Cardiovascular Research 67 (2005) 116 123

www.elsevier.com/locate/cardiores

Endomitosis and polyploidization of myocardial cells in the periphery

of human acute myocardial infarction
Patricia Cabeza Meckerta,c, Hernan Garca Rivelloa, Carlos Viglianoa, Pedro Gonzaleza,c,
Roberto Favalorob, Ruben Laguensa,*
a

Division of Pathology, Institute of Cardiology and Cardiovascular Surgery, Favaloro Foundation, Buenos Aires, Argentina
Division of Cardiothoracic Surgery and Heart-Lung Transplantation, Institute of Cardiology and Cardiovascular Surgery,
Favaloro Foundation, Buenos Aires, Argentina
c
Comision de Investigaciones Cientficas de la Provincia de Buenos Aires, La Plata, Argentina

Received 12 October 2004; received in revised form 14 February 2005; accepted 20 February 2005
Available online 31 March 2005
Time for primary review 20 days

Abstract
Objective: Although the genetic program for reinitiating DNA synthesis exists in post-mitotic cardiomyocytes, and it was reported that in
human acute myocardial infarction (AMI) a significant proportion of myocytes enter mitosis, the rule is that the lost tissue is replaced by a
collagen scar. The purpose of this study was to search for the basis of this discordance in order to devise future strategies to induce division of
myocytes into daughter cells that may replace the lost tissue with contractile cells.
Methods: In 15 human hearts with 1- to 21-day-old infarcts, the expression of the cell cycle proteins Ki67 antigen, cyclins D, A, and B1, the
presence of mitotic bodies, and the ploidy status were investigated with immunoenzymatic methods, light and laser confocal microscopy, and
densitometry in the myocytes surrounding the infarct area.
Results: In 7- to 13-day-old infarcts, 11.61 F 6.94% of the myocytes presented Ki67+ nuclei, and a lower proportion presented cyclins
D, A, and B. At earlier and later times, the proportion of Ki67+ myocytes was significantly lower. Although under confocal microscopy
and fluorescent labels, some of the Ki67+ myocytes appeared to be in different stages of mitosis, with Nomarski optics and hematoxylin
counterstaining, the condensed chromosomes, although arranged in metaphase and anaphase plates or split in sister chromatids, were
always located within a preserved nuclear envelope, indicating the presence of endomitosis. Conventional mitosis was exceptionally
observed. In the 14- and 21-day-old infarcts, the ploidy of the myocytes adjacent to the infarct was significantly higher than in distant
zones.
Conclusion: These observations indicate that in human infarcts, entrance of cardiomyocytes into the cell cycle is transient and that
endomitosis, leading to polyploidy, rather than mitosis, leading to karyokinesis, is the final fate of cycling cells. Both observations may
account for the discordance between the regenerative ability of myocytes and the lack of an efficient reparative process in human AMI.
D 2005 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
Keywords: Cell differentiation; Hypertrophy; Infarction; Myocytes; Remodeling

1. Introduction

T Corresponding author. Instituto de Cardiologia y Cirugia Cardiovascular, Fundacion Favaloro, Belgrano 1746, Buenos Aires 1092, Argentina.
Tel./fax: +54 11 43 78 13 15.
E-mail address: rlaguens@ffavaloro.org (R. Laguens).

In the last few years convincing evidence has been

published showing that the genetic program for reinitiating
DNA synthesis exists in post-mitotic cardiomyocytes [1], and
that in human acute myocardial infarction (AMI) a proportion
of adult myocytes enter in mitosis and proceed towards
cytokinesis with formation of daughter cells [2]. However,

0008-6363/$ - see front matter D 2005 European Society of Cardiology. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.cardiores.2005.02.017

