Synthesis and evaluation of cardiac glycoside mimics as potential anticancer drugs

Article abstract of DOI:10.1016/j.bmc.2011.02.016

The cardiac glycoside digitoxin, consisting of a steroid core linked to a labile trisaccharide, has been used for centuries for the treatment of congestive heart failure. The well known pharmacological effect is a result of the ability of cardiac glycosides to inhibit the Na⁺, K ⁺-ATPase. Within recent years cardiac glycosides have furthermore been suggested to possess valuable anticancer activity. To mimic the labile trisaccharide of digitoxin with a stabile carbohydrate surrogate, we have used sulfur linked ethylene glycol moieties of varying length (mono-, di-, tri- or tetra-ethylene glycol), and furthermore used these linkers as handles for the synthesis of bivalent steroids. The prepared compounds were evaluated for their potencies to inhibit the Na⁺, K⁺-ATPase and for their cytotoxic effect on cancerous MCF-7 cells. A clear trend is observed in both inhibition and cytotoxic effect, where the bioactivity decreases as the size increases. The most potent Na⁺, K⁺-ATPase inhibitors are the compounds with the shortest ethylene glycol chain (K_app 0.48 μM) and thiodigitoxigenin (K_app 0.42 μM), which both are comparable with digitoxigenin (K_app 0.52 μM). For the cancer cell viability assay the shortest mimics were found to have highest efficacy, with the best ligand having a monoethylene glycol unit (IC₅₀ 0.24 μM), which was slightly better than digitoxigenin (IC₅₀ 0.64 μM), while none of the novel cardiac glycoside mimics display an in vitro effect as high as digitoxin (IC₅₀ 0.02 μM).

Full text of DOI:10.1016/j.bmc.2011.02.016

Bioorganic & Medicinal Chemistry 19 (2011) 2407–2417
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of cardiac glycoside mimics as potential
anticancer drugs
Marie Jensen a, Steffen Schmidt b, Natalya U. Fedosova c, Jan Mollenhauer b, Henrik H. Jensen a,
⇑
a Department of Chemistry, Aarhus University, Langelandsgade 140, 8000 Aarhus C, Denmark
b Lundbeckfonden Center of Excellence NanoCAN and Molecular Oncology, Institute of Molecular Medicine, University of Southern Denmark, 5000 Odense C, Denmark
c Department of Physiology and Biophysics, Aarhus University, Ole Worms Allé 6, 8000 Aarhus C, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
The cardiac glycoside digitoxin, consisting of a steroid core linked to a labile trisaccharide, has been used
for centuries for the treatment of congestive heart failure. The well known pharmacological effect is a
result of the ability of cardiac glycosides to inhibit the Na+, K+-ATPase. Within recent years cardiac gly-
cosides have furthermore been suggested to possess valuable anticancer activity. To mimic the labile tri-
saccharide of digitoxin with a stabile carbohydrate surrogate, we have used sulfur linked ethylene glycol
moieties of varying length (mono-, di-, tri- or tetra-ethylene glycol), and furthermore used these linkers
as handles for the synthesis of bivalent steroids. The prepared compounds were evaluated for their poten-
cies to inhibit the Na+, K+-ATPase and for their cytotoxic effect on cancerous MCF-7 cells. A clear trend is
observed in both inhibition and cytotoxic effect, where the bioactivity decreases as the size increases. The
most potent Na+, K+-ATPase inhibitors are the compounds with the shortest ethylene glycol chain (Kapp
Received 20 October 2010
Revised 4 February 2011
Accepted 9 February 2011
Available online 16 February 2011
Keywords:
Digitoxin
Na+, K+-ATPase
Anti-proliferation
Ethylene glycol
0.48
0.52
l
l
M) and thiodigitoxigenin (Kapp 0.42
M). For the cancer cell viability assay the shortest mimics were found to have highest efficacy, with
M), which was slightly better than digitoxi-
M), while none of the novel cardiac glycoside mimics display an in vitro effect as high as
digitoxin (IC50 0.02 M).
lM), which both are comparable with digitoxigenin (Kapp
the best ligand having a monoethylene glycol unit (IC50 0.24
genin (IC50 0.64
l
l
l
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Cardiac glycosides seem to be beneficial for the treatment of
disorders like polyglutamine based diseases6 and cystic fibrosis,7
and their application in cancer therapy represents a new perspec-
tive. In 1979, Stenkvist et al.8 reported that 28 patients with breast
cancer and digitalis treatment had better survival rate and less
cancer recurrence than a control group of 114 patients. A 20 year
follow-up study showed that the death rate for digitalis treated pa-
tients was 6% compared to 34% in control group.9 Additionally, sev-
eral publications on epidemiological data indicate lower mortality
in cancer patients who received digitalis therapy compared to can-
cer patients not receiving digitalis. These key preliminary data
have paved the way for numerous published studies, which estab-
lish the anticancer effect of cardiac glycosides.10
Cardiac glycosides constitute a class of naturally occurring tox-
ins that are able to bind to and thereby inhibit the Na+, K+-ATPase.1
Among these are compounds like digitoxin and digoxin, which can
be extracted from the plants Digitalis purpurea and Digitalis lanata.
They are considered important medicinal agents and have been ap-
plied for the treatment of congestive heart failure for over
200 years,2 and were in 2005 used for 1.7 million patients in the
USA.3 The positive inotropic effect obtained by the treatment with,
for example, digitoxin has been explained by the ‘Na+ pump lag
hypotheses’, which proposes that the digitoxin effect on the heart
is associated with inhibition of the Na+, K+-ATPase.4 Besides being
an inhibitor of Na+, K+-ATPase ion pump, digitoxin is also believed
to serve as a signaling compound and its binding induces confor-
mational changes of the Na+, K+-ATPase that activate cascade reac-
tions leading to a variety of cellular responses.5 Activations of the
signaling cascades appear to occur at extremely low concentra-
tions, that is, involving only a few ATPase molecules without
affecting bulk Na+, K+-ATPase activity in the cell.
The glycon moiety of cardiac glycosides like digitoxin is known
to influence Na+, K+-ATPase inhibition.11 Comparable inhibitory
potencies are observed for digitoxin (1) and the aglycon digitoxi-
genin (4), while the di- and monoglycosylated steroids 2 and 3 ex-
hibit a three- to four-fold increased potency compared to 1 and 4.12
The binding affinities of the glycons 1, 2 and 3 are likewise analo-
gous, while 4 exhibit a sixfold decrease in affinity.12
López-Lázaro et al.13 have studied the effect of digitoxin (1) and
derivatives for growth inhibition activity in three human cancer
cell lines. A promising anticancer effect was observed for digitoxin
(1), and the effect was 20-fold lower for digitoxigenin (4).
⇑
Corresponding author. Tel.: +45 89423963.
E-mail address: hhj@chem.au.dk (H.H. Jensen).
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.02.016

2408
M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
Synthetic efforts directed towards manipulation on the carbo-
bohydrate mimics has previously been reported by Silva et al.18
Polyethylene glycol chains are often used to increase protein ser-
um half life19 mimicking one of the many effects of protein glyco-
sylation.20 The length of tetra ethylene glycol is modeled to be
14.3 Å (O-3 to terminal oxygen of tetraethylene glycol), which is
comparable to the length of tridigitoxoside modeled to be 15.2 Å
(O-3 to O-4 of terminal digitoxoside).21
We have therefore synthesized several digitoxin analogues (or
digitoxigenin derivatives), where the digitoxosides are replaced
by ethylene glycol moieties of varying lengths (Fig. 2). Furthermore
we have compared the biological activity of these compounds to-
gether with their bivalent analogues as Na+, K+-ATPase inhibitors
and as anticancer agents against the human breast cancer cell line
MCF-7.
hydrate fragment of cardiac glycosides have been conducted only
to a very small extent. With the focus of digitoxin, Rathore et al.
have studied the effect of carbohydrate stereochemistry on the
Na+, K+-ATPase inhibition.11 Additionally, non-natural and poten-
tially unstable glycosylated N,O-substituted hydroxylamines being
analogues of digitoxin have been evaluated as anticancer com-
pounds by Langenhan et al.14
As medicinal agents, cardiac glycosides like digitoxin (1) pos-
sess a labile glycon moiety, and the digitoxoside of digitoxin is
especially reactive (Fig. 1), due to it being 2,6-dideoxy glycoside
possessing an axial 3-OH.15 The metabolism of digitoxin produces
three major metabolites: digitoxigenin bisdigitoxoside 2, digitoxi-
genin monodigitoxoside 3 and the aglycon digitoxigenin (4), which
all have high binding affinities and inhibitory potencies of the Na+,
K+-ATPase. However, the pharmacokinetic profile strongly depends
on the presence of glycon moiety with the half life significantly
decreasing with loss of each digitoxoxide. The short half lives
and fast clearances of monodigitoxoside 3 and digitoxigenin (4)
have despite their potent Na+, K+-ATPase inhibition rendered these
two metabolites unsuitable for clinical use.16
2. Results and discussion
2.1. Chemistry
Using the chemistry developed by Langenhan et al.14a, Sawlewic
et al.22 and Gobbini et al.23 we decided to attach a sulfur atom to
facilitate easy linking of the ethylene glycol units (Scheme 1). Ini-
tially, treatment of digitoxin (1) under Jones oxidation conditions
resulted in glycoside cleavage and alcohol oxidation affording dig-
itoxigenon (13) in 75% yield. Subsequent reaction of ketone 13
with sodium borohydride at low temperature gave 3-epi-digitoxi-
genin 14 in 77% yield along with 5% of the undesired epimer, digi-
toxigenin (4). Although very selective, this reduction never
proceeded with complete stereocontrol contrary to previous
With monodigitoxoside being the most potent Na+, K+-ATPase
inhibitor among the digitoxigenin dervatives with varying degrees
of glycosylation, we speculate whether it will also have a superior
cytotoxic effect against cancer cells as suggested by Iyer et al.17
Given the short half life and high clearance of the cardiac ste-
roids with low degree of glycosylation we decided to study the
possibility of mimicking the glycon moiety with stable ethylene
glycol chains of varying length. The use of ethylene glycols as car-
O
O
O
O
OH
OH
18
Lactone
17
16
12
H
O
O
11
O
n
HO
19
13
14
9 H
1
4
OH
Digitoxigenin (4)
2
H 8 OH 15
10
5
n=3 Digitoxin (1)
3
7
6
2
n=2 Digitoxigenin bisdigitoxoside ( )
n=1 Digitoxigenin monodigitoxoside (3)
Glycon
H
Figure 1. Structures of digitoxin (1), and metabolites 2, 3 and 4. A general structure of cardiac glycosides is shown to the right.
