Cardiopulmonary Support and Physiology
EDITORIAL
Suzuki et al
This work was performed with the help of a
Project Grant from the British Heart
Foundation (PG 96085).
Received for publication Jan 17, 2001; revisions requested March 6, 2001; revisions
received March 21, 2001; accepted for publication March 26, 2001.
Address for reprints: Professor Sir Magdi H.
Yacoub, FRS, Department of Cardiothoracic
Surgery, National Heart and Lung Institute,
Imperial College School of Medicine,
Harefield Hospital, Harefield, Middlesex
UB9 6JH, United Kingdom (E-mail:
k.suzuki@ic.ac.uk).
J Thorac Cardiovasc Surg 2001;122:759-66
Copyright © 2001 by The American
Association for Thoracic Surgery
0022-5223/2001 $35.00 + 0 12/1/116210
doi:10.1067/mtc.2001.116210
Conclusions: We have generated connexin 43–overexpressing skeletal myoblast cell
lines that resulted in improved formation of functional intercellular gap junctions,
which could be relevant to synchronous contraction of grafted myoblasts in the
heart. In addition, these cells demonstrated more rapid differentiation, which would
also be advantageous in a graft for transplantation to the heart.
S
keletal myoblast transplantation is a promising strategy to treat
patients with end-stage heart failure.1-3 Success of this strategy
depends on the capacity of the grafted myoblasts to functionally
integrate with the preexisting myocardium. It is considered that this
integration would be dependent on the formation of gap junctional
intercellular communication (GJIC) between grafted myoblasts and
native cardiomyocytes, which is necessary for the synchronous contraction of
grafted cells and cardiomyocytes.3 Although the possible existence of such intercellular gap junctions has been suggested in several experimental models of
skeletal myoblast transplantation to the heart,1,2 this issue is still controversial.4
The Journal of Thoracic and Cardiovascular Surgery • Volume 122, Number 4 759
ACD
CSP
From the Department of Cardiothoracic
Surgery, National Heart and Lung Institute,
Imperial College School of Medicine at the
Heart Science Centre, Harefield Hospital,
Middlesex, United Kingdom.
Methods and Results: L6 rat skeletal myoblast cell lines overexpressing connexin
43 were generated by means of gene transfection and clonal selection. Connexin 43
overexpression of these myoblasts, which continued both in undifferentiated and
differentiated states (up to 17-fold greater protein level in comparison with controltransfected myoblasts, as measured with Western blotting), was observed on cell
surfaces where gap junctions should exist. Both dye microinjection and scrape loading with fluorescent dyes showed enhancement in intercellular dye transfer between
connexin 43–transfected myoblasts compared with that found in control-transfected
cells. Morphologically, these myoblasts fused and differentiated into multinucleated myotubes more rapidly, demonstrating a higher level of cellular creatine kinase
activity as a marker of myogenic differentiation throughout the culture period compared with that of control-transfected myoblasts.
ET
Objective: Skeletal myoblast transplantation is a promising strategy for treating
end-stage heart failure. One potential problem in the development of functional,
synchronously contracting grafts is the degree of intercellular communication
between grafted myoblasts and host cardiomyocytes. Thus it is expected that
enhancement of intercellular gap junction formation would result in improved efficiency of skeletal myoblast transplantation. In this study we investigated whether
myoblasts overexpressing connexin 43, a major cardiac gap junction protein, would
enhance this intercellular communication.
TX
Ken Suzuki, MD, PhD
Nigel J. Brand, PhD
Sean Allen, PhD
Mahboob A. Khan, PhD
Aldo O. Farrell
Bari Murtuza, FRCS
Reida El Oakley, FRCS
Magdi H. Yacoub, FRS
CPS
CHD
Overexpression of connexin 43 in skeletal myoblasts:
Relevance to cell transplantation to the heart
EDITORIAL
Cardiopulmonary Support and Physiology
Suzuki et al
CHD
CPS
ACD
Figure 1. Diagram of gap junction and differences among cardiomyocytes, skeletal myoblasts, and skeletal
myotubes. Gap junction is composed of connexons that are made of connexins, regulating intercellular passage
of molecules, including inorganic ions and second messengers. Skeletal muscle cells have several important differences in cellular characteristics from cardiomyocytes.
