side
chain.[11,12] Potent activation of the Hh pathway also depends
on the location and stereochemical orientation of the hydroxy
moiety. Whereas 20(S)-OHC is a potent Hh agonist, 20(R)-OHC
is significantly less active. In addition, the most active synthetic
OHC agonists of the Hh pathway disclosed to date (represent-
ed by Oxy34) all incorporate the 20(S)-hydroxy stereochemistry
and the C21 methyl group.
Naturally occurring OHCs are formed as metabolic byprod-
ucts of cholesterol oxidation and exert a range of physiological
effects through multiple cellular receptors.[8] One of the most
potent naturally occurring OHCs, 20(S)-OHC (1, Figure 1) has
been used to characterize the OHC class of sterols as activators
of Hh signaling that function through direct binding to
Smoothened (Smo), a seven-transmembrane G-protein-like re-
ceptor that serves as a key regulator of the Hh signaling cas-
cade. These studies have demonstrated that OHCs bind to the
extracellular cysteine-rich domain (CRD) of Smo.[9,10]
We previously performed a series of SAR studies focused on
the OHC side chain to further explore the stereo- and regiose-
lective specificity of Hh activation for this region of the scaf-
fold.[11] The most active compound to emerge from our study,
23(S)-OHC (4), was found to possess potency similar to that of
20(S)-OHC (EC50: 0.57 and 0.52 mm, respectively). Compared
with 20(S)-OHC, 23(S)-OHC also demonstrated greater selectivi-
ty for Hh agonism relative to the liver X receptor (LXR), which
has been identified as a negative regulator of the Hh pathway
and which maintains affinity for many endogenous OHC ago-
nists of Hh signaling.[13] OHC 4 was also found to induce osteo-
genic differentiation and osteoblast formation in cultured M2-
Previous structure–activity relationship (SAR) studies for OHC
agonists of Hh signaling have continually identified the impor-
[a] C. A. Maschinot, Dr. A. R. Corman, Dr. A. M. DeBerardinis,
Prof. M. K. Hadden
Department of Pharmaceutical Sciences, University of Connecticut
69 North Eagleville Road, Unit 3092, Storrs, CT, 06269-3092 (USA)
E-mail: kyle.hadden@uconn.edu
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cmdc.201500550.
ChemMedChem 2016, 11, 679 – 686
679
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
10B4 cells, highlighting its therapeutic potential; however, the
initial synthesis of 4 was low yielding (1.8%) and required
twelve steps from commercially available hyodeoxycholic
acid.[11] For this reason, we sought to develop a more direct
and higher-yielding procedure for the synthesis of 23(S)-OHC.
In addition, we used both our revised route and a slightly
modified procedure to synthesize and evaluate OHC analogues
that incorporate modifications to the C21 methyl group as
a means to explore the importance of this functionality and its
stereochemical orientation.
Table 1. Reduction of a,b-unsaturated aldehyde to provide compounds 9
and 10.
Entry Reagent(s)
Results
1
5% Pd/C, H2
Nonselective olefin
reduction
2
3
10% Pd/C, H2
PtO2
Complete saturation
Complete saturation
and deprotection
No reaction
No reaction
9, 60%
4
5
6
[(PPh3)CuH]6
(R)-Ru(OAc)2(BINAP)
(S)-(+)-2-(tert-butyl)-3-methyl-4-
imidazolidinone TFA, Hantzsch ester
(R)-(+)-2-(tert-butyl)-3-methyl-4-
imidazolidinone TFA, Hantzsch ester
7
10, 60%
Results and Discussion
Our revised synthetic route began with protection of commer-
cially available pregnenolone as the tetrahydropyran (THP)
ether (Scheme 1). Notably, the THP protecting group was
interest, given its ability to convert both E and Z alkenes into
a single b-stereogenic aldehyde.[15] Using the Hantzsch ester as
hydride source and (S)-imidazolidinone, the 8a/8b mixture
was converted into aldehyde 9 [20(R) configuration] in good
overall yield with no formation of stereoisomer 10. Performing
the same protocol on the 8a/8b mixture with (R)-imidazolidi-
none provided 10 [20(S) configuration] in good yield as the
singular reduction product.
We next looked to perform an enantioselective Grignard ad-
dition to aldehyde 9 to selectively obtain 4. Multiple attempts
were taken (see Supporting Information) following known pro-
cedures. The first series of asymmetric Grignard attempts used
Ti(OiPr)4, chiral BINOL, and a chelating agent (BDMAEE, bis[2-
(N,N-dimethylamino)ethyl] ether).[16] After multiple alterations
to the procedure with no success, we thought the Grignard re-
agent may be too deactivated with the use of the chelating
BDMAEE, making the reaction unsuccessful even with heating
due to the deactivated nature of the alkyl-aldehyde starting
material. Several other attempts were taken without the use of
BDMAEE following a second procedure.[17] With up to 10 equiv-
alents of Grignard used and no reaction observed for either
method, we believe that the enolizable proton of 9 may result
in its enolization by Ti(OiPr)4, preventing the Grignard addition
from occurring.
Scheme 1. Asymmetric synthesis of aldehydes 9 and 10. Reagents and condi-
tions: a) 3,4-dihydro-2H-pyran, p-TsOH, 12 h, RT, 96%; b) vinyl MgBr,
À788C!RT, 12 h, 97%; c) PCC, NaOAc, RT, 12 h, 64%.
chosen because it had previously resulted in a more straight-
forward chromatographic separation of the 23(S)- and 23(R)-
OHC stereoisomers. Following a previously reported proce-
dure,[14] Grignard addition of vinyl magnesium bromide gave
a mixture (1:9) of the tertiary allylic alcohols 7a and 7b. Treat-
ment of the mixture with pyridinium chlorchromate resulted in
oxidative rearrangement of the allylic alcohol to yield an 80:20
E/Z mixture of b-substituted-a,b-unsaturated aldehydes, 8a
and 8b.
A key requirement of this synthetic route was the ability to
reduce the a,b-unsaturation in 8a and/or 8b in a stereoselec-
tive fashion to provide both the R and S configurations of the
C21 methyl moiety. Initial attempts using standard palladium
or platinum catalysts resulted in nonselective reduction of the
side chain olefin or complete saturation of the OHC scaffold
(Table 1, Entries 1–3). Neither Stryker’s reagent (Entry 4) nor the
ruthenium BINAP complex (Entry 5) effected reduction of
either alkene in the scaffold. Following these unsuccessful at-
tempts, we turned to the iminium catalysts previously devel-
oped in the MacMillan research group, a reaction of particular
Following our unsuccessful attempts to prepare 23(S)-OHC
from aldehyde 9 in a diasteroselective fashion using enantio-
pure reagents, we used standard Grignard addition of isobutyl
magnesium bromide to either 9 or 10 to provide the corre-
sponding 23(S)- and 23(R)-OHC analogues as a 1:1 mixture that
was easily separable via standard silica gel chromatography
(Scheme 2). Final removal of the THP protecting group afford-
ed 23-hydroxylated OHCs 4, 13, and 16–17 in excellent yield.