O
O
OH
H
S
S
O
H
n-1
HO
O
O
n=1 (5), n=2, (6), n=3 (7) and n=4(8)
O
O
OH
S
H
n-1
O
HO
n=1 (9), n=2, (10), n=3 (11) and n=4 (12)
Figure 2. Structures of the prepared novel cardiac glycoside mimics, the mono-valent (9–12) and bivalent steroids (5–8).

M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
2409
O
O
O
O
OH
a
OH
H
O
O
O
1
13
O
3
OH
O
O
O
O
c
b
OH
OH
OH
14
H
15
17
H
OTs
OH
O
O
O
O
e
d
OH
16
AcS
HS
O
H
H
O
OH
f
S
H
H
S
18
HO
O
O
Scheme 1. Reagents and conditions: (a) CrO3, H2SO4/H2O, acetone, rt, 3 h, 75%; (b) NaBH4, dioxane/H2O, 0 °C, 30 min, 77%; (c) Ts2O, DMAP, pyridine, rt, o/n, 97%; (d) KSAc,
DMF, 60 °C, o/n, 69%; (e) NH3, MeOH/THF, rt, 2 h; (f) column chromatography, 46%.
reports.22,23 Tosylation of secondary alcohol 14 gave tosylate 15 in
O
O
near quantitative yield, which subsequently was substituted by
potassium thioacetate to afford the thioacetyl steroid 16 in 69%
yields. Thiodigitoxigenin 17 was obtained from 16 as described
by Gobbini et al.23 using NH3 in MeOH/THF (1:1). The free thiol
17 was found to be easily oxidized to disteroid disulfide 18 during
column chromatography (Scheme 1).
a
HO
OH
TBDMSO
OH
n-1
n-1
23
25
24
19,
20
n=2
n=1
n=3
(64%), n=2
(84%), n=4
(84%)
(73%)
n=1
26
n=3 21, n=4 22
O
b
TBDMSO
OMs
Ethylene glycol derivatives (27–30) bearing one TBDMS-
protected hydroxyl functionality and another as a mesylated
nucleofuge, were reacted with crude thiol 17 for the preparation
of mono-valent cardiac steroid mimics (9–12). The main limitation
in the choice of protecting group on the ethylene glycol unit was
the deprotection conditions, which was required to be neutral
since basic conditions (TBAF) is known to lead to epimerisation
at position 17,24 while acidic conditions can cause elimination of
the tertiary alcohol in position 14. Accordingly, the tert-butyldi-
methylsilyl protecting group was deprotected with a near neutral
fluoride ion source (Et3N/3HF) in good yields (Schemes 4 and 5).
For the preparation of ethylene glycol (vide supra), an excess
amount of ethylene glycol 19, 20, 21 or 22 was treated with
TBDMS-Cl in the presence of 4-(dimethylamino)pyridine (DMAP)
to afford mono TBDMS-protected 23, 24,25 2525 or 26, respectively.
Mesylation of the remaining alcohol afforded 2726, 28, 29 or 3027
(Scheme 2).
n-1
27
29
28
n=1
n=3
(62%), n=2
(59%)
(55%)
30
(96%), n=4
Scheme 2. Reagents and conditions: (a) TBDMS-Cl, DMAP, pyridine, rt, o/n; (b)
MsCl, Et3N, Et2O, 0 °C, 1 h.
O
O
a
HO
OH
MsO
OMs
n-1
n=2 20, n=3 21
22
n-1
n=2 31 (61%), n=3 32 (71%)
n=4 33 (47%)
n=4
O
b
I
I
n-1
n=2 35 (38%), n=3 36 (55%)
n=4 37 (60%)
For the synthesis of bivalent steroids dihalide ethylene glycol
derivatives were used. These (35–37) were prepared from their cor-
responding ethylene glycols 20, 21 or 22 via their dimesylated inter-
Scheme 3. Reagents and conditions: (a) MsCl, Et3N, Et2O, 0 °C to rt, 4 h; (b) NaI,
acetone, reflux, o/n.

2410
M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
O
O
O
TBDMSO
OMs
OH
n-1
27, 28 or 29
17
HS
a
H
O
O
OH
S
H
b
O
n-1
38
39
40
(36%), n=3 (26%)
R=TBDMS; n=1
(44%), n=2
RO
R=H; n=1 9 (85%), n=2 10 (77%), n=3 11 (83%)
Scheme 4. Reagents and conditions: (a) NaH, DMF, rt, 2 h, 26–44% over two steps; (b) Et3N:3HF, CH2Cl2, rt, o/n)
mediates (31–33).27 1,2-dibromo ethane (n = 1, 34) was used to pre-
O
O
pare the shortest bivalently linked cardiac steroids (Scheme 3).
Crude thiol 17 was reacted with the TBDMS-protected and
mesylated ethylene glycol linkers 27–30 in the presence of sodium
hydride to obtain the corresponding TBDMS-protected mono-
valent steroids 38–40 in moderate yields (Scheme 4). Desilylation
under neutral conditions uneventfully afforded the desired mono-
valent steroids 9, 10 and 11 in good yields. Attempts to synthesize
the mono-valently displaying target having the tetraethylene gly-
col unit (12) using the thiol 17 proved unsuccessful in accordance
with dropping yields as the linker size increased. The tetraethylene
glycol linked steroid (12), was, however, successfully prepared via
the intermediate 41 by reduction of the disulfide 18 in the
presence of 30 (Scheme 5).
OH
OH
S
H
H
H
S
18
O
O
HO
a
O
O
S
Difficulties arose in the formation of bivalent steroids (5–8,
Scheme 6). Despite full consumption of the crude thiol 17 was ob-
served, no product could be isolated. As a result, the crude thiol 17
was purified by column chromatography, which afforded the dis-
teroid disulfide derivative 18. The bivalent steroids were success-
fully obtained from the disulfide 18 under reductive condition in
the presence of a dihalide ethylene glycol linker (Scheme 6).
R=TBDMS 41 (58%)
R=H 12 (95%)
O
3
b
RO
Scheme 5. Reagents and conditions: (a) tetraethylene glycol linker 30, NaBH4,
absolute EtOH, rt, 20 h; (b) Et3N:3HF, CH2Cl2, rt, o/n.
O
O
OH
S
H
H
S
18
O
O
HO
O
O
a or b
OH
H
S
S
O
H
n-1
HO
n=1 5 (44%), n=2 6 (27%)
n=3 7 (49%), n=4 8 (58%)
O
O
Scheme 6. Reagents and conditions: (a) dihalide ethylene glycol linker 34, 36 or 37, NaBH4, absolute EtOH, rt, 3–17 h. Reagents and conditions for the formation of 6 (b)
diiodide diethylene glycol 35, NaBH4, dry DMF, rt, 1 h.

M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
2411
2.2. Inhibition of Na+, K+-ATPase
120
The cardiac glycoside mimics were tested for inhibition of Na+,
K+-ATPase activity (Figs. 3 and 4) and the apparent inhibitory con-
stants (Kapp) are listed in Table 1. The synthesized compounds ex-
hibit a clear trend; as the length of the ethylene glycol linker
increases the inhibitory power decreases. The bivalent steroids
5–8 display a significantly reduced inhibitory potency compared
to digitoxin (1), digitoxigenin (4) and the mono-valent steroids
(9–12). The impaired inhibitory potency of the bivalent cardiac
glycoside mimics could be due to the steric bulk of the extra
steroid.
100
80
60
40
20
0
The highest apparent inhibitory potency is observed for 9
(0.48 lM), which is close to the Kapp value 0.52 lM of digitoxigenin
-20
(4). The potency of 10 lies in the interval between digitoxin (1) and
digitoxigenin (4), while the largest mono-valent steroid 12 shows a
twofold decrease in potency compared to digitoxin (1). These
observations are consistent with the published data from Paula
et al.12, where the inhibition potency was found to increase when
0.01
0.1
1
10
100
Inhibitor (µM)
Figure 4. Inhibition of the Na+,K+-ATPase activity by digitoxigenin (filled circles),
digitoxin (filled squares), compound 12 (open circles) and compound 6 (open
squares). Residual activity is expressed in percent of that in the absence of inhibitor.
Na+, K+-ATPase is defined as ouabain-sensitive ATPase activity.
trimming of the
con is present. For cardiac glycosides including digitoxin (1),
b-glycon moieties are important for favorable interactions and the
c- and b-glycons and decreases again when no gly-
a- and
inhibition potency, while the c-glycone bear little significance for
Table 1
inhibition.12 The attached ethylene glycol moieties are suggested
to occupy a sterically restricted site, which have favorable polar
Mean Kapp values (
Mean IC50 values (
cell lines MCF-7. The mean values are from at least three determinations
l
l
M) of cardiac glycoside mimics against renal pig Na+, K+-ATPase.
M) for cardiac glycoside mimics against the human breast cancer
interactions to the a-saccharide unit.
The disulfide 18 was unstable and reduced to 3-mercapto digi-
toxigenin 17. The presented Kapp value for free thiol 17 is estimated
Compound
Inhibition
IC50 Values
(lM) in MCF-7
Kapp
(
lM)
to 0.42 lM.
