TVP1022 Protects Neonatal Rat Ventricular Myocytes against
Doxorubicin-Induced Functional Derangements
Alexandra Berdichevski, Gideon Meiry, Felix Milman, Irena Reiter, Oshra Sedan,
Sivan Eliyahu, Heather S. Duffy, Moussa B. Youdim, and Ofer Binah
Departments of Physiology (A.B., G.M., F.M., I.R., O.S., S.E., O.B.) and Pharmacology (M.B.Y.), Ruth and Bruce Rappaport
Family Institute for Research in the Medical Sciences, Ruth and Bruce Rappaport Faculty of Medicine, Technion-Israel Institute
of Technology, Haifa, Israel; and Cardiovascular Research, Cardiovascular Division, Beth Israel Deaconess Medical Center,
Boston, Massachusetts (H.S.D.)
Received September 2, 2009; accepted November 12, 2009
Our recent studies demonstrated that propargylamine deriva-
tives such as rasagiline (Azilect, Food and Drug Administration-
approved anti-Parkinson drug) and its S-isomer TVP1022
protect cardiac and neuronal cell cultures against apoptotic-
inducing stimuli. Studies on structure-activity relationship re-
vealed that their neuroprotective effect is associated with the
propargylamine moiety, which protects mitochondrial viability
and prevents apoptosis by activating Bcl-2 and protein kinase
C-? and by down-regulating the proapoptotic protein Bax.
Based on the established cytoprotective and neuroprotective
efficacies of propargylamine derivatives, as well as on our
recent study showing that TVP1022 attenuates serum starva-
tion-induced and doxorubicin-induced apoptosis in neonatal
rat ventricular myocytes (NRVMs), we tested the hypothesis
that TVP1022 will also provide protection against doxorubicin-
induced NRVM functional derangements. The present study
demonstrates that pretreatment of NRVMs with TVP1022 (1
?M, 24 h) prevented doxorubicin (0.5 ?M, 24 h)-induced ele-
vation of diastolic [Ca2?]i, the slowing of [Ca2?]irelaxation
kinetics, and the decrease in the rates of myocyte contraction
and relaxation. Furthermore, pretreatment with TVP1022 atten-
uated the doxorubicin-induced reduction in the protein expres-
sion of sarco/endoplasmic reticulum calcium (Ca2?) ATPase,
exchanger 1, and total connexin 43. Finally,
TVP1022 diminished the inhibitory effect of doxorubicin on gap
junctional intercellular coupling (measured by means of Lucifer
yellow transfer) and on conduction velocity, the amplitude of
the activation phase, and the maximal rate of activation (dv/
dtmax) measured by the Micro-Electrode-Array system. In sum-
mary, our results indicate that TVP1022 acts as a novel cardio-
protective agent against anthracycline cardiotoxicity, and
therefore potentially can be coadmhence, theinistered with
doxorubicin in the treatment of malignancies in humans.
Doxorubicin or adriamycin is a quinone-containing anthra-
cycline, and is of the most widely prescribed and effective
chemotherapeutic agent used in oncology. All anthracyclines
contain a common quinone moiety, readily participating
in oxidation-reduction reactions that ultimately generate
highly reactive oxygen species thought to be responsible for
the anthracycline-induced cardiotoxicity (Sarvazyan, 1996;
Menna et al., 2008). Doxorubicin is one of the most active
agents available for the treatment of breast cancer and other
indications, including Hodgkin’s and non-Hodgkin’s lympho-
mas, Ewing’s and osteogenic bone tumors, soft tissue sarco-
mas, and pediatric cancers such as neuroblastoma and
Wilms’ tumors. However, the clinical utility of doxorubicin is
limited by its cumulative, dose-related, potentially fatal, pro-
gressive, and often irreversible cardiac toxicity that may lead
to congestive heart failure (CHF) (Singal and Iliskovic, 1998;
Swain et al., 2003; Takemura and Fujiwara, 2007). The
chronic effects of doxorubicin expressed as CHF are invari-
ably associated with cumulative drug exposure (Singal and
Iliskovic, 1998; Swain et al., 2003; Takemura and Fujiwara,
2007). For example, at a cumulative dose not exceeding 450
to 500 mg/m2the incidence of CHF is ?4 to 5%, whereas at
550 to 600 mg/m2the incidence increases to 18% (Singal and
This work was supported by Alfred Mann Institute at the Technion [Grant
2009297], the Rappaport Family Institute for Research in the Medical Sciences
[Grant Rappaport2009], the Horowitz Foundation–Technion [Grant 2007454],
the Israeli Ministry of Science and Technology, MOST [Grant 2011198], and
the Israel Science Foundation [Grant 2011417].
Article, publication date, and citation information can be found at
ABBREVIATIONS: CHF, congestive heart failure; NRVM, neonatal rat ventricular myocyte; CSQ, calsequestrin; PLB, phospholamban; SERCA2,
sarco/endoplasmic reticulum calcium (Ca2?) ATPase 2; NCX1, Na?/Ca2?exchanger 1; total Cx43, total Connexin 43; NP Cx43, nonphosphory-
lated Connexin 43; RyR, ryanodine receptor; MEA, Micro-Electrode-Array; hESC-CM, human embryonic stem cell-derived cardiomyocyte; ROS,
reactive oxygen species; TBS, Tris-buffered saline; CBN, carbenoxolone; SR, sarcoplasmic reticulum.
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
JPET 332:413–420, 2010
Vol. 332, No. 2
Printed in U.S.A.
Iliskovic, 1998; Takemura and Fujiwara, 2007; Menna et al.,
2008). Nevertheless, despite these deleterious side effects,
the benefits of anthracyclines outweigh the risks, and thus
doxorubicin continues to serve as an important anticancer
Based on the established cytoprotective and neuroprotec-
tive efficacies of propargylamine derivatives (Youdim and
Weinstock, 2001; Youdim et al., 2003) as well as on our recent
study (Kleiner et al., 2008) showing that TVP1022 (the S-
isomer of rasagiline, Azilect, Food and Drug Administration-
approved anti-Parkinson drug) attenuates serum starvation-
induced and doxorubicin-induced apoptosis in neonatal rat
ventricular myocytes (NRVM), in the present study we tested
the hypothesis that TVP1022 will provide protection against
doxorubicin-induced functional derangements in NRVM. In-
deed, in support of this hypothesis, we demonstrated that
TVP1022 markedly attenuated the deleterious effects of
doxorubicin on the [Ca2?]itransients, the contractions and
intercellular coupling, rending this drug a potential cardio-
protective agent against anthracycline cardiotoxicity.