2.40
1.60
1.50
0.50
0.40
2.00
3.20
1.36
2.30
21.30
10.00
30.60
51.50
13.60
15.43
16.40
20.90
18.20
19.30
26.90
6.20
41.60
9.80
15.30
27.27
16.90
1.90
15.10
6.50
33.07
48.30
44.20
42.20
61.90
28.70
79.90
37.00
67.40
49.00
29.54
54.90
74.30
37.10
37.20
66.03
35.30
34.90
39.60
18.80
44.40
13.90
21.40
22.80
35.70
43.18
28.20
23.80
47.80
56.30
0.00
18.80
22.50
19.70
19.80
27.30
4.20
44.80
11.16
17.60
48.57
26.90
32.50
66.60
20.10
48.50
43.40
45.10
35.50
61.60
32.40
70.90
42.70
82.30
60.00
31.42
40.80
51.40
24.10
44.10
51.50
37.80
32.40
44.80
18.60
40.30
24.90
12.50
6.54
22.40
20.00
32.30
16.10
9.30
35.80
0.00
0.00
0.00
0.00
0.00
0.00
2.36
4.57
3.05
0.06
1.03
0.04
1.94
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.67
0.24
0.33
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.54
0.00
0.00
3.54
4.83
7.22
0.86
0.74
1.66
1.83
0.00
0.00
0.00
0.00
0.00
0.82
0.00
0.00
4.26
6.11
7.43
1.13
1.10
0.72
2.59
0.00
0.00
0.00
0.00
0.00
1.67
0.00
0.00
15.34
16.47
18.07
3.95
4.22
3.15
7.52
0.21
0.00
2.42
1
2
2
3
4
7
7
8
9
12
14
16
19
21
21
M
F
F
M
F
M
M
M
F
M
M
M
M
M
M

4C
2C
4C
2C
Cyclin A
Cyclin D
KI67

Evolution
(days)
Sex

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
A

Age
Patient

The hearts of 15 patients with left ventricle AMI were

obtained. Nine were collected after necropsies performed
within 7 h after death and six were hearts explanted from
transplanted patients.
Time of evolution of the condition was determined since
diagnosis, defined as CK elevation twice the normal value
or CK-MB N 24 UI/ml, plus ST-segment elevation of N 5 mV
in two or more consecutive leads. In no case the time
elapsed between onset of symptoms and diagnosis exceeded
8 h. Table 1 shows the patient characteristics and time of
evolution. The investigation conforms with the principles
outlined in the Declaration of Helsinki.
As controls we employed 5 human hearts that were
rejected for heart transplantation because of unsuitability of
the donor (n = 3) or excessive time from organ collection.
Donors were 1734 years old and three were men. Heart
weights ranged between 175 and 305 g, and no gross or
microscopic abnormalities were detected.

Table 1
Patients, cell cycle proteins, endomitosis and ploidy status

2.1. Patients

Group

2. Methods

Periinfarction positive cardiomyocytes (%)

Cyclin B

Endomitosis
(%)

Ploidy (periinfarct)

8C

Ploidy (distant)

8C

the rule is that in acute myocardial infarction the lost tissue is

replaced by a fibrous scar, and subsequent pathological
remodeling is one of the leading causes of heart failure [3,4].
This is the main reason for the increasing number of studies
attempting to replace the missing cardiomyocytes with
skeletal myoblasts or progenitor stem cells [5,6].
The mechanisms underlying the discordance between a
putative myocyte replication ability and the lack of an
efficient reparation are poorly understood. At present it is
not known if the newly formed cells have low viability
and die after completion of mitosis, if they enter in a
quiescent G0 state after a few replication cycles or if the
time of myocyte replication is very brief and the number
of newly formed cells is negligible as compared with the
whole myocardial mass. In addition, since it is known that
in the human heart the cardiomyocytes surrounding
fibrous scars display a high degree of polyploidization
and hypertrophy [7,8], the possibility that the final result
of the mitotic activity is the formation of polyploid
hypertrophic myocytes should also be considered.
With the aim of exploring which of these mechanisms
could be involved in this deficient reparative process, we
studied human hearts collected between 1 and 21 days after
diagnosis of an AMI.
In this report we present evidence that in human AMI
entrance of the myocytes into the cell cycle is transient and
that in these cells endomitosis, leading to polyploidy,
instead of conventional mitosis, leading to karyokinesis, is
the main mechanism of chromosome duplication. Both
observations may account for the discordance between the
regenerative ability of myocytes and the lack of an efficient
reparative process in human AMI.

117

50
65
25
57
19
61
72
62
47
79
47
56
56
65
42

Difference of 8c
(periinfarct distant)