This revised synthetic route afforded 23(S)-OHC (4) in an over-
all yield of 14.2% in six steps from commercially available pre-
gnenolone, 5, 28.3% overall for both isomers. Previous assign-
ments of the stereochemistry at C23 were based on a combina-
1
tion of thin-layer chromatography (TLC), H and 13C NMR spec-
troscopy, and circular dichroism analysis.[11] We used crystal
structure analysis of 4 and 13 to both unambiguously assign
the stereochemistry of the C23 hydroxy group and for use as
a base for assigning the stereochemistry of 16–17 and other
OHC analogues (Supporting Information).
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
680
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
sis of THP-protected intermediates 22 and 23 (Supporting In-
formation).
OHC analogues 4, 13, 16–17, and 24–25 were initially evalu-
ated for their ability to selectively activate Hh signaling in M2-
10B4 cells, a multipotent murine bone stromal cell line previ-
ously used as a model to evaluate small-molecule Hh agonists.
The ability of each analogue to up-regulate known Hh (GLI1)
or LXR (ABCA1) target genes was measured. For these studies,
DMSO was used as a control (set at 1.0), and all OHC ana-
logues were evaluated at 5 mm. Values in Table 2 represent the
Table 2. Activity of OHC analogues in M2-10B4 cells.
Scheme 2. Synthesis of 23-OHC analogues. Reagents and conditions:
a) iBuMgBr, THF, À788C!RT, 72–80%; b) 2n HCl, MeOH/THF, 96–97%.
OHC[a]
GLI1[b]
1.0
ABCA1[b]
Hh Selectivity[c]
DMSO
1
4
13
16
17
24
25
1.0
–
1.1
2.9
2.4
1.3
–
19.9Æ0.7
18.2Æ0.6
8.7Æ0.1
17.9Æ0.2
6.2Æ0.3
3.6Æ0.1
1.0Æ0.3
0.7Æ0.2
2.4Æ0.1
1.3Æ0.4
To complete our analysis of the importance of the C21
methyl functionality for Hh agonism, we prepared two OHC
analogues that maintain the C20 hydroxy moiety, but do not
incorporate the C21 methyl group. Preparation of these ana-
logues began with THP protection of commercially available
21-acetoxypregnenolone followed by lithium aluminum hy-
dride mediated reduction of the ester and ketone functionali-
ties to afford 20 (Scheme 3). Oxidative cleavage of the 1,2-diol
1.3Æ0.3
0.7Æ0.01
24.3Æ0.8
1.3Æ0.01
10.1
–
[a] All OHCs tested at 5 mm. [b] Values are the mean fold mRNA up-regula-
tion ÆSEM. [c] Hh selectivity determined as the ratio of GLI1/ABCA1. Data
are the average of three separate experiments performed in triplicate.
fold increase in mRNA expression over DMSO levels. The ability
of both 4 and 13 to up-regulate GLI1 mRNA closely mirrored
that reported from OHCs prepared through our previous syn-
thetic route.[11] Interestingly, inversion of the C21 methyl group
on the 23-hydroxylated OHCs (16 and 17) completely abolish-
ed their ability to up-regulate GLI1 expression. This was partic-
ularly interesting for analogue 16, which loses all Hh agonism
compared to 23(S)-OHC, when the C21 methyl is inverted.
These results suggest that inversion of the C21 methyl group
actively prevents the OHC scaffold from binding the Smo CRD.
This result is supported by a previous study demonstrating
that inversion of the C20 hydroxy group to 20(R)-OHC resulted
in complete loss of Hh agonism relative to natural 20(S)-OHC.[9]
Interestingly, removal of the C21 methyl moiety of 20(S)-
OHC, as in OHC 25, completely abolished the ability of the
scaffold to up-regulate GLI1 expression. Removal of the C21
methyl group for 20(R)-OHC resulted in OHC 24, which shows
potent agonism of Hh signaling, an unexpected result consid-
ering that 22(R)-OHC is a poor activator of Hh signaling.[11]
Taken together, these results strongly suggest stereoselective
interactions between the side chain of the OHC scaffold and
its binding site in the Smo CRD are essential for its ability to
activate Hh signaling.
Scheme 3. Synthesis of 21-desmethyl OHC analogues. Reagents and condi-
tions: a) 3,4-dihydro-2H-pyran, p-TsOH, 12 h, RT, 96%; b) LAH, THF, 08C!RT,
3 h, 95%; c) NaIO4, RT, 1 h, 93%; d) Mg, isopentyl bromide, THF, 508C, 55%;
e) 2n HCl, MeOH/THF, RT, 96%.
with sodium periodate provided aldehyde 21, which under-
went Grignard addition with in situ generated isopentyl mag-
nesium bromide to provide a mixture of C20 secondary alco-
hols, 22 and 23. Similar to our previous OHCs, use of the THP
protecting group proved advantageous for the separation of
the R and S isomers, 24 and 25, respectively. Assignment of
the stereochemical orientation of the C20 hydroxy moiety for
these compounds was performed through Mosher ester analy-
Based on its comparable potency and improved selectivity,
we evaluated OHC 24 in a series of secondary assays to further
probe its ability to activate the Hh signaling cascade. To con-
firm that GLI1 up-regulation with 24 is mediated through the
Hh pathway, a competition study was performed with cyclopa-
mine (Cyc), a potent Hh-selective inhibitor known to attenuate
GLI1 expression induced by OHCs.[5,9,11] As expected, co-admin-
istration with Cyc (5 mm) abolished GLI1 up-regulation, indicat-
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
681
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 2. Attenuation of 24-mediated GLI1 up-regulation with Cyc co-admin-
istration in M2-10B4 cells. OHC and Cyc evaluated at 5 mm. Data are the
mean ÆSEM from a representative experiment performed in triplicate.
Table 3. Dose-dependent pathway activation in M2-10B4 cells.
EC50 [mm]
OHC
GLI1[a]
PTCH1[a]
ABCA1[a]
1
4
24
0.52Æ0.04
0.57Æ0.1
1.33Æ0.1
0.72Æ0.2
0.65Æ0.06
1.29Æ0.01
4.67Æ0.9
1.53Æ0.2
2.76Æ0.1
[a] Values are the average ÆSEM of three separate experiments per-
formed in triplicate.
ing OHC 24 functions through Hh activation (Figure 2). Dose-
dependent evaluation of OHC 24 demonstrated that its ability
to up-regulate the key Hh target genes GLI1 and PTCH1 was
approximately twofold less than both 1 and 4 (Table 3). In ad-
dition, the EC50 value for up-regulation of ABCA1 by 24 was
similar to that obtained for the Hh pathway target genes, sug-
gesting removal of the C21 methyl group also affects its ability
to selectively activate Hh signaling.