Materials and Methods
Preparation of Neonatal Rat Ventricular Myocytes and the
Experimental Protocols. NRVM cultures were prepared from ven-
tricles of 1- to 2-day-old Sprague-Dawley rats as described previously
(Rubin et al., 1995), with minor modifications. In brief, the ventricles
of the excised hearts were dissociated with 0.1% RDB (Israel Insti-
tute for Biological Research, Ness Ziona, Israel) and resuspended in
F-10 culture medium containing 1 ?M CaCl2, 100 U/ml penicillin-
streptomycin, 5% fetal calf serum, 5% donor horse serum, and 25 mg
bromodeoxyuridine. Cell culture reagents were purchased from Bio-
logical Industries (Beit Haemek, Israel). The cells were preplated for
1 h to reduce the fibroblast content, and then seeded in 6-well plates
(1.6 ? 106cells/ml). For the Micro-Electrode-Array (MEA) experi-
ments, NRVM were plated on MEA plates at a density of 2 ? 106
cells/ml. Thereafter, the cultures were incubated at 37°C in a hu-
midified atmosphere containing 5% CO2. Unsettled cells were
washed out after 24 h, the medium was replaced, and then replaced
again on alternating days. Experiments were performed on days 4 to
6 after plating, on the following experimental groups: 1) untreated
cultures serving as control; 2) doxorubicin, 0.5 ?M, 24 h; 3) TVP1022
(Fig. 1) (Teva Pharmaceutical, Natanya, Israel), 1 ?M, 48 h; 4)
TVP1022, 24 h, followed by TVP1022 ? doxorubicin for an addi-
tional 24 h. For the MEA recordings, measurements were obtained
at 0 h (serving as control) and at 48 h. For all other experiments,
measurements were obtained 48 h after TVP1022 administration
or 24 h after doxorubicin administration.
In the present work the doxorubicin concentration (0.5 ?M) was
chosen based on our preliminary experiments (data not shown) and
on previous studies using a concentration range of 0.1 to 1 ?M
(Maeda et al., 1999; Shneyvays et al., 2001; Kleiner et al., 2008). To
determine the cardioprotective efficacy of TVP1022, a concentration
of 1 ?M was used based on our previous work (Kleiner et al., 2008)
and our findings (data not shown) illustrating that 0.01, 0.1, and 1
?M TVP1022 similarly affected the expression of the [Ca2?]i-han-
dling proteins (see details below), total and nonphosphorylated con-
nexin43 (Cx43), and intercellular coupling determined by Lucifer
yellow transfer. It is noteworthy that because TVP1022 is being
developed as a cardioprotective drug to be administered to cancer
patients before and during doxorubicin treatment, NRVM cultures
were pretreated with TVP1022 before exposure to doxorubicin.
Western Blot Analysis. Lysates were prepared from NRVM
cultures with use of radioimmunoprecipitation assay (20 mM Tris-
HCl, pH 7.4, 200 mM NaCl, 1% Triton X-100, 0.1% SDS, 0.2% sodium
deoxycholate, 5 mM EDTA, 1% phosphatase inhibitor) containing
cocktail protease inhibitor (Roche Diagnostics, Mannheim, Ger-
many). Protein concentration was determined by the bicinchoninic
acid assay. A 20- to 30-?g sample of total cellular protein was loaded
on 12% SDS-polyacrylamide gel electrophoresis, followed by blotting
into polyvinylidene difluoride membranes (Millipore, Billerica, MA).
Electrophoresis reagents were purchased from Invitrogen Corpora-
tion (Carlsbad, CA). Membranes were blocked with 5% dry milk in
0.05% Tween 20 in Tris-buffered saline (TBS) for 1 h. Antibodies
against sarco/endoplasmic reticulum calcium (Ca2?) ATPase 2
(SERCA2), calsequestrin (CSQ), phospholamban (PLB), and the ry-
anodine receptor (RyR) were purchased from Thermo Fisher Scien-
tific (Rockford, IL); Na?/Ca2?exchanger 1 (NCX1) was from Abcam
(Cambridge, UK); total Cx43 was from Millipore Bioscience Research
Reagents (Temecula, CA); and nonphosphorylated Cx43 (NP Cx43)
was from Zymed Laboratories (San Francisco, CA). ?-Actin antibody
and all other reagents were purchased from Sigma-Aldrich (St.
Louis, MO). Primary antibodies were diluted in 5% dry milk in TBS
containing 0.05% Tween 20 and incubated with membranes for 24 h
at 4°C followed by incubation (1 h at room temperature) in dilutions
of horseradish peroxidase-conjugated secondary antibodies in the
same buffer. After antibody incubations, membranes were washed in
0.05% Tween 20 in TBS. Detection was performed by use of Western
Healthcare, Little Chalfont Buckinghamshire, UK). Quantification
of the results was accomplished by measuring the optical density of
the labeled bands from the autoradiograms, using the computerized
imaging program Bio-1D (Vilber Lourmat Biotech, Bioprof, France).
The values were normalized to ?-actin intensity levels.
Measurement of [Ca2?]iTransients and Contractions.
[Ca2?]itransients and contractions were measured by means of
Fura-2 fluorescence and video edge detection, respectively, rou-
tinely used in our laboratory (Dolnikov et al., 2006; Sedan et al.,
2008). The following parameters were calculated: diastolic [Ca2?]i,
the rate of [Ca2?]iactivation (d[Ca2?]i,Act/dt), [Ca2?]irelaxation
time ([Ca2?]i,Relax), the maximal rate of contraction (dL/dtContrac),
the maximal rate of relaxation (dL/dtRelax), and contraction am-
Measuring Intercellular Coupling by Lucifer Yellow Trans-
fer. The scrape-loading technique was used to evaluate the degree of
gap junctional intercellular communication (Kavanagh et al., 1987).
In brief, NRVM cultures were bathed in phosphate-buffered saline
containing 0.5% Lucifer yellow (Sigma-Aldrich). Fine incisions on the
culture surface were made by use of a razor blade, allowing Lucifer
yellow to enter the scraped NRVM. After 2.5 min of incubation, the
cultures were rinsed three times with phosphate-buffered saline,
fixed in 4% p-formaldehyde (Sigma-Aldrich), and photographed. The
extent of Lucifer yellow transfer calculated by use of the Image J
software, was obtained by determining the maximal distances at
which Lucifer yellow fluorescence could be clearly detected perpen-
dicular to the incision.
Fig. 1. The chemical structure of TVP1022.
Berdichevski et al.
