Neuromusc. D~ord.,Vol. 3, No. 5/6, pp.

385-390, 1993
Pergamon Elsevier Science Ltd
Printed in Great Britain. All rights reserved
0960-8966/93 $6.00+ .00

BEHAVIOUR OF H U M A N ATRIAL MYOCYTES IN C U L T U R E IS

D O N O R AGE D E P E N D E N T

C. RCCKER-MART1N,* S. HATEM,* I. DUaUS,t L. MACE,* J. L. SAMUELt and J. J. MERCADIER*

*CNRS URA 1159, Le Plessis Robinson; tINSERM U 127, Paris, France

Abstract--The characteristics of cultured myocardial cells isolated from small mammals are well
documented, but there is a dearth of data on cultured human cardiocytes. The aim of this study
was to determine the main features of myocytes isolated from human atria and maintained in
culture in the presence of 10% fetal calf serum (FCS), according to the age of the donor. The
following characteristics were analysed: (!) yield and viability; (2) adhesive properties; and (3)
changes in cell morphology. Myocytes preferentially adhered to laminin-coated dishes and could
be maintained in culture for at least 2 weeks, whatever the age of the donor (which was from 6
days to 85 yr). Maintenance in culture induced morphologic changes characterized by myocyte
spreading and changes in myofibrillar organization. Interestingly, the time of onset of these
changes depended on the age of the donor: they occurred earlier in young atrial myocytes ( < 1
yr) than in older cells (> 13 yr).

Key words: Human atrial myocytes, myocyte culture, age.

INTRODUCTION MATERIALS AND METHODS

Cultures of myocardial cells from small mam- M y o c y t e isolation and culture

mals are routinely used for cardiac muscle
research (reviewed in [1]). They have been very
useful to determine the part played by various
neurohumoral agents and growth factors in the
changes of myocyte phenotype observed during
various myocardial diseases [2, 3]. Several studies
have also been performed using cultures of
human cardiocytes, most of them being grown
from fetal hearts [1]. Interestingly, behaviour of
rat cells during culture has been shown to depend
upon the developmental stage (newborn vs
adult) [1, 4]. Recently, Smith et al. [5] studying
cultures of adult human atrial myocytes, found a
relationship between the success rate of cell
cultures and donor age (38-81 yr).
This study was undertaken to answer several
questions raised by previous reports on the
characteristics of human cardiac cells in culture.
First, are the adhesive properties of human
cardiac myocytes similar to those described for
rat cardiac myocytes? Second, do the culture
conditions induce changes in the myocyte
structure? Third, do these adhesive properties
and potential changes vary according to the age
of the donor? To answer these questions, we have
studied, in addition to samples from adult
donors of a similar age range (35-81 yr) as those
studied by Smith et al., samples from younger
donors aged from 6 days to 30 yr.
All protocols for obtaining human cardiac
tissue were approved by the ethics committee of
our institution (GREBB, H6pital de Bic~tre,
Universit6 de Paris XI). Human right atrial
appendages were obtained from 14 patients (6
days-85 yr) undergoing surgery for congenital,
ischemic or valvular heart diseases. Myocytes
were isolated as described by Escande et al. [6]. A
first determination of the percentage of rod-
shaped cells was performed at the end of the
isolation procedure in the absence of calcium to
assess the yield of the procedure. Cells were then
suspended at a final density of 105 ml- ' of culture
medium [Dulbecco's modified Eagle's medium
( D M E M containing 2 mM Ca 2+ Gibco-BRL),
supplemented with 10% FCS (Boehringer
Mannheim, Germany), non-essential amino
acids, 1 nM insulin and antibiotics (100 IU ml-
penicillin and 0.1 /zg ml-l streptomycin)], and
plated in 35 mm plastic dishes (Falcon, Becton-
Dickinson) for 2 h; nonadherent cells (mainly
myocytes), were then transferred to dishes
pretreated with either 4% FCS or 20/zg ml-'
laminin (Boehringer Mannheim, Germany) [4].
A second determination of the percentage of rod-
shaped cells was performed to assess myocyte
viability (since myocytes which have been
damaged during the isolation procedure are
known to lose their typical rod-shape) with the

385
386 C. ROCKER-MARTIN et al.