Figure 3. Induction of osteogenic differentiation and osteoblast maturation
by OHCs at 24, 48, or 96 h; up-regulation of A) GLI1, B) OSX, and C) ALP.
Data are the mean ÆSEM from a representative experiment performed in
triplicate.
When the Hh pathway is functionally activated in M2-10B4
cells, osteogenic differentiation is promoted, resulting in
mature osteoblastic cells.[3,8] This differentiation process can be
followed by measuring the early- (osterox, OSX) and late-stage
(alkaline phosphatase, ALP) transcriptional markers of osteo-
genesis.[3,8] Previously, OHCs 1 and 3–4 demonstrated the abili-
ty to promote differentiation of M2-10B4 cells, and we evaluat-
ed 24 for its activity in this assay. Cells were treated with
either 1 or 24 (5 mm) for 24, 48, or 96 h to compare their ability
to both up-regulate GLI1 and induce cellular differentiation
over this time course (Figure 3). Both compounds demonstrat-
ed similar induction of GLI1 expression after 24 and 48 h; how-
ever, OHC 1 continued to increase GLI1 expression up to 96 h
post-treatment, whereas no additional increase in GLI1 expres-
sion was observed after 48 h for 24. After 24 h, both com-
pounds induced a modest increase in both OSX and ALP ex-
pression. Interestingly, while OHC 1 significantly increased OSX
expression up to 96 h, OSX expression was not up-regulated
by compound 24. In addition, increased expression of ALP, de-
noting mature osteoblasts, was demonstrated at both 48 and
96 h for OHC 1, but not until 96 h for OHC 24. Taken together,
these data support the ability of both OHCs to terminally dif-
ferentiate precursor cells via Hh activation, while also suggest-
ing that OHC 24 mediates its osteogenic action through
a mechanism distinct from OSX.
Conclusions
In summary, we have developed a new synthetic approach to
obtain 23(S)-OHC (4) from inexpensive commercially available
starting materials. This new route uses half the number of
steps previously required (six down from twelve), with only
two required chromatographic purifications (following the
Grignard coupling and the final deprotection), and a tenfold
increase in overall yield. The new approach also allowed the
synthesis and evaluation of two new diastereomers of 4, ana-
logues that did not activate Hh signaling. Based on the known
importance of the stereochemistry of the hydroxy moiety at
C20, we wanted to further explore SAR at this region of the
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
682
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
additional separation. H NMR (500 MHz, CDCl3): d=6.01 (dd, 1H),
5.37 (s, 1H), 5.17 (d, 1H), 4.99 (d, 1H), 4.75 (s, 1H), 3.94 (s, 1H),
3.53 (m, 2H) 2.30 (m, 2H), 2.13 (m, 1H), 1.99 (m, 1H), 1.89 (m, 3H),
1.76–1.64 (m, 5H), 1.58–1.39 (m, 14H) 1.36 (s, 3H), 1.28 (m, 1H),
1.13 (m, 2H), 1.04 (s, 3H), 0.96 (m, 1H), 0.86 ppm (s, 3H); 13C NMR
(126 MHz, CDCl3): d=146.14, 140.98, 121.43, 121.27, 110.22, 96.99,
96.85, 75.92, 75.73, 63.76, 62.90, 62.81, 59.48, 56.95, 56.83, 50.13,
44.03, 42.85, 40.21, 38.89, 37.47, 37.23, 36.84, 31.89, 31.55, 31.37,
29.71, 28.78, 28.01, 25.53, 24.51, 23.84, 23.22, 22.86, 20.90, 20.09,
19.40, 13.82, 13.24 ppm; IR (ATR FTIR): n˜ =3516, 2933, 1467, 1453,
scaffold by synthesizing and evaluating analogues of 20(S)-
OHC (1) and 20(R)-OHC (2) in which the C21 methyl group has
been removed. Interestingly, only 24 demonstrated the ability
to up-regulate Hh target genes at a level similar to 1 at 5 mm;
however, the EC50 value, selectivity, and capacity to functionally
differentiate mouse embryonic fibroblasts of 24 was signifi-
cantly decreased relative to both 1 and 4. Finally, our results
support the previous findings from our research group and
others that the proper stereochemical orientation of functional
groups on the OHC side chain is essential for Smo binding and
Hh agonism. Continued work detailing OHC binding to the
Smo CRD will provide further insight into the exact molecular
interactions that govern OHC-mediated Hh activation and pro-
vide for a structure-based approach to designing more potent
compounds.
1336, 1260, 1200, 1140, 1111, 1055, 1021, 976, 911, 866 cmÀ1
HRMS: [M+NH4]+, calcd: 445.3634, obsd: 446.3667.
;
(Z)-3-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)-2-methylbut-2-enal (8a) and
(E)-3-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)-2-methylbut-2-enal (8b). To
a suspension of pyridinium chlorochromate (5.2 g, 24.1 mmol) and
NaOAc (4.9 g, 59.7 mmol) in CH2Cl2 (45 mL) under argon, was
added the mixture of 7a/7b (5.2 g, 12.13 mmol). The mixture was
stirred at RT for 12 h. The mixture was filtered through Celite, fol-
lowed by copious rinsing with CH2Cl2 and concentrated. Purifica-
tion by flash chromatography (SiO2, 10% EtOAc in hexanes) provid-
ed an 80:20 E/Z mixture of 8a/8b as a white solid (3.3 g, 64%).
1H NMR (500 MHz, CDCl3): d=10.10 (d, 1H), 10.00 (d, 0.2H), 6.06 (d,
0.2H), 5.95 (d, 1H), 5.39 (m, 1H), 4.92 (m, 0.2H), 4.75 (m, 1H), 4.05
(m, 0.2H), 3.95 (m, 1H) 3.53 (m, 2H), 3.25(m, 0.2H), 2.4–2.26 (m,
2H), 2.23 (s, 3H), 2.01 (m, 2H), 1.87 (m, 6H), 1.75 (m, 4H), 1.61–
1.40 (m, 15H), 1.39–1.28 (m, 3H), 1.24–1.08 (m, 3H), 1.04 (s, 3H),
0.72 (s, 0.6H) 0.64 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=
191.3, 190.7, 164.3, 140.9, 131.5, 127.9, 121.2, 96.9, 75.9, 62.8, 60.2,
56.8, 56.2, 51.4, 50.2, 46.3, 45.1, 40.2, 38.7, 38.6, 37.4, 37.2, 36.8,
32.2, 31.8, 31.3, 29.6, 27.9, 25.5, 24.9, 24.4, 21.0, 20.0, 19.4, 19.3,
13.9, 13.1 ppm; IR (ATR FTIR): n˜ =2930, 1714, 1656, 1434, 1376,
Experimental Section
General information. Pregnenolone (5) and 21-acetoxypregneno-
lone (18) were purchased from Sigma–Aldrich. All other reagents
were purchased from commercial sources (Fisher Scientific or
Sigma–Aldrich) unless otherwise stated. Column chromatography
was performed with silica gel purchased from Sorbtech (Sorbent
1
Technologies). All H and 13C NMR data were collected on a Bruker
AVANCE 500 spectrometer and analyzed with MestReNova software
(ver. 9.1.0). HRMS data were gathered at the Mass Spectrometry Fa-
cility at the University of Connecticut, performed by Dr. You-Jun
Fu. FTIR analysis was performed on a Bruker Alpha Platinum ATR
instrument using OPUS software (ver. 7.2).