Recording Extracellular Electrograms from NRVM. Extra-
cellular electrograms were recorded from NRVM plated on the
MEAs, as described previously (Meiry et al., 2001; Zeevi-Levin et al.,
2005) by means of a setup by Multi Channel Systems (Reutlingen,
Germany). Drugs were introduced to the cultures as described above.
For the electrophysiological measurements, MEAs were removed
from the incubator and placed in the recording apparatus preheated
to 37°C; spontaneous electrical activity was recorded and stored for
off-line analysis. Conduction velocity, the voltage amplitude of the
wave form during activation (the QRS amplitude), and maximal rate
of activation (dV/dtmax) were calculated as described previously
(Meiry et al., 2001; Zeevi-Levin et al., 2005).
Human Embryonic Stem Cell-Derived Cardiomyocytes. Hu-
man embryonic stem cells (hESC) from clone H9.2 were grown on
mouse embryonic fibroblast feeder, and embryoid bodies prepared as
described previously (Sedan et al., 2008). The contracting areas
mostly composed of cardiomyocytes were carefully dissected out by a
microscalpel, and transferred to gelatin-coated 30-mm-diameter
glass slides suitable for fluorescent measurements (Thermo Fisher
Scientific). Contractions were recorded and analyzed as described
previously (Dolnikov et al., 2006; Sedan et al., 2008).
Statistical Analysis. Raw data were analyzed and expressed as
mean? S.E.M. Statistical differences between multiple groups were
determined using one-way analysis of variance, followed by the
Tukey post hoc test by use of Prism v.5.00 for Windows (GraphPad
Software Inc., San Diego, CA). In the MEA experiments the Stu-
dent’s paired t test was used. For all other experiments, independent
Student’s t test was used to compare the control and TVP1022
groups. A level of P ? 0.05 was accepted as statistically significant.
TVP1022 Attenuates the Deleterious Effects of Doxo-
rubicin on the [Ca2?]iTransients and Contractions. In
agreement with previous reports, this study shows that doxo-
rubicin adversely affects the [Ca2?]itransients and contrac-
tions of NRVM (Maeda et al., 1998, 1999; Mijares and Lopez,
2001; Shneyvays et al., 2001; Timolati et al., 2006). As shown
by a representative experiment (Fig. 2A) and by the sum-
mary of 8 to 10 experiments, doxorubicin elevated (P ?
0.0001) diastolic [Ca2?]i(Fig. 2B), decreased the maximal
rate of [Ca2?]iactivation (P ? 0.01) (Fig. 2C), and prolonged
(P ? 0.01) [Ca2?]irelaxation time (Fig. 2D). Accordingly,
doxorubicin decreased (P ? 0.05) the maximal rates of con-
traction and relaxation and the contraction amplitude (Fig. 3).
In support of our hypothesis, TVP1022 prevented the delete-
rious effects (except for the decline in the rate of [Ca2?]i
activation; Fig. 2C) of doxorubicin on the [Ca2?]itransients
(Fig. 2) and contractions of NRVM (Fig. 3).
TVP1022 Attenuates the Deleterious Effects of Doxo-
rubicin on the [Ca2?]iHandling Machinery. To decipher
the mechanisms underlying the protective capacity of
TVP1022 on the [Ca2?]itransients and contractions, we in-
vestigated the effects of doxorubicin alone, and TVP1022 ?
doxorubicin on the protein expression of PLB, CSQ, SERCA2,
Fig. 2. TVP1022 prevented doxorubicin-induced alterations in the [Ca2?]i
transients parameters in NRVM. NRVM cultures were preincubated with
TVP1022 (1 ?M) for 24 h before adding doxorubicin (0.5 ?M) for an
additional 24 h. [Ca2?]itransients were measured by Fura-2 fluorescence
as described under Materials and Methods. A, representative [Ca2?]i
transients recorded from a control culture and from cultures treated with
doxorubicin alone or with TVP1022 ? doxorubicin. B, diastolic [Ca2?]i
expressed as Fura-2 fluorescence ratio. C, the rate of [Ca2?]iactivation
(d[Ca2?]i,Act/dt). D, [Ca2?]irelaxation time ([Ca2?]i,Relax). The values are
mean? S.E.M. of eight to nine experiments. ???, P ? 0.001 versus control;
??, P ? 0.01 versus control. Dox, doxorubicin; TVP, TVP1022.
Fig. 3. TVP1022 prevented doxorubicin-induced alter-
ations in the contraction parameters in NRVM. Cultures
were treated as described in Fig. 2, and contraction param-
eters were measured by video edge detection as described
under Materials and Methods. A, representative contrac-
tions recorded from a control culture and from cultures
treated with doxorubicin alone or with TVP1022 ? doxo-
rubicin. B, the maximal rate of myocyte contraction (dL/
dtContrac). C, the maximal rate of myocyte relaxation (dL/
dtRelax). D, the contraction amplitude (LAmp). The values
are mean ? S.E.M. of five to seven experiments. ?, P ? 0.05
versus control. Dox, doxorubicin; TVP, TVP1022.
TVP1022 Attenuates Doxorubicin-Induced Cardiotoxicity
NCX1, and RyR. As depicted in Fig. 4, doxorubicin did not
affect the expression of PLB, CSQ, or RyR, but decreased the
expression of SERCA2 (P ? 0.01) and NCX1 (P ? 0.05). In
agreement with the protective efficacy of TVP1022, pretreat-
ment for 24 h prevented doxorubicin-induced decreased ex-
pression of SERCA2 and NCX1. Hence, these beneficial ef-
fects of TVP1022 can contribute to its protection against the
adverse effects of doxorubicin on the [Ca2?]itransients and
contractions of NRVMs.
TVP1022 Attenuates the Deleterious Effects of Doxo-
rubicin on Intercellular Communication and Cx43 Ex-
pression. Next, we investigated the effect of doxorubicin and
the protective efficacy of TVP1022 on Lucifer yellow transfer,
which is indicative of intercellular coupling via gap junctions
(Kavanagh et al., 1987; Matemba et al., 2006). As depicted in
Fig. 5A (top), Lucifer yellow transfer can be visualized by
means of fluorescence microscopy, and the distance is deter-
mined by use of the ImageJ software (Fig. 5A, bottom). As
seen in Fig. 5B, doxorubicin caused a 36% reduction (P ?
0.001) in Lucifer yellow transfer, which was mostly pre-
vented by pretreatment with TVP1022. Finally, to confirm
that the Lucifer yellow transfer method is responsive to
agents that affect intercellular coupling, we tested the effects
of the intercellular uncoupler carbenoxolone (CBN, 50 and
100 ?M) on levels of intercellular coupling between NRVM.
