Download Survivin Determines Cardiac Function by Controlling

Survivin Determines Cardiac Function by Controlling
Total Cardiomyocyte Number
Bodo Levkau, MD*; Michael Schäfers, MD*; Jeremias Wohlschlaeger, MD;
Karin von Wnuck Lipinski, PhD; Petra Keul, BS; Sven Hermann, MD; Naomasa Kawaguchi, MD;
Paulus Kirchhof, MD; Larissa Fabritz, MD; Jörg Stypmann, MD; Lars Stegger, MD; Ulrich Flögel, PhD;
Jürgen Schrader, MD; Jens W. Fischer, PhD; Patrick Hsieh, MD, PhD; Yen-Ling Ou, BS; Felix Mehrhof, MD;
Klaus Tiemann, MD; Alexander Ghanem, MD; Marek Matus, PhD; Joachim Neumann, MD;
Gerd Heusch, MD; Kurt W. Schmid, MD; Edward M. Conway, MD, PhD; Hideo A. Baba, MD
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Background—Survivin inhibits apoptosis and regulates cell division in many organs, but its function in the heart is
unknown.
Methods and Results—We show that cardiac-specific deletion of survivin resulted in premature cardiac death. The underlying
cause was a dramatic reduction in total cardiomyocyte numbers as determined by a stereological method for quantification
of cells per organ. The resulting increased hemodynamic load per cell led to progressive heart failure as assessed by
echocardiography, magnetic resonance imaging, positron emission tomography, and invasive catheterization. The reduction
in total cardiomyocyte number in ␣-myosin heavy chain (MHC)–survivin⫺/⫺ mice was due to an ⬇50% lower mitotic rate
without increased apoptosis. This occurred at the expense of DNA accumulation because survivin-deficient cardiomyocytes
displayed marked DNA polyploidy indicative of consecutive rounds of DNA replication without cell division. Survivin small
interfering RNA knockdown in neonatal rat cardiomyocytes also led to polyploidization and cell cycle arrest without
apoptosis. Adenoviral overexpression of survivin in cardiomyocytes inhibited doxorubicin-induced apoptosis, induced DNA
synthesis, and promoted cell cycle progression. The phenotype of the ␣MHC-survivin⫺/⫺ mice also allowed us to determine
the minimum cardiomyocyte number sufficient for normal cardiac function. In human cardiomyopathy, survivin was potently
induced in the failing heart and downregulated again after hemodynamic support by a left ventricular assist device. Its
expression positively correlated with the mean cardiomyocyte DNA content.
Conclusions—We suggest that the ontogenetically determined cardiomyocyte number may be an independent factor in the
susceptibility to cardiac diseases. Through its profound impact on both cardiomyocyte replication and apoptosis,
survivin may emerge as a promising new target for myocardial regeneration. (Circulation. 2008;117:1583-1593.)
Key Words: apoptosis 䡲 cardiomyopathy 䡲 heart-assist device 䡲 heart failure 䡲 myocardium
䡲 physiology 䡲 transplantation
H
human heart are unknown, a dynamic balance exists between
cardiomyocyte loss and replacement that controls cardiac fate
during life and disease.2
eart failure is associated with a high mortality rate and
poor quality of life. Its incidence is rapidly increasing,
accounting for 20% of all hospital admissions in individuals
⬎65 years of age in the United States.1 Apoptosis of
cardiomyocytes leading to loss of contractile units has been
implicated in the pathogenesis of heart failure. Although the
exact stimuli, mechanisms, and rate of apoptosis in the adult
Clinical Perspective p 1593
Caspases, the common executioners of the apoptotic program, are normally held in check by the inhibitor of apoptosis
Received August 15, 2007; accepted December 28, 2007.
From the Institut für Pathophysiologie (B.L., K.v.W.L., P. Keul, G.H.) and Institut für Pathologie (J.W., N.K., K.W.S., H.A.B.), Universitätsklinikum
Essen, Essen Germany; Klinik und Poliklinik für Nuklearmedizin (M.S., S.H., L.S.) and Medizinische Klinik und Poliklinik C (P. Kirchhof, L.F., J.
Stypmann), Universitätsklinikum Münster, Münster, Germany; Institut für Herz und Kreislaufphysiologie (U.F., J. Schroder) and Institut für
Pharmakologie und Klinische Pharmakologie (J.W.F.), Universität Düsseldorf, Düsseldorf, Germany; Institute of Clinical Medicine and Center of
Micro/Nano Science and Technology, National Cheng Kung University, Tainan City, Taiwan (P.H., Y.-L.O.); Medizinische Klinik mit Schwerpunkt
Kardiologie, Universitätsklinikum Charité, Berlin, Germany (F.M.); Medizinische Klinik und Poliklinik II, Universitätsklinikum Bonn, Bonn, Germany
(K.T., A.G.); Institut für Pharmakologie und Toxikologie, Universitätsklinikum Münster, Münster, Germany (M.M.); Institut für Pharmakologie und
Toxikologie, Universität Halle Wittenberg, Wittenberg, Germany (J.N.); and Center for Transgene Technology and Gene Therapy, Flanders
Interuniversity Institute for Biotechnology, University of Leuven, Leuven, Belgium (E.M.C.).
The online Data Supplement can be found with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.107.734160/DC1.
*The first 2 authors contributed equally to this work.
Correspondence to Bodo Levkau, Institute of Pathophysiology, University Hospital Essen, Hufelandstrasse 55, 45122 Essen, Germany. E-mail
[email protected]
© 2008 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org
DOI: 10.1161/CIRCULATIONAHA.107.734160
1583
1584
Circulation
March 25, 2008
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
proteins.3 Survivin is a member of the inhibitor of apoptosis
protein family with a unique structure devoid of the caspasebinding region present in all other inhibitor of apoptosis proteins.4 Nevertheless, it efficiently inhibits mitochondrial apoptosis by inhibiting caspases.5,6 In addition, survivin regulates
caspase-independent cell death and mitochondrial membrane
permeabilization.7 Apart from its role in cell death, survivin has
an important function in cell division: It controls multiple phases
of mitosis by regulating the spindle assembly checkpoint, microtubule stability, metaphase spindle formation, and the chromosomal passenger proteins aurora B kinase and INCENP.4,8
Accordingly, survivin is ubiquitously expressed during development and in a variety of human malignancies.4,8 The unfavorable
outcome associated with high survivin expression in clinical
oncology studies has made it a bona fide target for cancer drug
development.4
However, several differentiated adult cell types also use
survivin in physiological settings unrelated to cancer. Neutrophils require it for survival in acute inflammatory diseases9; T cells, for maturation10; and hematopoietic progenitor
cells, for proliferation.11 Survivin is also detectable in nonproliferating, terminally differentiated tissues such as the
myocardium, where it is induced in aged and failing hearts of
spontaneously hypertensive rats12 and upregulated after myocardial infarction in humans.13 However, the biological function of survivin in the heart is unknown. By generating
cardiomyocyte-specific survivin-deficient mice, we provide
evidence that survivin controls cardiomyocyte proliferation
and ploidy and that its loss leads to progressive heart failure
through a reduction in total cardiomyocyte numbers. In
addition, we identify unique roles of survivin in cardiomyocyte DNA content, proliferation, and death.
