Novel analogues of istaroxime, a potent inhibitor of Na+,K +-ATPase: Synthesis and structure-activity relationship

Article abstract of DOI:10.1021/jm800257s

We report the synthesis and biological properties of novel inhibitors of the Na⁺,K⁺-ATPase as positive inotropic compounds. Following our previously described model from which Istaroxime was generated, the 5α,14α-androstane skeleton was used as a scaffold to study the space around the basic chain of our lead compound. Some compounds demonstrated higher potencies than Istaroxime on the receptor and the (E)-3-[(R)-3- pyrrolidinyl]oxime derivative, 15, was the most potent; as further confirmation of our model, the E isomers of the oxime are more potent than the Z form. The compounds tested in the guinea pig model induced positive inotropic effects, which are correlated to the in vitro inhibitory potency on the Na ⁺,K⁺-ATPase. The finding that all tested compounds resulted less proarrhythmogenic than digoxin, a currently clinically used positive inotropic agent, suggests that this could be a feature of the 3-aminoalkyloxime derivative class of 5α,14α-androstane.

Full text of DOI:10.1021/jm800257s

J. Med. Chem. 2008, 51, 4601–4608
4601
Novel Analogues of Istaroxime, a Potent Inhibitor of Na+,K+-ATPase: Synthesis and
Structure-Activity Relationship†
Mauro Gobbini,*,‡ Silvia Armaroli,‡,| Leonardo Banfi,‡ Alessandra Benicchio,‡ Giulio Carzana,‡ Giorgio Fedrizzi,‡
Patrizia Ferrari,§ Giuseppe Giacalone,§ Michele Giubileo,‡ Giuseppe Marazzi,‡ Rosella Micheletti,§ Barbara Moro,§
Marco Pozzi,‡ Piero Enrico Scotti,‡ Marco Torri,‡ and Alberto Cerri‡
Departments of Medicinal Chemistry and CardioVascular Pharmacology, Prassis Istituto di Ricerche Sigma-Tau SpA, Via Forlanini 3,
20019 Settimo Milanese (MI), Italy
ReceiVed March 10, 2008
We report the synthesis and biological properties of novel inhibitors of the Na+,K+-ATPase as positive
inotropic compounds. Following our previously described model from which Istaroxime was generated, the
5R,14R-androstane skeleton was used as a scaffold to study the space around the basic chain of our lead
compound. Some compounds demonstrated higher potencies than Istaroxime on the receptor and the (E)-
3-[(R)-3-pyrrolidinyl]oxime derivative, 15, was the most potent; as further confirmation of our model, the
E isomers of the oxime are more potent than the Z form. The compounds tested in the guinea pig model
induced positive inotropic effects, which are correlated to the in vitro inhibitory potency on the Na+,K+-
ATPase. The finding that all tested compounds resulted less proarrhythmogenic than digoxin, a currently
clinically used positive inotropic agent, suggests that this could be a feature of the 3-aminoalkyloxime
derivative class of 5R,14R-androstane.
Chart 1. Structures of Digoxin and Istaroxime
Introduction
The digitalis cardiac glycoside digoxin (Chart 1) is currently
one of the most prescribed drugs for the treatment of congestive
heart failure (CHF)a due to its capability to increase the
contractile force of the cardiac muscle, known as positive
inotropic effect. The action of digitalis compounds, which
improve the pumping performance of the heart, is mainly due
to inhibition of the Na+,K+-ATPase. This is a membrane protein
that promotes the outward transport of Na+ and the inward
transport of K+ against their concentration gradients. Cardiac
glycosides reversibly bind to the extracellular side of the
Na+,K+-ATPase, thus blocking ATP hydrolysis and ion trans-
port. When the pump is inhibited, Na+ concentration inside the
cell is increased and, as a consequence, Ca2+ is introduced by
an exchange with Na+, through the Na+,Ca2+ exchanger. The
final result is the availability of higher Ca2+ concentrations
needed to activate contraction of the myocardium.1 Digoxin can
alleviate symptoms of CHF, improve exercise tolerance, and
reduce hospitalization, while having a neutral effect on mortal-
ity.2 A risk related to therapy with digitalis drugs is their
proarrhythmogenicity; evidence of digitalis toxicity emerges at
a two to three times higher serum concentration than the
therapeutic, and disturbances of conduction and cardiac arrhyth-
mia are characteristics of digitalis toxicity.3
The search for safer agents prompted a great deal of work.4
Recently, our group reported a model derived from superposition
of cassaine and digitoxigenin5 and the design of a new series
of compounds from a planar steroidal skeleton6 instead of the
bent skeleton typical of digitalis compounds (Figure 1). The
alkaloid cassaine, which was first isolated from the bark of
Erythrophleum guineense in 1935,7 shares many of the phar-
macologic actions of the cardiac glycosides but lacks the
structural characteristic typical of cardiac glycosides. By
introducing an O-(ω-aminoalkyl)oxime chain in position 3 and
oxo or hydroxy groups at positions 6 and 17 of the androstane
skeleton, we obtained compounds with inhibitory potency on
the Na+,K+-ATPase comparable to that of digoxin and positive
inotropic activity. Most importantly, some of those compounds
displayed a higher safety index than digoxin. From this effort,
Istaroxime (referred to as PST 2744; Chart 1), now in phase II
†
This paper is dedicated to the memory of Michele Giubileo
* To whom correspondence should be addressed. Phone: ++39.023357911.
Fax: ++39.0233500408. E-mail: mauro.gobbini@prassis.it. Address: De-
partment of Medicinal Chemistry, Prassis Istituto di Ricerche Sigma-Tau
SpA, Via Forlanini 3, 20019 Settimo Milanese (MI), Italy.
‡
Department of Medicinal Chemistry, Prassis Istituto di Ricerche Sigma-
Tau.
§
Department of Cardiovascular Pharmacology, Prassis Istituto di
Ricerche Sigma-Tau.
|
Current affiliation: Sigma-Tau Research Switzerland, P.O. Box 1823,
Via Motta 2/a, CH-6850 Mendrisio, Switzerland.
a Abbreviations: ATP, adenosine triphosphate; ATPase, adenosine triph-
osphatase; CHF, congestive heart failure; DIPEA, N,N-diisopropylethy-
lamine; ED80, effective dose (80%); Fmoc, 9-fluorenylmethoxycarbonyl;
IC50, inhibitory concentration (50%); rms, root-mean-square; SAR,
structure-activity relationship; SERCA2a sarco/endoplasmic reticulum
calcium ATPase isoform 2a; TBAF, tetra-n-butylammonium fluoride; TEA,
triethylamine; THF, tetrahydrofuran.