As shown in Fig. 5B, applying 50 ?M CBN for 30 min reduced
(P ? 0.001) Lucifer yellow transfer by 54%. Because 100 ?M
CBN did not further decrease Lucifer yellow transfer, indi-
cating that 50 ?M is sufficient to maximally block coupling,
only the effect of 50 ?M is shown. To determine whether the
effect of doxorubicin on Lucifer yellow transfer is additive to
that of carbenoxolone, NRVM were treated with doxorubicin
(0.5 ?M, 24 h) and carbenoxolone (50 ?M, 30 min). Figure 5B
shows that, in the doxorubicin ? carbenoxolone-treated
group, Lucifer yellow transfer was reduced by 48% (P ?
0.001), similarly to carbenoxolone alone, indicating that the
uncoupling due to doxorubicin is less efficacious than that
due to carbenoxolone, which maximally closes the channels
at this concentration. Finally, as shown in Fig. 5B, TVP1022
did not affect carbenoxolone-induced reduction in Lucifer
yellow transfer, indicating that the molecular mechanism by
which TVP1022 maintains coupling is not affecting the car-
benoxolone mechanism of channel closure. These data sug-
gest that carbenoxolone and doxorubicin are likely to have
different mechanisms of action on the closure of gap junction
Because intercellular coupling largely depends on gap
junctional functionality, we investigated whether doxorubi-
cin affects total and NP Cx43. Indeed, as illustrated by the
representative Western blots and the summary of six exper-
Fig. 4. The effects of TVP1022 and doxorubicin on the expression of
[Ca2?]i-handling proteins in NRVM. Cultures were treated as described
in Fig. 2, and the expression level of proteins was measured by Western
blotting. A–E (top), representative Western blots of PLB, RyR, SERCA2,
NCX1, and CSQ expression, respectively, in the control, doxorubicin and
the TVP1022 ? doxorubicin groups. Equivalency of loading was verified
with an antibody against ?-actin (A–E, middle). In A to E (bottom), bar
graphs depict the quantitative densitometric analysis of each protein.
Each value was divided by its corresponding ?-actin value. All values are
reported as mean ? S.E.M. of three to four experiments. ?, P ? 0.05
versus control; ??, P ? 0.01 versus control. Dox, doxorubicin; TVP,
Fig. 5. TVP1022 attenuated doxorubicin-induced decline in the intercel-
lular coupling in NRVM. Cultures were treated as described in Fig. 2.
Intercellular coupling was measured by Lucifer yellow transfer. A (up-
per), representative fluorescence pictures of a control, doxorubicin-
treated, and TVP1022 ? doxorubicin-treated cultures, after Lucifer yel-
low loading. A (bottom), the Image J output analysis of the pictures
shown in the top. B, Lucifer yellow transfer distance (?m). All values are
reported as mean ? S.E.M. of six to eight experiments. In the CBN
groups, n ? 3 to 4 experiments. ???, P ? 0.001 versus control. LY, Lucifer
yellow; Dox, doxorubicin; TVP, TVP1022; CBN, carbenoxolone.
Berdichevski et al.
iments, doxorubicin decreased the expression of total (P ?
0.01) and NP (P ? 0.001) Cx43 (Fig. 6). As seen in Fig. 6B,
whereas the seemingly beneficial effect of TVP1022 on the
reduction in NP Cx43 did not reach a level of significance, the
drug attenuated doxorubicin-induced reduction in total Cx43
(Fig. 6A), which can account for the ability of TVP1022 to
attenuate the deleterious effect of doxorubicin on intercellu-
TVP1022 Attenuates the Deleterious Effects of Doxo-
rubicin on Conduction Velocity and Activation of
NRVM. To determine whether the decrease in intercellular
coupling and Cx43 protein expression by doxorubicin and the
beneficiary effects of TVP1022 are correlated with corre-
sponding alterations in the activation properties and conduc-
tion velocity, we measured extracellular electrograms from
NRVM cultures by means of the MEA setup (Meiry et al.,
2001; Zeevi-Levin et al., 2005). These experiments showed
(Fig. 7) that only in the doxorubicin group, conduction veloc-
ity (P ? 0.01), QRS amplitude (P ? 0.001), and dV/dtmax(P ?
0.01) were significantly decreased at 48 h, compared with the
0-h time point. These novel findings demonstrate that,
whereas TVP1022 did not affect the conduction velocity and
activation properties, this drug attenuated the deleterious
effects of doxorubicin, in agreement with its previously
shown beneficial effects.
The Safety of TVP1022 Demonstrated in NRVM and
hESC-CM. Because TVP1022 is being developed as a new
cardioprotective drug against anthracycline cardiotoxicity,
we tested its effects on the [Ca2?]itransient and the con-
traction properties of human embryonic stem cell-derived
cardiomyocytes (hESC-CM), an in vitro model for human
cardiomyocytes. As seen in Fig. 8, TVP1022 at 0.01, 0.1,
and 1 ?M did not affect the contraction amplitude (LAmp),
the maximal rate of contraction (dL/dtContrac), and the
maximal rate of relaxation (dL/dtRelax), thus demonstrat-
ing the safety of TVP1022 in this model. The safety of
TVP1022 to cultured cardiomyocytes is further supported by
showing that the drug did not affect the [Ca2?]itransient and
contraction parameters (Fig. 9A), the expression of the
[Ca2?]ihandling proteins (Fig. 9B), intercellular coupling
(Fig. 9C), total Cx43 and NP Cx43 (Fig. 9D), as well as
conduction velocity and activation (Fig. 7). These findings are
of importance because of the therapeutic potential of
In the present work we investigated in NRVM the cardio-
protective efficacy of TVP1022 against functional derange-
ments caused by doxorubicin. The major findings are:
1) pretreatment with TVP1022 prevented doxorubicin-in-
duced elevation of diastolic [Ca2?]i, the slowing of [Ca2?]i
relaxation kinetic, the decrease in the maximal rates of con-
Fig. 6. The effects of TVP1022 on doxorubicin-induced alternations in
total and NP Cx43 expression in NRVM. Cultures were treated as de-
scribed in Fig. 2, and the expression level of proteins was measured by
Western blotting. A and B (top), representative Western blots of total
Cx43 and NP Cx43 expression, respectively, in the control, doxorubicin,
and TVP1022 ? doxorubicin groups. A and B (middle), equivalency of
loading was verified with an antibody against ?-actin. A and B (bottom),
bar graphs depict the quantitative densitometric analysis of total Cx43
and NP Cx43 expression, respectively. Each value was divided by its
corresponding ?-actin value. All the values are reported as mean ?