Methods
A detailed Material and Methods section is available in the onlineonly Data Supplement.
Generation of Mice With Cardiac-Specific Deletion
of Survivin
Mice homozygous for a floxed survivin allele10 were crossed with
heterozygous mice that express Cre recombinase under the control of
the ␣-myosin heavy chain (␣MHC-Cre).14 For genotyping of microdissected cardiomyocytes, deletion-specific polymerase chain reaction (PCR) was performed. For single-cell laser microdissection, a
PALM Robot-Microbeam (PALM GmbH, Bernried, Germany) was
applied, and 60 individual cardiomyocytes or interstitial cells were
separately dissected and pooled.
Mean Cardiomyocyte Diameter and Calculation of
Total Cardiomyocyte Number per Heart
The mean cardiomyocyte diameter and length were determined by
measuring 100 cardiomyocytes on periodic acid-Schiff–stained sections with an image analysis program (KS 300, Zeiss, Germany). To
calculate the absolute number of cardiomyocytes per heart, an
established 3-dimensional stereological method was used.15,16
Briefly, the volume fraction of cardiomyocytes (Vv Myo) in 10
randomly selected visual fields was determined by the principle of
Delesse (area density⫽volume density). A grid containing 513
points was laid over the images, and the points encountering
cardiomyocytes, blood vessels, and connective tissue were counted.
Vv Myo was calculated as a percentage. The mean cardiomyocyte
volume (V Myo) was calculated as follows: V Myo⫽␲ (mean
diametercardiomyocyte/2)2⫻mean lengthcardiomyocyte. The absolute number of
cardiomyocytes per left ventricle (N Myo) was calculated from the
following: N Myo⫽(Vv Myo⫻LV volume)/V Myo, where the total
tissue volume of the left ventricle (LV volume) was obtained by
dividing its weight by specific gravity (1.0048)15,16 and the cell
number is given in 106. In embryo studies, the whole embryos were
embedded and serial sections were performed parallel to the longitudinal axis, yielding ⬇60 sections. Cardiomyocytes were counted
on every fifth slide containing cardiac tissue.
Immunohistochemistry, Terminal Deoxynucleotidyl
Transferase–Mediated dUTP Nick-End Labeling,
Fibrosis, and Measurement of DNA Content
Immunohistochemistry for survivin was performed with polyclonal
(Acris, Hiddenhausen, Germany) and monoclonal (60.11, Novus,
Littleton, Colo) antibodies for mouse and human tissue, respectively.
Terminal deoxynucleotidyl transferase–mediated dUTP nick-end
labeling (TUNEL) was performed with ApopTaq Plus (Oncor,
Gaithersburg, Md) and FITC–anti-digoxigenin (Roche, Mannheim,
Germany). The percentage of TUNEL-positive nuclei was calculated
in 10 visual fields. Fibrosis (mean percentage of area) was determined with Sirius Red. DNA content was determined by automated
DNA cytometry.17 Briefly, the Feulgen reaction was applied to
nuclei isolated from three 50-␮m paraffin sections after hydrolysis,
stained with Schiff’s reagent, and analyzed with DNA cytometry
software (CYDOK, Fa. Hilgers, Königswinter, Germany).17
Magnetic Resonance Imaging, Positron Emission
Tomography, Echocardiography, and In Vivo
Hemodynamic Measurements
Magnetic resonance imaging (MRI) was performed with a Bruker
DRX 9.4-T wide-bore nuclear magnetic resonance spectrometer as
described.18 Positron emission tomography (PET) was performed
with [18F]FDG and a small animal camera (quadHIDAC, Oxford
Positron, Oxford, England). High-resolution echocardiography was
performed with an ultrasound device with frame rates up to 280 Hz
(Philips Medical Systems, Bothell, Wash). Left ventricular catheterization was performed in closed-chest mice as described.19 Increasing doses of dobutamine were perfused into the left jugular vein,
accompanied by measurements of heart rate, maximal left ventricular
pressure, and the first derivative of left ventricular pressure.19
Cardiomyocyte Culture, Small Interfering RNA
Transfection, and Adenoviral Infection
Rat neonatal cardiomyocytes were infected with adenovirus carrying
survivin or green fluorescent protein (GFP) for 24 hours as described.20 Apoptosis was induced with 1 ␮mol/L doxorubicin, and
DNA fragmentation was examined by flow cytometry. [3H]thymidine incorporation21 and small interfering RNA (siRNA) transfection22 were performed as described. Cell cycle profiles were analyzed in an EpicsXL flow cytometer (Beckman Coulter, Fullerton,
Calif).
Studies in Patients With Heart Failure and Left
Ventricular Assist Device Support
Ten male patients underwent left ventricular assist device (LVAD)
implantation for therapy of end-stage chronic heart failure as a bridge
to transplantation (7 patients received a Novacor N100 [Baxter
Healthcare Corp, Novacor, Oakland, Calif], 3 patients received a
DeBakey/NASA device [MicroMed Cardiovascular Inc, Houston,
Tex]). Four patients suffered from dilated cardiomyopathy and 6
from ischemic heart disease. The mean duration of LVAD support
was 129.9 days (median, 76 days; range, 17 to 298 days). Five donor
hearts served as controls. The numerical density of positive cells per
visual field of defined size was determined following the rules of the
forbidden and permitted lines in subepicardial, midendocardial, and
subendocardial areas in 7 randomly selected visual fields in a blinded
fashion. The present study was performed according to the Declaration of Helsinki. Written consent was obtained from each patient.
Levkau et al
A
flo
cardiomyocytes
Survivin Determines Cardiomyocyte Number In Vivo
d
ed
xe elet re
C
d
1585
B
500 425 389 bp
*
interstitial cells
wild type
C
wild type
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
MRI
D
PET
αMHC-survivin-/- wild type
αMHC-survivin-/-
αMHC-survivin-/-
MRI
PET
Figure 1. Generation of cardiomyocyte-specific survivin-deficient mice. A, Ablation of survivin in cardiomyocytes of ␣MHC-survivin⫺/⫺
mice is confirmed by single-cell PCR of laser-microdissected cardiomyocytes (top) and interstitial cells (bottom). Nuclei were labeled for
excision (red circles), and PCR was performed for floxed and deleted survivin (500-bp product when the allele was intact, 425-bp product when excised). PCR also was performed for Cre (lane 3). B, Heart pathology of a 30-month-old ␣MHC-survivin⫺/⫺ mouse and a
wild-type littermate control. Echocardiography shows long-axis view with a large thrombus (*) in the right atrium and tricuspid regurgitation with an enlarged pulmonary trunk (black arrow, right ventricular inflow; black arrowhead, LV inflow). Note the massive enlargement and dilation of both ventricles and atria with fresh and old organized thrombi. C, D, Representative MRI and 18F-FDG-PET images
of a 20-week-old ␣MHC-survivin⫺/⫺ mouse and a wild-type control.