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

118

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

2.2. Tissue preparation

The left ventricle was sectioned normally to the longer
axis in 1-cm-thick slices that were fixed for 72 h in 10%
phosphate buffered formaldehyde, divided in fragments and
embedded in paraffin. Four micrometer thickness tissue
sections were stained with hematoxylineosin, Gomoris
trichrome and Feulgens.
2.3. Immunohistochemistry
The tissue sections were deparaffinized and brought to
PBS, pH 7.2. After blocking endogenous peroxidase with
3% H2O2 in methanol and antigen retrieval pretreatment
with citrate buffer in a microwave oven, the slides were
incubated 1 h with monoclonal antibodies against the Ki67
antigen and cyclins D1, A and B1 (Novocastra Laboratories,
Newcastle upon Tyne), anti-CD68 (Dako, Carpinteria, CA,
USA) and anti-smooth muscle actin (Biogenex, San Ramon,
CA, USA) and post-treated with a biotinylated anti-mouse
immunoglobulin antiserum (Biogenex), followed by peroxidase-labeled avidin and revealed with AEC as chromogen.
Subsequently, the sections were incubated with an antisarcomeric a-actin antibody (Novocastra), post-treated with
the biotinylated antiserum followed by alkaline phosphatase-labeled streptavidin (Biogenex) and Fast Red as
chromogen, and counterstained with hematoxylin. Alternatively, after incubation with the anti-Ki67 and biotinylated
antibodies, sections were treated with fluorescein-labeled
streptavidin (Vector Laboratories, Peterborough, UK), followed by incubation with rhodamine-labeled phalloidin
(Sigma, St. Louis, MO). The tissue sections treated with
enzyme-labeled avidin were examined under Nomarski
optics or conventional light microscopy, and those with
the fluorescent reactants were examined with a confocal
microscope (Zeiss, Oberkochen, Germany). Specificity
tests, performed by omission of the specific antibody and
incubation with non-immune mouse serum, produced
negative results.
The determination of the proportion of cardiomyocyte
nuclei expressing cell cycle proteins was carried out in those
located in the three cell rows immediately adjacent to the
infarct area. The mean number of examined myocyte nuclei
in each heart ranged between 846 and 1138.
In order to control the accuracy of the Ki67 stain for
identification of cycling cells, tissue sections of two of the
hearts with the highest number of Ki67 positive myocytes
were incubated with a monoclonal antibody against
phosphorylated histone H3 (Sigma), an additional marker
for cell division [9], and processed for immunoenzymatic
study.
2.4. Determination of ploidy
For each case the nuclear DNA concentration of 200
cardiomyocytes located at the infarct border and 200 located

in distant areas was determined in Feulgen-stained tissue

sections with the aid of a digital analysis system (Vidas
Kontron, Zeiss, Germany). For assessing the accuracy of
measurements on tissue sections, myocytes were isolated
from small samples of the left ventricle free wall of two
control hearts by means of collagenase digestion with a
method adapted for human samples [10]. Isolated cells were
fixed in suspension in 4% buffered formaldehyde, washed in
PBS, dried on Vectabond (Vector) coated glass slides and
stained with Feulgen. The data of the nuclear DNA content
of these cells were compared with the DNA content of
myocyte nuclei examined in Feulgen-stained tissue sections
from the same hearts. For reference cells, the nuclear DNA
content of non-myocyte cells in the same tissue section, or
contaminating the smears of collagenase-isolated cells, that
were assumed to be diploid, was determined.
2.5. Statistics
Variables were expressed as mean F S.D. The statistical
significance of differences between groups was determined by one-way analysis of variance (ANOVA) and
Scheffes test as appropriate. P values b0.05 were
considered statistically significant. The data were analysed
using the SPSSR 6.1 statistical analysis software for
Windows (SPSS Inc).

3. Results
3.1. Patients and infarct pathology
The mean age was 53.5 F 15.6 years, range 1972 years.
Eleven were male and four female. In two young female
patients with Marfan syndrome infarction was due to
dissection of the left anterior descendent coronary artery.
In all cases the infarcts involved extensive areas of the left
ventricle and their histological patterns corresponded grossly
with the time of evolution. In three hearts collected at about 1
week after the onset of AMI, the infarct area was invaded by
inflammatory cells, mainly CD68-positive macrophages. In
recent infarcts (less than 48 h evolution) no inflammatory
reaction was seen, with the exception of a peripheral
infiltration of polymorphonuclear leukocytes, and in the
older infarcts, collected between 12 and 19 days after
diagnosis, smooth muscle actin-positive cells, interpreted
as proliferating myofibroblasts, were observed at the
periphery of the infarct area. Both the CD68 and the smooth
muscle actin positive cells displayed Ki67+ nuclei, mitotic
activity and occasional cytokinesis.
Tissue preservation and staining with immunologic
reactants was similar in hearts collected after necropsy or
after transplant. This could be ascribed to the short time
elapsed between collection of the material and fixation (less
than 7 h in the former and less than 60 min in cardiac
explants).