Digitoxin (DT)
Digitoxigenin (DTG)
Mono-valent Steroids
1
4
9
10
11
12
17
5
0.99
0.52
0.48
0.73
1.27
2.1
0.42
9.9
17.2
20.5
>53
0.02 0.0005
0.6 0.1
0.2 0.02
0.4 0.1
0.8 0.2
1.4 0.3
0.5 0.02
7.0 0.5
8.6 1.9
6.6 1.7
2.3. Antiproliferation effect on cancer cells
The compounds were also evaluated for their effect on the via-
bility of MCF-7 breast cancer cells. It was found that the inhibitory
trend for cell viability resembles the trend for Na+, K+-ATPase inhi-
bition activity, with a few noteworthy exceptions (Table 1 and
Fig. 3). In the activity assay a slightly higher inhibitory potency is
Thiol
Bivalent Steroids
6
7
8
13
6
observed for digitoxigenin 4 (0.52
(0.99 M), whereas the relationship is reversed in the viability as-
say, showing that digitoxin (1) reduces cell viability very potently
(IC50 0.02 M vs 0.64 M for digitoxigenin). The interpretation of
lM) compared to digitoxin 1
l
depend on the ratio of the extracellular to intracellular volumes
and intracellular protein concentration, since proteins are able to
non-specifically adsorb cardiotonic steroids.28 The ability of digi-
toxigenin to readily permeate membranes is widely used in studies
of sided membrane preparations.29
Generally, the bivalent steroids are significantly less effective
than mono-valent steroids. In the activity assay a high inhibitory
potency was observed for the smallest mimic 9 compared to digi-
toxigenin (4). No improvement in the viability assay was observed
l
l
the results of the viability essays rely on the effective concentra-
tions of the compounds in the incubation media. Most of the com-
pounds, including digitoxin, are not able to penetrate the cell
membranes and will remain in the extracellular media. Digitoxi-
genin, however, will distribute between the intracellular and
extracellular media, thus, its effective concentration on the extra-
cellular side (relevant for binding to ATPase) will be lower. The
apparently decreased inhibitory potency of the digitoxigenin will
Figure 3. The blue columns illustrates Àlog Kapp (M) of cardiac glycoside mimics against renal pig Na+, K+-ATPase, while the red columns illustrates Àlog IC50 (M) for cardiac
glycoside mimics against the human breast cancer cell lines MCF-7. Digitoxin (1, DT), digitoxigenin (4, DTG), thiol 17, mimics 9–12 are mono-valent mimics in increasing size,
mimics 5–8 are bivalent mimics in increasing size.

2412
M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
for any of the mimics compared to digitoxin (1). However, the
three smallest mimics 9, 10 and 11 exhibit IC50 values in the inter-
val between digitoxin (1) and digitoxigenin (4). The remaining
compounds 5–8, 12 and 17 have considerable lower anticancer
effects.
er-Blue reagent (Promega) according to manufacturer’s
instructions and the fluorescence signal was measured at the
wavelengths 560Ex/590Em nm with a Victor3 multilabel counter
(Perkin–Elmer).
4.3. Chemistry
3. Conclusion
All reagents, unless otherwise stated, were used as purchased
without further purification. Solvents were dried according to stan-
dard procedures. Columns for flash chromatography were packed
with silica gel (60 mesh). TLC plates (Kieselgel 60 F254) were visu-
alized by use of PMA (10 g phosphomolybdic acid in 100 mL. abso-
lute EtOH) and heated until spots appeared or by UV-irradiation.
1H and 13C NMR experiments were recorded on a Varian Mercury
400 NMR instrument. Mass spectral data were carried out as elec-
trospray experiments on a Micromass LC-TOF instrument. GC–MS
analyses were carried out on a Hewlett-Packard 5890A gas chro-
matograph equipped with a 5971A MSD mass-selective detector.
The GC column was an HP5 25 m with an internal diameter of
0.25 mm, the injection temperature was 250 °C, the He-flow set
to 1.0 mL/min, temperature program 40 °C for 5 min to 230 °C, rate
15 °C/min. Characterization of steroid NMR-data was performed on
the basis of published data of digitoxin and digitoxigenin from
Drakenberg et al.31
Novel digitoxin mimics had been synthesized by replacing of
the labile trisaccharide moiety with stable ethylene glycol chains
of different lengths. The compounds preserved their ability to inhi-
bit Na+, K+-ATPase, demonstrating that glycosylation is not a prere-
quisite for their binding to the enzyme. In actual fact,
monoethylene glycol 9 (Kapp 0.42
more potent than digitoxin (Kapp 0.99
l
M) was found to be slightly
M). We furthermore recon-
l
firmed that cardiac glycoside mimics are potent anti-cancer com-
pounds, though the compounds prepared in this study were less
powerful than digitoxin regarding the effects on the viability of
breast cancer cells.
In addition, using ethylene glycol linkers of varying lengths, we
have also prepared four compounds having a bivalent display of
the cardiotonic steroid core. These compounds were found to have
between 20 and 110-fold decreased inhibitory potency against the
Na+, K+-ATPase.
The monovalent cardiac glycoside mimics prepared in this
study could in principle lead to compounds with improved phar-
macokinetic properties optimizing their effect on cardiac contrac-
tility. The results from this study also suggest that ethylene
glycol linked cardiac steroids could be used as biochemical tools
for immobilisation and thereby purification of the Na+, K+-ATPase.
4.4. General procedure for the synthesis of TBDMS-protected
and mesylated ethylene glycol linkers
To a solution of ethylene glycol (19, 20, 21 or 22) (8–10 equiv)
and DMAP (cat.) in dry pyridine (5 mL) was added TBDMS-Cl
(1 equiv). The reaction mixture was stirred at room temperature
overnight and then diluted with CH2Cl2 before the organic phase
was washed twice with 1 M HCl, once with H2O and then satd
NaHCO3. The organic phase was dried over MgSO4, filtered and
concentrated under reduced pressure. The TBDMS-protected ethyl-
ene glycols (23, 24, 25 or 26) were used without further
purification.
To a cooled solution (0 °C) of TBDMS-protected ethylene gly-
col (23, 24, 25 or 26) (1 equiv) in dry Et2O (5–7 mL) was added
dry Et3N (2 equiv) followed by dropwise addition of MsCl
(1.5 equiv). The reaction mixture was stirred at 0 °C to room
temperature for 1 h and then diluted with CH2Cl2. The organic
phase was washed with mildly acidic H2O, H2O and satd NaCO3,
and dried over MgSO4, filtered and concentrated under reduced
pressure. The crude product was purified by column
chromatography.
4. Experimental section
4.1. Na+, K+-ATPase activity assay
Relative inhibitory potency of the compounds was estimated
under the steady-state conditions of ATPase reaction. Briefly, the
purified Na+, K+-ATPase from pig kidney (specific activity of 25–
30 U/mg protein)30 was incubated for 2 h with various concentra-
tions of the inhibitor in the reaction media containing 130 mM
NaCl, 20 mM KCl, 4 mM MgCl2, 30 mM histidine buffer, pH 7.4,
20 °C. Then 3 mM ATP was added and the reaction was allowed
to proceed for 5 min. The amount of inorganic phosphate released
was measured and plotted as function of inhibitor concentration.
The ouabain-insensitive ATPase activity was measured in a sepa-
rate experiment in the presence of 1 mM ouabain and subtracted
from each number. The inhibitory potency was evaluated by fitting
of the data to the hyperbolic function. Kapp value reflects both the
affinity of the compound to the enzyme and the velocity of its
interaction with the enzyme.
4.4.1. Monoethylene glycol linker (27)
Monoethylene glycol 19 (2.94 g, 47 mmol) was reacted with
TBDMS-Cl (699 mg, 4.64 mmol). The product 23 was obtained as
a clear oil without further purification (525 mg, 64%): Rf: 0.45
(EtOAc/pentane 1:9); 1H NMR (400 MHz, CDCl3): dH 2.68 (m, 2H,
H2), 2.62 (m, 2H), 2.21 (br s, 1H, OH), 0.89 (m, 9H, Si-C(CH3)3),
0.06 (m, 6H, Si-CH3); 13C NMR (100 MHz, CDCl3): dC 64.3, 63.8,
26.0 (Si-C(CH3)3), 18.4 (Si-C(CH3)3), À5.2 (Si-CH3); LRMS (ES):
Calcd for C8H20O2SiNa 199.1, found 199.1.
Compound 23 (411 mg, 2.35 mmol) was reacted with Et3N
(0.58 mL, 4.18 mmol) and MsCl (0.24 mL, 3.09 mmol). The crude
product was purified by column chromatography (EtOAc/pentane
1:9) to afford 27 as yellow oil (372 mg, 62%): Rf: 0.33 (EtOAc/pen-
tane 1:9); 1H NMR (400 MHz, CDCl3): dH 4.26 (t, 2H, J 4.8 Hz), 3.86
(t, 2H, J 4.8 Hz), 3.02 (s, 3H, S-CH3), 0.88 (s, 9H, Si-C(CH3)3), 0.07 (s,
6H, Si-CH3); 13C NMR (100 MHz, CDCl3): dC 71.4, 61.4, 37.8 (S-CH3),
25.9 (Si-C(CH3)3), 18.4 (Si-C(CH3)3), À5.3 (Si-CH3); HRMS (ES):
Calcd for C9H20O4SSiNa 277.0906, found 277.0903. NMR spectral
values for 27 are in accordance with reference.26
4.2. Viability assay
MCF-7 breast cancer cells were cultured in DMEM Glutamax
(Gibco) supplemented with 10% fetal calf serum, penicillin
(100 U/mL) and streptomycin (100 mg/mL) in a humidified atmo-
sphere with 5% CO2 at 37 °C.