10.1021/jm800257s CCC: $40.75
2008 American Chemical Society
Published on Web 07/19/2008

4602 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15
Gobbini et al.
drochlorides (see Supporting Information for the syntheses of
the latter compounds) in a THF/water solution at room tem-
perature (Scheme 1, see Experimental Section). The very good
regioselectivity of this reaction was confirmed6 because the
oxime at the 3 position was obtained with only traces of dioxime
derivatives even though in the presence of two or three keto
groups. Almost all the oximes obtained in the coupling reactions
were mixtures of E and Z isomers and, in most cases, not
resolved in the two components when their separation proved
to be impossible by crystallization or flash chromatography, as
reported in our previous papers. The required starting ketones
of formula 1a-c (Scheme 1) were prepared as previously
reported.6
Figure 1. A 3D model of digitoxigenin, a typical digitalis compound,
and compound 2, the E isomer of Istaroxime, showing the different
shapes between a bent and planar steroidal skeleton.
The separation of the E and Z diastereoisomers of Istaroxime
was carried out (Scheme 2) by protection of the amino group
as Fmoc derivative. Istaroxime was reacted with FmocCl in the
presence of TEA in CH2Cl2 to give, after separation by flash
chromatography, 27 (E isomer, 40% yield) and 28 (Z isomer,
40% yield). Each separated isomer was deprotected with TBAF
in THF to give the E (2, 80% yield) and Z (3, 83% yield)
isomers as pure compounds.
Only compound 14 was separated by crystallization in the
two E and Z isomers, 15 and 16, respectively, while compounds
18, 21, and 22 were isolated as E isomers only. All remaining
oximes were used in the pharmacological tests as mixtures.
Compound 8 was prepared from Istaroxime by reaction with
3,5-dimethyl-1-pyrazolylformamidinium nitrate (Scheme 3).
Results and Discussion
All compounds were tested in vitro for their inhibitory activity
on purified dog kidney Na+,K+-ATPase, as measured by the
32P-ATP hydrolysis method (see data in Table 1).9,10 Some
compounds, showing high inhibitory potency in vitro, were
investigated in vivo for their inotropic activity and lethal effect
by slow intravenous infusion in anesthetized guinea pigs (results
reported in Table 2). Digoxin was chosen as reference compound
because it is the most commonly prescribed cardiac glycoside
in the treatment of CHF, while Istaroxime is the lead compound
in this series of androstane derivatives.
Figure 2. Model of the superposition between cassaine (orange) and
2, the E isomer of Istaroxime (green).
Scheme 1a
The following Structure-Activity Relationships (SAR) are
based on the in vitro data shown in Table 1.
The Diastereoisomers of Istaroxime. One of the most
interesting features in our previously reported model6 focused
on cassaine was that the E, as opposed to the Z, isomers of the
oxime were predicted to be more active; some examples
supporting this prediction were already reported in that paper,
but the separated isomers of Istaroxime were not disclosed. E
isomer 2 was now found to be 11 times more potent on the
Na+,K+-ATPase than Z isomer 3, thus further confirming the
prediction and the model.
Modified Chains with Primary Amines. The introduction
of a methyl group in R position to the amine, both in the S (4)
and R (5) configuration, gave compounds with equal activity
and a three times lower potency than Istaroxime. Two methyl
groups in this position of the chain (6) reduced potency 260
times, when compared to our lead compound, and more than
70 times when compared with mono-methyl derivatives. Sur-
prisingly, two methyl groups in ꢀ position to the amino group
gave compound 7, which was five times more potent than
Istaroxime. The elongation of an alkylenic spacer between the
oximic group and the basic head gave compound 11, which
was 16 times less potent than Istaroxime. Taking the latter result
into account, as well as that of the previously reported compound
with a (CH2)3 alkylenic chain,6 about four times less active than
a Reagents and conditions. a: hydroxylamines, THF/water, room tem-
perature. 1a 6-oxo. 1b 6R-OH: OH directed downwards. 1c 6ꢀ-OH: OH
directed upwards.
clinical trials,8 emerged as a very promising positive inotropic
compound that may represent an innovative alternative in the
treatment of CHF. The rotated E isomer of the most potent
compound of that series (2) can be well superposed on cassaine,
thus confirming our model predicting that the optimal superposi-
tion between the basic chains of cassaine and our compounds
was expected to be reached with the E isomers of the oxime
group, as Figure 2 shows.
In the present paper, we report novel Istaroxime analogues
having modified oximic chains, with the aim of studying the
space around the basic group to obtain compounds with a higher
potency and better ratio between the active and toxic doses than
the parent compound.
Chemistry
The oximes listed in Table 1 (with the exception of compound
8), were synthesized from the corresponding ketones of formula
1a-c and the appropriate O-substituted hydroxylamine dihy-

NoVel Analogues of Istaroxime
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15 4603
Table 1. Structure and Na+,K+-ATPase Inhibition for Compounds 2-26
a Concentrations able to inhibit 50% of Na+,K+-ATPase enzyme activity; mean of two or three experiments.
Istaroxime, the length of the Istaroxime chain can be defined
as the optimal space available for the basic chain.
used to achieve more potent compounds. As a first instance,
the aminoalkyl chain of Istaroxime was “closed” to give a
3-pyrrolidinyl derivative. Among the numerous possibilities, this
cyclic amine originates from the conformational analysis of the
E isomer of Istaroxime (2):11 nine conformers were obtained
within 12.55 kJ mol-1 (3 kcal mol-1), a range of energy
compatible with the binding to a receptor (Figure 3a). As one
of the conformers is extended, eight of nine are obviously folded,
and specific/respective conformers superimposed almost per-
fectly on the specific conformer of the pyrrolidinyl oxime
derivative (for both the R and S diastereoisomers, in the E form,
there were five conformers); we considered the best superim-
position between the conformers of 2 and those of the two
pyrrolidinyl oxime derivatives on the basis of the rms. There
Secondary Amines on Straight Chains. While the dim-
ethylamine derivative of Istaroxime was already described and
found to be almost devoid of inhibitory activity on the Na+,K+-
ATPase,6 we here report some secondary amines on straight
chains. The monomethylamine derivative of Istaroxime, 9, had
a reduced potency of about six times; however, its homologue,
10, had a higher activity, if compared with 9, but still lower
than Istaroxime’s. A bulkier and more basic group, such as
guanidinyl in compound 8, caused an impressive drop in activity
of about 100 times.