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

3.2. Expression of cell cycle proteins and cardiomyocyte

mitosis
In some of the hearts with MI we observed numerous
myocytes, identified by the presence of sarcomeric a-actin
in their cytoplasm, their shape, size (more than 15 Am in
transversal diameter at the nuclear level) and presence of
sarcomeres, expressing Ki67 antigen in their nuclei (Fig. 1,
panel A). The pattern of Ki67 antigen stain was variable.
Usually it was diffuse, but at times rounded structures of
variable size could be observed (Fig. 1, panel B).
On account that almost all the Ki67+ myocytes were
dispersed in the two or three rows of surviving myocardium

119

immediately adjacent to the infarct area, and were rarely

observed in distant zones, we estimated their proportion
only in the periinfarct zone.
As can be seen in Table 1 and in Fig. 2A, the highest
proportion of periinfarct myocytes expressing the Ki67
antigen was observed at the second week after diagnosis
(11.61 F 6.94%), and was significantly higher ( P = 0.004)
than in the hearts collected at earlier (0.334 F 0.75%) or
later times (2.66 F 3.04%).
In tissue sections consecutive to those employed for
search of the Ki67 antigen, myocytes with similar characteristics and location were positive for markers of different
stages of the cell cycle, cyclin D (Panel E), cyclin A (Panel

Fig. 1. Ki67 and cyclin labeling of myocytes in an infarcted heart. In panel A, a low power view of the border of an 8-day-old infarct can be seen. Immediately
adjacent to the inflammatory infiltrate (asterisks) three myocytes (cytoplasm stained red after demonstration of sarcomeric a-actin) display Ki67+ nuclei
(arrows). Many of the non-myocyte cells present in the infarct area also show Ki67 positive nuclei (arrowheads). Panels B to E show cardiomyocyte nuclei
positive for the Ki67 antigen (panel B), phosphorylated histone-3 (panel C), cyclin A (panel D) and cyclin D1 (panel E). Panel F documents myocytes positive
for cyclin B1, either cytoplasmic (arrowheads) or translocated to the nucleus (arrow). Sarcomeric a-actin was not investigated in panels D, E and F in order to
detect the cytoplasmic presence of cyclins. Panels G and H show that myocytes in M phase of the cell cycle present different appearances according to the
method employed. The image in panel G, obtained with a confocal microscope and fluorescent labels, shows what can be considered a metaphase, with
advanced chromosome spiralization (arrow). Red fluorescence shows staining of myocyte cytoplasm by phalloidin, and green fluorescence shows
chromosomes labeled with the anti-Ki67 antibody. Panel H shows two myocytes examined under Normarski optics and immunoenzymatic methods. The redstained cytoplasm reveals the presence of sarcomeric a-actin and the condensed Ki67+ chromosomes are contained within the boundary or a preserved nuclear
envelope (arrows). The nucleoplasm appears blue after hematoxylin stain. Panels I and J illustrate myocytes displaying a transversally sectioned metaphase
plate (panel I) and two telophase plates (panel J). In both cells the cytoplasm is stained red with the anti-a-sarcomeric actin antibody and the Ki67+
chromosomes plates (arrows) are located within a blue stained nucleus. Panel K shows a myocyte (cytoplasm positive for sarcomeric a-actin) with a
conventional mitosis in early anaphase, with start of separation of the Ki67+ chromosomes (arrows).

120

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

at 78 days after infarction. At earlier and later times

endomitosis were exceptionally observed (Table 1).
When tissue sections consecutive to those observed with
Nomarski optics and immunoenzymatic methods were
examined after processing with fluorescent labels and under
laser confocal microscopy, numerous cells with phalloidin
stained cytoplasm and containing Ki67+ chromosomes
arranged in metaphase or anaphase plates were observed
(Fig. 1, panel G). Since with this approach the nuclear
envelope could not be identified, we could not determine if
chromosomes were disposed within that structure.
In the five control hearts no mitotic figures were
observed with both microscopic approaches, and the
proportion of Ki67+ myocyte nuclei was 0.43 F 0.22%.
On account of this finding, the expression of other cell cycle
proteins was not investigated.
3.3. Ploidy status of cardiomyocyte nuclei

Fig. 2. Panel A shows the mean (F2 S.E.M.) percent of Ki-67-labeled

myocyte nuclei in the periinfarct area at different times after diagnosis.
Panel B shows the difference (mean F 2 S.E.M.) between the proportion of
octaploid myocytes in the neighborhood of the infarct and in the remote
areas.