1-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-2-
yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cy-
clopenta[a]phenanthren-17-yl)ethanone, THP-pregnenolone (6).
To a solution of pregnenolone (5, 10.0 g, 31.6 mmol) in anhydrous
CH2Cl2 (150 mL) under argon, 3,4-dihydro-2H-pyran (39.5 mmol,
3.6 mL) and a catalytic amount of p-toluenesulfonic acid were
added. The mixture was stirred for 12 h at RT, washed with saturat-
ed NaHCO3 (2”75 mL), saturated NaCl (2”75 mL), dried over
Na2SO4, and concentrated. The mixture was purified by column
chromatography (SiO2, 10% EtOAc in hexanes) to give 6 as a white
solid (12.2 g, 96%). Characterization matched that previously de-
scribed.[16] 1H NMR (500 MHz, CDCl3): d=5.37 (m, 1H), 4.75 (m, 1H),
3.93 (m, 1H), 352 (m, 2H), 2.56 (m, 1H), 2.39 (m, 1H), 2.21 (m, 1H),
2.15 (s, 3H), 2.08–2.00 (2H), 1.88 (m, 3H), 1.77–1.64 (5H), 1.59–1.47
(11H), 1.29–1.09 (3H), 1.04 (s, 3H), 0.66 ppm (s, 3H).
1354, 1254, 1230, 1197, 1130, 1112, 1060, 1022, 972, 910, 864 cmÀ1
HRMS: [M+H]+, calcd: 427.3112, obsd: 427.3211.
;
(3R)-3-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)butanal (9). A solution of 8a/8b
(0.300 g, 0.706 mmol) in CHCl3 (10 mL) was cooled to 08C. The tri-
chloroacetic acid salt of (S)-(+)-2-(tert-butyl)-3-methylimidazolin-4-
one (0.038 g, 0.14 mmol) and Hantzsch ester (0.215 g, 0.85 mmol)
were added, and the mixture was stirred at 08C until the starting
material was consumed as evidenced by continuous TLC analysis
(~16 h). The solution was concentrated and directly purified by
flash chromatography (SiO2, 10% EtOAc in hexanes) to yield 9 as
a white solid (0.195 g, 64%). Characterization matched that previ-
ously reported by our research group.[11]
(2R)-2-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)but-3-en-2-ol (7a) and (2S)-2-
((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-2-
yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cy-
clopenta[a]phenanthren-17-yl)but-3-en-2-ol (7b). A solution of 6
(5.00 g, 12.5 mmol) in THF (100 mL) was cooled to À788C under
argon. Vinyl magnesium bromide (0.7m in THF, 44.8 mL,
31.4 mmol) was added dropwise. Upon complete addition, the
mixture was stirred for 1 h at À788C and then warmed to RT over
12 h. The mixture was re-cooled to 08C and quenched with satu-
rated aqueous NH4Cl and washed with Et2O (3”60 mL). The organ-
ic layers were combined and washed with saturated NaCl (1”
100 mL), dried with Na2SO4 and concentrated. Purification by flash
chromatography (SiO2, 12% EtOAc in hexanes) afforded a 1:9 mix-
ture of 7a/7b as a white solid (5.2 g, 97%) that was used without
(3S)-3-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)butanal (10). A solution of 8a/
8b (0.300 g, 0.706 mmol) in CHCl3 (10 mL) was cooled to 08C. The
trichloroacetic acid salt of (R)-2-(tert-butyl)-3-methylimidazolin-4-
one (0.038 g, 0.14 mmol) and Hantzsch ester (0.215 g, 0.85 mmol)
were added and the mixture was stirred at 08C until the starting
material was consumed as evidenced by continuous TLC analysis
(~16 h). The solution was concentrated and directly purified by
flash chromatography (SiO2, 10% EtOAc in hexanes) to yield 10 as
1
a white solid (0.195 g, 64%). H NMR (500 MHz, CDCl3): d=9.78 (s,
1H), 5.37 (m, 1H), 4.76 (m, 1H), 3.85 (m, 1H), 3.55 (m, 2H), 2.72 (m,
1H), 2.41–2.26 (m, 3H), 2.04 (m, 2H), 1.91 (m, 5H), 1.75 (m, 1H),
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
683
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1.64 (m, 2H), 1.57 (m, 6H), 1.51 (m, 2H), 1.37 (m, 1H), 1.27- 1.12
(m, 5H), 1.05 (s, 3H), 0.97 (d, 3H), 0.76 ppm (s, 3H); 13C NMR
(126 MHz, CDCl3): d=203.3, 140.9, 121.4, 96.9, 75.9, 62.9, 56.6, 55.6,
49.9, 42.4, 40.2, 39.7, 38.8, 37.4, 37.2, 36.8, 31.8, 31.3, 30.6, 29.7,
27.8, 25.5, 24.0, 21.0, 20.1, 19.9, 19.4, 12.3 ppm; IR (ATR FTIR): n˜ =
2930, 1715, 1439, 1372, 1258, 1199, 1135, 1112, 1056, 1021, 974,
912, 867 cmÀ1; HRMS: [M+H]+, calcd: 429.3368, obsd: 429.3360.
2.41 (m, 1H), 2.24 (m, 1H), 2.01 (m, 2H), 1.93–1.73 (7H), 1.66–1.45
(11H), 1.30 (m, 5H), 1.19–1.07 (5H), 1.05 (s, 3H), 0.97(dd, 6H), 0.90
(d, 3H), 0.73 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=141.1,
121.5, 97.0, 76.0, 68.6, 62.9, 56.5, 50.2, 46.1, 44.8, 42.4, 40.3, 39.9,
38.8, 37.5, 37.2, 36.8, 32.6, 31.9, 31.3, 29.7, 28.0, 27.5, 25.5, 24.5,
24.9, 21.7, 21.0, 20.1, 19.4, 18.9, 12.3 ppm; IR (ATR FTIR): n˜ =3519,
2931, 1465, 1375, 1260, 1113, 1075, 1055, 1019, 909, 868 cmÀ1
;
HRMS: [M+H]+, calcd: 487.4151, obsd: 487.4145; [MÀOH]+, calcd:
(2S,4S)-2-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahydro-2H-
pyran-2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahy-
dro-1H-cyclopenta[a]phenanthren-17-yl)-6-methylheptan-4-ol
(11) and (2S,4R)-2-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahy-
dro-2H-pyran-2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetra-
decahydro-1H-cyclopenta[a]phenanthren-17-yl)-6-methylheptan-
4-ol (12). A solution of isobutyl magnesium chloride (2.0m in THF,
0.84 mL, 1.68 mmol) was cooled to 08C in THF (5 mL) under argon.