Statistical Analysis
All data are expressed as mean⫾SEM and depicted as box plots
when appropriate. Student’s t test was used to evaluate statistical
significance between the different genotypes for all hemodynamic
and morphometry data, including heart weight. Statistical significance between different time points was determined by use of 1-way
ANOVA followed by multiple-comparison procedures according to
Duncan. Overall survival curves were estimated with the Kaplan–
Meier method, and differences between genotypes were compared
by the log-rank test. The nonparametric Wilcoxon test for paired
samples was used to evaluate statistical significance for survivin
immunoreactivity in the myocardium before and after LVAD support. Intergroup differences among samples before and after LVAD
and controls were calculated by 1-way ANOVA followed by
post-hoc analysis according to Duncan. Correlation analysis was
performed according to Spearman. A value of P⬍0.05 was considered significant.
The authors had full access to and take responsibility for the
integrity of the data. All authors have read and agree to the
manuscript as written.
Results
Progressive Heart Failure and Death in
␣MHC-Survivinⴚ/ⴚ Mice
Mice deficient for cardiac survivin were generated by crossing survivin flox/flox mice10 with mice expressing Cre
recombinase under the cardiac-specific ␣MHC promoter.14
Laser capture microdissection and single-cell PCR (Figure
1A) and immunohistochemistry (Figure 5) revealed that
survivin has been exclusively deleted in cardiomyocytes and
not in cardiac fibroblasts of ␣MHC-survivin⫺/⫺ mice. Although survivin-deficient mice were born with the expected
mendelian frequency and were indistinguishable from their
Cre-negative littermates at birth, they died prematurely with a
median survival of 34 weeks (Figure 2A). Before death,
␣MHC-survivin⫺/⫺ mice developed a characteristic syndrome
consisting of decreased activity, tachypnea, hunched posture,
and poor grooming. Serial echocardiography uncovered massive enlargement of all cardiac cavities, pericardial effusions,
and atrial thrombi (Figure 1B). On autopsy, all chambers of
the heart were enlarged (Figure 1B), and pathological findings consistent with decompensated heart failure such as
pericardial and pleural effusions, ascites, congested lungs,
and an enlarged liver were present (data not shown). MRI
confirmed the profound functional and structural cardiac
abnormalities of ␣MHC-survivin⫺/⫺ mice (Figure 1C and
1D). These abnormalities were not due to regional myocardial
viability defects as excluded by [18F]FDG-PET (Figure 1C
and 1D).
Quantitative analysis of cardiac function by MRI revealed
a progressive impairment of cardiac function in ␣MHCsurvivin⫺/⫺ mice at 18 and 36 weeks (Figure 2B). In particular, end-diastolic and end-systolic volumes were significantly higher in ␣MHC-survivin⫺/⫺ mice compared with
controls and increased progressively in an age-dependent
manner (Figure 2B). Whereas stroke volume and cardiac
output were still maintained in 18- and 36-week-old ␣MHC-
Circulation
March 25, 2008
A
C
Heart rate
650
1.1
wild type
*
600
0.9
heart rate
Heart[BPM]
Rat
Cumulative survivial
1.0
0.8
0.7
0.6
αMHC-survivin
-/-
0.5
*
E
*
115
110
105
*
550
Left ventricular pressure
120
LVP
[mmHg]
LVP
[mmHg
1586
500
450
100
95
90
85
80
0.4
400
wild type
αMHC-survivin-/-
0.3
0
10
20
30
75
350
weeks
* *
00
1
1
2
2
0
33
D
36 weeks
36
weeks
wild type
αMHC-survivin-/-
*
*°
20
20
0
0
8000
6000
*
* *
*
* * *
18 weeks
36 weeks
36 weeks
Max. rate of relaxation
-14000
-8000
-6000
-4000
wild type
αMHC-survivin-/-
2000
18 weeks
3
-10000
10000
4000
2
-12000
12000
18 weeks
18
weeks
1
F
Max. rate of contraction
14000
100
100
80
80
60
60
40
40
*
wild type
αMHC-survivin-/-
dobutamine [log(µg/kg)]
dP/dt(max)
[mmHg
dP/dt
min [mmHg]
140
120
120
ESV [µl]
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
40
40
20
20
0
0
*
dP/dt(max)
[mmHg
dP/dt
max [mmHg]
EDV [µl]
140
140
120
120
100
100
80
80
60
60
* * *
70
dobutamine [log(µg/kg)]
B
*
*
*
* *
*
* * *
* *
wild type
αMHC-survivin-/-
-2000
0
0
11
22
dobutamine [log(µg/kg)]
33
0
0
11
22
33
dobutamine [log(µg/kg)]
Figure 2. Functional characteristics. A, Survival analysis of 30 ␣MHC-survivin⫺/⫺ and 35 wild-type mice. B, End-diastolic (EDV) and
end-systolic (ESV) volumes of 18- and 36-week-old ␣MHC-survivin⫺/⫺ mice and controls as assessed by MRI (*P⬍0.05 vs wild type,
°P⬍0.05 between 18 and 36 weeks, Student’s t test). C, Heart rate; E, left ventricular pressure. D and F, Maximum (Max.) rate of contraction and relaxation, respectively, under basal conditions and in response to increasing dobutamine doses in 28-week-old ␣MHCsurvivin⫺/⫺ mice and controls as assessed by invasive hemodynamic measurements (*P⬍0.05, Student’s t test).
survivin⫺/⫺ mice, the mean ejection fraction was decreased by
17% and 29%, respectively, compared with controls, and
fractional shortening and wall thickening were progressively
reduced (Table). Invasive cardiac catheterization performed
to assess basal and dobutamine-stimulated left ventricular
pressures in vivo revealed that in ␣MHC-survivin⫺/⫺ mice
maximal left ventricular pressure and contraction/relaxation
rates were already lower under basal conditions compared
with controls (Figure 2C through 2F). When stimulated with
increasing doses of dobutamine, survivin-deficient hearts
developed higher heart rates, lower left ventricular pressures,
and lower maximum rates of contraction/relaxation compared
with controls (Figure 2C through 2F).
Polyploidy and Reduction in Total Cardiomyocyte
Numbers in ␣MHC-Survivinⴚ/ⴚ Mice
The relative heart weights of ␣MHC-survivin⫺/⫺ and control
mice were similar at any age examined (Figure 3A). However, there was a clear difference in mean cardiomyocyte
diameter in ␣MHC-survivin⫺/⫺ mice compared with controls
already at birth; this difference increased progressively with
age (from 13% higher diameters at birth to 80% at 270 days;
Figure 3B). There was also a 2- to 3-fold increase in
interstitial fibrosis with a reticular pattern and subendocardial
accentuation in survivin-deficient hearts beginning at 28 days
(Figure 3C).
However, the most striking morphological observation was
the marked enlargement of cardiomyocyte nuclei, along with
massive invaginations of the nuclear envelope in survivindeficient hearts (Figure 3D). The appearance of such gigantic
nuclei prompted us to determine whether they contained
higher amounts of DNA. Using automated DNA cytometry,
we calculated that survivin-deficient cardiomyocytes exhibited an accentuated polyploidy: ⬇70% of the cells displayed
a DNA content of ⱖ4n with a few individual cells reaching
even 16n (Figure 3E). The mean cardiomyocyte DNA content
in ␣MHC-survivin⫺/⫺ mice was ⬇2-fold higher compared
with controls at all ages examined (Figure 3F).