S.E.M. of six experiments. ??, P ? 0.01 versus control; ???, P ? 0.001
versus control. Dox, doxorubicin; TVP, TVP1022.
Fig. 7. TVP1022 prevented doxorubicin-induced alterations in the acti-
vation and conduction velocity in NRVM. Cultures were preincubated
with doxorubicin (0.5 ?M, 24 h), with or without prior incubation with
TVP1022 (1 ?M, 48 h). Untreated cultures served as controls. Activation
parameters and conduction velocity of spontaneously beating NRVM
were obtained by MEA recordings. Conduction velocity (A), dV/dtmax(B),
and QRS amplitude (C) in the control, doxorubicin, TVP1022, and
TVP1022?doxorubicin-treated cultures at 0 h and at 48 h. All the values
are reported as mean ? S.E.M. of five to seven experiments. For each
parameter, values were normalized to 0 h. ??, P ? 0.01 versus 0 h; ???,
P ? 0.001 versus 0 h. Dox, doxorubicin; TVP, TVP1022.
Fig. 8. The effects of TVP1022 on the contraction parameters of hESC-CM.
A, a representative experiment demonstrating that TVP1022 did not
affect the contraction amplitude of a 32-day-old hESC-CM paced at
0.5 Hz. EB, embryoid body. B, a summary of the effects of TVP1022 on the
contraction parameters on hESC-CM (n ? 5 embryoid bodies). In A and
B, the effects of TVP1022 are presented as the percentage of change of
their respective controls. For each drug concentration, the mean ? S.E.M.
TVP1022 Attenuates Doxorubicin-Induced Cardiotoxicity
traction and relaxation, and the decrease in contraction am-
plitude; 2) TVP1022 attenuated doxorubicin-induced reduc-
tion in the protein expression of SERCA2 and NCX1; 3)
TVP1022 attenuated doxorubicin-induced reduction in inter-
cellular coupling measured by Lucifer yellow transfer; ac-
cordingly, TVP1022 prevented doxorubicin-induced decrease
in total Cx43 protein expression; 4) TVP1022 attenuated the
doxorubicin-induced reduction in conduction velocity, QRS
amplitude, and dV/dtmaxof the extracellular electrograms;
and 5) TVP1022 did not affect the [Ca2?]itransients and
contractions of hESC-CM and NRVM, the expression of the
Ca2?-handling proteins, and the MEA-recorded conduction
and activation properties.
TVP1022 Attenuated the Deleterious Effects of Doxo-
rubicin on the [Ca2?]iTransients and Contractions of
NRVM. In the present study we showed that doxorubicin ele-
vated diastolic [Ca2?]i, decreased the maximal rate of [Ca2?]i
activation, and prolonged [Ca2?]irelaxation time. Concomi-
tantly, doxorubicin decreased the maximal rates of contraction
agreement with previous reports demonstrating that doxorubi-
cin adversely affects [Ca2?]ihandling and contractions in
NRVM (Maeda et al., 1999; Shneyvays et al., 2001) and in adult
cardiomyocytes from different species (Maeda et al., 1998; Mi-
jares and Lopez, 2001). In agreement with our recent report
showing that TVP1022 attenuated apoptosis induced by
doxorubicin in NRVM (Kleiner et al., 2008), this study
shows that TVP1022 diminished doxorubicin-induced del-
eterious effects on [Ca2?]itransients and contractions.
TVP1022 Attenuated the Deleterious Effects of Doxo-
rubicin on the [Ca2?]iHandling Proteins. To decipher
the mechanisms underlying the adverse effects of doxorubi-
cin on the [Ca2?]itransients and contractions, we measured
its effect on key [Ca2?]i-handling proteins. Doxorubicin de-
creased the expression of SERCA2 and NCX1, but did not
affect the expression of RyR, calsequestrin, or phospholam-
ban. Accordingly, previous studies showed that doxorubicin
decreased SERCA2 mRNA and protein expression in cul-
tured NRVM (Arai et al., 2000), adult rat ventricular cardio-
myocytes (Timolati et al., 2006), and rabbit in vivo model of
doxorubicin cardiotoxicity (Arai et al., 1998; Olson et al.,
2005). Furthermore, doxorubicin reduced NCX1 expression
(Fig. 4D), in agreement with previous findings showing sim-
ilar effects in atrial and ventricular preparations isolated
from doxorubicin-treated rabbits (Olson et al., 2005) and in
dog heart sarcolemmal vesicles (Caroni et al., 1981). Collec-
tively, the reduction in SERCA2 and NCX1 expression can
explain the elevated diastolic [Ca2?]iand the slowing of
[Ca2?]irelaxation kinetics, as well as reduced rate of myocyte
relaxation caused by doxorubicin. The increased diastolic
[Ca2?]ican also be accounted for by the findings that doxo-
rubicin increased [3H]ryanodine binding to the RyR, and the
increased RyR channel open probability (Pessah et al., 1990;
Saeki et al., 2002). These observations indicate that doxoru-
bicin alters RyR function, and not necessarily protein con-
tent, which is in agreement with our finding that doxorubicin
did not affect RyR protein expression [but contrasted with
the decrease in RyR protein expression in rabbit hearts
(Olson et al., 2005)]. Hence, the increased diastolic [Ca2?]i
and reduced rate of [Ca2?]iactivation may lead to the reduc-
tion in the contraction rate of myocytes. Our findings that
doxorubicin did not affect calsequestrin or phospholamban
expression are in partial agreement with those of Olson et al.
(2005) who found that the protein level of phospholamban
was unchanged and that of calsequestrin was decreased by
doxorubicin in rabbit heart model of cardiotoxicity. The dis-
similarities between our findings and those of Olson et al. can
be due to the difference in the experimental models: in vivo
versus in vitro, adult versus neonatal hearts, and rabbit
versus rat. Furthermore, Burke et al. (2002) reported that
neonatal rabbit sarcoplasmic reticulum (SR) is less sensitive
to anthracycline cardiotoxicity, which can further explain the
differences in the effects of doxorubicin in the diverse exper-
imental models. In agreement with the ability of TVP1022 to
attenuate the adverse effects of doxorubicin on the [Ca2?]i
transients and contractions, the drug also prevented the
decline in SERCA2 and NCX1 expression.