A solution of 9 (0.090 g, 0.210 mmol) in THF (5 mL) was added
dropwise, and the mixture was stirred at 08C for 90 min. The reac-
tion was quenched with saturated aqueous NH4Cl and washed
with Et2O (2”25 mL). The organic layers were dried (Na2SO4) and
concentrated. Purification by flash chromatography (SiO2, 8%
EtOAc in hexanes) provided two separable isomers as white solids
(80% overall total yield of both isomers). Characterization of frac-
tion 1 [11, THP-23(S)-OHC, Rf =0.39 in 8:1 hexanes/EtOAc] and frac-
tion 2 [12, THP-23(R)-OHC, Rf =0.29 in 8:1 hexanes/EtOAc] were as
previously described.[11]
469.4046, obsd: 469.4045.
(2S,4R)-2-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahydro-2H-
pyran-2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahy-
dro-1H-cyclopenta[a]phenanthren-17-yl)-6-methylheptan-4-ol
1
(15). Rf =0.30 in 8:1 hexanes/EtOAc. H NMR (500 MHz, CDCl3): d=
5.38 (m, 1H), 4.75 (m, 1H), 3.96 (m, 1H), 3.83 (m, 1H), 3.52 (m, 2H),
2.39–2.32 (2H), 2.01 (m, 2H), 1.88 (m, 4H), 1.76 (m, 4H), 1.63–1.35
(11H), 1.29–1.07 (9H), 1.04 (s, 3H), 0.96 (d, 6H), 0.92 (d, 3H),
0.74 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=140.9, 121.5, 97.0,
76.01, 67.4, 62.8, 56.8, 50.2, 48.1, 44.0, 42.4, 40.1, 38.8, 37.5, 37.2,
36.8, 31.9, 31.3, 29.7, 28.1, 28.0, 25.5, 24.7, 24.2, 23.4, 22.3, 21.1,
20.1, 19.4, 18.5, 12.2 ppm; IR (ATR FTIR): n˜ =3491, 2931, 1465, 1374,
1261, 1140, 1110, 1056, 1018, 978, 864 cmÀ1; HRMS: [M+H]+,
calcd: 487.4151, obsd: 487.4082; [MÀOH]+, calcd: 469.4046, obsd:
469.4035.
(3S,10R,13R,17R)-17-((2S,4S)-4-hydroxy-6-methylheptan-2-yl)-
10,13-dimethyl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahy-
dro-1H-cyclopenta[a]phenanthren-3-ol (16). The THP group was
removed from OHC 14 as described above for the preparation of
4. Purification by flash chromatography (SiO2, 20% EtOAc afforded)
16 as a white solid in excellent yield (0.035 g, 95%). 1H NMR
(500 MHz, CDCl3: d=5.39 (m, 1H), 3.82 (m, 1H), 3.57 (m, 1H), 2.30
(m, 2H), 2.03 (m, 2H), 1.88 (m, 3H), 1.79 (m, 1H), 1.62 (m, 3H),
1.58–1.48 (6H), 1.34 (m, 3H), 1.30 (s, 2H), 1.24–1.09 (4H), 1.05 (s,
3H), 0.97 (dd, 6H), 0.91 (d, 3H), 0.73 ppm (s, 3H); 13C NMR
(126 MHz, CDCl3): d=140.80, 121.65, 71.81, 68.61, 56.76, 56.54,
50.17, 46.09, 44.83, 42.41, 42.32, 39.89, 37.27, 36.54, 32.57, 31.91,
31.69, 29.72, 27.46, 24.50, 24.13, 23.97, 21.69, 21.06, 19.41, 18.92,
12.30 ppm; IR (ATR FTIR): n˜ =3247, 2928, 1463, 1435, 1373, 1140,
1112, 1056, 1019 cmÀ1; HRMS: [M+NH4]+, calcd: 420.3842, obsd:
420.3801; [MÀOH]+, calcd: 385.3470, obsd: 385.3434.
23(S)-hydroxycholesterol (4). OHC 11 (0.050 g, 0.103 mmol) was
dissolved in 5:1 MeOH/THF (6 mL), to which 2n HCl (2 mL) was
added dropwise. The mixture was stirred at RT for 3 h. The reaction
was quenched with H2O (5 mL) and washed with EtOAc (2”
20 mL). The organic layers were dried over Na2SO4, concentrated,
and purified as previously described to afford 4 as a white solid
(0.040 g, 97%). Characterization matched that previously pub-
lished.[11] 1H NMR (500 MHz, CDCl3): d=5.37 (m, 1H), 3.79 (m, 1H),
3.54 (m, 1H), 2.27 (m, 2H), 2.02 (m, 2H), 1.85 (m, 4H), 1.61–1.43
(11H), 1.28 (m, 4H), 1.20–1.07 (4H), 1.03 (s, 3H), 0.98 (d, 3H), 0.94
(dd, 6H), 0.71 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=140.8,
121.7, 71.8, 68.7, 56.9, 56.7, 50.2, 46.5, 45.2, 42.4, 42.3, 39.8, 37.3,
36.5, 33.9, 31.9, 31.6, 28.5, 24.5, 24.3, 23.9, 21.7, 21.1, 19.43, 19.42,
11.9 ppm.
23(R)-hydroxycholesterol (13). The THP group was removed from
OHC 12 as described above for the preparation of 4. Purification
by flash chromatography (SiO2, 20% EtOAc afforded) 13 as a white
solid in excellent yield (0.025 g, 94%). Characterization matched
that previously published.[11] 1H NMR (500 MHz, CDCl3): d=5.37 (m,
1H), 3.80 (m, 1H), 3.54 (m, 1H), 2.28 (m, 2H), 2.04 (m, 1H), 2.00 (m,
1H), 1.86 (m, 3H), 1.77 (m, 1H), 1.67 (m, 1H), 1.61–1.39 (11H), 1.28
(m, 2H), 1.20 (m, 2H), 1.12–1.05 (4H), 1.03 (s, 3H), 0.99 (d, 3H),
0.93 (dd, 6H), 0.74 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=
140.77, 121.67, 71.8, 67.0, 56.84, 56.78, 50.1, 48.0, 44.5, 42.4, 42.3,
39.8, 37.3, 36.5, 32.5, 31.9, 31.7, 28.5, 24.7, 24.3, 23.3, 22.3, 21.1,
19.4, 18.7, 11.9 ppm.