CSP
ET
TX
Gap junction is composed of connexons that are made of
connexins (Figure 1). This junction regulates intercellular
passage of molecules, including inorganic ions and second
messengers, thus achieving electrical, as well as metabolic,
coupling of the cells.5,6 Connexin 43 (Cx43) is the major gap
junction protein in the ventricular myocardium responsible
for GJIC between cardiomyocytes themselves5 and is therefore likely to have an important role in coupling between
grafted skeletal myoblasts and host cardiomyocytes.
Although gap junctions exist between undifferentiated
myoblasts with concurrent expression of Cx43, they are
absent from mature skeletal muscle with little Cx43 expression,4,7 as shown in Figure 1. It is reported that overexpression of Cx43 improves GJIC in various cancer cells, resulting
in their enhanced myogenic differentiation ability.8-10
Fibroblasts transfected with Cx43 antisense RNA, in contrast,
show a significant reduction in their GJIC.11 We therefore
hypothesised that overexpression of Cx43 would be advantageous for the skeletal myoblast-myotubes to form GJIC with
cardiomyocytes. The aim of this study was to generate skeletal myoblasts overexpressing Cx43, to investigate the role of
Cx43 in GJIC formation, and to characterize both proliferation and myogenic differentiation in skeletal myoblasts.
Methods
Cell Culture
The rat skeletal muscle cell line L6 (American Type Culture
Collection) was grown in Dulbecco’s modified Eagle’s medium
supplemented with 20% fetal calf serum (growth medium).
Skeletal myoblasts can be incubated without differentiation for a
long period if cultured in this medium at a low cell density.12 When
the medium is substituted with differentiation medium (Dulbecco’s
modified Eagle’s medium with 2% horse serum),l2 myoblasts start
differentiation into multinucleated myotubes (Figure 1).
Gene Transfer and Selection
Full-length rat Cx43 cDNA8 (kindly provided by Professor C. C.
Naus, Department of Anatomy and Cell Biology, University of
Western Ontario, Canada) was cloned into the mammalian expression vector pCI-neo (Promega). After gene transfection mediated
by SuperFect reagent (QIAGEN), the cells were plated at a limiting dilution (0.1 cell/well concentration) to obtain single-cell
colonies in the presence of G418 (Sigma), which kills all nontransfected and transiently transfected cells during a 4-week incubation. Surviving cells, which were stably transfected and thus
show long-term protein expression, were picked up, grown in number, and analyzed. Control cells were transfected by using pCI-neo
without the gene encoding Cx43 and selected by using the same
method.
760 The Journal of Thoracic and Cardiovascular Surgery • October 2001
EDITORIAL
Cardiopulmonary Support and Physiology
CHD
Suzuki et al
ACD
CPS
A
Western Blotting to Assess Cx43 Protein Level
Immunocytochemistry for Cx43
Time course of Cx43 protein expression was analyzed with
Western blotting. Cx43- or control-transfected cells were grown
according to the differentiation protocol described above. These
cultures were collected at day 1 to 7 (n = 5 in each point), and
each sample was loaded onto a sodium dodecylsulfate polyacrylamide gel for electrophoresis and then transferred onto a nitrocellulose membrane. The membrane was incubated with anti-rat
Cx43 monoclonal antibody (Chemicon) and then with horseradish peroxidase–conjugated secondary antibody (Sigma), followed by visualization with an ECL detection system
(Amersham). The films were scanned with a Molecular
Subcellular distribution of Cx43 expression was investigated with
immunocytochemistry. The Cx43- and control-transfected cells
were grown on 2-well chamber slides (Nunc), fixed, and incubated with anti-rat Cx43 monoclonal antibody (Chemicon). After
washing, this was followed by incubation in fluorescein isothiocyanate–conjugated secondary antibody (Dako). Fluorescein isothiocyanate can be detected under a fluorescent microscope. After
counterstaining with propidium iodide, cells were examined with
an Odyssey Laser Scanning Confocal Microscope (Noran
Instruments) to obtain image stacks of 0.5-µm optical sections of
regions of interest.
Dynamics 300A laser densitometer to determine Cx43 levels
with Quantity One software (P.D.I.). Densitometry readings of
the bands were used to calculate values relative to that on day 1
for control-transfected cells to obtain comparative data.