TVP1022 Attenuated the Adverse Effects of Doxoru-
bicin on Intercellular Coupling, Cx43 Expression, Ac-
tivation, and Conduction Velocity. A key ventricular
function adversely affected by doxorubicin is intercellular
coupling. As shown by the Lucifer yellow transfer experi-
ments, doxorubicin reduced intercellular coupling, as indi-
cated by a 36% decline in the Lucifer yellow transfer distance
(Fig. 5). It is noteworthy that doxorubicin did not amplify the
uncoupling effects of carbenoxolone, suggesting that carben-
oxolone maximally uncouples the cells and the doxorubicin
effect is swamped out. Furthermore, because TVP1022 did
not rescue the uncoupling due to carbenoloxone, but it res-
cued the doxorubicin-mediated loss of coupling, it is likely
that TVP1022 and carbenoxolone operate by different mech-
Hence, the decrease in intercellular coupling by doxorubi-
cin probably results from reduced Cx43 protein content, open
Fig. 9. The safety of TVP1022 is demonstrated in NRVM. Cultures were
treated with TVP1022 (1 ?M) for 48 h, and the effects were compared
with control cultures. A, [Ca2?]itransients (n ? 8–10 cultures) and
contraction parameters (n ? 5–6 cultures). B, the protein expression of
the [Ca2?]i-handling proteins (see details in Fig. 4) (n ? 3–4 cultures).
C, Lucifer yellow (n ? 6–8 cultures). D, total Cx43 and NP Cx43 expres-
sion (n ? 6 cultures). In A through C, the effects of TVP1022 are pre-
sented as the percentage of change of their respective controls. LY,
Berdichevski et al.
probability, and/or gap junctional permeability (Ek-Vitorin et
al., 2006). Furthermore, Valiunas et al. (2002) showed a
linear relation between gap junctional permeability and Lu-
cifer yellow transfer and conductance in Cx43-expressing
HeLa cells. Hence, the 39 and 56% reduction in total and NP
Cx43 protein expression, respectively (Fig. 6), shown here
can account (at least, in part) for the reduction in Lucifer
yellow transfer by doxorubicin.
As illustrated in Fig. 7, the reduction in the intercellular
coupling and Cx43 protein content was associated with a 50
to 60% decrease in the QRS amplitude, dV/dtmax, and con-
duction velocity at the cell network level. In agreement with
the alleviating effects of TVP1022 on NRVM function shown
previously, the drug attenuated doxorubicin-induced reduc-
tion in Lucifer yellow transfer, total Cx43 (but not NP CX43),
as well as the decrease in conduction velocity, QRS ampli-
tude, and dV/dtmax. In this regard, it was reported that the
relations between Cx43 phosphorylation state and intercel-
lular coupling are controversial, and that phosphorylation
modulates gap junctional conduction in both positive and
negative fashions, depending on the activity of different
kinases (Herve ´ and Sarrouilhe, 2002; Lampe and Lau,
2000; Lampe et al., 2000). The increase in total Cx43 in the
TVP1022 ? doxorubicin group (compared with the doxoru-
bicin group), but not in NP Cx43, led us to conclude that
TVP1022 elevated phosphorylated Cx43. Hence, the
TVP1022-induced augmentation in phosphorylated Cx43
probably contributed to the increased intercellular cou-
pling, resulting in increased activation and propagation
parameters. This conclusion is supported by the following
findings: 1) an increase in phosphorylated Cx43 protein
expression with no parallel increase in NP Cx43 is posi-
tively correlated with synchronized contraction, reflecting
better activation and propagation patterns in NRVM
(Oyamada et al., 1994); 2) we previously reported that, in
NRVM, increased Cx43 density is associated with higher
conduction velocity, QRS amplitude, and dV/dtmax(Meiry et
al., 2001); 3) slower conduction velocity was demonstrated in
myocytes from the Cx43 homozygote neonatal (Beauchamp et
al., 2004) and heterozygote neonatal and adult (Guerrero et
al., 1997) knockout mice compared with wild-type mice;
4) the conduction velocity in Cx43 mutant gene-modified
NRVM was more than 3-fold slower relative to controls,
indicating the central role of Cx43 in electrical impulse con-
duction (Kizana et al., 2007). However, in contrast with these
findings, it should be noted that Thomas et al. (2003) showed
that a 50% reduction in Cx43 levels did not result in a
reduced conduction velocity. In summary, based on our cur-
rent findings as well as on other reports, we suggest that by
increasing phosphorylated Cx43, TVP1022 alleviated the
adverse effects of doxorubicin on intercellular coupling,
activation, and conduction velocity. The mechanism
whereby TVP1022 increased phosphorylated Cx43 is yet to
Potential Protective Mechanisms of TVP1022
Against the Adverse Effects of Doxorubicin. The molec-
ular basis of the cardiotoxicity induced by doxorubicin re-
mains a matter of debate, and has been attributed to a large
number of effects. The proposed mechanisms of doxorubicin
cardiotoxicity include the generation of a secondary alcohol
metabolite doxorubicinol (doxol), which was found to dereg-
ulate calcium and iron homeostasis in cardiomyocytes (Mi-
notti et al., 2004; Menna et al., 2008), receptor-mediated and
direct effect on proteins (Pessah et al., 1990; Saeki et al.,
2002), and apoptosis (Kleiner et al., 2008, Ueno et al., 2006).
Although a number of mechanisms that lead to doxorubicin
cardiotoxicity have been proposed, most studies support the
view that a common trigger of doxorubicin cardiotoxicity is
linked to an oxidative stress caused by the production of
reactive oxygen species (ROS) (Doroshow, 1983; Sarvazyan,
1996; Singal et al., 2000; Takemura and Fujiwara, 2007). In
this regard, Kim et al. (2006) demonstrated in rat cardio-
myocytes that doxorubicin-mediated ROS formation in-
creased SR Ca2?release, which resulted in further ROS
generation. In addition, [Ca2?]ioverload in cardiac cells may
render mitochondrial Ca2?overloading, resulting in modu-
lated mitochondrial permeability by opening the mitochon-
drial permeability transition pores. This in turn leads to
cytochrome c release, which may culminate in apoptosis. Kim
et al. (2006) also showed that increased caspase 3 activity
was doxorubicin/ROS-mediated, via SR Ca2?release, indi-
cating the important role of Ca2?in cardiomyocyte apoptosis.
Respecting the cytoprotective efficacy of TVP1022, this mol-
ecule was shown to induce superoxide dismutase and cata-
lase in neuronal cultures and in vivo (Youdim et al., 2001), to
inhibit the apoptotic cascade and to protect mitochondrial
viability in NRVM and neuronal cells (Youdim et al., 2001,
2003; Youdim and Weinstock, 2001; Kleiner et al., 2008).