One possibility for the accumulation of polyploid DNA
may be the occurrence of DNA duplication without consecutive cell division. Thus, we looked for signs of its ultimate
consequence: the presence of a net reduction in total cardiomyocyte numbers per heart. Using an established
3-dimensional stereological approach,15,16 we compared the
total cardiomyocyte number per heart in control and ␣MHCsurvivin⫺/⫺ mice. Strikingly, ␣MHC-survivin⫺/⫺ mice had
34% fewer cardiomyocytes per heart than controls already at
birth (Figure 4A). This difference became even more pronounced during the following 4 weeks (⬇60% fewer cardiomyocytes in survivin-deficient hearts; Figure 4A). In addition, although the total cardiomyocyte numbers in control
Levkau et al
Table.
1587
Hemodynamic Parameters of Wild-Type and ␣MHC-Survivinⴚ/ⴚ Mice at 18 and 36 Weeks as Measured by Cardiac MRI
EDV, ␮L
Wild type, 18 wk
␣MHC-survivin⫺/⫺, 18 wk
CO, mL/min
FS, %
WT, %
V(myo), ␮L
70
19
39
36
104
85
19
66
78
33
46
90
95
73
16
57
78
28
46
62
107
84
25
59
70
32
40
65
114
62
12
50
80
24
48
88
18⫾2.1
55⫾3.8
75⫾2.1
27⫾2.6
44⫾1.9
68⫾9.9
69
98⫾7.9
87
20
67
77
34
46
110
82
97
39
58
59
25
31
44
100
114
48
67
58
25
31
62
112
89
34
55
62
20
33
53
105
0.02
39
36⫾4.6
0.008
45
58⫾4.0
NS
54
24
62⫾4.0
25⫾2.2
0.02
NS
27
34⫾3.1
0.025
34
61⫾13.2
NS
95
99⫾5.0
NS
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
66
21
45
68
25
38
45
125
93
35
58
62
25
33
39
112
92
20
73
79
34
47
72
111
63
20
43
69
22
38
66
90
86
29
57
66
20
36
46
99
80⫾6.4
␣MHC-survivin⫺/⫺, 36 wk
EF, %
44
85
P
SV, ␮L
19
95⫾5.3
Wild type, 36 wk
ESV, ␮L
63
73⫾4.9
25⫾3.1
55⫾5.2
69⫾2.7
25⫾2.3
38⫾2.3
54⫾6.5
107⫾6.1
105
59
45
43
19
21
21
141
104
38
27
16
12
12
120
116
64
53
45
21
23
5
182
127
77
49
39
24
19
45
142
93
44
49
53
23
27
51
120
135⫾12.9
116⫾8.4
P
Survivin Determines Cardiomyocyte Number In Vivo
0.09
70⫾10.1
0.009
47⫾2.6
41⫾4.3
20⫾1.4
20⫾2.4
27⫾9.2
NS
0.001
NS
0.001
0.04
110
NS
EDV indicates end-diastolic volume; ESV, end-systolic volume; SV, stroke volume; EF, ejection fraction; CO, cardiac output; WT, wall thickening; and V(myo),
myocardial volume.
mice continued to increase during the first 4 weeks after birth,
the number of survivin-deficient cardiomyocytes did not
(Figure 4A). There was an inverse correlation between
cardiomyocyte size and total cardiomyocyte number per heart
as reflected by an exponential increase in cell volume with
decreasing total cardiomyocyte number (which was striking
below an apparent threshold of ⬇5⫻106 cardiomyocytes per
heart; Figure 4B).
To identify the cause of the reduction in total cardiomyocyte
numbers in ␣MHC-survivin⫺/⫺ mice, we tested whether there
were alterations in the 2 major mechanisms that govern the
cellularity of an organ: cell death and cell division. Cardiomyocyte apoptosis was extremely low in both genotypes without
any differences as measured by TUNEL assays at all ages
(Figure 4C). However, there were clear differences in the
number of dividing cardiomyocytes in newborn mice; there was
an ⬇50% reduction in the number of mitotic figures in cardiomyocytes from ␣MHC-survivin⫺/⫺ mice compared with controls, together with evidence of aberrant mitosis (Figure 4D).
Pronounced nuclear abnormalities such as huge and irregularly
shaped nuclei were already present in survivin-deficient cardiomyocytes at embryonic day 10.5, along with the loss of survivin
immunostaining (Figure 5), suggesting that at this developmen-
tal stage, successful deletion of the gene had already occurred
and had led to defects in cell division (Figure 5).
Effects of Survivin Knockdown and
Overexpression on Ploidy, Apoptosis,
DNA Synthesis, and Cell Cycle in
Neonatal Cardiomyocytes
To address the biological mechanism behind the ␣MHCsurvivin⫺/⫺ phenotype in vitro, we knocked down survivin by
siRNA and overexpressed it by adenoviral infection in rat
neonatal cardiomyocytes. Survivin siRNA induced cell cycle
arrest in G2/M with 60% and 71% more cells accumulating in
G2/M phase of the cell cycle after 48 and 72 hours, respectively (P⬍0.05; Figure 6A). Knockdown of survivin also had
profound effects on nuclear morphology as evidenced by the
appearance of large and bizarrely shaped nuclei (Figure 6A).
These nuclei strongly resembled those of cardiomyocytes
from ␣MHC-survivin⫺/⫺ mice (Figure 3D and 5).
To test for effects of survivin overexpression, we infected
rat neonatal cardiomyocytes with adenovirus encoding the
survivin gene or a control virus carrying GFP (Figure 6C).
Survivin overexpression markedly protected cardiomyocytes
against doxorubicin-induced apoptosis compared with GFPinfected cells (Figure 6C). Furthermore, it induced DNA
Circulation
March 25, 2008
A
Heart weight (g)
B
Relative heart weight
0.03
Cardiomyocyte diameter (µm)
1588
wild type
αMHC-survivin-/-
0.02
0.01
0
N=
4
6
5
1
5
5
4
28
5
61
5
9
122
7
Myocyte diameter
45
wild type
αMHC-survivin-/-
40
*
25
20
15
*
10
5
0
N=
10 12
5
1
270
7
7
61
5
5
9
7
270
122
D
wild type
αMHC-survivin-/-
wild type
*
1.0
*
*
0.8
*
N
0.6
5˩m
0.4
αMHC-survivin-/0.2
0
N=
10 12
1
5
5
28
7
5
7
61
5
9
7
270
122
N
Age (days)
cardiomyocytes
fibroblasts
wild type
αMHC-survivin-/-
Cardiomyocyte DNA content
F
DNA content
Number of cells
E
Number of cells
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Fibrosis left ventricle (%)
5
28
Age (days)
Fibrosis
1.2
*
*
30
Age (days)
C
*
35
DNA content
5n
wild type
αMHC-survivin-/*
4n
*
*
3n
2n
28
61
122
Age (days)
2n 4n
8n
16n
Figure 3. Morphological characterization of ␣MHC-survivin⫺/⫺ mice. A, Relative heart weights (heart weight/body weight). B, Cardiomyocyte diameters. C, Cardiac fibrosis in mice at different ages. D, Representative hematoxylin-eosin staining and electron microscopy of
an 18-week-old ␣MHC-survivin⫺/⫺ mouse and a littermate control. The massive enlargement and bizarre nuclear morphology in cardiomyocytes from ␣MHC-survivin⫺/⫺ hearts are marked by arrows and shown in the 2 insets. Bars indicate diameter of individual cardiomyocytes. Electron microscopy depicts distinct nuclear enlargement (N) and bizarre nuclear membrane invaginations in ␣MHC-survivin⫺/⫺ cardiomyocytes. E, Representative DNA cytometry of a 28-day-old ␣MHC-survivin⫺/⫺ mouse and a littermate control shows
distribution of diploid (2n), tetraploid (4n), or polyploid cardiomyocytes. Cardiac fibroblasts were used as reference cells for 2n. F, Total
DNA content of knockout and control mice at different ages. Black box depicts the median. *Significant differences vs wild type.