For the toxicity screen, 3000 cells per well were seeded in 96-
well format as triplets and cultured overnight. Afterwards, the
medium was replaced with medium containing a dilution series
of the compounds. An equivalent dilution series of the solvent
DMSO served for normalization. The cells were exposed to the
compounds for 48 h followed by 72 h culture in fresh medium
without the compounds prior to the viability assay. Relative num-
bers of viable cells were determined with the fluorometric CellTit-

M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
2413
4.4.2. Diethylene glycol linker (28)
4.5. General procedure for the synthesis of di-iodo ethylene
glycol linkers (35–37)
Diethylene glycol 20 (3.43 g, 32 mmol) was reacted with
TBDMS-Cl (622 mg, 4.13 mmol). The product 24 (765 mg, 84%) as
clear oil was obtained without further purification: Rf: 0.33
(EtOAc/pentane 1:4); 1H NMR (400 MHz, CDCl3): dH 3.72 (m, 2H),
3.66 (m, 2H), 3.54 (m, 4H), 2.85 (br s, 1H, OH), 0.85 (m, 9H, Si-
C(CH3)3), 0.03 (m, 6H, Si-CH3); 13C NMR (100 MHz, CDCl3): dC
72.6, 72.5, 62.9, 61.8, 25.9 (Si-C(CH3)3), 18.4 (Si-C(CH3)3), À5.3
(Si-CH3); LRMS (ES): Calcd for C10H24O3SiNa 243.1392, found
243.1389. NMR spectral values for 24 are in accordance with
reference.25
Compound 24 (765 mg, 3.47 mmol) was reacted with Et3N
(0.87 mL, 6.3 mmol) and MsCl (0.35 mL, 6.3 mmol). The crude
product was purified by column chromatography (15% EtOAc in
pentane) to afford 28 as clear oil (605 mg, 59%): Rf: 0.33 (EtOAc/
pentane 1:9); 1H NMR (400 MHz, CDCl3): dH 4.34 (m, 2H), 3.74
(m, 4H), 3.55 (m, 2H), 3.03 (s, 3H, S-CH3), 0.86 (s, 9H, Si-C(CH3)3),
0.03 (s, 6H, Si-CH3); HRMS (ES): Calcd for C11H26OSSiNa
321.1168, found 321.1165.
A solution of ethylene glycol (20, 21 or 22) (1 equiv) in dry Et2O
(4–10 mL) was cooled to 0 °C and added dry Et3N (5 equiv) fol-
lowed by drop wise addition of MsCl (2.5 equiv). The reaction mix-
ture was stirred at 0 °C to room temperature for 4 h and then
diluted with CH2Cl2. The organic phase was washed with mildly
acidic H2O, H2O, satd NaHCO3 and dried over MgSO4, filtered and
concentrated under reduced pressure. The dimesylated ethylene
glycol (31, 32 or 33) was used without further purification.
To a stirred solution of dimesylated ethylene glycol (31, 32 or
33) (1 equiv) in dry acetone (5 mL) was added NaI (4 equiv). The
reaction mixture was stirred at reflux overnight before cooled to
room temperature and concentrated under reduced pressure. To
the residue was added CH2Cl2 and H2O, and the aqueous phase
was extracted four times with CH2Cl2. The combined organic
phases were washed with satd NaCl, dried over MgSO4, and con-
centrated under reduced pressure. The crude product was purified
by column chromatography to afford 35, 36 or 37.
4.4.3. Triethylene glycol linker (29)
4.5.1. Diethylene glycol linker (35)
Triethylene glycol 21 (4.95 g, 33 mmol) was reacted with
TBDMS-Cl (741 mg, 4.92 mmol). The product 25 (1.096 g, 84%)
was obtained as clear oil and used without further purification:
Rf: 0.36 (EtOAc/pentane 1:1); 1H NMR (400 MHz, CDCl3): dH 3.74
(t, 2H, J 5.4 Hz), 3.70–3.57 (m, 8H), 3.54 (t, 2H, J 5.2 Hz) 2.66 (br
s, 1H, OH), 0.87 (s, 9H, Si-C(CH3)3), 0.04 (s, 6H, Si-CH3); 13C NMR
(100 MHz, CDCl3): dC 72.8, 72.6, 70.9, 70.6, 62.8, 61.8, 26.0 (Si-
C(CH3)3), 18.4 (Si-C(CH3)3), À5.2 (Si-CH3); LRMS (ES): Calcd for
Diethylene glycol 20 (213 mg, 2.01 mmol) was reacted with
Et3N (1.4 mL, 10.02 mmol) and MsCl (0.4 mL, 5.16 mmol). The
product 31 was obtained without further purification as yellow
oil (319 mg, 61%): Rf: 0.57 (EtOAc); 1H NMR (400 MHz, CDCl3): dH
4.36 (m, 4H), 3.78 (m, 4H), 3.05 (s, 6H, S-CH3); 13C NMR
(100 MHz, CDCl3): dC 69.1, 68.9, 37.7 (S-CH3); LRMS (ES): Calcd
for C6H14O7S2Na 285.0, found 285.0.
Compound 31 (101 mg, 0.39 mmol) was reacted with NaI
(231 mg, 1.54 mmol). The crude product was purified by column
chromatography (1:4 CH2Cl2/pentane ? CH2Cl2 ? EtOAc) to afford
35 as clear oil (48 mg, 38%): Rf: 0.43 (CH2Cl2/pentane 1:4); 1H NMR
C
12H28O4SiNa 287.1655, found 287.0. NMR spectral values for 25
are in accordance with reference.25
Compound 25 (1.082 g, 4.09 mmol) was reacted with Et3N
(1.03 mL, 7.43 mmol) and MsCl (0.41 mL, 4.19 mmol). The product
29 (1.339 g, 96%) was obtained as pale yellow oil and used without
further purification: Rf: 0.51 (EtOAc/pentane 1:2); 1H NMR
(400 MHz, CDCl3): dH 4.36 (m, 2H), 3.75 (m, 4H), 3.64 (s, 4H),
3.53 (t, 2H, J 5.4 Hz), 3.05 (s, 3H, S-CH3), 0.88 (s, 9H, Si-C(CH3)3),
0.05 (s, 6H, Si-CH3); 13C NMR (100 MHz, CDCl3): dC 72.8, 70.8,
70.8, 69.4, 69.1, 62.8, 37.8 (S-CH3), 26.0 (Si-C(CH3)3), 18.5 (Si-
C(CH3)3), À5.2 (Si-CH3); HRMS (ES): Calcd for C13H30O6SSiNa
365.1430, found 365.1431.
(400 MHz, CDCl3): dH 3.77 (t, 4H, J 6.8 Hz), 3.26 (t, 4H, J 6.8 Hz): 13
C
NMR (100 MHz, CDCl3): dC 71.5, 2.6; GC–MS (EI): Calcd for C4H8I2O
326, found 326. NMR spectral values for 31 and 35 are in accor-
dance with reference.27
4.5.2. Triethylene glycol linker (36)
Triethylene glycol 21 (295 mg, 1.96 mmol) was reacted with
Et3N (1.4 mL, 10.0 mmol) and MsCl (0.4 mL, 5.16 mmol). The prod-
uct 32 appeared as a yellow oil and was used without further puri-
fication (424 mg, 71%): Rf: 0.46 (EtOAc); 1H NMR (400 MHz,
CDCl3): dH 4.35 (m, 4H), 3.75 (m, 4H), 3.66 (s, 4H), 3.05 (s, 6H, S-
CH3); 13C NMR (100 MHz, CDCl3): dC 70.6, 69.2, 69.1, 37.7 (S-
CH3); LRMS (ES): Calcd for C8H18O8S2Na 329.0, found 328.9.
Compound 32 (72 mg, 0.24 mmol) was reacted with NaI
(140 mg, 0.93 mmol). The crude product was purified by column
chromatography (CH2Cl2) to afford 36 as pale yellow oil (49 mg,
55%): Rf: 0.55 (CH2Cl2); 1H NMR (400 MHz, CDCl3): dH 3.77 (t, 4H,
J 6.8 Hz), 3.67 (s, 4H), 3.26 (t, 4H, J 6.8 Hz); 13C NMR (100 MHz,
CDCl3): dC 72.1, 70.3, 3.0; LRMS (ES): Calcd for C6H12I2O2Na
392.9, found 392.9. NMR spectral values for 32 and 36 are in accor-
dance with reference.27
4.4.4. Tetraethylene glycol linker (30)
Tetraethylene glycol 22 (6.144 g, 32 mmol) was reacted with
TBDMS-Cl (623 mg, 4.13 mmol). The product 26 (931 mg, 73%)
was obtained as pale yellow oil and used without further puri-
fication: Rf: 0.48 (EtOAc); 1H NMR (400 MHz, CDCl3): dH 3.66
(m, 2H), 3.61 (m, 2H), 3.56 (m, 8H), 3.50 (m, 2H), 3.46 (m,
2H), 3.11 (br s, 1H, OH), 0.80 (m, 9H, Si-C(CH3)3), À0.03 (m,
6H, Si-CH3); 13C NMR (100 MHz, CDCl3): dC 72.6, 72.5, 70.6,
70.6, 70.5, 70.3, 62.6, 61.5, 25.9 (Si-C(CH3)3), 18.3 (Si-C(CH3)3),
À5.3 (Si-CH3); HRMS (ES): Calcd for C14H32O5SiNa 331.1917,
found 331.1916.
Compound 26 (931 mg, 3.02 mmol) was reacted with Et3N
(0.75 mL, 5.43 mmol) and MsCl (0.31 mL, 3.99 mmol). The crude
product was purified by column chromatography (EtOAc/pentane
1:2 ? 1:1) to afford 30 as clear oil (644 mg, 55%): Rf: 0.39
(EtOAc/pentane 1:1); 1H NMR (400 MHz, CDCl3): dH 4.37 (m, 2H),
3.74 (m, 4H), 3.63 (m, 8H), 3.53 (t, 2H, J 5.4 Hz), 3.06 (d, 3H, J
1.2 Hz, S-CH3), 0.87 (d, 9H, J 0.8 Hz, Si-C(CH3)3), 0.05 (d, 6H, J
0.8 Hz, Si-CH3); 13C NMR (100 MHz, CDCl3): dC 72.8, 70.8, 70.7,
70.6, 69.4, 69.1, 62.8, 37.8 (S-CH3), 26.0 (Si-C(CH3)3), 18.5 (Si-
C(CH3)3, -5.2 (Si-CH3); HRMS (ES): Calcd for C15H34O7SSiNa
409.692, found 409.1689. NMR spectral values for 30 are in accor-
dance with reference.32
4.5.3. Tetraethylene glycol linker (37)
Tetraethylene glycol 22 (1.081 g, 5.56 mmol) was reacted with
Et3N (3.9 mL, 27.9 mmol) and MsCl (1.1 mL, 14.2 mmol). The prod-
uct 33 was obtained as a yellow oil and used without further puri-
fication (907 mg, 47%): Rf: 0.38 (EtOAc); 1H NMR (400 MHz, CDCl3):
dH 4.34 (m, 4H), 3.74 (m, 4H), 3.62 (m, 8H), 3.04 (s, 6H, S-CH3); 13
C
NMR (100 MHz, CDCl3): dC 70.7, 70.5, 69.3, 69.0, 37.7 (S-CH3);
HRMS (ES): Calcd for C10H22O9S2Na 373.0603, found 373.0601.