Cyclic Amines. The reduction in the degree of freedom of a
molecule is a well-known method in medicinal chemistry, often

4604 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15
Gobbini et al.
Scheme 2a
a Reagents and conditions. a: FmocCl, TEA, CH2Cl2, room temperature. b: flash chromatography. c: TBAF, THF, room temperature.
Scheme 3a
In previous papers, we demonstrated that tertiary amines
reduced activity by about 2 orders of magnitude, in comparison
to the parent primary amines;6,12,13 the N-methylpyrrolidinyl
analogue 18 was about 80 times less potent than the N-
unsubstituted parent compound, 15, in agreement with what was
already reported. Also, the N-methylation of the S-pyrrolidinyl
derivative, 13, giving compound 17, caused a potency reduction
of about 170 times. These cyclic tertiary amines had higher
potencies than the dimethylamine derivative of Istaroxime
(>100 µM); again, the cyclic nature of 17 and 18 had a
beneficial effect on interaction with the receptor.
a Reagents and conditions. a: 3,5-dimethyl-1-pyrazolylformamidinium
nitrate, DIPEA, EtOH, reflux.
The 2-pyrrolidinylmethyl substituent maintains the same
distance between the amino group and the oxygen of the oximic
group: in both the S and R derivatives potencies were consider-
ably reduced when compared to the 3-pyrrolidinyl analogues
(19 vs 13 and 20 vs 14) and Istaroxime.
Restriction of the cycle to four terms, maintaining the NH at
the same distance relative to the oxygen of the oximic group,
gave compound 12, with potency similar to that of Istaroxime.
The enlargement of the cycle to the 3-piperidinyl derivative 21
(diastereoisomeric mixture), again maintaining the NH at the
same distance relative to the oxygen of the oximic group,
reduced potency; activity was also reduced with 4-piperidinyl
22.
Table 2. Inotropic and Toxic Effects in Anesthetized Guinea Pig
a
b
c
Emax % increase EDmax
ED80
deads/
lethal
compound
in dP/dtmax
µmol/kg µmol/kg treated dose/ED80
digoxin
Istaroxime
2
3
14
15
16
23
25
26
127
182
260
235
223
186
114
77
0.97
5.74
6.79
12.1
8.09
2.65
43.5
8.95
9.62
0.41
1.82
1.03
3.36
2.01
0.75
10/10
7/8
8/8
4/5
1/3
3/3
0/3
2/2
4/5
2/2
3.2
25.6
20.8
22.5
12.0
8.0
27.7
>3.6
<17.0
21.0
49.1
216
256
1.64
3.92
44.8
a Maximal increase in dP/dtmax b Dose inducing maximum positive
.
inotropic effect. c Inotropic potency: dose increasing +dP/dtmax by 80%,
calculated from dose-response curves.
6-Hydroxy Derivatives. Consistent with our prediction,6 the
reduction of the keto group in position 6 led to diminished
potency on the Na+,K+-ATPase, with the ꢀ configuration in
this position more detrimental than the R stereochemistry; this
fact is confirmed in the present paper by the comparisons
between 23 and 10, 24 and 13, 25 and 14, and between 26 and
both 25 and 14.
was no way to choose a priori which of the two diastereoiso-
meric pyrrolidinyl derivatives would have higher inhibitory
potency, as differently folded conformers of Istaroxime super-
imposed equally well on R- or S-pyrrolidinyl derivatives (Figure
3b with the E R-pyrrolidinyl derivative 15; Figure 3c with the
E isomer of the S-pyrrolidinyl derivative 13). The two different
diastereoisomers were synthesized as E,Z mixtures, and the R
isomer, 14, proved nine times more potent than the S isomer,
13. The pure E isomer of 14, i.e., 15, was 15 times more potent
than the Z isomer, 16, further supporting our model. The E
isomer, 15, was found to be three times more potent than the E
isomer of Istaroxime, (2) and the Z isomer, 16, was twice as
potent as the Z isomer of Istaroxime (3). As shown above,
secondary amines were expected to give reduced potency in
comparison with primary amines; in this case, the reduction in
potency caused by this modification was counteracted by the
“closure” of the chain to give a cyclic, secondary amine.
In Vivo Activity in Guinea Pigs. Some compounds showing
high inhibitory potency on the Na+,K+-ATPase were investi-
gated in vivo in the anesthetized guinea pig (Table 2). In vitro
and in vivo potencies (ED80) were correlated (r2 ) 0.72, n )
8, with the exclusion of compound 16, which is an outlier, and
compound 23, which did not reach ED80), thus confirming the
predictivity of the in vitro test. Within the homogeneous series
of the androstane derivatives, which excludes digoxin, the
correlation is higher (r2 ) 0.87, n ) 7).
Even though none of the tested compounds showed inotropic
potencies (ED80) comparable to digoxin, all displayed, most
importantly, safety ratios higher than digoxin, which could be
an intrinsic property of this class of cardiac androstane deriva-

NoVel Analogues of Istaroxime
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15 4605
Figure 3. (a) A 3D model of conformers of compound 2. (b) Superposition of compounds 2 (red) and 15 (green); the basic nitrogen atom of the
chains are colored in cyan. (c) Superposition of compound 2 (red) and the E isomer of compound 13 (green); the basic nitrogen atom of the chains
are colored in cyan.
After 1.5 h, NaCl (7.5 equiv) was added and the mixture stirred
for 10 min. The phases were separated and the aqueous phase
extracted with THF (2×). The combined organic extracts were dried
over Na2SO4, filtered, and evaporated.
tives. The reported ability of low digoxin concentrations to
stimulate Ca2+ release from the sarcoplasmic reticulum14 with
high efficacy may partly account for its proarrhythmic properties.
On the other hand, we showed that Istaroxime increases
SERCA2a (Sarco/Endoplasmic Reticulum Calcium ATPase
Isoform 2a) affinity for Ca2+, enhancing Ca2+ reuptake.15 The
contribution of these mechanisms to in vivo behavior (inotropic
potency, safety ratio) is not known. Investigation of SERCA2a
interaction of Istaroxime’s analogues was beyond the scope of
the present work.
(E)-3-(2-Aminoethoxyimino)androstane-6,17-dione fumarate (2).