Table 1. Cell yield and percentage of rod-shaped myocytes. N u m b e r s of ceils were determined before plating and the
percent of rod-shaped cells was determined before and after returning to physiologic calcium concentration (see text
for details). Values are means ± S.E. The number of cultures is indicated in brackets

D o n o r age N u m b e r of cells Rod-shaped cells

(10~) (%)

Per preparation Per g of tissue Before Ca 2+ After Ca 2+

< 13 months 1.0 + 0.2 (4) 6.0 + 0.0 (2) 57 ± 5 (4) 30 + 4 (4)
13-30 yr 0.8 ± 0.5 (4) 1.2 :~ 0.4 (3) 39 ± 6 (4) 18 ± 1 (3)
> 35 yr 0.4 ± 0.1 (6) 1.2 ± 0.3 (3) 42 ± 4 (6) 25 • 4 (5)

physiological Ca 2+ concentration present in the 100-

culture medium. The culture medium was then
renewed every 3 days. Plating efficiency and i 80-

myocyte viability during culture were deter-

60-
mined by the number of cells attached to the
dishes and cell morphology (cell shape and
striations), and their alterations during culture.
20-
Double indirect immunofluorescence
To check for the specificity of our antibodies, a 0 I I I i I I I

sample of right atrium was mounted, frozen in 0 1 2 3 4 5 6

cold isopentane and maintained at - 70°C until
cryosectioning (5/ma). Cultured cells were fixed
with methanol for 10 rain at -200C. After
washing in PBS, tissue sections and cell cultures
were incubated with monoclonal antibodies
directed against pan-myosin (1/10, Amersham),
followed by horse biotinylated anti-mouse IgG
(Vector Laboratories), and streptavidin-Texas
red (Amersham), then with polyclonal anti-
bodies directed against fibronectin (1/200, Euro-
medex Chemicon), followed by fluoresce±n-
conjugated donkey anti-rabbit IgG (Amer-
sham), and mounted after a final wash.
RESULTS
Cell yield (Table 1)
The yield of myocyte isolation was 5-fold
greater in young patients (6 days-13 months)
than in older patients (> 13 yr). A decrease of
approximately 50% in the percentage of rod-
shaped cells was observed after 2 h in the culture
medium containing 2 mM Ca 2+, whatever the
age of the donor.
Adhesive properties of atrial cardiomyocytes
and cell behaviour in culture
During the first few days of culture, atrial
myocytes from adult patients attached more
efficiently to laminin than to FCS: after 6 days,
> 50% and < 25% of the cells initially plated
Days~/euitme
Fig. 1. Adhesion of adult atrial myocytes on laminin (20/Jg
m l - ' , solid circles) or FCS (4%, open circles) pretreated
dishes. Values are means q- S.E. of three different cultures.
For each culture, three different dishes have been analysed, *
and **, respectively, indicate p < 0.05 and 0.01 for laminin
values vs FCS. Adhesion of young atrial myocytes on laminin
(20/1g ml ', solid triangles), was not significantly different
from that of adult myocytes.
remained present on laminin- and on FCS-
coated dishes, respectively (Fig. 1). Adhesion of
atrial myocytes from young patients to laminin-
coated dishes was similar to that of adult
myocytes. Subsequent cultures were thus per-
formed on laminin-coated dishes.
Myocytes could be maintained in culture for 2
weeks whatever the age of the donor. Daily
phase-contrast examination revealed marked
changes during the first 10 days, consisting of
myocyte spreading and spontaneous beating.
However, the time-course of these changes was
clearly related to donor age. Most myocytes from
adult samples spread and started spontaneous
beating at the end of the first week. These
contractions were transient (approximately 48
h). By contrast, myocytes from young patients
( < 13 months), spread and beat spontaneously as
early as 2 days after plating. Cultures from adult
samples never reached confluence, but those
from young samples did so before the end of the
second week; interestingly, contractions still
occurred within cellular islets.
 Culture of Human Atrial Myocytes
Immunocytochemical analysis
Atrial sections allowed us to check for the
specificity of our antibodies (not shown). As
expected, anti-myosin immunoglobulins
strongly labelled myocytes, which were not
labelled with anti-fibronectin immunoglobulins.
The opposite pattern was seen with anti-
fibronectin immunoglobulins which labelled
fibronectin distributed in the interstitial tissue
and basal membranes.
At day 5, rod-shaped myocytes from adult
atria still exhibited the typical cross-striated