D) and cyclin B1 (Panel F). In the hearts collected at earlier

and later times, the number of cardiomyocyte nuclei positive
for cell cycle proteins was lower or nil (Table 1).
In two hearts collected at the second week after diagnosis
we observed numerous myocytes located at the infarct
border with their nuclei stained with the anti-phosphorylated
histone H3 antibody. The staining pattern was similar to that
observed with the anti-Ki67 antigen antibody, and condensed chromosomes were seen within the boundary of an
intact nuclear envelope (Fig. 1, Panel C). The proportion of
myocytes with positive phosphorylated histone H3 nuclei
was lower than the Ki67 antigen-positive cells, 9.33% in
patient 7 and 14.22% in patient 8.
Despite a thorough search, we encountered only three
classical mitotic bodies in two of the examined hearts, one
in an explanted heart collected 7 days after the acute episode
(Fig. 1, panel K), and two in a heart from a necropsy
performed 12 days after the acute episode. However, in
hearts collected between 7 and 12 days after diagnosis, we
observed myocyte nuclei with condensed Ki67+ chromosomes, occasionally spliced in sister chromatids, dispersed
within the nucleus or arranged in structures similar to a
metaphase plate (Panel I), or disposed in clumps located at
opposite poles of the nucleus, resembling late anaphase or
telophase (Panel J) of conventional mitosis, always within
the boundary of an intact nuclear envelope (Fig. 1, panel H).
As occurred with the Ki67+ myocytes, the number of
endomitosis, although lower, peaked in the hearts collected

As can be seen in Table 1, although there was a wide

variation in the ploidy status for each patient, in the samples
collected at the third week after diagnosis, the difference
between the proportion of octaploid myocytes in the
neighborhood of the infarct and in the remote areas was
significantly higher ( P = 0.02) than in the hearts collected at
earlier times (Fig. 2, panel B).
In the five control hearts 21.93 F 6.62 (S.D.) of the
myocyte nuclei were diploid, 67.81 F 6.62 were tetraploid
and 10.3 F 2.88 were octaploid. The comparative study
carried out on 200 myocyte nuclei in tissue sections and in
isolated cells of two normal hearts rendered similar results,
and no significant difference was found between the data
collected with both approaches.

4. Discussion
Our study shows that in human AMI, at the second week
after diagnosis, more than 10% of the cardiomyocytes
located at the border of the infarct are undergoing the cell
cycle, as evidenced by the presence in their nuclei of the
Ki67 antigen [11,12]. Given that in the normal control
hearts the proportion of Ki67 positive cardiomyocyte nuclei
was lower than 0.5%, a result similar to that of other report
[13], it can be assumed that AMI induces the entrance of a
significant number of periinfarct cardiomyocytes into the
cell cycle. Apparently, this phenomenon is transient,
because in hearts collected at earlier and later times after
AMI the proportion of Ki67+ myocytes at the infarct border
was significantly lower. This may be one of the reasons for
the inefficiency of heart regeneration.
The transience of the described phenomenon has been
also observed in rat infarcts, where the appearance of
cardiomyocytes in cell cycle or mitosis is a temporal event
lasting only a few days [14]. However, in the rat model the
cycling and dividing myocytes were detected at earlier times

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

than in human infarcts, and the number of Ki67+ cells was

at least 10 times lower. This difference could be ascribed to
the well-known fact that in the rat most adult myocytes are
post-mitotic cells unable to replicate their DNA [15,16], a
condition not present in humans. In fact, myocyte hypertrophy secondary to overwork is usually associated with an
increase in the ploidy status, a situation that requires
previous DNA replication [17].
The presence of sarcomeric a-actin in cycling or
endomitotic cells provides strong evidence of their myocytic
origin [2]. However, the highest number of myocyte Ki67+
nuclei as well as maximal chromosome condensation were
observed at a time when heavy macrophage infiltration
occurs and myofibroblast proliferation starts. Therefore,
concomitant with the detection of the Ki67 antigen, we used
immunohistochemical markers of those cells. Although we
observed numerous Ki67+ macrophages and myofibroblasts, as well as mitotic bodies within them, myocytes
were easily recognized on the basis of their sarcomeric aactin reaction, in addition to their characteristic shape and
the length of their short axis diameter. The possibility that
the cycling myocytes could be originated from putative
resident myocyte progenitors, recently described for the
human heart [18], or from stem cells homing in the heart
[19], appears unlikely on account of the large size of the
cycling myocytes observed in our specimens. However, it
cannot be discarded that cycling and endomitotic cardiomyocytes result from fusion with other cell types, such as
adult stem cells, endothelial cells and fibroblasts, that retain
their replicative ability, as has been recently suggested for
murine hearts [20].
Although expression of the Ki67 antigen and phospho
histone-3 indicates that the cardiomyocytes are in the cell
cycle, it does not allow to establish if they are detained in
the G1 phase or have progressed towards more advanced
stages. Thus, we investigated in the 713 day-old infarcts,
those with the highest number of Ki67+ myocytes, the
presence of proteins that are expressed at specific
moments of the cell cycle [21,22]. We observed that
many myocytes located at the infarct border expressed
cyclin D1 (present in the G1-S boundary), cyclin A
(increased in the S phase) and cyclin B1 (essential for
progression towards the G2-M phase).
However, despite the presence of proteins involved in the
completion of the cell cycle and progression towards mitosis
was documented, we could detect only three conventional
cardiomyocyte mitosis in all the studied hearts. Since this
observation was not coincident with a recent report claiming
that in human AMI myocytes proliferate with relatively high
mitotic indexes [2], we wondered which could be the reason
for this discrepancy. Although the mean age of patients was
10 years lower, and the hearts were fixed earlier after death,
the mean number of Ki67+ myocytes for all the hearts in
both series was almost identical (4.84 vs.4). We thus
speculated that the discrepancy could be ascribed to the
different techniques employed for identification of mitosis