(3S,10R,13R,17R)-17-((2S,4R)-4-hydroxy-6-methylheptan-2-yl)-
10,13-dimethyl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahy-
dro-1H-cyclopenta[a]phenanthren-3-ol (17). The THP group was
removed from OHC 15 as described above for the preparation of
4. Purification by flash chromatography (SiO2, 20% EtOAc in hex-
anes) afforded 17 as a white solid in excellent yield (0.034 g, 96%).
1H NMR (500 MHz, CDCl3): d=5.40 (m, 1H), 3.84 (m, 1H), 3.56 (m,
1H), 2.30 (m, 2H), 2.05 (m, 2H), 1.88 (m, 3H), 1.76 (m, 3H), 1.64–
1.32 (m, 11 H), 1.24–1.12 (m, 7H), 1.05 (s, 3H), 0.96 (d, 6H), 0.93 (d,
3H), 0.76 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=140.8, 121.7,
71.8, 67.4, 56.9, 56.8, 50.1, 48.0, 44.0, 42.4, 42.3, 40.1, 37.3, 36.5,
31.9, 31.8, 31.7, 28.1, 24.7, 24.2, 23.4, 22.3, 21.1, 19.4, 18.5,
12.2 ppm; IR (ATR FTIR): n˜ =3332, 2936, 1460, 1433, 1382, 1305,
Preparation of 14 and 15. Grignard addition of isobutyl magnesi-
um chloride to 10 was analogous to that described above for addi-
tion to 9. Purification by flash chromatography (SiO2, 8% EtOAc in
hexanes) provided two separable isomers as white solids (72%
overall total yield of both isomers).
1055, 952 cmÀ1
;
HRMS: [M+NH4]+, calcd: 420.3842, obsd:
420.3839; [MÀOH]+, calcd: 385.3470, obsd: 385.3434.
3-THP-21-acetoxypregnenolone (19). THP protection of 21-acet-
oxypregnenolone (18) was carried out as described above for the
preparation of 6. Purification by flash chromatography (10% EtOAc
in hexanes) afforded 19 as a white solid in excellent yield (1.16 g,
95%). 1H NMR (500 MHz, CDCl3): d=5.37 (m, 1H), 4.73 (m, 2H),
4.56 (d, 2H), 3.94 (m, 1H), 3.54 (m, 2H), 2.53 (m, 1H), 2.38 (m, 1H),
2.24 (m, 1H), 2.19 (s, 3H), 2.06 (m, 2H), 1.88 (m, 3H), 1.73 (m, 3H),
(2S,4S)-2-((3S,10R,13R,17R)-10,13-dimethyl-3-((tetrahydro-2H-
pyran-2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahy-
dro-1H-cyclopenta[a]phenanthren-17-yl)-6-methylheptan-4-ol
1
(14). Rf =0.41 in 8:1 hexanes/EtOAc. H NMR (500 MHz, CDCl3): d=
5.39 (m, 1H), 4.76 (m, 1H), 3.96 (m, 1H), 3.81 (m, 1H), 3.55 (m, 2H),
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
684
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1.64 (m, 1H), 1.58 (m, 7H), 1.49m, 1H), 1.42 (m, 1H), 1.31 (m, 1H),
1.21–1.07 (2H), 1.03 (s, 3H), 0.99 (m, 1H), 0.69 ppm (s, 3H);
13C NMR (126 MHz, CDCl3): d=203.7, 170.2, 141.1, 121.1, 96.9, 75.9,
69.2, 62.8, 59.3, 57.1, 49.9, 44.7, 40.2, 38.6, 37.4, 37.2, 36.8, 31.9,
31.3, 29.6, 27.9, 25.5, 24.6, 22.8, 21.0, 20.5, 20.0, 19.4, 13.1 ppm; IR
(ATR FTIR): n˜ =2936, 1745, 1722, 1417, 1370, 1229, 1198, 1112,
1074, 1056, 1029, 975, 904, 837 cmÀ1; HRMS: [M+H]+, calcd:
459.3111, obsd: 459.3186.
1H), 4.76 (s, 1H), 3.97 (m, 1H), 3.58 (m, 3H), 2.41 (m, 2H), 2.13 (m,
1H), 2.02 (m, 1H), 1.91 (m, 3H), 1.76 (m, 1H), 1.68 (m, 2H), 1.57 (m,
13H), 1.40–1.20 (m, 12H), 1.06 (s, 3H), 1.00 (m, 1H), 0.92 (d, J=
7.5 Hz, 6H), 0.82 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=141.2,
121.3, 96.8, 76.0, 74.4, 62.8, 56.7, 56.2, 50.2, 42.4, 40.3, 40.0, 39.1,
37.3, 37.2, 36.8, 31.9, 31.8, 31.3, 28.03, 25.54, 25.50, 24.6, 22.9, 22.7,
22.5, 21.0, 20.1, 19.4, 12.4 ppm; IR (ATR FTIR): n˜ =3466, 2933, 1464,
1376, 1364, 1259, 1134, 1112, 1093, 1077, 1057, 1022, 973, 910,
867 cmÀ1
;
HRMS: [M+H]+, calcd: 473.3995, obsd: 473.4140.
[MÀOH]+, calcd: 455.3889, obsd: 455.3848.
1-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-2-
yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cy-
clopenta[a]phenanthren-17-yl)ethane-1,2-diol (20). A solution of
19 (1.0 g, 2.18 mmol) in anhydrous THF (20 mL) under argon was
cooled to 08C. LiAlH4 (1.0m in THF, 8.8 mmol) was added dropwise,
and the mixture was stirred for 4 h at 08C. The reaction was
quenched using the Fieser method[18] and concentrated. Purifica-
tion by flash chromatography (SiO2, 30% EtOAc in hexanes) afford-
ed 20 as a white solid in excellent yield (1.02 g, 96%). 1H NMR
(500 MHz, CDCl3): d=5.36 (m, 1H), 4.74 (s, 1H), 3.94 (m, 1H), 3.67
(d, 2H), 3.51 (m, 2H), 3.38 (m, 1H), 2.38 (m, 1H), 2.22–2.11 (m, 3H),
2.00 (m, 1H), 1.87 (m, 3H), 1.74 (m, 2H), 1.66 (m, 2H), 1.55–1.47 (m,
10H), 1.27–1.11 (4H), 1.04 (s, 3H), 0.81 ppm (s, 3H); 13C NMR
(126 MHz, CDCl3): d=141.2, 121.3, 96.9, 75.9, 74.6, 66.5, 62.8, 55.9,
52.4, 50.2, 42.5, 40.2, 39.7, 38.8, 37.2, 36.8, 31.9, 31.8, 31.3, 29.7,
27.9, 25.5, 24.6, 20.9, 20.1, 19.4, 12.4 ppm; IR (ATR FTIR): n˜ =3338,
2931, 1438, 1351, 1305, 1199, 1135, 1114, 1060, 1027, 972, 909,
868 cmÀ1; HRMS: [M+H]+, calcd: 419.3161, obsd: 419.3139.