Collectively, these activities probably caused ROS reduction
and improved myocyte viability, thereby attenuating doxoru-
bicin-induced cardiotoxicity. However, despite this support-
ing evidence, the interaction between TVP1022 and ROS
resulting in attenuation of doxorubicin cardiotoxicity was not
determined directly in this study, and other possible mecha-
nisms of action may account for the beneficial effects of
The Safety of TVP1022. As clearly depicted in this study,
TVP1022 did not affect the measured functional properties of
NRVM or hESC-CM. These findings bare significant clinical
relevance because they imply that TVP1022 may potentially
be safely administered to the patients treated with anthra-
cyclines. In further support of the suitability of TVP1022 for
providing cardioprotection to doxorubicin-treated cancer pa-
tients, we recently found that TVP1022 (0.01, 0.1, and 1 ?M)
neither caused proliferation of the human cancer cell lines
HeLa and MDA-231 nor diminished the anticancer efficacy of
doxorubicin (Kleiner et al., 2008).
Study Limitation. This study demonstrated the cardio-
protective efficacy of TVP1022 against doxorubicin toxicity in
vitro in NRVM. A key limitation of this, as well as many
other in vitro studies investigating disease settings, is that
most pathologies in humans develop over time, whereas in
vitro experiments in cell culture are naturally short-termed
(in our study, 24 h). In contrast to the in vitro scenario, the
clinical signs of anthracycline-induced cardiotoxicity are as-
sociated with cumulative drug exposure, and develop on a
time scale of months/years. In this regard, even animal mod-
els of doxorubicin-induced cardiotoxicity are somewhat lim-
ited because cardiac damage develops within days/weeks.
Hence, in view to these limitations, the results of this study
should be carefully interpreted.
In summary, this study demonstrated that TVP1022 at-
tenuated doxorubicin-induced adverse effects on functional
properties of NRVM. Based on these encouraging findings,
TVP1022 Attenuates Doxorubicin-Induced Cardiotoxicity
TVP1022 can be considered as a novel cardioprotective mol- Download full-text
ecule against anthracycline cardiotoxicity, and may thus be
administered in combination with doxorubicin in the treat-
ment of different malignancies in humans.
Arai M, Tomaru K, Takizawa T, Sekiguchi K, Yokoyama T, Suzuki T, and Nagai R
(1998) Sarcoplasmic reticulum genes are selectively down-regulated in cardio-
myopathy produced by doxorubicin in rabbits. J Mol Cell Cardiol 30:243–254.
Arai M, Yoguchi A, Takizawa T, Yokoyama T, Kanda T, Kurabayashi M, and Nagai
R (2000) Mechanism of doxorubicin-induced inhibition of sarcoplasmic reticulum
Ca2?-ATPase gene transcription. Circ Res 86:8–14.
Beauchamp P, Choby C, Desplantez T, de Peyer K, Green K, Yamada KA, Weingart
R, Saffitz JE, and Kle ´ber AG (2004) Electrical propagation in synthetic ventricular
myocyte strands from germline connexin43 knockout mice. Circ Res 95:170–178.
Burke BE, Mushlin PS, Cusack BJ, Olson SJ, Gambliel HA, and Olson RD (2002)
Decreased sensitivity of neonatal rabbit sarcoplasmic reticulum to anthracycline
cardiotoxicity. Cardiovasc Toxicol 2:41–51.
Caroni P, Villani F, and Carafoli E (1981) The cardiotoxic antibiotic doxorubicin
inhibits the Na?/Ca2?exchange of dog heart sarcolemmal vesicles. FEBS Lett
Dolnikov K, Shilkrut M, Zeevi-Levin N, Gerecht-Nir S, Amit M, Danon A, Itskovitz-
Eldor J, and Binah O (2006) Functional properties of human embryonic stem
cell-derived cardiomyocytes: intracellular Ca2?handling and the role of sarcoplas-
mic reticulum in the contraction. Stem Cells 24:236–245.
Doroshow JH (1983) Effect of anthracycline antibiotics on oxygen radical formation
in rat heart. Cancer Res 43:460–472.
Ek-Vitorin JF, King TJ, Heyman NS, Lampe PD, and Burt JM (2006) Selectivity of
connexin 43 channels is regulated through protein kinase C-dependent phosphor-
ylation. Circ Res 98:1498–1505.
Guerrero PA, Schuessler RB, Davis LM, Beyer EC, Johnson CM, Yamada KA, and
Saffitz JE (1997) Slow ventricular conduction in mice heterozygous for a
connexin43 null mutation. J Clin Invest 99:1991–1998.
Herve ´ JC and Sarrouilhe D (2002) Modulation of junctional communication by
phosphorylation: protein phosphatases, the missing link in the chain. Biol Cell
Kavanagh TJ, Martin GM, el-Fouly MH, Trosko JE, Chang CC, and Rabinovitch PS
(1987) Flow cytometry and scrape-loading/dye transfer as a rapid quantitative
measure of intercellular communication in vitro. Cancer Res 47:6046–6051.
Kim SY, Kim SJ, Kim BJ, Rah SY, Chung SM, Im MJ, and Kim UH (2006)
Doxorubicin-induced reactive oxygen species generation and intracellular Ca2?
increase are reciprocally modulated in rat cardiomyocytes. Exp Mol Med 38:535–
Kizana E, Chang CY, Cingolani E, Ramirez-Correa GA, Sekar RB, Abraham MR,
Ginn SL, Tung L, Alexander IE, and Marba ´n E (2007) Gene transfer of connexin43
mutants attenuates coupling in cardiomyocytes: novel basis for modulation of
cardiac conduction by gene therapy. Circ Res 100:1597–1604.
Kleiner Y, Bar-Am O, Amit T, Berdichevski A, Liani E, Maor G, Reiter I, Youdim
MB, and Binah O (2008) TVP1022 and propargylamine protect neonatal rat
ventricular myocytes against doxorubicin-induced and serum starvation-induced
cardiotoxicity. J Cardiovasc Pharmacol 52:268–277.
Lampe PD and Lau AF (2000) Regulation of gap junctions by phosphorylation of
connexins. Arch Biochem Biophys 384:205–215.
Lampe PD, TenBroek EM, Burt JM, Kurata WE, Johnson RG, and Lau AF (2000)
Phosphorylation of connexin43 on serine368 by protein kinase C regulates gap
junctional communication. J Cell Biol 149:1503–1512.
Maeda A, Honda M, Kuramochi T, and Takabatake T (1998) Doxorubicin cardiotox-
icity: diastolic cardiac myocyte dysfunction as a result of impaired calcium han-
dling in isolated cardiac myocytes. Jpn Circ J 62:505–511.