pattern of myosin immunolabelling, although
the myofibrils partially lost their registered
alignment (Fig. 2a). While spreading out,
myocytes developed numerous thin fibres which
were poorly reactive to anti-myosin immunoglo-
bulins and negative for fibronectin (Fig. 2b, c).
The nonmuscle cells present in the culture were
unreactive to anti-myosin immunoglobulins but
positive with anti-fibronectin immunoglobulins.
The aspect of myocytes isolated from patients
< 13 months of age at day 5 was clearly different
(Fig. 3a); indeed, the anti-myosin immunoglobu-
lins revealed, in spread cells, myofibrils with the
typical sarcomeric organization which have
completely lost their registered alignment,
together with stress fibres which were also
labelled but not striated (Fig. 3a). Interestingly,
the sarcomeric organization was maintained
until the end of the second week of culture (Fig.
3b). As with adult cells, cardiac myocytes from
young patients did not label with anti-fibronec-
tin immunogiobulins (Fig. 3c).
Whatever the parameter considered (cell yield
and beating, changes in cell morphology, myofi-
brillar organization), no difference was observed
according to the diagnosis.
DISCUSSION
To our knowledge, this study is the first to
report the main features of cultures of myocytes
isolated from atria of donors ranging from 6 days
to 85 yr. The data presented here show clear
differences in muscle cell yield according to the
age of the donor and no differential behaviour
with respect to the adhesive properties. More
importantly, they also indicate that human atrial
myocytes undergo marked changes in overall
morphology and myofibrillar structure in culture
in the presence of 10% FCS, and that the onset of
these changes varies according to the age of the
donor.
387
There have been few descriptions of human
cardiac cells in culture (see review in [1]), and
only one [5] has concerned cells other than fetal
cells. In their pioneering work on adult atrial
myocardium, Smith et al. [5] showed that donor
age (38-81 yr) was an important factor in
successful culture: the best results were obtained
with atrial samples from the youngest patients.
Our findings are in slight contrast with this study
in that we obtained successful cultures even from
the oldest donors. Moreover, our study extends
the early observation to much younger patients
and allows common, but also distinct, features to
be established according to donor age. Indeed,
myocytes from young and adult atrial tissue
share some properties when cultured in unde-
fined conditions, as previously reported for rat
cells [1], such as preferential adhesion to laminin,
and several features of dedifferentiation (spread-
ing, beating and reorganization of the myofibril-
lar apparatus). By contrast, they clearly differ in
several aspects. First, the isolation yield was five-
fold higher with samples from the young
patients. This finding may be attributed to an
increase in the density of the extracellular matrix
occurring with age. Second, morphological
changes such as spreading and spontaneous
beating occurred more rapidly with cells from the
young patients. These results suggest that
cardiocytes from young patients have a greater
capacity to develop in undefined conditions, than
those from older patients. However, these muscle
cells did not undergo complete differentiation,
since they did not express fibronectin, a
glycoprotein of the extracellular matrix only
synthesized by non-muscle and smooth muscle
cells [7, 8].
In the present study, no difference in cell yield,
cell morphology, and myofibrillar organization
could be observed according to the diagnosis.
This may be related to the limited number of
samples of each pathology group studied and
does not preclude that more subtle phenotypic
differences may exist, such as differences in the
to fl-myosin heavy chain ratio [9], or in the level
of expression of the atrial natriuretic factor gene
[7]. Assessment of such differences and their
potential changes in culture might be the subject
of other studies.
Well-defined growth medium lacking serum
has been shown to delay these phenotypic
changes significantly in rat cardiac myocytes and
to limit the proliferation of nonmuscle cells [1,4].
We are currently investigating the conditions
required for long term culture of human cardiac
388 C. ROCKER-MARTIN et al,

....