Compound 33 (376 mg, 1.07 mmol) was reacted with NaI
(642 mg, 4.28 mmol). The crude product was purified by column
chromatography (CH2Cl2 ? 2.5% Et2O in CH2Cl2) to afford 37 as

2414
M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
clear oil (267 mg, 60%): Rf: 0.25 (EtOAc/pentane 1:9); 1H NMR
(400 MHz, CDCl3): dH 3.74 (t, 4H, J 6.8 Hz), 3.65 (s, 8H), 3.25 (t,
4H, J 6.8 Hz); 13C NMR (100 MHz, CDCl3): dC 72.0, 70.8, 70.3, 3.1;
HRMS (ES): Calcd for C8H16I2O3Na 436.9087, found 436.9084.
NMR spectral values for 33 and 37 are in accordance with
reference.27
2.45 (s, 3H, SO2C6H4CH3), 2.15 (m, 2H), 1.96–1.16 (m, 18H), 0.99
(m, 1H), 0.88 (s, 3H, CH3), 0.85 (s, 3H, CH3); 13C NMR (100 MHz,
CDCl3): dC 175.0, 174.7 (C20, C23), 144.6 (Ar), 134.5 (Ar), 129.8
(Ar), 127.6 (Ar), 117.6 (C22), 85.2 (C14), 82.8 (C3), 73.6 (C21),
50.9 (C13 or C17), 49.6 (C17 or C13), 41.7, 41.6, 39.8, 36.1, 34.7,
34.6, 33.1, 33.0, 27.7, 26.9, 26.7, 23.0, 21.7, 21.3, 20.9, 15.8; LRMS
(ES): Calcd for C30H40O6SNa 551.2, found 551.3. NMR spectral val-
ues are in accordance with reference.31
4.5.4. Digitoxigenon (13)
Jones reagent (CrO3 (2.03 g, 20.3 mmol) in H2O/H2SO4 (8 mL/
2 mL)) was slowly added to a suspension of digitoxin 1 (994 mg,
1.29 mmol) in acetone (40 mL) at 0 °C. The reaction mixture was
stirred at room temperature for 3 h and added MeOH (5 mL) and
stirred further 20 min. The mixture was diluted with H2O and ex-
tracted four times with CH2Cl2. The combined organic phases were
washed once with satd NaHCO3 and twice with H2O before dried
over Na2SO4, filtered and concentrated under reduced pressure to
afford 13 (361 mg, 75%) as white foam. The product obtained
was used without further purification: Rf: 0.40 (pentane/EtOAc
1:3); 1H NMR (400 MHz, CDCl3): dH 5.83 (s, 1H, H22), 4.97 (dd,
1H, J17,21a 1.8 Hz, Jgem 18.2 Hz, H21a), 4.78 (dd, 1H, J17,21b 1.4 Hz,
Jgem 18.2 Hz, H21b), 2.77 (m, 1H, H17), 2.59 (t, 1H, J 14.4 Hz),
2.31 (dt, 1H, J 5.2 Hz, 14.4 Hz), 2.16–1.20 (m, 19H), 0.98 (s, 3H,
CH3), 0.86 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 212.9 (C3),
174.9 (C20 or C23), 174.7 (C23 or C20), 117.6 (C22), 85.1 (C14),
73.6 (C21), 50.8, 49.7 (C13, C17), 43.7, 42.1, 41.5, 39.7, 37.11,
36.7, 36.6, 35.2, 33.0, 26.9, 26.5, 22.5, 21.2, 21.0, 15.8; LRMS (ES):
Calcd for C23H32O4Na 395.2, found 395.4. NMR spectral values
are in accordance with reference.31
4.5.7. 3-Deoxy-3-mercaptoacetyl-digitoxigenin (16)
To a solution of 15 (369 mg, 0.698 mmol) in dry DMF (10 mL)
was added potassium thioacetate (202 mg, 177 mmol). The reac-
tion mixture was stirred at 50 °C for 22 h before cooled to room
temperature and diluted with EtOAc and washed four times with
H2O and brine. The organic phase was dried over MgSO4, filtered
and concentrated under reduced pressure before the crude product
was purified by column chromatography (EtOAc/pentane
1:3 ? 1:2) to afford 16 as white solid (209 mg, 69%): Rf: 0.31
(EtOAc/pentane 1:2); 1H NMR (400 MHz, CDCl3): dH 5.84 (s, 1H,
H22), 4.98 (dd, 1H, J17,21a 1.6 Hz, Jgem 18.0 Hz, H21a), 4.79 (dd,
1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 4.03 (br s, 1H, H3), 2.75
(m, 1H, H17), 2.29 (s, 3H, SCOCH3), 2.13 (m, 2H), 1.89–1.14 (m,
19H), 0.92 (s, 3H, CH3), 0.85 (s, 3H, CH3); 13C NMR (100 MHz,
CDCl3): dC 195.7 (SCO), 174.8, 174.7 (C20, C23), 117.7 (C22), 85.5
(C14), 73.6 (C21), 50.9, 49.7 (C13, C17), 42.5, 41.9, 40.0, 39.1,
36.1, 35.6, 33.2, 32.8, 31.9, 31.1, 26.9, 26.8, 26.5, 23.8, 21.6, 21.2,
15.9; HRMS (ES): Calcd for
C25H36O4SNa 455.2232, found
455.2249. NMR spectral values are in accordance with reference.31
4.5.5. 3-epi-digitoxigenin (14)
4.5.8. 3-Mercapto digitoxigenin (17)
To a solution of digitoxigenon 13 (361 mg, 0.969 mmol) in H2O/
dioxane (1:7, 24 mL) at 0 °C was added NaBH4 (92 mg, 2.4 mmol).
The reaction mixture was stirred at 0 °C for 1 h and diluted with
CH2Cl2. The mixture was washed once with H2O before the aque-
ous phase was extracted twice with CH2Cl2. The combined organic
phases were washed with satd NaCl, dried over MgSO4, filtered and
concentrated under reduced pressure. The crude product was puri-
fied by column chromatography (EtOAc/pentane 1:1 ? EtOAc) to
afford 14 as white solid (280 mg, 77%): Rf: 0.48 (EtOAc); 1H NMR
(400 MHz, DMSO-d6): dH 5.91 (s, 1H, H22), 4.97 (dd, 1H, J17,21a
1.8 Hz, Jgem 18.4 Hz, H21a), 4.88 (dd, 1H, J17,21b 1.6 Hz, Jgem
18.4 Hz, H21b), 4.43 (br s, 1H, OH), 4.08 (s, 1H, H3), 3.38 (br s,
1H, H3), 3.31 (s, 1H, OH), 2.74 (m, 1H, H17), 2.05 (m, 2H), 1.83–
1.05 (m, 18H), 0.92 (dt, J 3.2 Hz, J 14.0 Hz, 1H) 0.85 (s, 3H, CH3),
0.77 (s, 3H, CH3); 13C NMR (100 MHz, DMSO-d6): dC 176.3, 173.8
(C20, C23), 116.2 (C22), 83.7 (C14), 73.1 (C21), 69.8, 50.2, 49.4,
41.3, 41.0, 36.2, 35.5, 34.9, 34.5, 32.2, 30.5, 26.9, 26.4, 23.2, 21.2,
20.6 (CH3), 15.7 (CH3); LRMS (ES): Calcd for C23H34O4Na 397.2,
found 397.3. NMR spectral values are in accordance with
reference.31
Thioacetate 16 (270 mg, 0.624 mmol) was dissolved in dry
MeOH/THF (1:1, 15 mL) and the mixture was degassed with N2.
Ammonia was then gently bubbled through the reaction mixture
at room temperature for 2 h. The reaction mixture was concentrated
under reduced pressure to afford crude 17 (303 mg, 124%), which
was sufficiently pure for further reactions. Rf: 0.30 (EtOAc/pentane
1:2); 1H NMR (400 MHz, CDCl3): dH 5.85 (s, 1H, H22), 4.98 (dd, 1H,
J17,21a 1.2 Hz, Jgem 18.0 Hz, H21a), 4.79 (dd, 1H, J17,21b 1.6 Hz, Jgem
18.0 Hz, H21b), 3.58 (br s, 1H, H3), 2.76 (m, 1H, H17), 2.14 (m, 2H),
1.92–1.15 (m, 19H), 0.96 (s, 3H, CH3), 0.85 (s, 3H, CH3); HRMS (ES):
Calcd for C23H34O3SNa 413.2126, found 413.2132 NMR spectral val-
ues are in accordance with reference.31
4.5.9. 1,2-Bis(3-digitoxigenin)disulfane (18)
The crude 17 was purified by column chromatography (EtOAc/
pentane 1:3 ? 1:2) to afford 18 as white solid (226 mg, 46%): Rf:
0.30 (EtOAc/pentane 1:2); 1H NMR (400 MHz, CDCl3): dH 5.85 (s,
1H, H22), 4.98 (dd, 1H, J17,21a 1.2 Hz, Jgem 18.0 Hz, H21a), 4.79
(dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.58 (br s, 1H, H3),
2.76 (m, 1H, H17), 2.14 (m, 2H), 1.92–1.15 (m, 18H), 0.96 (s, 3H,
CH3), 0.85 (s, 3H, CH3).