(1S)-1-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)-5-methylhexan-1-ol (23). Rf =
1
0.23 in 10:1 hexanes/EtOAc. H NMR (500 MHz, CDCl3): d=5.39 (s,
1H), 4.76 (s, 1H), 3.96 (m, 1H), 3.58 (m, 3H), 2.38–2.24 (2H), 2.13
(m, 1H), 2.02 (m, 1H), 1.91 (m, 5H), 1.75–1.19 (30H), 1.05 (s, 3H),
0.92 (d, 6H), 0.74 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=140.9,
121.5, 97.0, 76.0, 73.2, 62.9, 56.6, 56.3, 50.3, 41.7, 40.3, 39.1, 38.8,
37.5, 37.2, 36.9, 31.9, 31.6, 31.3, 29.7, 28.1, 25.5, 24.8, 24.2, 23.2,
22.7, 22.6, 20.8, 20.1, 19.4, 12.7 ppm; IR (ATR FTIR): n˜ =3500, 2923,
1460, 1384, 1353, 1260, 1199, 1109, 1053, 1020, 973, 911, 864 cmÀ1
;
HRMS: [M+H]+, calcd: 473.3995, obsd: 473.4028; [MÀOH]+, calcd:
455.3889, obsd: 455.3873.
(3S,10R,13S,17S)-17-((R)-1-hydroxy-5-methylhexyl)-10,13-dimeth-
yl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclo-
penta[a]phenanthren-3-ol (24). The THP group was removed from
OHC 22 as described above for the preparation of 4. Purification
by flash chromatography (SiO2, 20% EtOAc in hexanes) afforded 24
(3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-2-
yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cy-
clopenta[a]phenanthrene-17-carbaldehyde (21). To a solution of
20 (0.75 g, 1.79 mmol) in 2:1 THF/H2O (45 mL) was added NaIO4
(1.2 g, 5.61 mmol), and the mixture was stirred at RT for 1 h. The
mixture was extracted with EtOAc (3”50 mL), and the organic
layer was dried over Na2SO4 and concentrated. Purification by flash
chromatography (10% EtOAc in hexanes) afforded 21 as a white
1
as a white solid in excellent yield (0.063 g, 96%). H NMR (500 MHz,
CDCl3): d=5.40 (s, 1H), 3.60 (m, 2H), 2.30 (m, 2H), 2.15 (m, 1H),
2.02 (m, 1H), 1.89 (m, 2H), 1.69 (m, 2H), 1.56 (m, 8H), 1.43 (m, 2H),
1.30–1.14 (12H), 1.07 (s, 3H), 0.92 (d, 6H), 0.82 ppm (s, 3H);
13C NMR (126 MHz, CDCl3): d=140.9, 121.6, 74.4, 71.8, 56.7, 56.2,
50.2, 42.4, 42.3, 40.0, 39.1, 37.3, 37.2, 36.6, 31.9, 31.8, 31.7, 28.0,
25.5, 24.6, 22.9, 22.7, 22.5, 21.0, 19.4, 12.4 ppm; IR (ATR FTIR): n˜ =
3298, 2925, 1729, 1463, 1373, 1351, 1260, 1056, 1023, 954,
912 cmÀ1; HRMS: [M+NH4]+, calcd: 406.3685, obsd: 406.3661;
[MÀOH]+, calcd: 371.3314, obsd: 371.3302.
1
solid (0.64 g, 92%). H NMR (500 MHz, CDCl3): d=5.37 (s, 1H), 4.74
(s, 1H), 3.94 (m, 1H), 3.54 (m, 2H), 3.38 (m, 1H), 2.35 (m, 2H), 2.04
(m, 2H), 1.88 (m, 3H), 1.76 (m, 3H), 1.29 (m, 1H), 1.15 (m, 2H), 1.04
(s, 3H), 0.79 ppm (s, 3H); 13C NMR (126 MHz, CDCl3): d=204.9,
141.1, 121.2, 96.9, 75.9, 62.84, 56.5, 50.2, 44.7, 40.2, 38.3, 37.2, 36.8,
31.8, 31.5, 29.7, 27.9, 25.5, 24.9, 21.1, 20.6, 20.1, 19.4, 13.7 ppm; IR
(ATR FTIR): n˜ =2938, 2695, 1719, 1438, 1352, 1260, 1198, 1112,
1054, 1021, 973, 911, 866 cmÀ1; HRMS: [MÀH]+, calcd: 385.2743,
obsd: 385.2742; [M+H]+, calcd: 387.2899, obsd: 387.2867.
(3S,10R,13S,17S)-17-((S)-1-hydroxy-5-methylhexyl)-10,13-dimeth-
yl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclo-
penta[a]phenanthren-3-ol (25). The THP group was removed from
OHC 23 as described above for the preparation of 4. Purification
by flash chromatography (SiO2, 20% EtOAc in hexanes) afforded 25
1
as a white solid in excellent yield (0.031 g, 95%). H NMR (500 MHz,
Grignard addition to 21. To a suspension of acid-activated Mg
turnings (0.360 g, 14.8 mmol) in anhydrous THF (5 mL) under
argon at RT was added a catalytic amount of Br2Et (3 drops) and
heated at 358C. The suspension was stirred until bubbling of the
Mg turnings was observed. A solution of bromo-4-methylpentane
(1.6 mL, 10.9 mmol) in anhydrous THF (6 mL) was added dropwise
over 20 min to the Mg suspension and heated at 558C. After stir-
ring for 2 h, the mixture was cooled to 08C. A solution of 21
(0.700 g, 1.81 mmol) in anhydrous THF (6 mL) was added and
stirred at 08C for 2 h. The reaction was quenched with saturated
NH4Cl (15 mL) and washed with Et2O (2”25 mL). The combined or-
ganic layers were dried over Na2SO4 and concentrated. Purification
by flash chromatography (10% EtOAc in hexanes) afforded 22 and
23 as easily separable white solids (55% overall conversion).
CDCl3): d=5.40 (s, 1H), 3.60 (m, 2H), 2.31 (m, 2H), 2.04 (m, 1H),
1.90 (m, 4H), 1.68 (m, 2H), 1.56 (m, 9H), 1.42 (m, 4H), 1.30–1.12
(8H), 1.06 (s, 3H), 0.92 (d, 6H), 0.75 ppm (s, 3H); 13C NMR
(126 MHz, CDCl3): d=140.8, 121.6, 73.2, 71.8, 56.6, 56.2, 50.2, 42.3,
41.7, 39.1, 38.8, 37.3, 37.2, 36.6, 31.9, 31.7, 31.6, 28.1, 24.8, 24.2,
23.2, 22.7, 22.6, 20.8, 19.4, 12.8 ppm; IR (ATR FTIR): n˜ =3290, 2926,
1715, 14622, 1378, 1365, 1320, 1238, 1058, 1028, 953, 936 cmÀ1
;
HRMS: [M+NH4]+, calcd: 406.3685, obsd: 406.3658; [MÀOH]+,
calcd: 371.3314, obsd: 371.3300.