121

rather than to differences in age or sample fixation. To test

for this possibility, we studied successive tissue sections
stained with fluorescent labels under confocal microscopy,
and with immunoenzymatic methods under light microscopy. With fluorescent techniques we observed myocytes
with condensed chromosomes arranged in metaphase and
anaphase plates, indicative of outgoing mitosis, but when
we employed immunoenzymatic methods and light microscopy, although we also observed many myocytes with
condensed Ki67+ chromosomes presenting a distribution
corresponding to the prophase, metaphase and anaphase
stages of mitosis, the chromosomes were always contained
within the boundary of an intact nuclear envelope. Since
breakdown of this structure marks the end or prophase and
initiation of metaphase, the presence of chromosomes
arranged in metaphase and anaphase plates contained within
a preserved nuclear envelope indicates that these cells are in
endomitosis rather than in true mitosis. On account that the
Ki67 antigen may occasionally adopt a punctuate or
granular pattern that may be confused with condensed
chromosomes, we considered that myocytes were in
endomitosis only when the chromosomes were split in
sister chromatids or were arranged in metaphase or anaphase
plates, despite this requirement could lead to an underestimation of the real number of myocyte endomitosis.
Intranuclear mitosis and chromosome displacements with
formation of chromosomes plates within an intact nuclear
membrane have been described in some lower organisms
[23], and in human trophoblast [24] and cancer cells [25].
As opposed to mitosis, endomitosis is an alternative
mode of replication and distribution of the genetic material,
and is one of the mechanisms responsible for polyploidy
[26,27]. The most characteristic example of endomitosis in
humans is the bone marrow megakaryocyte, a cell that
leaves the diploid (2C) state to differentiate, synthesizing 4
to 64 times the normal DNA content within a single nucleus
[28,29].
Our results, providing evidence of cardiomyocyte endomitosis, pose the possibility that this may be a frequent
mechanism for replication of their genetic material, and that
polyploidization is the consequence of this process, as has
been hypothesized almost 30 years ago [30].
This is supported by the fact that in the samples collected
after 2 weeks of evolution, when the number of myocytes in
the cell cycle was very low, the nuclear DNA content of the
myocytes adjacent to the infarct area was significantly
higher than in those located in distant areas.
The facts that in the normal human left ventricle most of
the myocyte nuclei are tetraploid [31,32], and that polyploidization is the rule in cardiomyocyte hypertrophy [33
36] and in cells surrounding myocardial scars [7,8], support
the possibility that endomitosis could be a frequent
phenomenon in the human heart.
The biological significance of polyploidy in hypertrophic
cardiomyocytes is largely unknown. In addition to provide
increased genetic material able to cope with the increased

122

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

metabolic demands of a larger cell, it is known that ploidy

regulates gene expression, and thus hyperploidy may
control cell physiology, morphology and behavior by
permitting the expression of genes that otherwise should
not be expressed in diploid cells [37].
On the basis that the genetic program for reinitiating
DNA synthesis exists in post-mitotic cardiomyocytes [1],
that components of the cell-cycle machinery are involved in
hypertrophic growth [38], and that, despite the discordance
about its frequency, adult cardiomyocytes may enter in
mitosis, if endomitosis is the usual replication mechanism of
human adult cardiomyocytes, we think that efforts to induce
karyokinesis and division into daughter cells should be
addressed to determine why these cells, that apparently may
arrive to the M phase, do not proceed towards a conventional mitosis, with formation of a mitotic spindle and
breakdown of the nuclear envelope, but proceed instead
towards endomitosis, with polyploidy as the final result.
Although this may explain the absence of an efficient
reparative process in the human heart, it also indicates that
there is a potential to manipulate cell division, repair and
proliferation. This is supported by recent reports from our
laboratory, showing that VEGF gene transfer in a pig model
of chronic ischemia induces an increase in the myocyte
mitotic index [39] and myocyte hyperplasia [40].