4.5.6. 3-epi-O-p-toluenesulfonyl digitoxigenin (15)
To a solution of 3-epi-digitoxigenin 14 (269 mg, 0.718 mmol) in
dry pyridine (15 mL) at 0 °C was added Ts2O (592 mg, 1.81 mmol)
and DMAP (cat.) The reaction mixture was stirred at room temper-
ature overnight before quenched with H2O (10 mL). The solution
was stirred for 10 min and diluted with CH2Cl2 and washed twice
with satd NaHCO3. The aqueous phases were extracted thrice with
CH2Cl2 and the combined organic layers dried over MgSO4, filtered
and concentrated under reduced pressure. The crude product was
purified by column chromatography (EtOAc/pentane 2:5) to afford
15 as white foam (369 mg, 97%): Rf: 0.38 (EtOAc/pentane 1:1); 1H
NMR (400 MHz, CDCl3): dH 7.79 (d, 2H, Jortho 8.2 Hz, ArH), 7.33 (d,
2H, Jortho 8.2 Hz, ArH), 5.87 (s, 1H, H22), 4.97 (dd, 1H, J17,21a
1.2 Hz, Jgem 18.40 Hz, H21a), 4.79 (dd, 1H, J17,21b 1.8 Hz, Jgem
18.2 Hz, H21b), 4.45 (sp, 1H, J 5.2 Hz, H3), 2.77 (m, 1H, H17),
4.6. General synthesis of TBDMS-protected 3-(ethylene glycol)-
digitoxigenin-monomer (n = 1, 2 or 3)
Sodium hydride (60%, 1.5 equiv) was added to a solution of
crude thiol 17 (50–75 mg, 1 equiv) and mesylate 27, 28 or 29
(3 equiv) in dry DMF (2 mL) under inert atmosphere). The mixture
stirred at room temperature for 1–2 h before transferred to a sep-
aration funnel containing ice water and CH2Cl2, and the aqueous
phase was extracted four times with CH2Cl2 before the combined
organic phases were concentrated under reduced pressure and
the residue was then dissolved in EtOAc and washed five times
with H2O. The organic phase was dried over MgSO4, filtered and
concentrated under reduced pressure. The crude product was puri-
fied by column chromatography.

M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
2415
4.6.1. tert-Butyldimehylsilyl 2-(digitoxigenin-3-ylthio)ethanol
(38)
fore diluted with brine. The aqueous phase was extracted four
times with CH2Cl2 and the combined organic phases were dried
over MgSO4, filtered and concentrated under reduced pressure.
The crude product was purified by column chromatography
(EtOAc/CH2Cl2 1:9) and afforded 41 as clear oil (25 mg, 58%): Rf:
0.19 (EtOAc/CH2Cl2 1:4); 1H NMR (400 MHz, CDCl3): dH 5.86 (s,
1H, H22), 4.98 (dd, 1H, J17,21a 1.6 Hz, Jgem 18.4 Hz, H21a), 4.79
(dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.75 (t, 2H, J30,31
5.6 Hz, H31), 3.64 (m, 10H, H25, H26, H27, H28, H29), 3.54 (t,
2H, J 5.6 Hz, H30), 3.23 (br s, 1H, H3), 2.77 (m, 1H, H17), 2.67 (t,
2H, J24,25 7.0 Hz, H24), 2.13 (m, 2H), 1.90–1.15 (m, 19H), 0.93 (s,
3H, CH3), 0.88 (s, 9H, (Si-C(CH3)3), 0.86 (s, 3H, CH3), 0.05 (s, 6H,
Si-CH3); 13C NMR (100 MHz, CDCl3): dC 174.7 (C20, C23), 117.8
(C22), 85.7 (C14), 73.6 (C21), 72.8, 71.2, 70.8, 70.8, 70.7, 70.4
(C25, C26, C27, C28, C29, C30), 62.8 (C31), 51.0, 49.7 (C13, C17),
43.9, 42.0, 40.1, 37.1, 36.1, 35.7, 33.3, 31.8, 31.1, 30.9, 26.9, 26.8,
26.2, 26.1 (Si-C(CH3)3), 23.8, 21.7, 21.2, 18.5, 15.9, -5.1 (Si-
(CH3)2); HRMS (ES): Calcd for C37H64O7SSiNa 703.4040, found
703.4057.
Compound 17 (52 mg) was reacted with 27 (89 mg,
0.349 mmol) in dry DMF (2 mL) in the presence of NaH (7 mg,
60%, 0.175 mmol). The crude product was purified by column chro-
matography (EtOAc/pentane 1:4 ? 1:1) to afford 27 mg of 38 as
thick clear oil (0.049 mmol, 44% over 2 steps): Rf: 0.26 (EtOAc/pen-
tane 1:3); 1H NMR (400 MHz, CDCl3): dH 5.87 (s, 1H, H22), 4.98 (dd,
1H, J17,21a 1.6 Hz, Jgem 18.0 Hz, H21a), 4.80 (dd, 1H, J17,21b 2.0 Hz,
Jgem 18.0 Hz, H21b), 3.73 (t, 2H, J24,25 7.2 Hz, H25), 3.25 (br s, 1H,
H3), 2.78 (m, 1H, H17), 2.60 (t, 2H, J24,25 7.2 Hz, H24), 2.08 (m,
2H), 1.90–1.17 (m, 19H), 0.94 (s, 3H, CH3), 0.90 (s, 9H, Si-
C(CH3)3), 0.86 (s, 3H, CH3), 0.06 (s, 6H, Si-CH3); 13C NMR
(100 MHz, CDCl3): dC 174.7 (C20, C23), 117.8 (C22), 85.7 (C14),
73.6 (C21), 63.6 (C25), 51.1, 49.7 (C13, C17), 43.9, 42.1, 40.2,
37.1, 36.1, 35.8, 34.1, 33.3, 31.9, 30.9, 27.0, 26.9, 26.2, 26.1 (Si-
C(CH3)3), 23.8, 21.7, 21.2, 18.5, 15.9, À5.1 (Si-CH3); HRMS (ES):
Calcd for C31H52O4SSiNa 571.3253, found 571.3289.
4.6.2. tert-Butyldimehylsilyl 4-(digitoxigenin-3-ylthio)
diethylene glycol (39)
4.7. General procedure for TBDMS-deprotection (n = 1, 2, 3 or 4)
Compound 17 (75 mg) was treated with 28 (181 mg,
0.573 mmol) in the presence of NaH (12 mg, 60%, 0.300 mmol).
The crude product was purified by column chromatography
(EtOAc/pentane 1:4) to afford 39 as thick clear oil (28 mg, 36% over
2 steps): Rf: 0.20 (EtOAc/pentane 1:3); 1H NMR (400 MHz, CDCl3):
dH 5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.2 Hz, Jgem 18.0 Hz, H21a),
4.80 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.75 (t, 2H, J26,27
5.2 Hz, H27), 3.63 (t, 2H, J24,25 7.2 Hz, H25), 3.53 (t, 2H, J26,27
5.2 Hz, H26), 3.25 (br s, 1H, H3), 2.78 (m, 1H, H17), 2.67 (t, 2H,
J24,25 7.2 Hz, H24), 2.04 (m, 2H), 1.90–1.18 (m, 19H), 0.94 (s, 3H,
CH3), 0.89 (s, 9H, Si-C(CH3)3), 0.86 (s, 3H, CH3), 0.06 (s, 6H, Si-
CH3); 13C NMR (100 MHz, CDCl3): dC 174.7(C20, C23), 117.8
(C22), 85.7 (C14), 73.6 (C21), 72.6, 71.4 (C25, C26), 62.9 (C27),
51.1, 49.7 (C13, C17), 43.9, 42.1, 40.2, 37.2, 36.1, 35.7, 33.3, 31.8,
31.3, 30.9, 27.0, 26.9, 26.2, 26.1 (Si-C(CH3)3), 23.8, 21.7, 21.2,
18.5, 15.9, À5.1(Si-CH3); HRMS (ES): Calcd for C33H56O5SSiNa
615.3515, found 615.3524.
Silylether 38, 39, 40 or 41 (1 equiv) was dissolved in dry CH2Cl2
(2–3 mL), cooled to 0 °C and added Et3N:3HF (5–8 equiv). The reac-
tion was stirred at room temperature for 20–24 h before diluted
with toluene and concentrated under reduced pressure. The crude
product was purified by column chromatography.
4.7.1. 2-(Digitoxigenin-3-ylthio)ethanol (9)
Compound 38 (27 mg, 0.049 mmol) was reacted with Et3N:3HF
(48 lL, 0.295 mmol). The crude product was purified by column
chromatography (pentane/EtOAc 1:2) to afford 9 as clear oil
(18 mg, 85%): Rf: 0.25 (EtOAc/pentane 1:2); 1H NMR (400 MHz,
CDCl3): dH 5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.6 Hz, Jgem
18.0 Hz, H21a), 4.80 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b),
3.70 (t, 2H, J24,25 6.0 Hz, H25), 3.23 (br s, 1H, H3), 2.77 (m, 1H,
H17), 2.71 (t, 2H, J24,25 6.0 Hz, H24), 2.14 (m, 2H), 1.92–1.17 (m,
19H), 0.95 (s, 3H, CH3), 0.87 (s, 3H, CH3); 13C NMR (100 MHz,
CDCl3): dC 174.6 (C20, C23), 117.8 (C22), 85.7 (C14), 73.6 (C21),
60.6 (C25), 51.0, 49.7 (C13, C17), 43.3, 42.1, 40.1, 37.3, 36.1, 35.8,
35.1, 33.3, 31.9, 31.0, 27.0, 26.9, 26.3, 23.8, 21.7, 21.2, 15.9; HRMS
(ES): Calcd for C25H38O4SNa 457.2389, found 457.2389.
4.6.3. tert-Butyldimehylsilyl 6-(digitoxigenin-3-ylthio)
triethylene glycol (40)
Compound 17 (51 mg) was reacted with 29 (155 mg,
0.453 mmol) in the presence of NaH (7 mg, 60%, 0.175 mmol).