Maeda A, Honda M, Kuramochi T, and Takabatake T (1999) A calcium antagonist
protects against doxorubicin-induced impairment of calcium handling in neonatal
rat cardiac myocytes. Jpn Circ J 63:123–129.
Matemba SF, Lie A, and Ransjo ¨ M (2006) Regulation of osteoclastogenesis by gap
junction communication. J Cell Biochem 99:528–537.
Meiry G, Reisner Y, Feld Y, Goldberg S, Rosen M, Ziv N, and Binah O (2001)
Evolution of action potential propagation and repolarization in cultured neonatal
rat ventricular myocytes. J Cardiovasc Electrophysiol 12:1269–1277.
Menna P, Salvatorelli E, and Minotti G (2008) Cardiotoxicity of antitumor drugs.
Chem Res Toxicol 21:978–989.
Mijares A and Lo ´pez JR (2001) L-Carnitine prevents increase in diastolic [Ca2?]
induced by doxorubicin in cardiac cells. Eur J Pharmacol 425:117–120.
Minotti G, Recalcati S, Menna P, Salvatorelli E, Corna G, and Cairo G (2004)
Doxorubicin cardiotoxicity and the control of iron metabolism: quinone-dependent
and independent mechanisms. Methods Enzymol 378:340–361.
Olson RD, Gambliel HA, Vestal RE, Shadle SE, Charlier HA Jr, and Cusack BJ
(2005) Doxorubicin cardiac dysfunction: effects on calcium regulatory proteins,
sarcoplasmic reticulum, and triiodothyronine. Cardiovasc Toxicol 5:269–283.
Oyamada M, Kimura H, Oyamada Y, Miyamoto A, Ohshika H, and Mori M (1994)
The expression, phosphorylation, and localization of connexin 43 and gap-
junctional intercellular communication during the establishment of a synchro-
nized contraction of cultured neonatal rat cardiac myocytes. Exp Cell Res 212:351–
Pessah IN, Durie EL, Schiedt MJ, and Zimanyi I (1990) Anthraquinone-sensitized
Ca2?release channel from rat cardiac sarcoplasmic reticulum: possible receptor-
mediated mechanism of doxorubicin cardiomyopathy. Mol Pharmacol 37:503–514.
Rubin Y, Kessler-Icekson G, and Navon G (1995) The effect of furosemide on calcium
ion concentration in myocardial cells. Cell Calcium 18:135–139.
Saeki K, Obi I, Ogiku N, Shigekawa M, Imagawa T, and Matsumoto T (2002)
Doxorubicin directly binds to the cardiac-type ryanodine receptor. Life Sci 70:
Sarvazyan N (1996) Visualization of doxorubicin-induced oxidative stress in isolated
cardiac myocytes. Am J Physiol 271:H2079–H2085.
Sedan O, Dolnikov K, Zeevi-Levin N, Leibovich N, Amit M, Itskovitz-Eldor J, and
Binah O (2008) 1,4,5-Inositol trisphosphate operated [Ca2?]istores and angioten-
sin-II/endothelin-1 signaling pathway are functional in human embryonic stem
cell-derived cardiomyocytes. Stem Cells 26:3130–3138.
Shneyvays V, Mamedova L, Zinman T, Jacobson K, and Shainberg A (2001) Activa-
tion of A(3)adenosine receptor protects against doxorubicin-induced cardiotoxicity.
J Mol Cell Cardiol 33:1249–1261.
Singal PK and Iliskovic N (1998) Doxorubicin-induced cardiomyopathy. N Engl
J Med 339:900–905.
Singal PK, Li T, Kumar D, Danelisen I, and Iliskovic N (2000) Adriamycin-induced
heart failure: mechanism and modulation. Mol Cell Biochem 207:77–86.
Swain SM, Whaley FS, and Ewer MS (2003) Congestive heart failure in patients
treated with doxorubicin: a retrospective analysis of three trials. Cancer 97:2869–
Takemura G and Fujiwara H (2007) Doxorubicin-induced cardiomyopathy from the
cardiotoxic mechanisms to management. Prog Cardiovasc Dis 49:330–352.
Thomas SP, Kucera JP, Bircher-Lehmann L, Rudy Y, Saffitz JE, and Kle ´ber AG
(2003) Impulse propagation in synthetic strands of neonatal cardiac myocytes with
genetically reduced levels of connexin43. Circ Res 92:1209–1216.
Timolati F, Ott D, Pentassuglia L, Giraud MN, Perriard JC, Suter TM, and Zup-
pinger C (2006) Neuregulin-1 beta attenuates doxorubicin-induced alterations of
excitation-contraction coupling and reduces oxidative stress in adult rat cardio-
myocytes. J Mol Cell Cardiol 41:845–854.
Ueno M, Kakinuma Y, Yuhki K, Murakoshi N, Iemitsu M, Miyauchi T, and Yamagu-
chi I (2006) Doxorubicin induces apoptosis by activation of caspase-3 in cultured
cardiomyocytes in vitro and rat cardiac ventricles in vivo. J Pharmacol Sci 101:
Valiunas V, Beyer EC, and Brink PR (2002) Cardiac gap junction channels show
quantitative differences in selectivity. Circ Res 91:104–111.
Youdim MB, Amit T, Falach-Yogev M, Bar Am O, Maruyama W, and Naoi M (2003)
The essentiality of Bcl-2, PKC and proteasome-ubiquitin complex activations in
the neuroprotective-antiapoptotic action of the anti-Parkinson drug, rasagiline.
Biochem Pharmacol 66:1635–1641.
Youdim MB, Wadia A, Tatton W, and Weinstock M (2001) The anti-Parkinson drug
rasagiline and its cholinesterase inhibitor derivatives exert neuroprotection unre-
lated to MAO inhibition in cell culture and in vivo. Ann N Y Acad Sci 939:450–
Youdim MB and Weinstock M (2001) Molecular basis of neuroprotective activities of
rasagiline and the anti-Alzheimer drug TV3326 [(N-propargyl-(3R)aminoindan-5-
YL)-ethyl methyl carbamate]. Cell Mol Neurobiol 21:555–573.
Zeevi-Levin N, Barac YD, Reisner Y, Reiter I, Yaniv G, Meiry G, Abassi Z, Kostin S,
Schaper J, Rosen MR, et al. (2005) Gap junctional remodeling by hypoxia in
cultured neonatal rat ventricular myocytes. Cardiovasc Res 66:64–73.
Address correspondence to: Dr. Ofer Binah, Department of Physiology,
Ruth and Bruce Rappaport Faculty of Medicine, PO Box 9649, Haifa 31096,
Israel. E-mail: Binah@tx.technion.ac.il
Berdichevski et al.