References
[1] Engel FB, Hauck L, Cardoso MC, Leonhardt H, Dietz R, von
Harsdorf R. A mammalian myocardial cell-free system to study cell
cycle reentry in terminally differentiated cardiomyocytes. Circ Res
1999;85:294 301.
[2] Beltrami AP, Urbanek K, Kajstura J, Yan SM, Finato N, Bussani R, et
al. Evidence that human cardiac myocytes divide after myocardial
infarction. N Engl J Med 2001;344:1750 7.
[3] Pfeffer MA, Braunwald E. Ventricular remodeling after myocardial
infarction: experimental observations and clinical implications.
Circulation 1990;81:1161 72.
[4] St. John Sutton MG, Sharpe N. Left ventricular remodeling after
myocardial infarction. Pathophysiology and therapy. Circulation
2000;101:2981 8.
[5] Reffelmann T, Kloner RA. Cellular cardiomyoplastycardiomyocytes, skeletal myoblasts, or stem cells for regenerating myocardium
and treatment of heart failure? Cardiovasc Res 2003;58:358 68.
[6] Wollert KC, Meyer GP, Ringes-Lichtenberg S, Lippolt P, Breidenbach
C, Fichtner S, et al. Intracoronary autologous bone-marrow cell
transfer after myocardial infarction: the BOOST randomized controlled clinical trial. Lancet 2004;364:141 8.
[7] Ebert L, Pfitzer P. Nuclear DNA of myocardial cells in the periphery
of infarctions and scars. Virchows Arch B Cell Path 1977;24:209 17.
[8] Herget GW, Neuburger M, Plagwitz R, Adler CP. DNA content,
ploidy level and number of nuclei in the human heart after myocardial
infarction. Cardiovasc Res 1997;36:45 51.
[9] Hans F, Dimitrov S. Histone H3 phosphorylation and cell division.
Oncogene 2001;20:3021 7.
[10] Li RK, Weisel RD, Williams WG, Mickle DAG. Method of
culturing cardiomyocytes from human pediatric ventricular myocardium. J Tissue Cult Methods 1992;14:93 100.
[11] Scholzen T, Gerdes J. The Ki-67 protein: from the known and the
unknown. J Cell Physiol 2000;182:311 22.