Levkau et al
B
Cardiomyocyte number
30
wild type
αMHC-survivin-/-
20
*
*
10
*#
*#
*#
10 12
5
1
5
5
5
5
61
28
5
100
R = 0.85
80
60
R = -0.85
40
20
0
6
9
122
wild type
αMHC-survivin-/-
120
0
N=
1589
Cell number vs cell size
Cardiomyocyte volume [pl]
Cardiomyocyte number (*106)
A
Survivin Determines Cardiomyocyte Number In Vivo
270
0
5
10
15
20
Cardiomyocyte number [*106]
Age (days)
C
Mitosis
20
wild type
αMHC-survivin-/-
Mitotic counts (%)
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
1.5
TUNEL-positivity (%)
D
Apoptosis
1.0
0.5
0.0
wild type
αMHC-survivin-/-
15
wild type
10
αMHC-survivin-/*
5
0
N=
6
5
1
7
5
61
5
5
122
9
7
270
Age (days)
1
Age (days)
Figure 4. Cardiomyocyte numbers and apoptosis. A, Total cardiomyocyte numbers per heart. B, Correlation between cardiomyocyte
volume and cardiomyocyte number in both genotypes together according to this equation: cardiomyocyte volume⫽6.51⫹74.97 Recip(cardiomyocyte number[⫻106]). C, TUNEL. D, Quantification of mitotic figures in cardiomyocytes in newborn control and survivindeficient hearts. Right, Cardiomyocyte mitosis in an ␣MHC-survivin⫺/⫺ mouse and a wild-type control. Data are expressed as box plots.
°Outliers; *significant difference vs wild type; #significant difference vs 1-day-old ␣MHC-survivin⫺/⫺.
synthesis as measured by [3H]thymidine incorporation and
promoted cell cycle progression as evidenced by the increased number of cells in the G2/M phase of the cell cycle
(Figure 6C).
Survivin Is Upregulated in the Failing Human
Heart and Decreases Again After Hemodynamic
Support With a LVAD
Survivin was detectable at extremely very low levels in the
normal human heart, but its expression was dramatically
increased in the hearts of patients with terminal heart failure
resulting from ischemic or dilative cardiomyopathy (Figure
7). Failing hearts showed ⬇8-fold more survivin-positive
cardiomyocytes than donor hearts used as control (median,
3.3; range, 1.8 to 4.7 versus median, 0.4; range, 0.0 to 0.9;
Figure 7). Remarkably, hemodynamic support through an
LVAD (for the average of 130 days) resulted in a distinct
⬇60% decrease in the number of survivin-expressing cardiomyocytes when matched samples of identical hearts we
examined (median, 3.3; range, 1.8 to 4.7 versus median, 1.4;
range, 0.7 to 5.1; P⬍0.007, Wilcoxon test for paired samples;
Figure 7). This suggests that survivin is reversibly regulated
by the hemodynamic load affecting the failing heart. Furthermore, survivin expression correlated significantly with the
mean DNA content in all examined hearts, including control,
failing, and supported hearts (R⫽0.48, P⬍0.05, correlation
according to Spearman; Figure 8).
Discussion
Survivin is a key regulator of mitotic progression, and its
inactivation in mammalian cells or that of its orthologues in
lower organisms causes severe mitotic defects.8 Deletion of
survivin has been shown to be embryonically lethal with 3
characteristic abnormalities: macronucleation and
multinucleation with polyploidization, reduced cellularity,
and increased apoptosis.23 In our study, we have observed 2
of these features, marked polyploidy and reduced cell numbers, in vivo and in vitro but found no evidence of enhanced
apoptosis. In fact, it has been shown that apoptosis must not
necessarily follow survivin suppression despite the occurrence of pronounced cell division defects.4,7 Some cell types
fail to arrest after survivin suppression,24 whereas others exit
1590
Circulation
March 25, 2008
αMHC-survivin-/-, E10.5
200x
600x
1000x
200x
600x
1000x
Figure 5. Immunostaining for survivin in ␣MHCsurvivin⫺/⫺ and control hearts at embryonic day
10.5. Note the enlarged, bizarrely shaped cardiomyocyte nuclei devoid of survivin immunostaining in ␣MHC-survivin⫺/⫺ mice.
wild type, E10.5
A
B
G1(2n)
52.2 ± 3.1 %
S
29.2 ± 6.7 %
G2/M (4n) 18.6 ± 3.7 %*
*p<0.05
G1(2n)
56.5 ± 3.4 %
S
31.8 ± 5.8 %
G2/M (4n) 11.6 ± 2.7 %
48h
48h
2n
2n
4n
54.8 ± 2.6 %
G1(2n)
S
35.2 ± 4.5 %
G2/M (4n) 10.0 ± 2.1 %
4n
Apoptosis (% of control)
survivin siRNA
C
Cardiomyocyte
apoptosis
**
4n
non-silencing siRNA
1000x
+
+
Dox
Ad-surv-infected cardiomyocytes
72h
2n
+
v FP
l
ur
ro
nt d-s d-G
co
A
A
Ad-GFP
4
2n
Ad-GFP-infected cardiomyocytes
600x
-
51.3 ± 1.7 %
G1(2n)
31.5 ± 5.2 %
S
G2/M (4n) 17.1 ± 3.7 %*
*p<0.05
72h
Cardiomyocyte
cell cycle profile
40
Ad-surv
4
40
x1010 VP/mL
survivin
4n
survivin siRNA
1000x
70
Cardiomyocyte
DNA synthesis
**
P=0.016
60
*
50
% Cells
non-silencing siRNA
DNA synthesis (% of control)
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
for regular cardiomyocyte proliferation and mitosis. However, they also suggest that its loss is not sufficient to
completely abrogate DNA replication in vivo because a
subpopulation of cells exhibited an 8n and even a 16n DNA
mitosis to reform single tetraploid nuclei25; in both cases,
apoptosis does not take place.
In our study, the marked DNA polyploidization and G2/M
arrest in vivo and in vitro suggest that survivin is necessary
Ad-GFP
Ad-surv
40
30
20
10
0
l
ro surv GFP FBS
nt
co Ad
Ad
subG1
G1
S
G2/M
Figure 6. Effect of survivin siRNA knockdown and adenoviral overexpression on cell cycle, proliferation, and apoptosis in rat neonatal
cardiomyocytes. A, DNA profiles of cardiomyocytes transfected with siRNA against survivin and nonsilencing siRNA at different times
after transfection. Parallel cultures were immunostained for desmin, and nuclei were counterstained with DAPI 72 hours after transfection. B, Survivin overexpression prevents apoptosis (top) and induces DNA synthesis (bottom). DNA fragmentation was quantified by
flow cytometry after 24 hours of doxorubicin (Dox) treatment (**P⬍0.01, Student’s t test). DNA synthesis was measured by [3H]thymidine after adenoviral overexpression of survivin (Ad-surv) or GFP (Ad-GFP). Fetal bovine serum (FBS) was used as a control. Western
blot analysis shows expression of survivin in infected cardiomyocytes (viral particles [VP]). C, DNA profiles of cardiomyocytes adenovirally overexpressing survivin and their respective GFP controls 48 hours after adenoviral infection.
Levkau et al
Survivin Determines Cardiomyocyte Number In Vivo
Failing human heart pre-LVAD
B
P<0.007
6
5
Figure 7. Induction of survivin in the failing human
heart and its regulation by hemodynamic unloading by LVAD. Representative immunohistochemistry and quantitative analysis of survivin in control
human hearts and intraindividually matched samples from failing human hearts before and after
LVAD support. Arrows indicate survivin-positive
cardiomyocyte nuclei.
4
3
2
1
0
Control Failing heart
pre-LVAD
heart
content, indicative of several consecutive rounds of DNA
replication without cell division. The missing increase in
apoptosis in survivin-deficient hearts indicates that this
polyploidization had not activated any of the cell cycle
checkpoints that normally lead to apoptosis in the case of
abnormal DNA content. The data also suggest that survivin is
necessary for the actual cardiomyocyte division that follows
DNA replication because deletion of survivin resulted in a
dramatic reduction in cardiomyocyte numbers. This is in line
with gene knockout studies showing reduced cellularity in
tissues deleted for survivin, regardless of whether the cells
concomitantly survive10 or die.23 With apoptosis being unaffected, the reduction in cardiomyocyte numbers in ␣MHCsurvivin⫺/⫺ mice could thus be explained as being a conse-
Survivin-positive cardiomyocytes
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Failing human heart post-LVAD
Survivin-positive cardiomyocytes
Control human heart
1591
Failing heart pre-LVAD
Failing heart post-LVAD
Controls
R = 0.48
P < 0.05
Mean cardiomyocyte DNA content
Figure 8. Correlation between mean cardiomyocyte DNA content and survivin-immunopositive cardiomyocytes in supported
and nonsupported failing and control human hearts. The correlation includes all values.
Failing heart
post-LVAD
quence of cell division defects that impair cardiomyocyte
proliferation, which takes place up to several weeks after
birth.2
Why downregulation of survivin leads to perturbations of
mitosis in all cell types but to an increase in apoptosis in only
certain ones remains a tantalizing question. This disparity is
evident from all tissue-specific survivin knockouts, including
ours; whereas deletion of survivin in the brain or endothelium
is lethal as a result of massive apoptosis of the respective cell
types,26,27 its deletion in T cells impairs homeostatic expansion but does not induce apoptosis.10 In our study, deletion of
survivin was not sufficient to drive cardiomyocytes into
apoptosis, but its overexpression potently protected them,
which is in line with its function as an apoptosis inhibitor.
The induction of survivin that we have seen in the failing
human heart may serve the same purpose: to protect the
cardiomyocyte against apoptosis. Remarkably, survivin is
downregulated again after hemodynamic relief by LVAD,
implying its potential involvement in the cardiac reverse
remodeling process.28 The increase in [3H]thymidine incorporation and cells within G2/M after survivin overexpression
in cardiomyocytes in vitro is remarkable in that it suggests
that enforced expression of the gene suffices to induce both
DNA synthesis and cell cycle progression. Thus, survivin
may have a potentially regenerative function in vivo, a notion
that remains to be addressed by the overexpression of
survivin in the heart in vivo. Interestingly, in the failing
human heart, only individual cardiomyocytes showed strong
induction of survivin, whereas the majority did not. Have
these cardiomyocytes induced the gene as a protective measure against apoptosis, or are they possibly synthesizing
DNA? There is a large body of literature on the question of
whether the adult cardiomyocyte has the ability to proliferate.
In contrast, the existence of pronounced cardiomyocyte
polyploidization in the failing human heart has been known
since the 1960s, suggesting the existence of active DNA
synthesis.29 We also have seen polyploidization in the failing
1592
Circulation
March 25, 2008
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
human heart but, even more remarkably, its dramatic disappearance after LVAD support in the same heart (data not
shown). Whether this reduction in DNA content is the result
of successful completion of cell division or that of DNA
degradation during hemodynamic relief is unknown. However, the mean cardiomyocyte DNA content correlated significantly with survivin immunopositivity in all examined
hearts. Thus, survivin may serve as a marker of DNA
synthesis, polyploidization, or possibly even cell cycle traverse in the failing human heart.
Loss of cardiomyocytes through cell death regardless of
the underlying cause has been implied in the pathogenesis of
cardiomyopathies.30 Elegant proof has been provided in
transgenic mice expressing low levels of active caspase-8 and
Mst-1 in the heart.31,32 In both models, a gradual loss of
cardiomyocytes resulting from apoptosis has been implied to
be the cause of heart failure and death. Although the apoptotic
rate was 0.023% and 0.3%, respectively, neither of the studies
had determined the actual numerical extent of cardiomyocyte
reduction. Our study may close the gap by providing the
actual number of total cardiomyocytes required to maintain
normal lifelong cardiac function. We have defined this
number as ⬇5⫻106 cardiomyocytes per heart because below
this threshold the individual cardiomyocytes were hypertrophied, reflecting an adaptive response to increased workload.
Cellular hypertrophy exponentially increased with declining
cardiomyocyte numbers, resulting in up to 7-fold larger cells.
The extremely efficient compensation of cell number decline
by an increase in cell size may explain why survivin-deficient
hearts exhibiting cardiomyocyte numbers below the threshold
did not succumb immediately to heart failure but were able to
maintain a satisfactory function for months.
Organ size is confined within boundaries considered normal for each species, and it is determined by the number of
cells per organ rather than their size.33 The signaling pathways determining cell number during organogenesis are
highly conserved and remarkably few.33,34 Among species,
the heart weights differ by ⬎1000-fold (⬇0.16 g in mice and
⬇210 g in humans), and although cardiomyocyte size is
rather similar (between ⬇13-␮m diameter in mice and 16-␮m
diameter in humans), the heart of a mouse contains ⬎300fold fewer cardiomyocytes than that of a human (8⫻106
versus 2600⫻106).35 How utterly important the number of
cells in an organ is for its function has been elegantly
established for the nervous system, where the number of
neurons has been shown to crucially influence brain size,
complexity, and possibly enlargement during evolution.36 A
similarly crucial role has been proposed for the number of
nephrons in the pathogenesis of hypertension; patients with
essential hypertension appear to have 50% fewer nephrons
than healthy individuals.16 We propose that, correspondingly,
the individual number of “cardiomyocyte working units” per
heart may have an impact on the onset, extent, and progression of cardiac diseases by codetermining cardiac function.
Because the cardiomyocyte number of each individual is
determined during ontogenesis,37 all genetic or environmental
factors that influence cardiac development also may influence
cardiac cellularity. Such environmental factors may be, for
example, alterations in intrauterine nutrition that have been
implied in the enhanced susceptibility to cardiovascular
diseases in later life.38 Although such “perinatal programming“ has been suggested to affect nephron number and thus
hypertension,39 from our work, we would suggest that it also
affects cardiomyocyte number and thus cardiac function.
Conclusion
Survivin plays a crucial role in controlling cardiomyocyte
number during embryonic development and adult life through
its profound impact on cardiomyocyte replication and may
thus emerge as a new target for myocardial regeneration.
Acknowledgments
We thankfully acknowledge the technical assistance of K. Abouhamed,
S. Mersmann, V. Brinkmann, D. Möllmann, and C. Bätza. We are
grateful to Professor G. Mall for discussing the cell number determination and Dr J. Pollerberg for the DNA cytometry equipment.
Sources of Funding
This study was supported in part by the Deutsche Forschungsgemeinschaft (LE940/3-1, BA1730/9-1), SFB656 (A1, A3, C3, Z2),
SFB612 (Z2), IZKF Münster (ZPG 4a/b), European Commission
FP6 Project DiMI, LSHB-CT-2005-512146, the FWO, Belgium, the
Belgian Federation Against Cancer, a postdoctoral fellowship of
Peter and Traudl Engelhorn Stiftung (to Dr von Wnuck Lipinski),
and the Deichmann Foundation for Atherosclerosis Research.
Disclosures
None.
References
1. Jessup M, Brozena S. Heart failure. N Engl J Med. 2003;348:2007–2018.
2. Nadal-Ginard B, Kajstura J, Leri A, Anversa P. Myocyte death, growth,
and regeneration in cardiac hypertrophy and failure. Circ Res. 2003;92:
139 –150.
3. Vaux DL, Silke J. IAPs, RINGs and ubiquitylation. Nat Rev Mol Cell
Biol. 2005;6:287–297.
4. Altieri DC. Molecular circuits of apoptosis regulation and cell division
control: the survivin paradigm. J Cell Biochem. 2004;92:656 – 663.
5. Marusawa H, Matsuzawa S, Welsh K, Zou H, Armstrong R, Tamm I,
Reed JC. HBXIP functions as a cofactor of survivin in apoptosis suppression. EMBO J. 2003;22:2729 –2740.
6. Song Z, Yao X, Wu M. Direct interaction between survivin and Smac/
DIABLO is essential for the anti-apoptotic activity of survivin during
taxol-induced apoptosis. J Biol Chem. 2003;278:23130 –23140.
7. Castedo M, Perfettini JL, Roumier T, Andreau K, Medema R, Kroemer
G. Cell death by mitotic catastrophe: a molecular definition. Oncogene.
2004;23:2825–2837.
8. Okada H, Mak TW. Pathways of apoptotic and non-apoptotic death in
tumour cells. Nat Rev Cancer. 2004;4:592– 603.
9. Altznauer F, Martinelli S, Yousefi S, Thurig C, Schmid I, Conway EM,
Schoni MH, Vogt P, Mueller C, Fey MF, Zangemeister-Wittke U, Simon
HU. Inflammation-associated cell cycle-independent block of apoptosis
by survivin in terminally differentiated neutrophils. J Exp Med. 2004;
199:1343–1354.
10. Xing Z, Conway EM, Kang C, Winoto A. Essential role of survivin, an
inhibitor of apoptosis protein, in T cell development, maturation, and
homeostasis. J Exp Med. 2004;199:69 – 80.
11. Fukuda S, Mantel CR, Pelus LM. Survivin regulates hematopoietic progenitor cell proliferation through p21WAF1/Cip1-dependent and
-independent pathways. Blood. 2004;103:120 –127.
12. Abbate A, Scarpa S, Santini D, Palleiro J, Vasaturo F, Miller J, Morales
C, Vetrovec GW, Baldi A. Myocardial expression of survivin, an apoptosis inhibitor, in aging and heart failure: an experimental study in the
spontaneously hypertensive rat. Int J Cardiol. 2006;113:371–376.
13. Santini D, Abbate A, Scarpa S, Vasaturo F, Biondi-Zoccai GG, Bussani
R, De Giorgio F, Bassan F, Camilot D, Di Marino MP, Feroce F, Baldi
F, Silvestri F, Crea F, Baldi A. Surviving acute myocardial infarction:
survivin expression in viable cardiomyocytes after infarction. J Clin
Pathol. 2004;57:1321–1324.
Levkau et al
Survivin Determines Cardiomyocyte Number In Vivo
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
14. Agah R, Frenkel PA, French BA, Michael LH, Overbeek PA, Schneider
MD. Gene recombination in postmitotic cells: targeted expression of Cre
recombinase provokes cardiac-restricted, site-specific rearrangement in
adult ventricular muscle in vivo. J Clin Invest. 1997;100:169 –179.
15. Buzello M, Boehm C, Orth S, Fischer B, Ehmke H, Ritz E, Mall G,
Amann K. Myocyte loss in early left ventricular hypertrophy of experimental renovascular hypertension. Virchows Arch. 2003;442:364 –371.
16. Keller G, Zimmer G, Mall G, Ritz E, Amann K. Nephron number in
patients with primary hypertension. N Engl J Med. 2003;348:101–108.
17. Haroske G, Giroud F, Reith A, Bocking A. 1997 ESACP consensus report
on diagnostic DNA image cytometry, part I: basic considerations and
recommendations for preparation, measurement and interpretation:
European Society for Analytical Cellular Pathology. Anal Cell Pathol.
1998;17:189 –200.
18. Flogel U, Laussmann T, Godecke A, Abanador N, Schafers M, Fingas
CD, Metzger S, Levkau B, Jacoby C, Schrader J. Lack of myoglobin
causes a switch in cardiac substrate selection. Circ Res. 2005;96:
e68 – e75.
19. Engelhardt S, Hein L, Wiesmann F, Lohse MJ. Progressive hypertrophy
and heart failure in beta1-adrenergic receptor transgenic mice. Proc Natl
Acad Sci U S A. 1999;96:7059 –7064.
20. Hsieh PC, Davis ME, Gannon J, MacGillivray C, Lee RT. Controlled
delivery of PDGF-BB for myocardial protection using injectable selfassembling peptide nanofibers. J Clin Invest. 2006;116:237–248.
21. Schulze PC, de Keulenaer GW, Kassik KA, Takahashi T, Chen Z, Simon
DI, Lee RT. Biomechanically induced gene iex-1 inhibits vascular
smooth muscle cell proliferation and neointima formation. Circ Res.
2003;93:1210 –1217.
22. von Wnuck Lipinski K, Keul P, Ferri N, Lucke S, Heusch G, Fischer JW,
Levkau B. Integrin-mediated transcriptional activation of inhibitor of
apoptosis proteins protects smooth muscle cells against apoptosis induced
by degraded collagen. Circ Res. 2006;98:1490 –1497.
23. Uren AG, Wong L, Pakusch M, Fowler KJ, Burrows FJ, Vaux DL, Choo
KH. Survivin and the inner centromere protein INCENP show similar
cell-cycle localization and gene knockout phenotype. Curr Biol. 2000;
10:1319 –1328.