To a stirred solution of 27 (2.00 g, 3.43 mmol) in dry THF (9.6
mL) at 0 °C, 1 M tetrabutylammonium fluoride in THF (4.1 mL,
4.1 mmol) was added and the mixture stirred at room temperature
for 3 h. The mixture was loaded onto a flash chromatography
column (SiO2, CH2Cl2/MeOH/26% NH4OH 90/10/1). The fractions
containing 2 as a base were combined and concentrated under
vacuum at 30 °C. MeOH was added and evaporated again to
eliminate ammonia. Fumaric acid (0.40 g, 3.44 mmol) in MeOH
(10 mL) was added. After addition of EtOAc (20 mL) and Et2O
(150 mL), the precipitate was filtered to give the title compound 2
Conclusions
The compounds reported in this paper support interest in the
androstane skeleton, with a proper aminoalkyloxime in position
3 as a pharmacophore for positive inotropic activity, with the
likely advantage of higher safety, as compared to classic digitalis
compounds. The importance of the amino group at a proper
distance in the oxime chain in position 3 is demonstrated and
the higher potency of the E, as opposed to the Z, isomers is
confirmed; further, some of the oximic chain with a cyclic
amine, here introduced for the first time, have been found to
confer a higher potency than the oximic chain with a linear
amine. All these confirmations or achievements encourage
additional exploitation of the androstane skeleton in the search
for new, positive inotropic agents, possibly safer than digoxin.
1
(1.30 g, 80%). H NMR (300 MHz, DMSO-d6, ppm from TMS):
δ 8.1 (bb, 4H, NH3+, COOH), 6.45 (s, 2H, CHdCH), (m, 2H,
CH2-O), 3.08 (m, 1H, H-2eq), 2.98 (m, 2H, CH2-N), 0.79 (s,
3H, CH3), 0.78 (s, 3H, CH3); white solid. Anal. (C21H32N2O3 ·
C4H4O4) C, H, N.
(Z)-3-(2-Aminoethoxyimino)androstane-6,17-dione fumarate (3).
The title compound 3 was prepared in 83% yield from 28, following
1
the procedure described above for the preparation of 2. H NMR
+
(300 MHz, DMSO-d6, ppm from TMS): δ 7.95 (bb, 4H, NH3
,
COOH), 6.45 (s, 2H, CHdCH), 4.04 (m, 2H, CH2-O), 3.02 (m,
1H, H-4eq), 2.96 (m, 2H CH2-N), 0.79 (s, 3H, CH3), 0.77(s, 3H,
CH3); white solid. Anal. (C21H32N2O3 ·C4H4O4) C, H, N.
(E,Z)-3-[(S)-2-Amino-1-propoxyimino]androstane-6,17-dione hy-
drochloride (4). Prepared in 81% yield from 1a and (S)-2-amino-
1-propoxyamine dihydrochloride (Supporting Information, I-a). The
title compound 4 was obtained from the crude after washing the
residue with EtOAc and filtering. 1H NMR (300 MHz, DMSO-d6,
ppm from TMS): δ 8.00 (bb, 3H, NH3+), 3.97 (m, 2H, CH2-O),
3.40 (m, 1H, CH–N), 3.11 (m, 0.5H, H-2eq E isomer), 3.05 (m,
0.5H, H-4eq Z isomer), 1.16 (d, 3H, CH3), 0.79 (s, 3H, CH3), 0.78
Experimental Section
General Details. General experimental details regarding chem-
istry and pharmacology are reported elsewhere.12
Chemistry. General Procedure for the Reaction of Scheme
1. To a stirred solution of the appropriate ketone (1 equiv) in THF
(0.15 M), a solution of the appropriate hydroxylamines dihydro-
chlorides (1 equiv) in H2O (0.30 M) was rapidly added dropwise.

4606 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15
Gobbini et al.
(s, 1.5H, CH3), 0.77 (s, 1.5H, CH3); white solid. Anal. (C22H34N2O3 ·
(E,Z)-3-(4-Aminobutoxyimino)androstane-6,17-dione hydro-
chloride (11). Prepared in 64% yield from 1a and 4-aminobu-
toxyamine dihydrochloride.12 The title compound 11 was obtained
from the crude after washing the residue with EtOAc and filtering.
1H NMR (300 MHz, DMSO-d6, ppm from TMS): δ 7.74 (bb, 3H,
NH3+), 3.92 (m, 2H, CH2-O), 2.99 (m, 0.5H, H-2eq E isomer),
2.94 (m, 0.5H, H-4eq Z isomer), 2.72 (m, 2H, CH2-N), 0.79 (s,
3H, CH3), 0.78 (s, 1.5H, CH3), 0.77 (s, 1.5H, CH3); white solid.
Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-(3-Azetidinyloxyimino)androstane-6,17-dione fumarate
(12). Prepared in 80% yield from 1a and 3-azetidinyloxyamine
dihydrochloride (Supporting Information, I-g). The crude product
was purified by flash chromatography (SiO2, CH2Cl2:MeOH:NH3
9:1:0.1). To the concentrated fractions, fumaric acid in EtOAc was
added. After dilution with Et2O, the solid was collected by filtration
to give the title compound 12. 1H NMR (300 MHz, DMSO-d6,
ppm from TMS): δ 8.00 (bb, 4Η, NH3+, COOH), 6.45 (s, 2H,
CHdCH), 4.88 (m, 1H, CH-O), 4.13 (m, 2H, CH2-N), 3.87
(m,2H, CH2-N), 3.05 (m, 0.5H, H-2eq E isomer), 2.99 (m, 0.5H,
H-4eq Z isomer), 0.79 (s, 3H, CH3), 0.79 (s, 3H, CH3); white solid.
Anal. (C22H32N2O3 ·C4H4O4) C, H, N.
(E,Z)-3-[3-(S)-Pyrrolidinyloxyimino]androstane-6,17-dione hy-
drochloride (13). Prepared in 78% yield from 1a and 3-(S)-
pyrrolidinyloxyamine dihydrochloride (Supporting Information,
I-h). The title compound 13 was obtained from the crude, after
washing the residue with EtOAc, Et2O, and filtering. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 9.23 (bb, 2H, NH2+), 4.74
(m, 1H, CH–O), 3.15 (m, 4H, CH2-N–CH2), 3.01 (m, 0.5H, H-2eq
E isomer), 2.96 (m, 0.5H, H-4eq Z isomer), 0.79 (s, 3H, CH3),
0.78 (s, 3H, CH3); white solid. Anal. (C23H34N2O3 ·HCl) C, H, N,
Cl.
(E,Z)-3-[3-(R)-Pyrrolidinyloxyimino]androstane-6,17-dione hy-
drochloride (14). Prepared in 72% yield from 1a and 3-(R)-
pyrrolidinyloxyamine dihydrochloride (Supporting Information, I-i).