Biological assays
Cell culture and reagents: The murine cell line, M2-10B4, was pur-
chased from American Type Culture Collection (ATCC, Manassas,
VA, USA). M2-10B4 cells were cultured in RPMI 1640 medium
(Gibco) supplemented with 10% fetal bovine serum (FBS) and 1%
penicillin/streptomycin. Cells were maintained using the media de-
scribed above (denoted “growth” media). Media denoted as “low
(1R)-1-((3S,10R,13S,17S)-10,13-dimethyl-3-((tetrahydro-2H-pyran-
2-yl)oxy)-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-
cyclopenta[a]phenanthren-17-yl)-5-methylhexan-1-ol (22). Rf =
1
0.28 in 10:1 hexanes/EtOAc. H NMR (500 MHz, CDCl3): d=5.40 (s,
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
685
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
FBS” contained 0.5% FBS and the same percentage of other sup-
plements as specified for growth media. Cells were grown in Corn-
ing Cell Culture, canted neck T75 or T150 flasks (Fisher Scientific)
in an Autoflow IR water-jacket CO2 incubator. Experiments were
performed in BD Falcon sterile 24-well tissue culture treated plates.
Dimethyl sulfoxide (DMSO) was used as solvent to prepare all drug
solutions, and the final DMSO concentration did not exceed 0.3%.
20(S)-Hydroxycholesterol was purchased from Sigma–Aldrich. Cy-
clopamine was purchased from LC Labs.
Keywords: Hedgehog · oxysterols · signal transduction ·
stereochemistry · structure–activity relationships
[1] A. P. McMahon, P. W. Ingham, C. J. Tabin, Curr. Top. Dev. Biol. 2003, 53,
1–114.
[2] T. Dellovade, J. T. Romer, T. Curran, L. L. Rubin, Annu. Rev. Neurosci.
2006, 29, 539–563.
[3] P. V. Giannoudis, H. Dinopoulos, E. Tsirisdis, Injury 2005, 36, S20–S27.
[4] R. B. Corcoran, M. P. Scott, Proc. Natl. Acad. Sci. USA 2006, 103, 8408–
8413.
[5] J. R. Dwyer, N. Sever, M. Carlson, S. F. Nelson, P. A. Beachy, F. Parhami, J.
Biol. Chem. 2007, 282, 8959–8968.
[6] H. T. Kha, B. Basseri, D. Shouhed, J. Richardson, S. Tetradis, T. J. Hahn, F.
Parhami, J. Bone Miner. Res. 2004, 19, 830–840.
[7] T. L. Aghaloo, C. M. Amantea, C. M. Cowanm, J. A. Richardson, B. M. Wu,
F. Parhami, S. Tetradis, J. Orthop. Res. 2007, 25, 1488–1497.
[8] V. A. Olkkonen, O. Beaslas, E. Nissila, Biomolecules 2012, 2, 76–103.
[9] S. Nachtergaele, L. K. Mydock, K. Krishnan, J. Rammohan, P. H. Schlesing-
er, F. D. Covey, R. Rohatgi, Nat. Chem. Biol. 2012, 8, 211 –220.
[10] D. Nedelcu, J. Liu, Y. Xu, C. Jao, A. Salic, Nat. Chem. Biol. 2013, 9, 557–
564.
[11] A. Corman, A. M. DeBerardinis, M. K. Hadden, ACS Med. Chem. Lett.
2012, 3, 828–833.
[12] M. K. Hadden, ChemMedChem 2014, 9, 27–37.
[13] W. K. Kim, V. Meliton, K. W. Park, C. H. Hong, P. Tontonoz, P. Niewiadom-
ski, J. A. Waschek, S. Tetradis, F. Parhami, Mol. Endocrinol. 2009, 23,
1532–1543.
[14] J. Drew, M. Letellier, P. Morand, A. G. Szabo, J. Org. Chem. 1987, 52,
4047–4052.
[15] S. G. Ouellet, J. B. Tuttle, D. W. C. MacMillan, J. Am. Chem. Soc. 2005,
127, 32–33.
General procedure for analysis of Hh target gene regulation: Cells
were grown to confluence in T75 flasks. Once cells reached conflu-
ence, they were plated into 24-well plates at concentration of
50000 cells in 500 mL growth media per well. After 24 h, the
growth media was removed and replaced with low FBS media
(500 mL per well). Cells were treated with corresponding com-
pound: DMSO, OHC, or OHC and cyclopamine. Cells were incubat-
ed (378C, 5% CO2) for the indicated time period, at which time
mRNA was isolated and evaluated as described previously.[11]
General procedure for RT-PCR analysis: Following treatment and in-
cubation, total RNA was extracted using Ambionꢁ by Life Technolo-
gies Taqmanꢁ Fast Cells-to-CTTM kit following the manufacturer’s in-
structions. Synthesis of cDNA was performed using a BioRad MyCy-
cler. Quantitative PCR was performed on an ABI 7500 system using
the following Taqman Gene Expression Primer/Probe solutions
(ABI):
mouse
ActB
(Mm00607939s1),
mouse
GLI1
(Mm00494645m1), mouse PTCH1 (Mm00436926m1), mouse ABCA1
(Mm00442646m1), mouse Alp1 (Mm00475834m1), mouse sp7
(Mm04209856m1). Relative gene expression levels were computed
by the DDCt method. Values represent mRNA expression relative
to DMSO control (vehicle, set at 1.00). Data were analyzed using
GraphPad Prism 5, and EC50 values are the mean ÆSEM for at least
two separate experiments performed in triplicate.
[16] R. Katoch, S. S. Korde, K. D. Deodhar, G. K. Trivedi, Tetrahedron 1999, 55,
1741–1754.
[17] E. Fernµndez-Mateous, B. Maciµ, M. Yus, Adv. Synth. Catal. 2013, 355,
1249–1254.
[18] L. F. Fieser, M. Fieser, Reagents for Organic Synthesis Vol. 1, Wiley, New
York, 1967, p. 564.
Acknowledgements
The authors thank Victor W. Day (University of Kansas, USA) for
crystallography work, as well as the US National Science Founda-
tion (NSF) Major Research Instrumentation (MRI) Program (CHE-
0923449) for financial support of the X-ray diffractometer and
software used in this study.
Received: November 23, 2015
Revised: February 12, 2016
Published online on February 25, 2016
ChemMedChem 2016, 11, 679 – 686
www.chemmedchem.org
686
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim