Download Cellular Biology - Circulation Research

Cellular Biology
Regulation of Cardiomyocyte Polyploidy and
Multinucleation by CyclinG1
Zhipei Liu,* Shijing Yue,* Xiaobo Chen, Thomas Kubin, Thomas Braun
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Rationale: Polyploidy and multinucleation are characteristic features of mammalian cardiomyocytes, which develop
shortly after birth when most differentiated cardiomyocytes become acytokinetic. Cardiac overload and
hypertrophy further increase the degree of polyploidy of cardiomyocytes, suggesting a role in cell type–specific
responses to physiological and pathological stimuli.
Objective: We sought to study the function of cyclinG1 in the regulation of polyploidy and multinucleation in
cardiomyocytes.
Methods and Results: We found that expression of cyclinG1, a transcriptional target of p53, coincides with arrest
of cardiomyocyte proliferation and onset of polyploidization. Overexpression of cyclinG1 promoted DNA
synthesis but inhibited cytokinesis in neonatal cardiomyocytes leading to an enlarged population of binuclear
cardiomyocytes. Reciprocally, inactivation of the cyclinG1 gene in mice lowered the degree of polyploidy and
multinucleation in cardiomyocytes. Moreover, lack of cyclinG1 prevented the increase of polynucleated
cardiomyocytes in response to pressure overload and hypertrophy.
Conclusions: CyclinG1 is an important player for the regulation of polyploidy and multinucleation in cardiomyocytes probably by inhibition of apoptosis caused by checkpoint activation. (Circ Res. 2010;106:1498-1506.)
Key Words: cyclinG1 䡲 cardiomyocytes 䡲 polyploidy 䡲 multinucleation
D
nucleated with either 2N (⬇3%) or 4N (⬇5%), and 10%
are octaploid (4N⫻2). The remaining cells show an even
higher ploidy of up to 32N. These numbers differ depending on the techniques used and individual mouse strains,
which have led to some discrepancies in the literature.3
Because polyploidization of ventricular cardiomyocytes
occurs within the first three weeks of life during the same
time when cardiomyocytes loose the ability to undergo
cytokinesis it has been speculated that both events are caused
by similar mechanism and represent two sides of the same
coin. According to this view the increase of polyploidy might
be a specific reaction of cardiomyocytes to respond to strong
mitotic stimuli because they lack parts of the cell division
machinery and therefore might adopt an abortive cell cycle
mode. This hypothesis is also supported by the additional
increase of the polyploidy of cardiomyocytes, which occurs
in some pathological situations such as ischemic injury or end
stage of heart failure that provide strong cues for cell
division.2 The accurate balance of polyploidization in different species makes it likely that polyploidization is controlled
by a specific cellular program that has evolved to avoid the
tetraploidy checkpoint, acquisition of multiple centromers,
aberrant mitosis, and chromosomal instability, which will
uring embryonic and fetal life, myocardial mass of
mammals increases by proliferation of cardiomyocytes, which synthesize DNA and undergo mitosis in the
presence of myofibrils that are partially disassembled
because of the resorption of Z-discs. Shortly after birth, the
ability for cytokinesis ceases and is virtually absent in
adult cardiomyocytes.1 The lack of cytokinesis, however,
should not be confused with a failure of cardiomyocytes to
synthesize DNA. A large fraction of cardiomyocytes is
able to complete a full round of DNA synthesis under
conditions of cardiac overload and hypertrophy. This
property is particularly evident in primates, which are able
to increase the ploidy of cardiomyocytes under pathological conditions. In rodents, the ability to respond to damage
by polyploidization is mostly restricted to the atrium. All
mammalian species studied so far seem to own a program
that allows generation of cardiomyocytes with different
degrees of polyploidy after birth but the exact reason and
the regulatory mechanisms which drive physiological
polyploidization remain unknown.2 In adult mice, the
majority of ventricular cardiomyocytes (ca. 80%) are
tetraploid and carry two nuclei (2N⫻2, where N represents
the haploid DNA content per nucleus), ⬇8% are mono-
Original received October 23, 2009; revision received March 19, 2010; accepted March 22, 2010.
From the Max-Planck-Institut for Heart und Lung Research, Bad Nauheim, Germany. Present address for Z.L.: Union Gene Test & Health Management
Center, Tianjin, People’s Republic of China. Present address for X.C.: Union Stem Cell & Gene Engineering Co, Tianjin, People’s Republic of China.
*Both authors contributed equally to this work.
Correspondence to Thomas Braun, Max-Planck-Institut for Heart und Lung Research, Parkstrasse 1, D-61231 Bad Nauheim, Germany. E-mail
[email protected]
© 2010 American Heart Association, Inc.
Circulation Research is available at http://circres.ahajournals.org
DOI: 10.1161/CIRCRESAHA.109.211888
1498
Liu et al
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
trigger apoptosis or uncontrolled proliferation.4 Such a program might enable cardiomyocytes to adapt to different
conditions such as hypertrophic growth, which will help to
avoid potentially dangerous effects of cell divisions in the
working myocardium that has to maintain a high blood
pressure during the whole lifetime.
Acquisition of postmitotic fate and polyploidization of
cardiomyocytes during the first weeks after birth are characterized by downregulation of cyclins and cyclin-dependent
kinases (CDKs) related to G1/S and G2/M transition and
upregulation of G1 phase–related cyclins and CDKs.5 Cyclins
belong to a family of proteins, which control the progression
of cells through the cell cycle by activating cyclin-dependent
kinases. Several different cyclins exist, which are active in
different parts of the cell cycle. There are also several
“orphan” cyclins for which no Cdk partner has been identified including cyclinG1, which is a transcriptional target of
p53 and involved in several p53 related processes such as
apoptosis, growth control and checkpoint regulation in
response to DNA damage.6 The physiological function of
cyclinG1, which is expressed in adult muscle tissues
(including the heart),7 is still largely unknown. No gross
defects were reported for cyclinG1 deficient mouse strains
so far, which have been established independently by two
different groups.6,8 Here, we describe the role of cyclinG1
for the regulation of cardiomyocyte polyploidy and
binucleation.
Methods
CyclinG1 knockout mice were kindly provided by Prof Thorgeirsson
(NIH) and maintained on a C57BL6 background (⬎9 back-crossings
to C57BL6).8 Transaortic constriction (TAC) and cardiac infarction
was achieved based on previously described procedures.9 Adenoviruses expressing E2F2, LacZ and cyclinG1 were generated as
described previously.10
Cardiomyocytes from rats and mice were isolated and cultured
using standard procedures. DNA content measurements by FACS
were performed as described previously.10,11 Quantitative realtime PCR, immunofluorescence, protein isolation, Western Blot
analysis, and tritiated thymidine incorporation were done using
established protocols, which are described in detail in the expanded Methods section, available in the Online Data Supplement
at http://circres.ahajournals.org.
Regulation of Cardiomyocyte Polyploidy
1499
Non-standard Abbreviations and Acronyms
CDK
d
DAPI
E
FACS
LAD
TAC
VSMC
WT
cyclin-dependent kinase
postnatal day
4⬘,6-diamidino-2-phenylindole dihydrochloride
embryonic day
fluorescence-activated cell sorting
left anterior descending
transaortic constriction
vascular smooth muscle cell
wild type
Results
The Expression of CyclinG1 Coincides With the
Onset of Polyploidization and Hypertrophic
Growth During Postnatal Rat
Cardiomyocyte Development
Cardiac development and growth might be roughly divided
into 2 distinct phases, which encompass a proliferation phase
during prenatal stages and a hypertrophic phase that covers
the postnatal to adult period. In rodents, the switch between
both phases, which is also accompanied by binucleation and
polyploidization of cardiomyocytes, is completed within the
first 2 weeks after birth. To identify the molecules that might
be responsible for this switch, we investigated transcriptional
profiles of rat hearts from embryonic day (E)18 as well as
from postnatal (postnatal day [d]2) and adult (7 weeks) stages
using Affymetrix DNA microarray analysis. We found that
the expression of several cyclins and CDKs (cyclinA2,
cyclinB1, Cdk2, and Cdk6), and mitosis-activating genes
(Mad2, Plk1) were significantly downregulated in 7 weeks
hearts compared to E18. In contrast, only minor changes were
detected between E18 and d2 (data not shown). One of the
cell-cycle related genes, which showed significant expression
differences between E18 and 7 weeks, attracted our attention,
because it was up- and not downregulated at 7 weeks. We
found that the mRNA level of cyclinG1 increased from
relatively low levels at E18 and d2 to much higher levels in
7 weeks hearts (Figure 1A). This observation was also
confirmed by real-time PCR analysis (data not shown) and
Figure 1. The onset of polyploidization
and hypertrophic growth during postnatal rat cardiomyocyte development coincides with cyclinG1 expression. A and B,
Expression of cyclinG1 at E18 (embryonic),
d2 (postnatal), and 7-week-old (week) rat
hearts. RNA and protein levels were determined by Affymetrix microarray (A) and
Western blot (B) analysis. RNA levels of
cyclinG1 in E18 hearts were set to 100%. C
and D, Expression of cyclinG1 mRNA and
protein in postnatal rat hearts measured by
quantitative real-time PCR (C) and Western
blot (D) analysis. GAPDH and ␣-actinin were
used as loading controls. The RNA levels of
cyclinG1 in d2 hearts were set to 100%.
*P⬍0.05; **P⬍0.01 by Student’s t test (n⫽3).
1500
Circulation Research
May 14, 2010
Figure 2. Increased DNA synthesis and
G1/S transition in neonatal rat cardiomyocytes after overexpression of
cyclinG1. Cardiomyocytes were synchronized in serum free medium before virus
infections. A, Representative distribution
of cell cycle phases (n⫽3) of cardiomyocytes 36 hours after infection with different viruses. B, Labeling of cardiomyocytes with 3H-thymidine for 12 hours to
measure DNA synthesis. 3H incorporation
is displayed as fold-increase to mockinfected cells. A significant increase of
DNA synthesis was observed after
cyclinG1 expression in comparison to
expression of the LacZ control. *P⬍0.05;
**P⬍0.01 by Student’s t test. C, Representative Western blot analysis (n⫽3) of
the expression of biomarkers for G1/S
transition.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
immunoblotting (Figure 1B). A more detailed analysis revealed that the cyclinG1 mRNA level increased strongly
between d2 and d4 and was maintained at this level thereafter
(Figure 1C). CyclinG1 protein was first detected at d4 in the
heart and strongly increased by d8 (Figure 1D). The distinct
upregulation of cyclinG1 during the early postnatal period
suggested a role in cardiomyocyte maturation, which includes
polyploidization, binucleation, and cessation of proliferation.
Overexpression of CyclinG1 Moderately
Accelerates G1/S Transition in Neonatal
Rat Cardiomyocytes
To investigate whether cyclinG1 affects cell cycle exit,
polyploidization, and binucleation of cardiomyocytes we
overexpressed cyclinG1 using adenovirus mediated gene
transfer into primary neonatal (d2) rat cardiomyocytes. We
found that cyclinG1 stimulated S-phase entry of cardiomyocytes (n⫽3) 36 hours after infection. 31% of all cyclinG1
overexpressing cardiomyocytes were in S phase compared to
8% infected with a LacZ control virus (Figure 2A). The
efficiency of cyclinG1 to induce S-phase entry was similar to
E2F2 (29% cardiomyocytes in S phase), which was used as a
positive control. We also wanted to know whether and to
what extent cyclinG1 might modulate proproliferative signals. Therefore, neonatal rat cardiomyocytes were coinfected
with cyclinG1 and E2F2, which we have described previously
to strongly induce proliferation of neonatal cardiomyocytes
without induction of apoptosis.10 Interestingly, a reduced
number of cardiomyocytes (15%) remained in S phase,
whereas 38% of all coinfected cardiomyocytes accumulated
in G2/M phase, indicating that coexpression of cyclinG1 and
E2F2 pushed cells quickly through the S phase but restricted
progression beyond the G2/M checkpoint (Figure 2A). Analysis of DNA synthesis of cardiomyocytes infected with LacZ,
cyclinG1, E2F2 and cyclinG1/E2F2 adenoviruses using 3Hthymidine incorporation for 12 hours confirmed these findings (Figure 2B). Again, cyclinG1 overexpression significantly increased up-take of 3H-thymidine compared to the
LacZ control virus (n⫽3, P⬍0.05) although the rate of
3
H-thymidine incorporation achieved by E2F2 was much
higher in this assay. Coexpression of cyclinG1 with E2F2
resulted in a further increase of 3H-thymidine incorporation
(Figure 2B). The quantitative differences between the FACS
analysis and the 3H-thymidine incorporation are probably
attributable to different timings in both experimental settings
and to the relatively long time, which it takes for neonatal
cardiomyocytes to complete a cell cycle.
Western blot analysis of cell cycle regulators confirmed the
induction of S-phase entry (Figure 2C). We detected a strong
stimulation of the expression of cyclin D1 and induction of
phosphorylation of Rb (Ser795) by cyclinG1, whereas E2F2
induced phosphorylation of Rb at (Ser807/811), suggesting
that cyclinG1 and E2F2 increased DNA synthesis via distinct
pathways. (Ser807/811) phosphorylation of RB disrupts c-abl
binding (but not E2F binding), whereas (Ser795) phosphorylation is important to abolish E2F binding.12 Both cyclinG1
and E2F2 enhanced the expression of cyclinE1. In contrast,
cyclinG1 but not E2F2 induced phosphorylation of p38 and
Erk, which might be responsible for the cyclinG1 mediated
phosphorylation of Rb at (Ser795) and the activation of cyclin
D1 expression.13 We also noted a significant lower phosphorylation level of Rb at (Ser807/811) in E2F2/cyclinG1 coinfected cells compared to cardiomyocytes infected with E2F2
alone (Figure 2C), suggesting that cyclinG1 partially interfered with E2F2 mediated effects. Apparently, cyclinG1mediated phosphorylation of Rb at (Ser795) prevented E2Finduced Ser807/811 phosphorylation in coexpression
experiments hence reducing the release of nuclear c-abl. Both
the release of c-abl and E2F will promote G1/S transition
albeit c-abl has also been involved in the regulation of cell
cycle arrest and apoptosis depending on the cellular context.14
Overexpression of CyclinG1 Delays Mitosis and
Induces Multinucleation in
Neonatal Cardiomyocytes
Initiation of DNA-synthesis and S-phase entry does not
necessarily indicate that a cell undergoes mitosis and cytokinesis.15 To investigate whether cyclinG1 stimulates or
suppresses proliferation of cardiomyocytes, we measured
various parameters of late mitosis and cytokinesis. Western
Liu et al
Regulation of Cardiomyocyte Polyploidy
1501
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Figure 3. Inhibition of mitosis and cytokinesis of neonatal rat cardiomyocytes
after expression of cyclinG1. Cardiomyocytes were synchronized in serum free
medium for 24 hours before virus infections. A, Representative Western blot
analysis (n⫽3) of the expression of different mitosis markers. Ral A was used as a
loading control. B, Quantification of immunoblots to indicate the ratio of phosphorylated cofilin and total cofilin. The
increased ratio in cyclinG1 expressing
cells indicates the inability to complete
mitosis and cytokinesis. The ratio in
mock-infected cardiomyocytes was set to
100%. **P⬍0.01 by Student’s t test
(n⫽3). C, Number of cardiomyocytes in
cytokinesis as determined by immunofluorescence staining. D, Cardiomyocytes
in different mitotic phase as determined
by Aurora B kinase staining. Cardiomyocytes were identified by myosin heavy
chain staining (green). Nuclei or DNA
were identified by DAPI staining (blue). E,
Cardiomyocytes in different mitotic phase
as determined by ␣-tubulin staining. Cardiomyocytes were identified by troponin I
staining (red). Nuclei were identified by
DAPI staining (blue). *P⬍0.05; **P⬍0.01
by Student’s t test.
blot analysis revealed that the mitosis markers auroraB,
phosphorylated histone H3 (Ser10), Plk1, and Mad2 were all
significantly repressed after overexpression of cyclinG1 in
neonatal rat cardiomyocytes compared to uninfected cells and
cell infected with a LacZ control virus. Moreover, concomitant expression of cyclinG1 and E2F2 efficiently inhibited
E2F2 mediated upregulation of aurora B, phosphorylated
histone H3 (Ser10), Plk1, and Mad2 (Figure 3A). We also
determined the ratio of phosphorylated cofilin to total cofilin
by Western blot analysis because dephosphorylation and
reactivation of cofilin is critical for late stages of mitosis and
cytokinesis.16 Expression of cyclinG1 increased the ratio even
in the presence of E2F2, indicating that cyclinG1 imposed a
block on cytokinesis (Figure 3B and Online Figure I). Finally,
we performed immunofluorescence staining using antibodies
directed against auroraB and ␣-tubulin to directly visualize
different stages of mitosis and cytokinesis in neonatal cardiomyocytes (Figure 3D and 3E). Forced expression of cyclinG1
reduced the number of cells in cytokinesis to 0.57% compared to 0.92% observed after infection with the control virus
LacZ virus. Concomitant expression of cyclinG1 and E2F2
reduced the number of cells in cytokinesis to 1.19% compared to 1.89% after infection with the E2F2 virus alone
(Figure 3C).
A detailed analysis of cell cycle profiles by FACS analysis
conducted 48 hours after expression of cyclinG1 revealed an
accumulation of cardiomyocytes in G2/M (Figure 4A). In the
absence of additional cell proliferation signals, the increase of
the number of cells in G2/M was relatively modest, reaching
37% in comparison to 33% infected with the lacZ control
virus and 21% infected with the E2F2 virus. Coexpression of
E2F2 and cyclinG1, however, led to a massive increase of
cardiomyocytes in G2/M phase from 21% (E2F2 alone) to
47% indicating that high levels of cyclinG1 lead to a G2/M
arrest when proproliferative signals are present.
It is well known that an arrest at the G2/M check-point is
unstable; normally G2/M-arrested cells will either undergo
apoptosis or overcome the arrest and complete mitosis. To
investigate whether cardiomyocytes arrested in G2/M upregulate apoptosis markers we studied the expression of p53 and
activated caspases and performed TUNEL assays. No evidence for an upregulation of apoptosis markers or an increase
of the number of apoptotic cells were found (data not shown),
suggesting that cyclinG1-mediated G2/M arrest does not
favor apoptosis. In contrast, the level of phosphorylated
histone H3 (Ser10) and Plk1 increased in cyclinG1 overexpressing cells albeit at a later time point, indicating that
cyclinG1 delayed rather than inhibited mitosis (Figure 4B and
Online Figure II).
To further characterize the role of cyclinG1 for the regulation of mitosis and cytokinesis we used the flow cytometric
bromodeoxyuridine–Hoechst assay, which allowed us to
distinguish newly generated daughter cells from those that
1502
Circulation Research
May 14, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Figure 4. Delayed mitosis and induction
of polyploidization in neonatal rat cardiomyocytes after expression of cyclinG1.
Cardiomyocytes were synchronized in
serum free medium for 24 hours before
virus infection. A, Representative distribution of cell cycle phases of cardiomyocytes
48 hours after infection with different viruses
(n⫽3). B, Quantification of immunoblots
revealed a belated increase of the M-phase
marker p-histone H3 (ser 10) and the G2/M
marker PLK1 after cyclinG1 overexpression
indicating that high levels of cyclinG1 delay
mitosis even in the presence of proproliferative molecules such as E2F2. C, Proliferation of cardiomyocytes as determined by
the flow cytometric bromodeoxyuridine–
Hoechst assay at 48 hours after infection.
The induction measured after infection with
the LacZ control virus was set to 1-fold. D,
Number of binucleated cardiomyocytes as
determined by fluorescence staining. Cardiomyocytes were identified by myosin heavy
chain staining (green). Nuclei were identified by DAPI staining (shown in red color).
E, Representative immunofluorescence
image of cardiomyocytes after expression
of cyclinG1. The white arrow indicates a
binucleated cardiomyocyte. *P⬍0.05;
**P⬍0.01 by Student’s t test.
synthesized new DNA but did not divide.10,11 The effects
of cyclinG1 were compared to rat cardiomyocytes infected
with a LacZ control virus, in which the ratio of newly
generated daughter cells was set to 100% (Figure 4C).
Overexpression of cyclinG1 diminished the number of
newly generated neonatal rat cardiomyocytes to 39%,
whereas E2F2 increased the number to 175%. Coexpression of E2F2 and cyclinG1 reduced the ratio of newly
generated cardiomyocytes from 175% to 59%, which
illustrated the dominant role of cyclinG1 in the control of
cell cycle arrest.
Mitosis is not always followed by cytokinesis in cardiomyocytes but might also result in the formation of bior
multinucleated cells. We therefore assessed directly the
number of binucleated cardiomyocytes, which occurred in
response to cyclinG1 and E2F2 by immunofluorescence
staining. Comparison to neonatal rat cardiomyocytes,
which were infected with a LacZ control virus, revealed a
cyclinG1-dependent increase of binuclear cardiomyocytes
of 16%, whereas expression of E2F2 led to a reduction.
Moreover, concomitant expression of both E2F2 and
cyclinG1 resulted in 50% more binucleated cardiomyocytes compared to cardiomyocytes treated with E2F2 alone
(n⫽3, 700 cardiomyocytes from 10 randomly chosen
areas) (Figure 4D and 4E). Taken together, our results
suggested that cyclinG1 had a 2-sided effect on neonatal
rat cardiomyocytes: (1) it promoted DNA synthesis and
G1/S transition; and (2) it delayed mitosis causing an
accumulation in the G2/M phase and inhibited cytokinesis.
Yet, important quantitative differences of both effects
exist: the effects of cyclinG1 in imposing a G2/M block
were stronger than induction of S phase.
Knockout of CyclinG1 Causes an Early
Arrest of DNA Synthesis and Mitosis
and Reduces Polyploidization of Postnatal
Mouse Cardiomyocytes After
Pressure-Induced Hypertrophy
To investigate the role of cyclinG1 in cell cycle regulation
and polyploidization in vivo we analyzed the physiological
transition from proliferative to hypertrophic growth in the
postnatal myocardium of wild-type (WT) and cyclinG1
knockout mice and in response to cardiac overload and
hypertrophy. Western blot analysis of isolated cardiomyocytes revealed a more rapid decrease of the expression of
PCNA and phosphorylated Rb between d2 and d14 in
cyclinG1 knockout mice (Figure 5A and Online Figure
III). Similarly, we observed a rapid downregulation of
survivin, auroraB, and mad2, indicating that the loss of
cyclinG1 led to precocious cell cycle exit of cardiomyocytes (Figure 5A and Online Figure III). We also detected
an upregulation of Cks1, a Cdc2 adapter protein that
promotes cyclinB1degradation, between d2 and d10, which
was more pronounced in WT than in cyclinG1 knockout
mice (Figure 5A and Online Figure III). This finding
corresponds nicely to previous findings that overexpression Cks1 alters the status of the mitotic spindle cell cycle
checkpoint leading to mitotic exit in the absence of
chromosome segregation and cytokines and induction of
polyploidization in vascular smooth muscle cells.17,18 It did
not escape our attention that overexpression of cyclinG1 in
vitro and inactivation of cyclinG1 in vivo sometimes
caused similar effects (f. e. decreased levels of auroraB).
We assume that some differences were caused by inherent
divergences between both systems used. For our experiments we used cultures of synchronized neonatal cardio-
Liu et al
Regulation of Cardiomyocyte Polyploidy
1503
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Figure 5. Precocious arrest of DNA
synthesis and mitosis and reduction of
polyploidization in postnatal cyclinG1
mutant mouse cardiomyocytes. A,
Quantification of expression levels of different mitosis and cytokinesis markers in
postnatal mouse heart ventricles (n⫽4 for
WT, n⫽4 for mutants at each time point).
Loading of protein was normalized to
expression of pan-actin. B, Representative distribution of ploidy of cardiomyocyte nuclei at d14 (n⫽3). Results are displayed as a FACS scan (left) and as a bar
chart (right). C, Quantification of expression levels of histone H3 in postnatal
mouse heart ventricles at postnatal days
d2 to d14 (n⫽4 for WT, n⫽4 for mutants
at each time point). Protein loading was
normalized to pan-actin. *P⬍0.05;
**P⬍0.01 by Student’s t test. D, Representative immunofluorescence image of
isolated cardiomyocytes used for the
analysis of ploidy in cyclinG1 mutant
mice. A tetra-nucleated cardiomyocyte
from WT mice is shown (myosin heavy
chain and dystrophin staining in green;
DAPI staining in red).
myocytes whereas cardiac tissue contains cardiomyocytes
at all stages of the cell cycle. Moreover, cultured neonatal
cardiomyocytes, unlike their counterparts in vivo, undergo
only a single round of mitosis.
Next, we determined the degree of polyploidy of cardiomyocytes in 14-day-old and adult mouse hearts. We
observed a significant decrease of the DNA content of
cyclinG1 knockout cardiomyocytes from 6N to 4N by
FACS analysis (Figure 5B). This observation was corroborated by a decreased ratio of histone H3 to pan-actin, in
cyclinG1 mutant compared to WT mice, which reflects an
altered nucleus/cytoplasm relationship (Figure 5C). To
assess the number of nuclei in adult cardiomyocytes we
isolated myosin heavy chain positive cardiomyocytes and
stained the nuclei with DAPI (Figure 5D). We observed a
significant reduction of polynuclear (Nu⬎2) and a relative
increase of the number of binuclear (Nu⫽2) cardiomyocytes in cyclinG1 knockout mice (n⫽6, 600 cardiomyocytes from 10 randomly chosen areas) (Figure 6A), suggesting an impaired ability of cyclinG1 mutant
cardiomyocytes to undergo multinucleation. The relative
decrease of nuclear material in cardiomyocytes of cyclinG1 mutants was also reflected by the reduction of the
ratio of histone 3 to pan-actin (Figure 6C). Next, we
determined the number of cardiomyocytes in cyclinG1 and
WT mice per mm2 and measured the diameter of cardiomyocytes to rule out that the inactivation of the cyclinG1
gene, which also reduces the level of survivin (Online
Figure III), affects the number and size of cardiomyocytes.
In addition, we performed a number of Western blot
experiments to determine the ratio of cardiomyocytespecific proteins to proteins found in noncardiomyocytes
in the heart. In both experiments we did not find a
significant difference between WT and mutant hearts
suggesting that the number and size of cardiomyocytes was
not significantly altered in cyclinG1 mutant compared to
WT mice (Online Figure IV).
Induction of cardiac overload and hypertrophy in WT mice
by TAC revealed an increase of cardiomyocytes with multiple nuclei (Nu⬎2) from 9.5⫾0.4% to 10.9⫾0.1% (P⬍0.05)
and a decrease of mononucleated cardiomyocytes from
5.2⫾0.3% to 4.2⫾0.1% (P⬍0.05), whereas the number of
binucleated cardiomyocytes remained constant (85.3⫾0.4%
before and 85.0⫾0.1 after TAC) (n⫽6, 600 cardiomyocytes
from 10 randomly chosen areas) (Figure 6A and 6D). In
contrast, cyclinG1 mutant cardiomyocytes were characterized
by a decrease of mononucleated from 6.8⫾0.3% to
4.0⫾0.1% (P⬍0.05) and polynucleated (Nu⬎2) cardiomyocytes from 3.8⫾0.2% to 2.3⫾0.1% (P⬍0.05), whereas the
number of binucleated cardiomyocytes increased considerably (89.4⫾0.4% before and 93.7⫾0.1 after TAC, P⬍0.05)
(n⫽6, 600 cardiomyocytes from 10 randomly chosen areas)
(Figure 6A and 6D). We reasoned that the impaired ability of
cyclinG1 cardiomyocytes to undergo polyploidization prevented transition to a multinucleated state in response to
pressure overload and cardiac hypertrophy. The relative
decrease of the number of multinucleated (Nu⬎2) cardiomyocytes (Figure 6B and 6E) was most likely attributable to
1504
Circulation Research
May 14, 2010
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Figure 6. Reduced ploidy of adult
cyclinG1 mutant mouse cardiomyocytes under physiological conditions and after pressure overload.
A, Number of nuclei in adult cardiomyocytes. Cardiomyocytes were isolated
from 10-week-old mice, stained with
antibodies against myosin heavy chain
and dystrophin, and reacted with
DAPI (n⫽6 for each group, ⬎600 cardiomyocytes from 20 randomly chosen
fields were counted for each individual
animal). B and E, Representative immunofluorescence images of cardiomyocytes isolated from cyclinG1 mutant (B)
and WT mice (E) 3 weeks after TAC
(myosin heavy chain and dystrophin
staining in green; DAPI staining in red).
C, Quantification of expression levels
of histone H3 in adult mouse heart
ventricles (10 weeks) (n⫽6). Protein
loading was normalized to pan-actin.
D, Number of nuclei in adult cardiomyocytes 3 weeks after TAC (n⫽6 for each
group, ⬎600 cardiomyocytes from 20
randomly chosen fields were counted
for each individual animal). Note the
decreased number of polynucleated cardiomyocytes in cyclinG1 knockout mice. *P⬍0.05; **P⬍0.01 by Student’s t test for
A, C, and D.
the increase of binucleated cardiomyocytes that occurred in
TAC-operated cyclinG1 mutants. We would like to point out
that changes in the stoichiometry of bi- and multinucleated do
not necessarily require proliferation or ablation of specific
cell populations because our measurements compared the
relative distribution of different cell populations and not
absolute numbers in the heart.
To investigate the response of cyclinG1 mutant hearts to
loss of contractile tissue we ligated the left anterior
descending artery (LAD) in cyclinG1 and WT control mice
and measured cardiac function by MRI. We found that the
end-diastolic and end-systolic volumes were significantly
increased in cyclinG1 mutant compared to WT mice,
which corresponded to a significantly reduced ejection
fraction in mutant mice (Online Figure V). Apparently, the
lack of cyclinG1 impaired the ability of infarcted hearts to
cope with the loss of contractile tissue after infarction. It
was also interesting to note that the reduced ejection
fraction of cyclinG1 mutant mice, which show a reduced
expression of survivin (Online Figure III), corresponds to
impaired cardiac function observed in survivin deficient
hearts.19
Taken together, our results clearly demonstrate that cyclinG1 is instrumental for the acquisition of the regular
degree of polyploidization in postnatal cardiomyocytes in
vivo. Lack of cyclinG1 induced a precocious exit of maturating cardiomyocytes from the cell cycle causing a reduced
DNA content and a smaller number of nuclei in young and
adult cardiomyocytes. Moreover, lack of cyclinG1 prevented
the increase of polynucleated cardiomyocytes in response to
pressure overload and hypertrophy.
Discussion
Polyploidy is a characteristic feature of some cells types in
the mammalian body and coincides most often with terminal
differentiation ie, with the acquisition of specialized cellular
properties.3 In addition, polyploidy can be induced by stress
and injury in the cardiovascular system,18 suggesting that
polyploidization serves specific functions in these cells.4 In
fact, it has been described that increased polyploidy alters
cellular physiology and enables cells to respond to specific
needs.20 Polyploidization might be particularly relevant for
cells such as cardiomyocytes, which are generally unable to
replicate and hence possess only limited number of options to
deal with increased metabolic and biophysical stress.21 So far,
very little is known about the molecules, which activate and
mediate the process of polyploidization in cardiomyocytes.
Here, we have demonstrated that cyclinG1 is an important
component of the machinery that controls polyploidization of
cardiomyocytes based on 4 major arguments. (1) The expression of cyclinG1 during postnatal maturation of cardiomyocytes correlates with the onset of polyploidization. (2) Overexpression of cyclinG1 in primary neonatal rat
cardiomyocytes increased DNA synthesis and delayed mitosis, which interfered with proliferation but enlarged the
polynucleated subpopulation of cardiomyocytes. (3) Inactivation of cyclinG1 in mice caused cardiomyocytes to withdraw
faster from cell cycling resulting in reduced numbers of
polynucleated cells. (4) Lack of cyclinG1 prevented the
increase of polynucleated cardiomyocytes in response to
pressure overload and hypertrophy. The loss of cyclinG1,
however, did not block polyploidization in cardiomyocytes
completely, which is not surprising given the complex array
of different factors and mechanisms that affect cell cycle
control. Clearly, the process of polyploidization of cardiomyocytes, which involves both endoreduplication (inhibition
of mitosis) and multinucleation (inhibition of cytokinesis), is
controlled by multiple factors that fine tune the process while
responding to physiological and pathophysiological requirements. A complete block of polyploidization might disrupt
Liu et al
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
cell cycle regulation in cardiomyocytes and hence might be
incompatible with life or would require thorough cellular
reprogramming.
The generation of polyploid cardiomyocytes might be
viewed as a specialized type of cell cycle control, which
uncouples DNA replication from cell division. During this
process, cells skip mitosis and proceed through one or
several rounds of DNA replication, resulting in autopolyploid cells, which do not proliferate any further.
Because checkpoint activation sometimes does not persist
cells might “sneak” past the arrest and produce a tetraploid
cell if apoptosis is inhibited. Interestingly, cyclinG1 has
been identified as a part of a negative feedback loop that
down-modulates p53 activity via ARF and MDM-2.22
Thus, it is tempting to speculate that a part of the function
of cyclinG1 in polyploidization is attributable to inhibition
of apoptosis caused by checkpoint activation. Further
support for this hypothesis comes from the results of our
myocardial infarction experiments in cyclinG1 mutant
mice. The reduced ejection fraction of cyclinG1 mutant
mice after MI compared to WT resembles the phenotype of
mutants, which lack antiapoptotic molecules such as survivin19 and GDF-5.23 Another molecule, which might be
involved in cyclinG1-mediated polyploidization is cyclin
D1. Because cyclin D1 was significantly upregulated by
overexpression of cyclinG1 in neonatal cardiomyocytes it
seems reasonable to conclude that several effects of
cyclinG1 such as increased numbers of cardiomyocytes in
S phase as well as increased concentrations of phosphorylated Rb, E2F2, and cyclin E1 were caused by cyclin D1.
This view is also supported by previous reports, which
described that overexpression of cyclin D1 in cardiomyocytes induced multinucleation.24
Our finding that cyclinG1 is instrumental for the proper
control of polyploidization in cardiomyocytes is the first to
uncover specific functions of a member of the cyclin family
of genes for polyploidization. We propose that cyclinG1 is
instrumental to convert incoming cues, which would cause
proliferation in other cell types, into signals for polyploidization. In cardiomyocytes proproliferative signals often cause
hypertrophy, which is tightly linked to polyploidy, suggesting
that both processes are part of a program that enables
cardiomyocytes to respond to specific physiological requirements.15 Interestingly, molecules of the PI3K/Akt pathway,
which plays a major role in cardiac hypertrophy,25 have
recently been described to control also the generation of
binucleated tetraploid liver cells26 and polyploidization of
vascular smooth muscle cells (VSMC).17 Although the mechanism of polyploidization in hepatocytes, VSMCs and cardiomyocytes differ this coincidence is intriguing and further
illustrates the link between hypertrophy and polyploidization.
At present, it is unclear whether polyploidization represents an advantage or a disadvantage for cardiomyocytes
although polyploidization has been linked to the increase of
cell viability and protection from apoptosis in cardiomyocytes and hepatocytes.27 More detailed knowledge about this
process will open new ways to stimulate or prevent
polyploidization in an effort to interfere with cellular decompensation and improve cardiomyocyte function.
Regulation of Cardiomyocyte Polyploidy
1505
Sources of Funding
This work was supported by the Max-Planck-Society, Deutsche
Forschungsgemeinschaft grant SFB 547, the Kerckhoff-Foundation,
the Excellence Initiative “Cardiopulmonary System,” and the University of Giessen and Marburg Lung Center.
Disclosures
None.
References
1. Clubb FJ Jr, Bishop SP. Formation of binucleated myocardial cells in the
neonatal rat. An index for growth hypertrophy. Lab Invest. 1984;50:
571–577.
2. Meckert PC, Rivello HG, Vigliano C, Gonzalez P, Favaloro R, Laguens
R. Endomitosis and polyploidization of myocardial cells in the periphery
of human acute myocardial infarction. Cardiovasc Res. 2005;67:
116 –123.
3. Brodsky VY. Cell ploidy in the mammalian heart. In: Oberpriller JO, ed.
The Development and Regenerative Potential of Cardiac Muscle.
London, UK: Harwood Academic Publishers; 1991;253–292.
4. Storchova Z, Pellman D. From polyploidy to aneuploidy, genome instability and cancer. Nat Rev Mol Cell Biol. 2004;5:45–54.
5. Poolman RA, Brooks G. Expressions and activities of cell cycle regulatory molecules during the transition from myocyte hyperplasia to hypertrophy. J Mol Cell Cardiol. 1998;30:2121–2135.
6. Kimura SH, Ikawa M, Ito A, Okabe M, Nojima H. Cyclin g1 is involved
in g2/m arrest in response to DNA damage and in growth control after
damage recovery. Oncogene. 2001;20:3290 –3300.
7. Horne MC, Goolsby GL, Donaldson KL, Tran D, Neubauer M, Wahl AF.
Cyclin g1 and cyclin g2 comprise a new family of cyclins with contrasting tissue-specific and cell cycle-regulated expression. J Biol Chem.
1996;271:6050 – 6061.
8. Jensen MR, Factor VM, Fantozzi A, Helin K, Huh CG, Thorgeirsson SS.
Reduced hepatic tumor incidence in cyclin g1-deficient mice. Hepatology.
2003;37:862–870.
9. Ebelt H, Jungblut M, Zhang Y, Kubin T, Kostin S, Technau A, Oustanina
S, Niebrugge S, Lehmann J, Werdan K, Braun T. Cellular cardiomyoplasty: Improvement of left ventricular function correlates with the
release of cardioactive cytokines. Stem Cells. 2007;25:236 –244.
10. Ebelt H, Hufnagel N, Neuhaus P, Neuhaus H, Gajawada P, Simm A,
Muller-Werdan U, Werdan K, Braun T. Divergent siblings: E2f2 and e2f4
but not e2f1 and e2f3 induce DNA synthesis in cardiomyocytes without
activation of apoptosis. Circ Res. 2005;96:509 –517.
11. Rabinovitch PS, Kubbies M, Chen YC, Schindler D, Hoehn H. Brduhoechst flow cytometry: A unique tool for quantitative cell cycle analysis.
Exp Cell Res. 1988;174:309 –318.
12. Knudsen ES, Wang JY. Differential regulation of retinoblastoma protein
function by specific cdk phosphorylation sites. J Biol Chem. 1996;271:
8313– 8320.
13. Garnovskaya MN, Mukhin YV, Vlasova TM, Grewal JS, Ullian ME,
Tholanikunnel BG, Raymond JR. Mitogen-induced rapid phosphorylation
of serine 795 of the retinoblastoma gene product in vascular smooth
muscle cells involves erk activation. J Biol Chem. 2004;279:
24899 –24905.
14. Yoshida K, Yamaguchi T, Natsume T, Kufe D, Miki Y. Jnk phosphorylation of 14 –3-3 proteins regulates nuclear targeting of c-abl in the
apoptotic response to DNA damage. Nat Cell Biol. 2005;7:278 –285.
15. Ahuja P, Sdek P, MacLellan WR. Cardiac myocyte cell cycle control in
development, disease, and regeneration. Physiol Rev. 2007;87:521–544.
16. Amano T, Kaji N, Ohashi K, Mizuno K. Mitosis-specific activation of lim
motif-containing protein kinase and roles of cofilin phosphorylation and
dephosphorylation in mitosis. J Biol Chem. 2002;277:22093–22102.
17. Hixon ML, Muro-Cacho C, Wagner MW, Obejero-Paz C, Millie E, Fujio
Y, Kureishi Y, Hassold T, Walsh K, Gualberto A. Akt1/pkb upregulation
leads to vascular smooth muscle cell hypertrophy and polyploidization.
J Clin Invest. 2000;106:1011–1020.
18. Hixon ML, Obejero-Paz C, Muro-Cacho C, Wagner MW, Millie E, Nagy
J, Hassold TJ, Gualberto A. Cks1 mediates vascular smooth muscle cell
polyploidization. J Biol Chem. 2000;275:40434 – 40442.
19. Levkau B, Schafers M, Wohlschlaeger J, von Wnuck Lipinski K, Keul P,
Hermann S, Kawaguchi N, Kirchhof P, Fabritz L, Stypmann J, Stegger L,
Flogel U, Schrader J, Fischer JW, Hsieh P, Ou YL, Mehrhof F, Tiemann
K, Ghanem A, Matus M, Neumann J, Heusch G, Schmid KW, Conway
1506
Circulation Research
May 14, 2010
EM, Baba HA. Survivin determines cardiac function by controlling total
cardiomyocyte number. Circulation. 2008;117:1583–1593.
Galitski T, Saldanha AJ, Styles CA, Lander ES, Fink GR. Ploidy regulation of gene expression. Science. 1999;285:251–254.
Rubart M, Field LJ. Cardiac regeneration: Repopulating the heart. Annu
Rev Physiol. 2006;68:29 – 49.
Harris SL, Levine AJ. The p53 pathway: Positive and negative feedback
loops. Oncogene. 2005;24:2899 –2908.
Zaidi SH, Huang Q, Momen A, Riazi A, Husain M. Growth differentiation factor 5 regulates cardiac repair after myocardial infarction. J Am
Coll Cardiol. 2010;55:135–143.
Soonpaa MH, Koh GY, Pajak L, Jing S, Wang H, Franklin MT, Kim KK,
Field LJ. Cyclin d1 overexpression promotes cardiomyocyte DNA syn-
20.
21.
22.
23.
24.
thesis and multinucleation in transgenic mice. J Clin Invest. 1997;99:
2644 –2654.
25. Heineke J, Molkentin JD. Regulation of cardiac hypertrophy by
intracellular signalling pathways. Nat Rev Mol Cell Biol. 2006;7:
589 – 600.
26. Celton-Morizur S, Merlen G, Couton D, Margall-Ducos G, Desdouets C.
The insulin/akt pathway controls a specific cell division program that
leads to generation of binucleated tetraploid liver cells in rodents. J Clin
Invest. 2009;119:1880 –1887.
27. Anatskaya OV, Vinogradov AE. Genome multiplication as adaptation to
tissue survival: Evidence from gene expression in mammalian heart and
liver. Genomics. 2007;89:70 – 80.
Novelty and Significance
What Is Known?
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
●
●
●
Polyploidization of cardiomyocytes occurs mainly during the first
three weeks of postnatal life in mice and rats and during the first
postnatal decade in human beings.
Cardiac remodeling and hypertrophy might lead to polyploidization
and multinucleation of cardiomyocytes during adult life.
CyclinG1 is a transcriptional target of p53 and is involved in several
p53 related processes such as apoptosis, growth control, and
checkpoint regulation in response to DNA damage.
What New Information Does This Article Contribute?
●
●
●
CyclinG1 was identified as an important component of the molecular
machinery controlling polyploidization and multinucleation of
cardiomyocytes.
Directed expression of cyclinG1 in neonatal cardiomyocytes promotes
G1/S cell cycle transition but inhibits cytokinesis thereby promoting cardiomyocyte polyploidy.
Adult cardiomyocytes of cyclinG1 knockout mice show a reduced
number of nuclei per cell both under physiological conditions and
after pressure-induced hypertrophy.
Polyploidization or the increase of the cellular DNA content is an
intrinsic program of cardiomyocyte development. Yet, relatively
little is known about the molecular mechanisms that control
polyploidization and multinucleation. Similarly, the reasons why
genome duplication occurs during postnatal heart development
and why the DNA content of cardiomyocytes increases under
disease conditions are enigmatic. Here, we show that overexpression of cyclinG1 promotes DNA synthesis but inhibits
division of neonatal cardiomyocytes leading to an enlarged
population cardiomyocytes with two or more nuclei. Furthermore, we demonstrate that inactivation of the cyclinG1 gene in
mice lowers the degree of polyploidy and multinucleation in
cardiomyocytes and prevents an increase of polynucleated
cardiomyocytes in response to pressure overload and hypertrophy. Our results define cyclinG1 as an important component of
the machinery that controls the DNA content of cardiomyocytes.
We propose that our study opens new ways to understand and
manipulate polyploidization. Because polyploidization has been
linked to an increase of cell viability and protection from
apoptosis a better knowledge of the underlying molecular
mechanisms might help to improve cardiomyocyte function and
prevent cellular decompensation.
Regulation of Cardiomyocyte Polyploidy and Multinucleation by CyclinG1
Zhipei Liu, Shijing Yue, Xiaobo Chen, Thomas Kubin and Thomas Braun
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Circ Res. 2010;106:1498-1506; originally published online April 1, 2010;
doi: 10.1161/CIRCRESAHA.109.211888
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 2010 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/106/9/1498
Data Supplement (unedited) at:
http://circres.ahajournals.org/content/suppl/2010/04/01/CIRCRESAHA.109.211888.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation Research 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 Research is online at:
http://circres.ahajournals.org//subscriptions/
Supplemental material
Full Materials and methods
Animals, surgery and MRI analysis
Cyclin G1 knockout mice were kindly provided by Prof. Thorgeirsson (NIH) and maintained
on a C57BL6 background (>9 backcrossings to C57BL6)1. Wild type C57BL6 mice and
Sprague Dawley rats were purchased from Harlan Winkelman GmbH (Borchen, Germany).
For TAC (transaortic constriction) male C57BL/6 mice or Cyclin G1 mutant mice were
intubated, ventilated with oxygen, and anesthetized with inhaled 1.5 – 2.0% Isoflurane. To
induce reproducible pressure overload of the left ventricle, a constriction (approx. 66%) of the
proximal aorta was applied, based on a previously described banding procedure 2. After right
sided upper thoracotomy the outflow tract of the left heart was exposed and the aorta was
freed from surrounding tissues before a Weck® haemoclip was put around the aorta resulting
in TAC. The haemoclip applicater was endued with an adjuster bolt, allowing only a
predefined partial closure of the clip. Cardiomyocytes were isolated 21d after operation using
previously published protocols 3, 4.
Myocardial infarction was accomplished by ligation of the the left anterior descending artery
(LAD) as described
5, 6
. In brief, mice were anesthetized with isofluorane and ventilated
mechanically, left sided thoracotomy was performed in the 4th ICR, and the LAD was ligated
proximal to its main bifurcation.
Cardiac magnetic resonance imaging (MRI) was performed at time points 0d (before OP) and
21d (3 weeks post OP) on a Bruker Pharmascan 7.0 T, equipped with a 300mT/m gradient
system, using a custom-built circularly polarized birdcage resonator and the Early Access
Package for self-gated cardiac Imaging (Intragate, Bruker, Ettlingen, Germany).
Measurements were based on the gradient echo method (repetition time= 6.3 ms; echo time=
1.5 ms; flip angle = 15°; field of view= 2.20x2.20 cm; slice thickness= 1.0 mm;
matrix=128x128; repetitions=100; resolution = 0.0172 cm/pixel). The imaging plane was
localized using scout images showing the 4- and 2-chamber view of the heart, followed by
acquisition in short axis view, orthogonal on the septum in both scouts. Multiple contiguous
short-axis slices consisting of 6 to 8 slices were acquired for complete coverage of the left
ventricle. All MRI data were analyzed using Qmass digital imaging software (Medis, Leiden,
Netherlands). Mice were anaesthesized using isoflurane (1.7 - 2.0 %) and the body
temperature was maintained throughout the measurements.
All animal experimentations were performed in accordance with German legislation on
protection of animals and were approved by the local governmental animal care committee
(Regierungspraesidium Darmstadt).
Generation of adenoviruses, cell culture, and FACS
Adenoviruses expressing E2F2 and LacZ were generated as described previously 3. The
mouse cyclin G1 cDNA (generous gift from Dr. Mary C. Horne, University of Iowa, USA)
was cloned into pShuttle-CMV and used to generate a recombinant adenovirus using the
AdEasy system following instructions of the manufacturer (Stratagene).
Cardiomyocytes from rats and mice were isolated using standard procedures. Neonatal cells
were seeded in DMEM/199 (4/1) medium containing 1% FCS. Cardiomyocytes were
synchronized in serum free medium for 24 hours before infection with adenoviruses carrying
LacZ, E2F2 or cyclin G1. Infection was achieved by incubation of cells with DMEM/199
medium containing adenovirus (50 pfu/cell) and insulin-transferrin-sodium selenite media
supplement (Sigma) for 36 hours.
DNA content measurements by FACS were performed as described previously 3, 7. Cell cycle
profiles were calculated using the ModFit LT software (Verity). Proliferation of
cardiomyocytes was analyzed using the BrdU-Hoechst method 3, 8. Briefly, cells were cultures
with BrdU for 48 hrs. Once BrdU is incorporated in DNA, the Hoechst but not the PI staining
will be quenched. Thus, newly generated cells will present as Hoechst staining 1N and PI
staining 2N cells. Cells, which have not replicated/proliferated and stayed in G1, will be
detected as Hoechst 2N and PI 2N. The ratio of proliferated cells was calculated as the ratio of
replicated cells to the total number of cells. All FACS analyses were restricted to myosin
heavy chain-positive cells, to prevent contamination from other cell types.
Quantitative Real-Time PCR
Total RNA was isolated from hearts using RNeasy kit (Qiagen) following the instructions of
the manufacturer. cDNA was synthesized with oligo-dT primers. Real-time PCR was
performed using the iCycler system (Bio-Rad) using the following primers: cyclin G1 forward
5'-gttccaggacagctggagag-3’, reverse 5’- tgttgctgctgctgtaatcc-3’; GAPDH forward 5’gatgacatcaagaaggtggtga-3’, reverse 5’-accctgttgctgtagccatatt-3’.
Immunofluorescence
Cardiomyocytes were seeded in 2-well chamber slides using standard procedures. For
immunofluorescence cells were stained with antibodies against myosin heavy chain (MF20),
auroraB, α-tubulin, collagen VI, cardiac troponin I and dystrophin (Sigma) as described
previously
4, 9
. Nuclei were visualized by 4′, 6-Diamidino-2-phenylindole dihydrochloride
(DAPI) staining.
Tritiated thymidine (³H-thymidine) incorporation
To quantify DNA synthesis cells were pulse labeled with ³H-thymidine for 12 hours before
harvest. Cell lysates were prepared and analyzed as described previously 10.
Protein Isolation, Western Blots
Protein isolation and Western blot analysis were performed using standard protocols. Protein
samples (15 µg) were separated in 4% - 12% SDS-PAGE gradient gels (Invitrogen) and
transferred onto nitrocellulose membranes. Proteins were identified using the Super-Signal
West Femto (Pierce) detection solutions, and images were acquired using the VersaDoc
system (Bio-Rad). The following antibodies were used: anti-cyclin G1 (Santa Cruz Biotech);
anti-α-actinin (Sigma); anti-cyclin E and anti-E2F2 (both from Abcam); anti-proliferating cell
nuclear antigen (PCNA), anti-mitotic arrest deficient 2 (Mad2) and anti-Ral A (all from BD
Pharmingen); anti-phosphorylated Histone H3, anti-Histone H3, anti-cyclin D1, anti-cyclin
D3, anti-phosphorylated cofilin, anti-cofilin, anti-phosphorylated retinoblastoma protein (Rb,
s-795 and s-807/811), anti-polo-like kinase 1 (Plk1), anti-phosphorylated Erk1/2, antiphosphorylated p38MAPK, anti-pan-actin and anti-survivin (all from Cell Signaling
Technology); and anti-atrial natriuretic peptide (ANP, Milipore).
Morphometric analysis of WT and cyclinG1 mutant hearts
Hearts were isolated from from age-matched adult cyclinG1 mutant and WT mice (3 month),
fixed, and processed as described previously 6. Hearts were sectioned completely at 7μm,
yielding approximately 100 slides per heart with each slide carrying 8-10 tissue sections. 37
slides were selected from different parts of the heart for each genotype and stained with
collagen VI and myosin heavy chain antibodies. The number of cardiomyocytes were counted
from five random fields from each slide and used to calculate the diameter of cardiomyocytes.
The diameter was defined as the average of two vertical diameters of every cardiomyocyte.
Statistical analysis was done using student's t-test.
References:
;7
<7
=7
>7
?7
@7
A7
B7
C7
;:7
','4-(+4'-(33"4%"'4.!4!(+ "+,,('7.
!)-"-.&(+"'"'"'2%"' ;8""'-&"77<::=5=A6B@<8
BA:
($&'4'(4(,,4(',4+"&"4!+ /4(,,4+74!"'
7(%.%+')!2,"(%( "%%-+-"(',"'&.+"'/'-+".%+2,.'-"('7
7;CC>5C;6<@C>8<@CB
%-4.' %4.!.,4.!.,4#04"&&4.%%+8+'
4+'4+.'7"/+ '-,"%"' ,6<<'<>.-'(-<;'<=
"'.,2'-!,","'+"(&2(2-,0"-!(.--"/-"('()()-(,",77
<::?5C@6?:C8?;A
$!+.,!/4&(%$4#04(,-"'4(-- +4."'4+.'4
(+7"+-A"'+,,,-+,,+,",-'(+"(&2(2-,')+/'-,
)()-(,",'"'%&&-(+2+"(&2()-!2"'&"77<::B5;:<6A:=8A;:
%-4.' %.-4!' 4."'4(,-"'4!'.4.,-'"'4
"+. 4!&''4+'4+.'7%%.%++"(&2()%,-26
&)+(/&'-(%-/'-+".%+.'-"('(++%-,0"-!-!+%,(
+"(-"/2-($"',77<::A5<?6<=@8<>>
%&84!"4+-"+4(,-"'4+.'7""'-!(&"' (
&.%-")(-'-.%-&,'!2&%,-&%%,)',('+(.'-8&"-
%.,-+"' (+<77<::B5<6?@@8?A?
.",4"'(/"-!7%(02-(&-+"'%2,",(-(+,0!"!"'%.'
-!+.+8!(!,-*.'!"' -"'.%-"/-!.&'"+(%,-,'
%2&)!(2-,77;CB=5=6<A@8<B;
"'(/"-!4.",4!'4!"'%+4(!'7+.8!(!,-%(0
2-(&-+26.'"*.-((%(+*.'-"--"/%%2%'%2,",77
;CBB5;A>6=:C8=;B
!.%34%&84!'.4+.'7,'!2&%,-&%%,+
++."--(,-+"-&.,%2'-9"%8>8&"-%%.,"('77
<::?5;C6;ABA8;ACB
$"$24.+"'4',('4%+4+/%!(4."'4+(4
+"'!"+"47+(&"''-%",!&'"'-" '-!-,-"&.%-,!.&'
)+")!+%%((&('('.%+%%,-(1)+,,-!;8-2)2-($"')+("%'-(
)+(."'-+%.$"';<77;CC?5;B;6;?<A8;?=A
Online Figure I: Western blot analysis of the expression of phosphorylated cofilin and
total cofilin in neonatal rat cardiomyocytes. Cardiomyocytes were synchronized in serum
free medium for 24 hours before expression of LacZ, cyclin G1 (CyG1), E2F2, and a
combination of E2F2 and cyclin G1 (E2F2-CyG1). Mock-infected cells were used as a
control.
Online Figure II: Western blot analysis of the expression of phosphorylated histone H3,
PLK1, and pan-actin. Cardiomyocytes were synchronized in serum free medium for 24
hours before expression of LacZ, cyclin G1 (CyG1), E2F2, and a combination of E2F2 and
cyclin G1 (E2F2-CyG1). Pan-actin was used as a loading control.
Online Figure III: Western blot analysis of the expression of different cell cycle markers.
Proteins were isolated from WT and cyclin G1 mutant hearts at different time points as
indicated. Pan-actin was used as a loading control.
Online Figure IV: CyclinG1 mutant mice do not show changes in the number and size of
cardiomyocytes. (A, C) Western blot based analysis of the ratio of cardiomyocyte specific
proteins (cardiac-troponin) to non-cardiomyocyte proteins (vimentin, collagenVI, α-SM
actin). Proteins were isolated from WT and cyclin G1 mutant hearts at different time points as
indicated. Pan-actin was used as a loading control. No significant differences between WT
and cyclinG1 mutant mice were detected by student's t-test. (B) No significant differences
were found between the heart weight/body weight ratios of WT and cyclinG1 mutant mice.
(C) Morphometric analysis of the number and diameter of cardiomyoctes in adult WT and
cyclinG1 mutant mice. No significant difference was detected by student's t-test.
Online Figure V: Reduced cardiac function of cyclinG1 mutant mice compared to WT
mice after cardiac infarction. Mid-ventricular (A, B) and coronal (C, D) MRI images of WT
(A, C) and cyclinG1 mutant mice (B, D) 3 weeks after ligation of the LAD. (E, F)
Quantitative evaluation of MRI data revealed an increased end-diastolic (EDV) and endsystolic volume (ESV) in cyclinG1 mutant compared to WT mice 3 weeks after ligation of
the LAD, which corresponded to a reduced ejection fraction (EF) in cyclinG1 mutant mice
(F). *, p < 0.01 by student's t-test.