The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 14. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 9.20 (bb, 2H, NH2+), 4.75
(m, 1H, CH–O), 3.18 (m, 4H, CH2-N–CH2), 3.03 (m, 0.5H, H-2eq
E isomer), 2.96 (m, 0.5H, H-4eq Z isomer), 0.79 (s, 3H, CH3),
0.78 (s, 3H, CH3); white solid. Anal. (C23H34N2O3 ·HCl) C, H, N,
Cl.
(E)-3-[3-(R)-Pyrrolidinyl]oxyiminoandrostane-6,17-dione hy-
drochloride (15). (E,Z) 3-[3-(R)-pyrrolidinyloxyimino]androstane-
6,17-dione hydrochloride (14, 0.65 g) was suspended in EtOAc
(150 mL) and stirred for 3 h. After filtration, the procedure was
repeated on the solid to give the title compound 15 (0.30 g, 46%).
1H NMR (300 MHz, DMSO-d6, ppm from TMS): δ 9.26 (bb, 2H,
NH2+), 4.74 (m, 1H, CH–O), 3.18 (m, 4H, CH2-N–CH2) 3.03 (m,
1H, H-2eq), 0.79 (s, 3H, CH3), 0.78 (s, 3H, CH3); white solid. Anal.
(C23H34N2O3 ·HCl) C, H, N, Cl.
(Z)-3-[3-(R)-Pyrrolidinyl]oxyiminoandrostane-6,17-dione hydro-
chloride (16). The mother liquor of the first filtration reported for
the preparation of compound 15 was evaporated to dryness. The
residue was dissolved in EtOH, filtered on charcoal, and the filtrate
concentrated to small volume; the solid was collected by filtration
to give the title compound 16 (0.25 g, 38%). 1H NMR (300 MHz,
DMSO-d6, ppm from TMS): δ 9.22 (bb, 2H, NH2+), 4.75 (m, 1H,
CH–O), 3.18 (m, 4H, CH2-N–CH2), 2.96 (m, 1H, H-4eq), 0.79
(s, 3H, CH3), 0.78 (s, 3H, CH3); white solid. Anal. (C23H34N2O3 ·
HCl) C, H, N, Cl.
(E,Z)-3-[3-(S)-(1-Methyl)pyrrolidinyloxyimino]androstane-6,17-
dione hydrochloride (17). Prepared in 45% yield from 1a and 3-(S)-
(1-methyl)pyrrolidinyloxyamine dihydrochloride (Supporting In-
formation,I-j).Thecrudeproductwaspurifiedbyflashchromatography
(SiO2, CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions,
5 M HCl in EtOAc was added. After dilution with Et2O, the solid
was collected by filtration to give the title compound 17. 1H NMR
(300 MHz, DMSO-d6, ppm from TMS): δ 10.58 (bb, 1H, NH+),
4.76 (m, 1H, CH–O), 3.30 (m, 4H, CH2-N–CH2), 3.05 (m, 0.5H,
HCl) C, H, N, Cl.
(E,Z)-3-[(R)-2-Amino-1-propoxyimino]androstane-6,17-dione fu-
marate (5). Prepared in 46% yield from 1a and (R)-2-amino-1-
propoxyamine dihydrochloride (Supporting Information, I-b). The
crude product was purified by flash chromatography (SiO2, CH2Cl2:
MeOH:NH3 9:1:0.1). To the concentrated fractions fumaric acid
in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 5. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 8.1 (bb, 4H, NH3+, COOH),
6.45(s, 2H, CHdCH), 3.95 (m, 2H, CH2-O), 3.38 (m, 1H, CH–N),
3.11 (m, 0.5H, H-2eq E isomer), 3.03 (m, 0.5H, H-4eq Z isomer),
1.14 (d, 3H, CH3), 0.78 (s, 3H, CH3), 0.77 (s, 3H, CH3); white
solid. Anal. (C22H34N2O3 ·C4H4O4) C, H, N.
(E,Z)-3-(2-Amino-2-methyl-1-propoxyimino)androstane-6,17-di-
one hydrochloride (6). Prepared in 48% yield from 1a and 2-amino-
2-methyl-1-propoxyamine dihydrochloride (Supporting Information,
I-c). The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 6. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 7.84 (bb, 3H, NH3+), 3.91
(s, 2H, CH2-O), 3.16 (m, 0.5H, H-2eq E isomer), 3.08 (m, 0.5H,
H-4eq Z isomer), 1.21 (s, 6H, C–(CH3)2), 0.79 (s, 3H, CH3), 0.78
(s, 3H, CH3); white solid. Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-(3-Amino-2-methyl-2-propoxyimino)androstane-6,17-di-
one fumarate (7). Prepared in 48% yield from 1a and 1-amino-2-
methyl-2-propoxyamine dihydrochloride (Supporting Information,
I-d). The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, fumaric
acid in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 7. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 8.1 (bb, 4H, NH3+, COOH),
6.41 (s, 2H, CHdCH), 3.07 (m, 0.5H, H-2eq E isomer), 3.02 (m,
0.5H, H-4eq Z isomer), 2.94 (m, 2H, CH2-N), 1.22 (s, 3H, CH3),
1.21 (s, 3H, CH3), 0.79 (s, 3H, CH3), 0.78 (s, 1.5H, CH3), 0.77 (s,
1.5H, CH3); white solid. Anal. (C23H36N2O3 ·C4H4O4) C, H, N.
(E,Z)-3-(2-Guanidinoethoxyimino)androstane-6,17-dione nitrate
(8). Istaroxime hydrochloride6 (0.50 g, 1.26 mmol) and 3,5-
dimethyl-1-pyrazolylformamidinium nitrate (0.30 g, 1.49 mmol)
were dissolved in EtOH (7.5 mL), then DIPEA (0.43 mL, 2.47
mmol) was added and the mixture refluxed for 1.5 h. After cooling,
the resulting slurry was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 90:10:1): the solid obtained was washed with
water and then dried at 40 °C under vacuum to give the title
1
compound 8 (0.27 g, 46%). H NMR (300 MHz, DMSO-d6, ppm
from TMS): δ 7.38 (t, 1H, NH), 7.02 (m, 4H, C-N2H4+), 4.00
(m, 2H, CH2O), 3.35 (m, 2H, CH2N), 3.02 (m, 0.5H, H-2eq E
isomer), 2.97 (m, 0.5H, H-4eq Z isomer), 0.79 (s, 3H, CH3), 0.78
(s, 3H, CH3); white solid. Anal. (C22H34N4O3 ·HNO3) C, H, N.