Fig. 2. Changes in myocytes isolated from adult atria (45 58 yr), during 2 weeks in
culture. Myocytes cultured for 6 days (a) and 15 days (b, c) were incubated with
anti-myosin (a, b) and anti-fibronectin (c) immunoglobulins. During culture, the
typical striated pattern of myofibrils (a) progressively disappeared from myocytes
(b) while anti-myosin immunolabelling appeared localized at the border of the cell.
Note that the nonmuscle cell was labelled with anti-fibronectin and not with anti-
myosin immunoglobulins. (a) x 600; (b, c) x 225.
 Culture of Human Atrial Myocytes 389

Fig. 3. Changes in myocytes isolated from young atria ( 2 ~ months), during 2

weeks in culture. Myocytes cultured for 6 days (a) and 15 days (b, c) were incubated
as described in Fig. 2. Sarcomeric organization was maintained in spite of cell
spreading (a, b). Note that in (c), the culture has reached confluence. The myocyte
in the centre of the micrograph (m) is not labelled with anti-fibronectin
immunoglobulins, in contrast to the surrounding nonmuscle cells and extracellular
matrix. (a) x 300; (b) x 825; (c) x 225.
390 C. RUCKER-MARTIN et al.
myocytes with minimal dedifferentiation as a
tool for research in cardiovascular pathophysi-
ology and therapeutics, including DNA transfec-
tion experiments [10].
Acknowledgements--The authors wish to thank Patricia
Oliviero for her excellent technical assistance. This work was
supported in part by a grant from the F~d~ration Franqaise
de Cardiologie. Catherine Rficker-Martin was supported by
a grant of the Group¢ de R~flexion sur la Recherche
Cardiovasculaire (GRRC) of the Soci6t~ Franqaise de
Cardiologie.
REFERENCES
1. Piper H M. Cell Culture Techniques in Heart and Vessel
Research. Berlin: Springer, 1990.
2. Schneider M D, Parker T G. Cardiac myocytes as
targets for the action of peptide growth factors.
Circulation 1989; 80: 219-233.
Chien K R, Knowlton K U, Zhu H, Chien S.
Regulation of cardiac gene expression during myocar-
dial growth and hypertrophy: molecular studies of an
adaptive physiologic response. FASEB J 1991; 5: 3037-
3046.
4. Dubus I, Rappaport L, Barrieux A, Lompr6 A M,
Schwartz K, Samuel J L. Contractile protein gene
expression in serum-free cultured adult rat cardiac
myocytes. Pfliigers Arch 1993; 423: 455~461.
5. Smith D A, Glover J L, Townsend L E, Maupin D E. A
method for the harvest, culture, and characterization of
human adult atrial myocardial cells: correlation with
age of donor. In Vitro CellDev Bio11991; 27A: 914-920.
6. Escande D, Coulombe A, Faivre J F, Coraboeuf E.
Characteristics of the time-dependent slow inward
current in adult human atrial single myocytes. J Mol
Cell Cardiol 1986; 18: 547-551.
7. Swynghcdauw B. Research in Cardiac Hypertrophy and
Failure. Paris: INSERM/John Libbey Eurotext, 1990.
8. Samuel J L, Barrieux A, Dufour S, et al. Accumulation
of fetal fibronectin mRNAs during the development of
rat cardiac hypertrophy induced by pressure overload. J
Clin Invest 1991; 88: 1737-1746.
9. Mercadier J J, de la Bastie D, M6naschb P, et al. Alpha-
myosin heavy chain isoform and atrial size in patients
with various types of mitral valve dysfunction: a
quantitative study. J Am Coil Cardiol 1987; 9: 1024-
1030.
10. Kirshenbaum L A, MacLellan W R, Mazur W, French
B A, Schneider M D. Highly efficient gene transfer into
adult ventricular myocytes by recombinant adenovirus.
J Clin Invest 1993; 92: 381-387.