The crude product was purified by column chromatography
(EtOAc/pentane 1:3) to afford 40 as thick clear oil (19 mg, 26% over
2 steps): Rf: 0.20 (EtOAc/pentane 1:2); 1H NMR (400 MHz, CDCl3):
dH 5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.6 Hz, Jgem 18.0 Hz, H21a),
4.80 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.76 (t, 2H, J28,29
5.2 Hz, H29), 3.64 (m, 6H, H25, H26, H27), 3.56 (t, 2H, J 5.2 Hz,
H28), 3.25 (br s, 1H, H3), 2.78 (m, 1H, H17), 2.68 (t, 2H, J24,25
7.2 Hz, H24), 2.12 (m, 2H), 1.91–1.19 (m, 19H), 0.94 (s, 3H, CH3),
0.89 (s, 9H, Si-C(CH3)3), 0.87 (s, 3H, CH3), 0.06 (s, 6H, Si-CH3); 13C
NMR (100 MHz, CDCl3): dC 174.6 (C20, C23), 117.9 (C22), 85.7
(C14), 73.6 (C21), 72.9, 71.3, 70.9, 70.6 (C25, C26, C27, C28), 62.9
(C29), 51.1, 49.7 (C13, C17), 43.9, 42.1, 40.2, 37.2, 36.2, 35.8,
33.3, 31.9, 31.2, 30.9, 27.0, 26.9, 26.2, 26.1 (Si-C(CH3)3), 23.8,
21.7, 21.2, 18.5, 15.9, À5.1(Si-CH3); HRMS (ES): Calcd for
4.7.2. 4-(Digitoxigenin-3-ylthio)diethylene glycol (10)
Compound 39 (28 mg, 0.047 mmol) was treated with Et3N:3HF
(55 lL, 0.337 mmol). The crude product was purified by column
chromatography (pentane/EtOAc 1:2) to afford 10 as clear oil
(17 mg, 77%): Rf: 0.45 (EtOAc); 1H NMR (400 MHz, CDCl3): dH
5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.2 Hz, Jgem 18.0 Hz, H21a),
4.80 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.73 (t, 2H, J26,27
4.6 Hz, H27), 3.65 (t, 2H, J24,25 6.6 Hz, H25), 3.58 (t, 2H, J26,27
4.6 Hz, H26), 3.26 (br s, 1H, H3), 2.78 (m, 1H, H17), 2.70 (t, 2H,
J24,25 6.6 Hz, H24), 2.14 (m, 2H), 1.91–1.18 (m, 19H), 0.94 (s, 3H,
CH3), 0.87 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 174.6
(C20, C23), 117.8 (C22), 85.7 (C14), 73.6 (C21), 72.1, 70.8 (C25,
C26), 61.9 (C27), 51.1, 49.7 (C13, C17), 44.1, 42.1, 40.2, 37.2, 36.1,
35.8, 33.3, 31.8, 31.5, 30.9, 27.0, 26.9, 26.2, 23.8, 21.7, 21.2, 15.9;
HRMS (ES): Calcd for C27H42O5SNa 501.2651, found 501.2648.
C35H60O6SSiNa 659.3778, found 659.3760.
4.6.4. tert-Butyldimehylsilyl 8-(digitoxigenin-3-ylthio)
tetraethylene glycol (41)
4.7.3. 6-(Digitoxigenin-3-ylthio)triethylene glycol (11)
Compound 40 (19 mg, 0.030 mmol) was reacted with Et3N:3HF
To a solution of 18 (25 mg, 0.032 mmol) in absolute EtOH
(1 mL) was added NaBH4 (5 mg, 0.132 mmol). The mixture was
stirred at room temperature for 10 min, followed by addition of a
solution of 30 (34 mg, 0.088 mmol) in absolute EtOH (0.5 mL).
The reaction mixture was stirred at room temperature for 20 h be-
(40 lL, 0.245 mmol). The crude product was purified by column
chromatography (MeOH/CH2Cl2, 1:19) and afforded 11 as clear
oil (13 mg, 83%): Rf: 0.51 (MeOH/CH2Cl2 1:19); 1H NMR
(400 MHz, CDCl3): dH 5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a
1.2 Hz, Jgem 18.0 Hz, H21a), 4.80 (dd, 1H, J17,21b 0.8 Hz, Jgem

2416
M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
18.0 Hz, H21b), 3.74 (m, 2H, H29), 3.64 (m, 8H, H25, H26, H27,
H28), 3.25 (br s, 1H, H3), 2.78 (m, 1H, H17), 2.69 (t, 2H, J24,25
7.0 Hz, H24), 2.38 (br s, 1H, C29OH), 2.13 (m, 2H), 1.91–1.18 (m,
19H), 0.94 (s, 3H, CH3), 0.87 (s, 3H, CH3); 13C NMR (100 MHz,
CDCl3): dC 174.6 (C20, C23), 117.8 (C22), 85.7 (C14), 73.6 (C21),
72.6, 71.2, 70.5, 70.5 (C25, C26, C27, C28), 61.9 (C29), 51.1, 49.7
(C13, C17), 44.0, 42.1, 40.2, 37.2, 36.1, 35.8, 33.3, 31.8, 31.2, 30.9,
27.0, 26.9, 26.2, 23.8, 21.7, 21.2, 15.9; HRMS (ES): Calcd for
MgSO4, filtered and concentrated under reduced pressure. The
crude product was purified by column chromatography (EtOAc/
CH2Cl2 1:2) to afford 6 as white solid (24 mg, 27%): Rf: 0.24
(EtOAc/CH2Cl2 1:2); 1H NMR (400 MHz, CDCl3): dH 5.87 (s, 1H,
H22), 4.98 (dd, 1H, J17,21a 1.6 Hz, Jgem 18.0 Hz, H21a), 4.80 (dd,
1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.60 (t, 2H, J24,25 7.0 Hz,
H25), 3.25 (br s, 1H, H3), 2.77 (m, 1H, H17), 2.67 (t, 2H, J24,25
7.0 Hz, H24), 2.12 (m, 2H), 1.90–1.17 (m, 19H), 0.94 (s, 3H, CH3),
0.86 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 174.7 (C20,
C23), 117.8 (C22), 85.7 (C14), 73.6 (C21), 70.9 (C25), 51.0, 49.7
(C13, C17), 44.0, 42.0, 40.1, 37.2, 36.1, 35.7, 33.3, 31.8, 31.3, 30.9,
27.0, 26.9, 26.2, 23.8, 21.7, 21.2, 15.9; HRMS (ES): Calcd for
C29H46O6SNa 545.2913, found 545.2922.
4.7.4. 8-(Digitoxigenin-3-ylthio)tetra ethylene glycol (12)
Compound 41 (25 mg, 0.037 mmol) was treated with Et3N:3HF
(29
lL, 0.181 mmol). The crude product was purified by column
C50H74O7S2Na 873.4773, found 873.4814.
chromatography (MeOH/CH2Cl2 1:19) to afford 12 as clear oil
(20 mg, 95%): Rf: 0.46 (MeOH/CH2Cl2 1:19); 1H NMR (400 MHz,
CDCl3): dH 5.86 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.2 Hz, Jgem
18.2 Hz, H21a), 4.79 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.2 Hz, H21b),
3.72 (m, 2H, H31), 3.63 (m, 12H, H25, H26, H27, H28, H29, H30),
3.24 (br s, 1H, H3), 2.77 (m, 1H, H17), 2.68 (t, 2H, J24,25 7.2 Hz,
H24), 2.19–2.07 (m, 3H), 1.89–1.25 (m, 19H), 0.93 (s, 3H, CH3),
0.86 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 174.7 (C20,
C23), 117.8 (C22), 85.7 (C14), 73.6 (C21), 72.6, 71.2, 70.7, 70.8,
70.5, 70.4 (C25, C26, C27, C28, C29, C30), 61.9 (C31), 51.0, 49.7
(C13, C17), 43.9, 42.0, 40.2, 37.2, 36.1, 35.7, 33.3, 31.8, 31.1, 30.9,
29.8, 27.0, 26.8, 26.2, 23.8, 21.7, 21.2, 15.9; HRMS (ES): Calcd for
4.8.3. 1,6-(1,6-Dimercapto-1,6-dideoxy-triethylene
glycol)bis(digitoxigenin-3-ylsulfane) (7)
Disulfide 18 (18 mg, 0.023 mmol) was treated with NaBH4
(3 mg, 0.079 mmol) in absolute EtOH (1 mL) for 10 min before a
solution of 36 (12 mg, 0.032 mmol) in absolute EtOH (0.5 mL)
was added. The reaction mixture was stirred for 3 h at room tem-
perature before the crude product was purified by column chroma-
tography (EtOAc/pentane 1:2) to afford 7 as white solid (10 mg,
49%): Rf: 0.25 (pentane/EtOAc 1:2); 1H NMR (400 MHz, CDCl3):
dH 5.87 (s, 1H, H22), 4.98 (dd, 1H, J17,21a 1.2 Hz, Jgem 18.0 Hz,
H21a), 4.80 (dd, 1H, J17,21b 1.6 Hz, Jgem 18.0 Hz, H21b), 3.62 (m,
4H, H25, H26), 3.25 (br s, 1H, H3), 2.78 (m, 1H, H17), 2.69 (t, 2H,
J24,25 7.0 Hz, H24), 2.19 (m, 2H), 1.91–1.17 (m, 19H), 0.94 (s, 3H,
CH3), 0.87 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 174.5
(C20, C23), 117.9 (C22), 85.7 (C14), 73.6 (C21), 71.3, 70.5 (C25,
C26), 51.1, 49.7 (C13, C17), 43.9, 42.1, 40.2, 37.2, 36.2, 35.8, 33.3,
31.9, 31.2, 30.9, 27.0, 26.9, 26.2, 23.8, 21.7, 21.2, 15.9; HRMS
(ES): Calcd for C52H78O8S2Na 917.5036, found 917.5029.
C31H50O7SNa 566.3277, found 566.3281.
4.8. General procedure for the synthesis of bivalent steroids (5, 7
and 8)
To a solution of 18 (1.1 equiv) in absolute EtOH (2 mL) was
added NaBH4 (4 equiv). The mixture was stirred at room tempera-
ture for 10 min, followed by addition of dihalide linker (1 equiv).
The reaction mixture was stirred at room temperature for 1–17 h
and diluted with brine. The aqueous phase was extracted four
times with CH2Cl2 and the combined organic phases were dried
over MgSO4, filtered and concentrated under reduced pressure.
The crude product was purified by column chromatography.