(E,Z)-3-(2-N-Methylaminoethoxyimino)androstane-6,17-dione hy-
drochloride (9). Prepared in 64% yield from 1a and 2-N-methy-
laminoethoxyamine dihydrochloride (Supporting Information, I-e).
The title compound 9 was obtained from the crude, after washing
the residue with EtOAc and crystallizing from 96% EtOH. 1H NMR
(300 MHz, DMSO-d6, ppm from TMS): δ 8.66 (bb, 2H, NH2+),
4.14 (m, 2H, CH2-O), 3.14 (m, 2H, CH2-N), 3.08 (m, 0.5H, H-2
E isomer), 3.00 (m, 0.5H, H-4 Z isomer), 2.55 (s, 1.5H, NCH3),
2.55 (s, 1.5H, NCH3), 0.79 (s, 3H, CH3), 0.78 (s, 1.5H, CH3), 0.77
(s, 1.5H, CH3); white solid. Anal. (C22H34N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-(3-N-Methylaminopropoxyimino)androstane-6,17-di-
one hydrochloride (10). Prepared in 63% yield from 1a and 3-N-
methylaminopropoxyamine dihydrochloride (Supporting Informa-
tion, I-f). The title compound 10 was obtained from the crude after
washing the residue with EtOAc and filtering. 1H NMR (300 MHz,
DMSO-d6, ppm from TMS): δ 8.56 (bb, 2H, NH2+), 3.97 (m, 2H,
CH2-O), 3.00 (m, 0.5H, H-2eq E isomer), 2.93 (m, 0.5H, H-4eq
Z isomer), 2.88 (m, 2H, CH2-N), 2.51 (s, 3H, CH3-N), 0.79 (s,
3H, CH3), 0.78 (s, 1.5H, CH3), 0.77(s, 1.5H, CH3); white solid.
Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.

NoVel Analogues of Istaroxime
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15 4607
H-2eq E isomer), 2.97 (m, 0.5H, H-4eq Z isomer), 2.75 (s, 3H,
CH3-N), 0.79 (s, 3H, CH3), 0.78 (s, 1.5H, CH3), 0.77 (s, 1.5H,
CH3); white solid. Anal. (C24H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-[3-(S)-Pyrrolidinyl]oxyimino-6r-hydroxyandrostane-17-
one hydrochloride (24). Prepared in 70% yield from 1b and 3-(S)-
pyrrolidinyloxyamine dihydrochloride (Supporting Information,
I-h). The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 24. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 9.12 (bb, 2H, NH2+), 4.72
(m, 1H, CH–O), 4.54 (d, 1H, OH), 3.40 (m, 0.5H, H-4eq Z isomer),
3.25 (m, 5H, CH2-N–CH2 and CH–O), 2.99 (m, 0.5H, H-2eq E
isomer), 0.86 (s, 1.5H, CH3), 0.85 (s, 1.5H, CH3), 0.77 (s, 3H, CH3);
white solid. Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-[3-(R)-Pyrrolidinyl]oxyimino-6r-hydroxyandrostane-17-
one hydrochloride (25). Prepared in 70% yield from 1b and 3-(R)-
pyrrolidinyloxyamine dihydrochloride (Supporting Information, I-i).
The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 25. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 9.15 (bb, 2H, NH2+), 4.73
(m, 1H, CH–O), 4.52 (d, 1H, OH), 3.40 (m, 0.5H, H-4eq Z isomer),
3.25 (m, 5H, CH2-N–CH2 and CH–O) 2.99 (m, 0.5H, H-2eq E
isomer), 0.87 (s, 1.5H, CH3), 0.85 (s, 1.5H, CH3), 0.77 (s, 3H, CH3);
white solid. Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.
(E)-3-[3-(R)-(1-Methyl)pyrrolidinyloxyimino]androstane-6,17-di-
one (18). Prepared in 85% yield from 1a and 3-(R)-(1-methyl)pyr-
rolidinyloxyamine dihydrochloride (Supporting Information, I-k).
The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). The solid residue from the concen-
trated fractions was suspended in EtOAc and then collected by
filtration to give the title compound 18. 1H NMR (300 MHz,
DMSO-d6, ppm from TMS): δ 4.57 (m, 1H, CH–O), 3.00 (m, 1H,
H-2eq), 2.19 (s, 3H, CH3-N), 0.78 (s, 3H, CH3), 0.77 (s, 3H, CH3);
light yellow solid. Anal. (C24H36N2O3) C, H, N.
(E,Z)-3-[2-(S)-Pyrrolidinylmethoxyimino]androstane-6,17-di-
one hydrochloride (19). Prepared in 57% yield from 1a and 2-[(S)-
pyrrolidinyl]methoxyamine dihydrochloride (Supporting Informa-
tion, I-l). The crude product was purified by flash chromatography
(SiO2, CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions,
5 M HCl in EtOAc was added. After dilution with Et2O, the solid
was collected by filtration to give the title compound 19. 1H NMR
(300 MHz, DMSO-d6, ppm from TMS): δ 9.55 (bb, 1Η, ΝΗ+),
8.79 (bb, 1H, NH+), 4.12 (m, 2H, CH2-O), 3.69 (m, 1H, CH–N),
3.12 (m, 2H, CH2-N), 3.10 (m, 0.5H, H-2 E isomer), 3.01 (m,
0.5H, H-4 Z isomer), 0.79 (s, 3H, CH3), 0.78 (s, 3H, CH3); white
solid. Anal. (C24H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-[2-(R)-Pyrrolidinylmethoxyimino]androstane-6,17-di-
one hydrochloride (20). Prepared in 57% yield from 1a and 2-[(R)-
pyrrolidinyl]methoxyamine dihydrochloride (Supporting Informa-
tion, I-m). The crude product was purified by flash chromatography
(SiO2, CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions,
5 M HCl in EtOAc was added. After dilution with Et2O, the solid
was collected by filtration to give the title compound 20. 1H NMR
(300 MHz, DMSO-d6, ppm from TMS): δ 9.40 (bb, 1H, NH+),
8.84 (bb, 1H, NH+), 4.11 (m, 2H, CH2-O), 3.71 (m, 1H, CH–N),
3.10 (m, 2H, CH2-N), 3.08 (m, 0.5H, H-2 E isomer), 3.02 (m,
0.5H, H-4 Z isomer), 0.79 (s, 3H, CH3), 0.78 (s, 3H, CH3); white
solid. Anal. (C24H36N2O3 ·HCl) C, H, N, Cl.