24. Lens SM, Wolthuis RM, Klompmaker R, Kauw J, Agami R, Brummelkamp T, Kops G, Medema RH. Survivin is required for a sustained
spindle checkpoint arrest in response to lack of tension. EMBO J. 2003;
22:2934 –2947.
25. Carvalho A, Carmena M, Sambade C, Earnshaw WC, Wheatley SP.
Survivin is required for stable checkpoint activation in taxol-treated HeLa
cells. J Cell Sci. 2003;116:2987–2998.
1593
26. Jiang Y, de Bruin A, Caldas H, Fangusaro J, Hayes J, Conway EM,
Robinson ML, Altura RA. Essential role for survivin in early brain
development. J Neurosci. 2005;25:6962– 6970.
27. Zwerts F, Lupu F, De Vriese A, Pollefeyt S, Moons L, Altura RA, Jiang
Y, Maxwell PH, Hill P, Oh H, Rieker C, Collen D, Conway SJ, Conway
EM. Lack of endothelial cell survivin causes embryonic defects in angiogenesis, cardiogenesis, and neural tube closure. Blood. 2007;109:
4742– 4752.
28. Wohlschlaeger J, Schmitz KJ, Schmid C, Schmid KW, Keul P, Takeda A,
Weis S, Levkau B, Baba HA. Reverse remodeling following insertion of
left ventricular assist devices (LVAD): a review of the morphological and
molecular changes. Cardiovasc Res. 2005;68:376 –386.
29. Sandritter W, Scomazzoni G. Deoxyribonucleic acid content (Feulgen
photometry) and dry weight (interference microscopy) of normal and
hypertrophic heart muscle fibers. Nature. 1964;202:100 –101.
30. Garg S, Narula J, Chandrashekhar Y. Apoptosis and heart failure: clinical
relevance and therapeutic target. J Mol Cell Cardiol. 2005;38:73–79.
31. Wencker D, Chandra M, Nguyen K, Miao W, Garantziotis S, Factor SM,
Shirani J, Armstrong RC, Kitsis RN. A mechanistic role for cardiac
myocyte apoptosis in heart failure. J Clin Invest. 2003;111:1497–1504.
32. Yamamoto S, Yang G, Zablocki D, Liu J, Hong C, Kim SJ, Soler S,
Odashima M, Thaisz J, Yehia G, Molina CA, Yatani A, Vatner DE,
Vatner SF, Sadoshima J. Activation of Mst1 causes dilated cardiomyopathy by stimulating apoptosis without compensatory ventricular myocyte
hypertrophy. J Clin Invest. 2003;111:1463–1474.
33. Hidalgo A, ffrench-Constant C. The control of cell number during central
nervous system development in flies and mice. Mech Dev. 2003;120:
1311–1325.
34. Leevers SJ, McNeill H. Controlling the size of organs and organisms.
Curr Opin Cell Biol. 2005;17:604 – 609.
35. Adler CP, Friedburg H, Herget GW, Neuburger M, Schwalb H. Variability of cardiomyocyte DNA content, ploidy level and nuclear number
in mammalian hearts. Virchows Arch. 1996;429:159 –164.
36. Chenn A, Walsh CA. Regulation of cerebral cortical size by control of
cell cycle exit in neural precursors. Science. 2002;297:365–369.
37. Meilhac SM, Esner M, Kelly RG, Nicolas J-F, Buckingham ME. The
clonal origin of myocardial cells in different regions of the embryonic
mouse heart. Dev Cell. 2004;6:685– 698.
38. Armitage JA, Taylor PD, Poston L. Experimental models of developmental programming: consequences of exposure to an energy rich diet
during development. J Physiol. 2005;565:3– 8.
39. Ingelfinger JR. Is microanatomy destiny? N Engl J Med. 2003;348:
99 –100.
CLINICAL PERSPECTIVE
Cardiomyocyte death leading to loss of contractile units has been implicated in the pathogenesis of all forms of heart
failure. Although the exact stimuli, mechanisms, and rate of apoptosis in the adult human heart are unknown, a dynamic
balance exists between cardiomyocyte loss and replacement during life and disease. We found that the small antiapoptotic
molecule survivin controls both cell death and cell division during cardiac development; its deletion in a cardiomyocytespecific fashion in mice led to progressive heart failure resulting from a profound reduction in total cardiomyocyte numbers
per heart. By systematically counting total cardiomyocytes, we were able to determine the minimal cardiomyocyte number
sufficient for normal lifelong cardiac function. Survivin overexpression in cultured cardiomyocytes inhibited apoptosis,
induced DNA synthesis, and promoted cell cycle progression. In the failing human heart, survivin was potently induced
and decreased again after hemodynamic relief through a mechanical left ventricular assist device. We observed that the
long-known existence of polyploidy of cardiomyocytes in the failing human heart was remarkably decreased after
hemodynamic support and correlated with the level of survivin expression at any time. Thus, survivin may be a marker of
DNA synthesis, polyploidy, or possibly even cell cycle traverse in the failing human heart, making it a bona fide target for
myocardial regeneration therapies. We suggest that the individual number of “cardiomyocyte working units” that are under
the control of survivin may be an independent factor in the susceptibility to cardiac diseases independently of their cause.
Survivin Determines Cardiac Function by Controlling Total Cardiomyocyte Number
Bodo Levkau, Michael Schäfers, Jeremias Wohlschlaeger, Karin von Wnuck Lipinski, Petra
Keul, Sven Hermann, Naomasa Kawaguchi, Paulus Kirchhof, Larissa Fabritz, Jörg Stypmann,
Lars Stegger, Ulrich Flögel, Jürgen Schrader, Jens W. Fischer, Patrick Hsieh, Yen-Ling Ou,
Felix Mehrhof, Klaus Tiemann, Alexander Ghanem, Marek Matus, Joachim Neumann, Gerd
Heusch, Kurt W. Schmid, Edward M. Conway and Hideo A. Baba
Downloaded from http://circ.ahajournals.org/ by guest on July 28, 2017
Circulation. 2008;117:1583-1593; originally published online March 10, 2008;
doi: 10.1161/CIRCULATIONAHA.107.734160
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2008 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circ.ahajournals.org/content/117/12/1583
An erratum has been published regarding this article. Please see the attached page for:
/content/117/24/e497.full.pdf
Data Supplement (unedited) at:
http://circ.ahajournals.org/content/suppl/2008/03/14/CIRCULATIONAHA.107.734160.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial
Office. Once the online version of the published article for which permission is being requested is located,
click Request Permissions in the middle column of the Web page under Services. Further information about
this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/
Correction
In the article, “Survivin Determines Cardiac Function by Controlling Total Cardiomyocyte
Number” by Levkau et al that appeared in the March 25, 2008, issue (Circulation.
2008;117:1583–1593), Dr Fischer’s middle initial was omitted in the byline. Jens Fischer, PhD,
should have been listed as Jens W. Fischer, PhD. The current online version of the article has been
corrected.
The authors regret this error.
DOI: 10.1161/CIRCULATIONAHA.108.190032
(Circulation. 2008;117:e497.)
© 2008 American Heart Association, Inc.
Circulation is available at http://circ.ahajournals.org
e497