4.8.4. 1,8-(1,8-Dimercapto-1,8-dideoxy-tetraethylene
glycol)bis(digitoxigenin-3-ylsulfane) (8)
Disulfide 18 (52 mg, 0.067 mmol) was treated with NaBH4
(17 mg, 0.449 mmol) in absolute EtOH (2 mL) for 10 min before a
solution of 37 (30 mg, 0.072 mmol) in absolute EtOH (1 mL) was
added. The reaction mixture was stirred for 17 h and the crude
product was purified by column chromatography (EtOAc/CH2Cl2
1:3) to afford 8 as white solid (39 mg, 58%): Rf: 0.26 (EtOAc/CH2Cl2
1:1); 1H NMR (400 MHz, CDCl3): dH 5.86 (s, 1H, H22), 4.98 (dd, 1H,
J17,21a 1.2 Hz, Jgem 18.0 Hz, H21a), 4.79 (dd, 1H, J17,21b 1.2 Hz, Jgem
18.0 Hz, H21b), 3.61 (m, 6H, H25, H26, H27), 3.23 (br s, 1H, H3),
2.77 (m, 1H, H17), 2.67 (t, 2H, J24,25 7.0 Hz, H24), 2.13 (m, 2H),
1.90–1.16 (m, 19H), 0.94 (s, 3H, CH3), 0.86 (s, 3H, CH3); 13C NMR
(100 MHz, CDCl3): dC 174.7, 174.7 (C20, C23), 117.8 (C22), 85.6
(C14), 73.6 (C21), 71.2, 70.7, 70.4 (C25, C26, C27), 51.0, 49.7
(C13, C17), 43.9, 41.9, 40.1, 37.1, 36.1, 35.7, 33.2, 31.8, 31.1, 30.9,
27.0, 26.8, 26.1, 23.8, 21.7, 21.2, 15.9; HRMS (ES): Calcd for
4.8.1. 1,2-Bis(digitoxigenin-3-ylthio)ethane (5)
Compound 18 (59 mg, 0.076 mmol) was treated with NaBH4
(11 mg, 0.29 mmol) in absolute EtOH (2 mL). 1,2-Dibromoethane
(34) (6 lL, 0.070 mmol) was added after 10 min and the reaction
mixture was stirred for 16 h. The crude product was purified by
column chromatography (EtOAc/CH2Cl2 1:2) to afford 5 as white
solid (26 mg, 46%): Rf: 0.30 (EtOAc/CH2Cl2 1:2); 1H NMR
(400 MHz, CDCl3): dH 5.86 (s, 1H, H22), 4.98 (dd, 1H, J17,21a
1.2 Hz, Jgem 18.0 Hz, H21a), 4.79 (dd, 1H, J17,21b 1.6 Hz, Jgem
18.0 Hz, H21b), 3.24 (br s, 1H, H3), 2.77 (m, 1H, H17), 2.67 (m,
2H, H24), 2.13 (m, 2H), 1.90–1.18 (m, 19H), 0.93 (s, 3H, CH3),
0.86 (s, 3H, CH3); 13C NMR (100 MHz, CDCl3): dC 174.7, 174.6
(C20, C23), 117.8 (C22), 85.7 (C14), 73.6 (C21), 51.0, 49.7 (C13,
C17), 43.8, 42.0, 40.1, 37.2, 36.1, 35.7, 33.3, 32.1, 31.8, 30.9, 27.0,
26.8, 26.2, 23.8, 21.7, 21.2, 15.9; HRMS (ES): Calcd for C48H70O6S2-
Na 829.4512, found 829.4517.
C54H82O9S2Na 961.5298, found 961.5291.
Acknowledgements
We are thankful to B. Bjerring Jensen for excellent technical
assistance. We also gratefully appreciate financiel support from
The Lundbeck Foundation, The Carlsberg Research Foundation,
The Danish Agency for Science, Technology and Innovation; and
the OChem Graduate School for MJ and HHJ.
4.8.2. 1,4-(1,4-Dimercapto-1,4-dideoxy-diethylene
glycol)bis(digitoxigenin-3-ylsulfane) (6)
Disulfide 18 (82 mg, 0.105 mmol) was treated with NaBH4
(15 mg, 0.39 mmol) in dry DMF (1 mL) for 10 min before a solution
of 35 (34 mg, 0.104 mmol) in dry DMF (1 mL) was added. The reac-
tion mixture was stirred at room temperature for 1 h and then di-
luted with mildly acidic H2O. The aqueous phase was extracted
four times with CH2Cl2 before the combined organic phases were
concentrated in vacuo. The residue was dissolved in EtOAc and
washed five times with H2O. The organic phase was dried over
References and notes
1. Schatzmann, H. J. Helv. Physiol. Pharm. A. 1953, 11, 346.
2. Hauptman, P. J.; Kelly, R. A. Circulation 1999, 99, 1265.
3. Gheorghiade, M.; Adams, K. F.; Colucci, W. S. Circulation 2004, 109, 2959.
4. (a) Langer, G. A. J. Mol. Cell. Cardiol. 1970, 1, 203; (b) Erdmann, E.; Philipp, G.;
Scholz, H. Biochem. Pharmacol. 1980, 29, 3219.

M. Jensen et al. / Bioorg. Med. Chem. 19 (2011) 2407–2417
2417
5. (a) Bagrov, A. Y.; Shapiro, J. I.; Fedorova, O. V. Pharmacol. Rev. 2009, 61, 9; (b)
Newman, R. A.; Yang, P.; Pawlus, A. D.; Block, K. I. Mol. Interventions 2008, 8,
36.
16. (a) Graves, P. E.; Fenster, P. E.; Macfarland, R. T.; Marcus, F. I.; Perrier, D. Clin.
Pharmacol. Ther. 1984, 36, 601; (b) Belz, G. G.; Breithaupt-Grögler, K.; Osowski,
U. Eur. J. Clin. Invest. 2001, 31, 10.
6. Piccioni, F.; Roman, B. R.; Fischbeck, K. H.; Taylor, J. P. Hum. Mol. Genet. 2004, 13,
437.
7. Srivastava, M.; Eidelman, O.; Zhang, J.; Paweletz, C.; Caohuy, H.; Yang, Q. F.;
Jacobson, K. A.; Heldman, E.; Huang, W.; Jozwik, C.; Pollard, B. S.; Pollard, H. B.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 7693.
8. Stenkvist, B.; Bengtsson, E.; Eriksson, O.; Holmquist, J.; Nordin, B.; Westman-
Naeser, S.; Eklund, G. Lancet 1979, 313, 563.
9. Stenkvist, B. Oncol. Rep. 1999, 6, 493.
10. (a) Haux, J. Med. Hypotheses 1999, 53, 543; (b) Haux, J.; Klepp, O.; Spigset, O.;
Tretli, S. BMC Cancer 2001, 1; (c) Mijatovic, T.; Van Quaquebeke, E.; Delest, B.;
Debeir, O.; Darro, F.; Kiss, R. Biochim. Biophys. Acta, Rev. Cancer 2007, 1776, 32.
11. Rathore, H.; From, A. H. L.; Ahmed, K.; Fullerton, D. S. J. Med. Chem. 1986, 29,
1945.
12. Paula, S.; Tabet, M. R.; Ball, W. J. Biochemistry 2005, 44, 498.
13. Lopez-Lazaro, M.; Pastor, N.; Azrak, S. S.; Ayuso, M. J.; Austin, C. A.; Cortes, F. J.
Nat. Prod. 2005, 68, 1642.
17. Iyer, A. K. V.; Zhou, M.; Azad, N.; Elbaz, H.; Wang, L.; Rogalsky, D. K.;
Rojanasakul, Y.; O’Doherty, G. A.; Langenhan, J. M. ACS Med. Chem. Lett. 2010, 1,
326.
18. Silva, D. J.; Kraml, C. M.; Kahne, D. Bioorg. Med. Chem. 1994, 2, 1251.
19. Harris, J. M.; Martin, N. E.; Modi, M. Clin. Pharmacokinet. 2001, 40, 539.
20. Dwek, R. A. Chem. Rev. 1996, 96, 683.
21. MacroModel; version 9.7, S. L., New York, NY, 2009.
22. Sawlewic, L.; Weiss, E.; Linde, H. H. A.; Meyer, K. Helv. Chim. Acta 1972, 55,
2452.
23. Gobbini, M.; Benicchio, A.; Padoani, G.; Torri, M.; Melloni, P. Bioorg. Med. Chem.
Lett. 1997, 7, 469.
24. Staroske, T.; Hennig, L.; Welzel, P.; Hofmann, H. J.; Muller, D.; Hausler, T.;
Sheldrick, W. S.; Zillikens, S.; Gretzer, B.; Pusch, H.; Glitsch, H. G. Tetrahedron
1996, 52, 12723.
25. Ishizone, T.; Han, S.; Okuyama, S.; Nakahama, S. Macromolecules 2002, 36, 42.
26. Pospíšil, J.; Markó, I. E. Org. Lett. 2006, 8, 5983.
14. (a) Langenhan, J. M.; Peters, N. R.; Guzei, I. A.; Hoffmann, F. M.; Thorson, J. S.
Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 12305; (b) Langenhan, J. M.; Engle, J. M.;
Slevin, L. K.; Fay, L. R.; Lucker, R. W.; Smith, K. R.; Endo, M. M. Bioorg. Med.
Chem. Lett. 2008, 18, 670.
15. (a) Namchuk, M. N.; McCarter, J. D.; Becalski, A.; Andrews, T.; Withers, S. G. J.
Am. Chem. Soc. 2000, 122, 1270; (b) Bols, M.; Liang, X.; Jensen, H. H. J. Org. Chem.
2002, 67, 8970; (c) Jensen, H. H.; Lyngbye, L.; Bols, M. Angew. Chem., Int. Ed.
2001, 40, 3447.
27. Provencher-Mandeville, J.; Descôteaux, C.; Mandal, S. K.; Leblanc, V.; Asselin,
É.; Bérubé, G. Bioorg. Med. Chem. Lett. 2008, 18, 2282.
28. Lukas, D. S.; Demartin, Ag. J. Clin. Invest. 1969, 48, 1041.
29. Frank, J. S.; Philipson, K. D.; Beydler, S. Circ. Res. 1984, 54, 414.
30. Klodos, I.; Esmann, M.; Post, R. L. Kidney Int. 2002, 62, 2097.
31. Drakenberg, T.; Brodelius, P.; McIntyre, D. D.; Vogel, H. J. Can. J. Chem. 1990, 68,
272.
32. Bertozzi, C. R.; Bednarski, M. D. J. Org. Chem. 1991, 56, 4326.