(E)-3-[3-(R,S)-Piperidinyloxyimino]androstane-6,17-dione hy-
drochloride (21). Prepared in 76% yield from 1a and 3-(RS)-
piperidinyloxyamine dihydrochloride (Supporting Information, I-n).
The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 21. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 8.68 (bb, 2H, NH2+), 4.21
(m, 1H, CH–O), 3.10 (m, 5H, CH2-N–CH2 + H-2eq), 0.79 (s,
3H, CH3), 0.79 (s, 3H, CH3); white solid. Anal. (C24H36N2O3 ·HCl)
C, H, N, Cl.
(E,Z)-3-[3-(R)-Pyrrolidinyl]oxyimino-6ꢀ-hydroxyandrostane-17-
one hydrochloride (26). Prepared in 80% yield from 1c and 3-(R)-
pyrrolidinyloxyamine dihydrochloride (Supporting Information, I-i).
The crude product was purified by flash chromatography (SiO2,
CH2Cl2:MeOH:NH3 9:1:0.1). To the concentrated fractions, 5 M
HCl in EtOAc was added. After dilution with Et2O, the solid was
collected by filtration to give the title compound 26. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 9.25 (bb, 2H, NH2+), 4.73
(m, 1H, CH–O), 4.45 (d, 0.5H, OH), 4.43 (d, 0,5H, OH), 3.65 (m,
1H, CH–O), 3.17 (m, 4H, CH2-N–CH2), 3.00 (m, 0.5H, H-2eq E
isomer), 2.76 (m, 0.5H, H-4eq Z isomer), 0.96 (s, 3H, CH3), 0.78
(s, 3H, CH3); white solid. Anal. (C23H36N2O3 ·HCl) C, H, N, Cl.
[(E)-2-(6,17-Dioxoandrostane-3-ylideneaminooxy)ethyl]carbam-
ic Acid 9H-Fluoren-9-ylmethyl ester (27) and [(Z)-2-(6,17-Dioxoan-
drostane-3-ylideneaminooxy)ethyl]carbamic Acid 9H-fluoren-9-
ylmethyl ester (28). A solution of 9-fluorenylmethoxycarbonyl
chloride (10.18 g, 39.3 mmol) in CH2Cl2 (40 mL) was added
dropwise to a stirred solution of Istaroxime hydrochloride6 (13.00
g, 32.7 mmol) in CH2Cl2 (90 mL) and Et3N (10.0 mL, 71.7 mmol)
in an ice bath. After completing the addition, the mixture was stirred
in an ice bath for 10 min. Water (200 mL) was carefully added
dropwise and the mixture extracted with CH2Cl2 (2 × 100 mL).
The organic phase was washed with 0.1N HCl, 5% aqueous
NaHCO3, H2O, dried over Na2SO4, and evaporated to dryness. The
residue was purified by flash chromatography and divided into four
columns (SiO2; CH2Cl2/acetone 95/5) to give, as overall result,
(E)-3-(4-Piperidyloxyimino)androstane-6,17-dione hydrochlo-
ride (22). Prepared in 60% yield from 1a and 4-piperidyloxyamine
dihydrochloride (Supporting Information, I-o). The crude product
was purified by flash chromatography (SiO2, CH2Cl2:MeOH:NH3
9:1:0.1). To the concentrated fractions, 5 M HCl in EtOAc was
added. After dilution with Et2O, the solid was collected by filtration
to give the title compound 22. 1H NMR (300 MHz, DMSO-d6,
ppm from TMS): δ 8.68 (bb, 2H, NH2+), 4.17 (m, 1H, CH–O),
3.05 (m, 5H, CH2-N–CH2 and H-2eq) 0.79 (s, 3H, CH3), 0.78 (s,
3H, CH3); white solid. Anal. (C24H36N2O3 ·HCl) C, H, N, Cl.
(E,Z)-3-(3-N-Methylaminopropoxyimino)-6r-hydroxyandrostane-
17-one hydrochloride (23). Prepared in 57% yield from 1b and 3-N-
methylaminopropoxyamine dihydrochloride (Supporting Informa-
tion, I-f). The title compound 23 was obtained from the crude after
washing the residue with EtOAc, Et2O, and filtering. 1H NMR (300
MHz, DMSO-d6, ppm from TMS): δ 8.75 (bb, 2H, NH2+), 4.55
(d, 0.5H, OH E isomer), 4.51 (d, 0.5H, OH Z isomer), 3.96 (m,
2H, CH2-O), 3.41 (m, 0.5H, H-4eq Z isomer), 3.29 (m, 1H, CH–O),
2.97 (m, 0.5H, H-2eq E isomer), 2.89 (m, 2H, CH2-N), 2.51 (s,
3H, CH3-N), 0.86 (s, 1.5H, CH3), 0.85 (s, 1.5H, CH3), 0.77 (s,
1.5H, CH3); white solid. Anal. (C23H38N2O3 ·HCl) C, H, N, Cl.
1
compounds 27 (7.60 g, 40%) and 28 (7.60 g, 40%). 27: H NMR
(300 MHz, acetone-d6, ppm from TMS) δ 7.90-7.25 (m, 8H,
aromatics), 6.53 (bt, 1H, NH), 4.28 (m, 3H, CH-CH2-O), 4.05
(m, 2H, CH2-O), 3.41 (m, 2H, CH2-N), 3.16 (m, 0.5H, H-2eq),
0.85 (s, 3H, CH3), 0.84 (s, 3H, CH3). 28: 1H NMR (300 MHz,
acetone-d6, ppm from TMS) δ 7.90-7.25 (m, 8H, aromatics), 6.56
(bt, 1H, NH), 4.25 (m, 3H, CH-CH2-O), 4.05 (m, 2H, CH2-O),
3.41 (m, 2H, CH2-N), 3.11 (m, 0.5H, H-4eq), 0.85 (s, 3H, CH3),
0.84 (s, 3H, CH3).
Conformational Energy Calculations. The conformational analy-
ses of 2, 15, and the E isomer of 13 were performed using the
OPLS2005 all-atom force field as implemented in the MacroModel
9.1 program; PR conjugate gradient was used in all the minimization
steps, with the derivative convergence set to 0.05 (kJ/mol)/Å, with
a maximum of 5000 iterations. The Monte Carlo multiple minimum
method was used in the conformational search (OPLS2005, 5000
steps, torsion rotations allowed on the amino chain). All conforma-
tions within 50 kJ/mol of the identified lowest energy conformer
were minimized again (Multiple Minimization routine in the
program) using OPLS2005 in conjunction with the GB/SA con-
tinuum model to simulate water solvation.