[12] Brown DC, Gatter KC. Ki67 protein: the immaculate deception?
Histopathology 2002;40:2 11.
[13] Hein S, Arnon E, Kostin S, Schonburg M, Elsasser A, Polyakova V, et
al. Progression from compensated hypertrophy to failure in the
pressure-overloaded human heart: structural deterioration and compensatory mechanisms. Circulation 2003;107:984 91.
[14] Yuasa S, Fukuda K, Tomita Y, Fujita J, Ieda M, Tahara S, et al.
Cardiomyocytes undergo cells division following myocardial infarction is a spatially and temporally restricted event in rats. Mol Cell
Biochem 2004;259:177 81.
[15] Soonpaa MH, Field LJ. Survey of studies examining mammalian
cardiomyocyte DNA synthesis. Circ Res 1998;83:15 26.
[16] Pasumarthi KBS, Field LJ. Cardiomyocyte cell cycle regulation. Circ
Res 2002;90:1044 54.
[17] von Harsdorf R, Philip A, Poole-Wilson PA, Dietz D. Regenerative
capacity of the myocardium: implications for treatment of heart
failure. Lancet 2004;363:1306 13.
[18] Urbanek K, Quaini F, Tasca G, Torella D, Castaldo C, NadalGinard B, et al. Intense myocyte formation from cardiac stem cells
in human cardiac hypertrophy. Proc Natl Acad Sci U S A 2003;
100:10440 5.
[19] Oh H, Bradfute SB, Gallardo TD, Nakamura T, Gaussi V, Mishina Y,
et al. Cardiac progenitor cells from adult myocardium: homing,
differentiation, and fusion after infarction. Proc Natl Acad Sci U S A
2003;100:12315 8.
[20] Matsuura K, Wada H, Nagai T, Iijima Y, Minamino T, Sano M, et al.
Cardiomyocytes fuse with surrounding noncardiomyocytes and
reenter the cell cycle. Cell Biol 2004;167:351 63.
[21] Nasmyth K. Viewpoint: putting the cell cycle in order. Science
1996;274:1643 5.
[22] Li JM, Brooks G. Cell cycle regulatory molecules (cyclins, cyclindependent kinases and cyclin-dependent kinase inhibitors) and the
cardiovascular system: potential targets for therapy? Eur Heart J
1999;20:406 20.
[23] Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P, editors.
Molecular biology of the cell. Fourth edition. Electronic version at:
htttp://www.ncbi.nlm.nih.gov/books/. Garland Publishing; 2002.
Chapter 18: The Mechanics of Cell Division. Cytokinesis. Fig. 41.
[24] Sarto GE, Stubblefield PA, Therman E. Endomitosis in human
trophoblast. Hum Genet 1982;62:228 32.
[25] Therman E, Sarto GE, Kuhn EM. The course of endomitosis in human
cells. Cancer Genet Cytogenet 1986;19:301 10.
[26] Nagl W. Cdc2-kinases, cyclins, and the switch from proliferation to
polyploidization. Protoplasma 1995;188:143 50.
[27] Datta NS, Williams JL, Caldwell J, Curry AM, Ashcraft EK, Long
MW. Novel alterations in CDK1/Cyclin Bi kinase complex. Formation occur during the acquisition of a polyploid DNA content. Mol
Biol Cell 1996;7:209 23.
[28] Vitrat N, Cohen-Solal K, Pique C, Le Couedic JP, Norol F, Larsen
AK, et al. Endomitosis of human megakaryocytes are due to abortive
mitosis. Blood 1998;91:3711 23.
[29] Zimmet J, Ravid K. Polyploidy: occurrence in nature, mechanisms
and significance for the megakaryocyteplatelet system. Exp Hematol
2000;28:3 16.
[30] Sandritter W, Adler CP. Polyploidization of heart muscle nuclei as a
prerequisite for heart growth and numerical hyperplasia in heart
hypertrophy. Recent Advan Stud Cardiac Struct Metab 1976;12:
115 27.
[31] Pfitzer P. Nuclear DNA content of human myocardial cells. Curr Top
Pathol 1971;54:125 68.
[32] Adler CP, Friedburg H. Myocardial DNA content, ploidy level and
cell number in geriatric hearts: post-mortem examination of human
myocardium in old age. J Mol Cell Cardiol 1986;18:39 53.
[33] Vliegen HW, Eulderink F, Bruschke AV, van der Laarse A, Cornelisse
CJ. Polyploidy of myocyte nuclei in pressure overloaded hearts: a
flow cytometric study in left and right ventricular myocardium. Am J
Cardiovasc Pathol 1995;5:27 31.

P.C. Meckert et al. / Cardiovascular Research 67 (2005) 116123

[34] Brodsky VYa, Sarkisov DS, Arefyeva AM, Panova NW. DNA and
protein relations in cardiomyocytes. Growth reserve in cardiac muscle.
Eur J Histochem 1993;37:199 206.
[35] Brodsky VYa, Sarkisov DS, Arefyeva AM, Panova NW, Gvasava IG.
Polyploidy in cardiac myocytes of normal and hypertrophic human
hearts; ranges of values. Virchows Arch 1994;424:429 35.
[36] Vliegen HW, Bruschke AV, Van der Laarse A. Different response of
cellular DNA content to cardiac hypertrophy in human and rat heart
myocytes. Comp Biochem Physiol A 1990;95:109 14.
[37] Galitsky AG, Saldanha AJ, Styles CA, Lander ES, Fink GR. Ploidy
regulation of gene expression. Science 1999;285:251 4.

123

[38] Li JM, Poolman RA, Brooks G. Role of G1 phase cyclins and cyclindependent kinases during cardiomyocyte hypertrophic growth in rats.
Am J Physiol 1998;275:H814 22.
[39] Laguens R, Cabeza Meckert P, Vera Janavel G, Del Valle H, Lascano
E, Negroni J, et al. Entrance in mitosis of adult cardiomyocytes in
ischemic pig hearts after plasmid-mediated rhVEGF165 gene transfer.
Gene Ther 2002;9:1676 81.
[40] Laguens R, Cabeza Meckert P, Vera Janavel G, De Lorenzi A,
Lascano E, Negroni J, et al. Cardiomyocyte hyperplasia after plasmidmediated vascular endothelial growth factor gene transfer in pigs with
chronic myocardial ischemia. J Gene Med 2004;6:222 7.