4608 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 15
Gobbini et al.
the Safety and Tolerability of Istaroxime and Specific Effects on
Electrocardiographic and Hemodynamic Parameters in Patients with
Chronic Heart Failure with Reduced Systolic Function. Am. J. Cardiol.
2007, 99 (2A), 47A-55A.
Supporting Information Available: Elemental analysis results
for target compounds, experimental details of the synthesis and
characterization data for the required hydroxylamines. This material
is available free of charge via the Internet at http://pubs.acs.org.
(9) Jørghensen, P. L. Purification and Characterization of Na+,K+-ATPase.
III. Purification from the Outer Medulla of Mammalian Kidney after
Selective Removal of Membrane Components by Sodium Dodecyl-
sulphate. Biochim. Biophys. Acta 1974, 356, 36–52.
References
(10) Noel, F.; Godfraind, T. Heterogeneity of Ouabain Specific Binding
Sites and Sodium-Potassium ATPase Inhibition in Microsomes from
Rat Heart. Biochem. Pharmacol. 1984, 33, 47–53.
(1) Gobbini, M.; Cerri, A. Digitalis-like Compounds: the Discovery of
the O-aminoalkyloxime Group as a Very Powerful Substitute for the
Unsaturated γ-Butyrolactone Moiety. Curr. Med. Chem. 2005, 12,
2343–2355, and references cited therein.
(2) The Digitalis Investigation Group. The Effect of Digoxin on Mortality
and Morbidity in Patients with Heart Failure. N. Engl. J. Med. 1997,
336, 525-533.
(3) Hoffman, B. F.; Bigger, J. T. Digitalis and Allied Cardiac Glycosides
In The Pharmacological Basis of Therapeutics, 8th ed.; Goodman
Gilman, A., Nies, A. S., Rall, T. W., Taylor, P. Eds.; Pergamon Press:
New York, 1990, pp 814-839.
(4) (a) Thomas, R.; Gray, P.; Andrews, J. Digitalis: its Mode of Action,
Receptor and Structure-Activity Relationships. In AdVances in Drug
Research; Academic Press: New York, 1990, Vol. 19, pp 311-562.
(b) Repke, K. R. H.; Sweadner, K. J.; Weiland, J.; Megges, R.; Scho¨n,
R. In Search of Ideal Inotropic Steroids: Recent Progress. Prog. Drug
Res. 1996, 47, 9–52.
(5) De Munari, S.; Barassi, P.; Cerri, A.; Fedrizzi, G.; Gobbini, M.;
Mabilia, M.; Melloni, P. A New Approach to the Design of Novel
Inhibitors of Na+,K+-ATPase: 17R-Substituted Seco-D-5ꢀ-Androstane
as Cassaine Analogues. J. Med. Chem. 1998, 41, 3033–3040.
(6) De Munari, S.; Cerri, A.; Gobbini, M.; Almirante, N.; Banfi, L.;
Carzana, G.; Ferrari, P.; Marazzi, M.; Micheletti, R.; Schiavone, A.;
Sputore, S.; Torri, M.; Zappavigna, M. P.; Melloni, P. Structure-Based
Design and Synthesis of Novel Potent Na+,K+-ATPase Inhibitors
Derived from a 5R,14R-Androstane Scaffold as Positive Inotropic
Compounds. J. Med. Chem. 2003, 46, 3644.
(7) Dalma, G. New Alkaloids in Erythrophleum guineense. Ann. Chim.
Appl. (Rome) 1935, 25, 569–571.
(8) Ghali, J. K.; Smith, W. B.; Torre-Amione, G.; Haynos, W.; Rayburn,
B. K.; Amato, A.; Zhang, D.; Cowart, D.; Valentini, G.; Carminati,
P.; Gheorghiade, M. A Phase 1-2 Dose-Escalating Study Evaluating
(11) MacroModel, version 9.1. Schro¨dinger, L. L. C.; Mohamadi, F.;
Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield,
C.; Chang, G.; Hendrikson, T.; Still, W. C. MacroModel, an Integrated
Software System for Modeling Organic and Bioorganic Molecules
Using Molecular Mechanics. J. Comput. Chem. 1990, 11, 440–467.
(12) Cerri, A.; Almirante, N.; Barassi, P.; Benicchio, A.; Fedrizzi, G.;
Ferrari, P.; Micheletti, R.; Quadri, L.; Ragg, E.; Rossi, R.; Santagostino,
M.; Schiavone, A.; Serra, F.; Zappavigna, M. P.; Melloni, P. 17ꢀ-O-
Aminoalkyloximes of 5ꢀ-Androstane-3ꢀ,14ꢀ-diol with Digitalis-Like
Activity: Synthesis, Cardiotonic Activity, Structure-Activity Relation-
ships, and Molecular Modeling of the Na+,K+-ATPase Receptor.
J. Med. Chem. 2000, 43, 2332–2349.
(13) Gobbini, M.; Barassi, P.; Cerri, A.; De Munari, S.; Fedrizzi, G.;
Santagostino, M.; Schiavone, A.; Torri, M.; Melloni, P. 17R-O-
Aminoalkyloxime Derivatives of 3ꢀ,14ꢀ-Dihydroxy-5ꢀ-androstane and
3ꢀ-Hydroxy-14-oxo-seco-D-5ꢀ-androstane as Inhibitors of the Digitalis
Receptor on Na+,K+-ATPase. J. Med. Chem. 2001, 44, 3821–3830.
(14) Rocchetti, M.; Besana, A.; Mostacciuolo, G.; Micheletti, R.; Ferrari,
P.; Sarkosi, S.; Szegedi, C.; Jona, I.; Zaza, A. Modulation of
Sarcoplasmic Reticulum Function by Na+/K+ Pump Inhibitors with
Different Toxicity: Digoxin and PST2744 [(E,Z)-3-((2-Aminoethoxy-
)imino)androstane-6,17-dione Hydrochloride]. J. Gen. Physiol. 2005,
126 (1), 58A-59A.
(15) Micheletti, R.; Palazzo, F.; Barassi, P.; Giacalone, G.; Ferrandi, M.;
Schiavone, A.; Moro, B.; Parodi, O.; Ferrari, P.; Bianchi, G. Istaroxime,
a Stimulator of Sarcoplasmic Reticulum Calcium Adenosine Triph-
osphatase Isoform 2a Activity, as a Novel Therapeutic Approach to
Heart Failure. Am. J. Cardiol. 2007, 99 (Suppl. 1), S24-S32.
JM800257S