The Journal of Thoracic and Cardiovascular Surgery • Volume 122, Number 4 761
ET
Figure 3. Immunocytochemical staining for Cx43. A, Immunocytochemical study with anti-Cx43 monoclonal antibody demonstrated low-level Cx43 expression (shown by green fluorescence)
in cultures of control myoblasts. B, In contrast, Cx43(19) displayed
obvious Cx43 overexpression (green fluorescence) on the cell surfaces. Cell nuclei are stained orange by propidium iodide. Bar = 20
µm.
TX
Figure 2. Cx43 overexpression. A, Six clones of L6 skeletal
myoblasts genetically engineered to overexpress Cx43, named
Cx43(8), Cx43(11), Cx43(15), Cx43(16), Cx43(19), Cx43(21), were generated by means of gene transfection and clonal selection. B,
Control cells were transfected with pCI-neo vector only. Whereas
Cx43 expression in the control cells was downregulated to a minimum level after differentiation, Cx43(15) and Cx43(19) cells continued to express significantly larger amounts of Cx43 protein
throughout the period. Data are shown as relative levels of Cx43,
assigning a value of 1.0 to the band density from day 1 sample of
control-transfected L6 cells. *P < .05 versus control-transfected L6
cells (n = 5 in each point). Values are expressed as means ± SEM.
CSP
B
EDITORIAL
Cardiopulmonary Support and Physiology
Suzuki et al
B
C
D
E
F
CHD
A
CPS
ACD
CSP
ET
Figure 4. Intercellular communication. Scrape loading with Lucifer yellow dye demonstrated that the dye spread
more quickly and widely throughout the cell monolayer up to 7 or 9 cell diameters away from the scrape cut in
Cx43(19) cells (D) compared with control cells (C). A and B are pictures of reverse-phase contrast microscopic
observation of the corresponding fields. Microinjection of 6-CF into single cells showed that injected dye spread
to cells well beyond immediate neighbors in Cx43-transfected cells, reaching cells up to 7 cell diameters away
(F), whereas dye permeated only into the immediate neighbors in control L6 cells (E). Arrows show microinjected
cells with dye. Bar for A, B, C, and D = 250 µm. Bar for E and F = 20 µm.
Dye-transfer Experiments with Scrape Loading and
Microinjection
TX
To test for functional intercellular coupling through gap junctions,
we performed scrape-loading and microinjection experiments8,13,14 with fluorescent dyes that transfer between cells
through gap junctions but never permeate or diffuse into or from
cells through intact cell membranes. Short scrape lines were made
by using a scrape blade on a confluent cell layer, and cells were
incubated in 0.5 mg/mL Lucifer yellow dye (dilithium salt, Sigma)
in phosphate-buffered saline. The dye permeates into damaged
cells through broken cell membranes by scraping and transfers to
adjacent cells through gap junctions. The dye solution was
removed after 3 minutes, and GJIC permeability was estimated 10
minutes after scraping by taking photomicrographs on an Axiovert
25 microscope (Zeiss).
Microinjections of fluorescent dye into a single cell were per-
762 The Journal of Thoracic and Cardiovascular Surgery • October 2001
Cardiopulmonary Support and Physiology
A
EDITORIAL
Suzuki et al
CHD
B
ACD
CPS
C
Figure 5. Myogenic differentiation and proliferation. A and B, Cells were fed with growth medium for the first 2
days and thereafter with differentiation medium. Morphologically, an enhancement of fusion and differentiation
into multinucleated myotubes is evident in Cx43(19) at day 4 by comparison with control myoblasts. Bar = 100 µm.
C, CK activity was significantly higher in Cx43(15) and Cx43(19) myoblasts than in control myoblasts. D, Cell proliferation after plating of 1% 104 cells was significantly more rapid in control cells than in Cx43(15) and Cx43(19)
cells. *P < .05 versus control-transfected L6 cells (n = 5 for CK measurement and n = 6 for proliferation in each
point). Values are expressed as means ± SEM.
Assessment of Myogenic Differentiation and
Proliferation
Cells grown according to the differentiation protocol were collected at day 1 to 7 for creatine kinase (CK) activity measurement and
morphologic study (n = 5 for CK activity and n = 2 for morphology at each point) to study differentiation capacity. Although CK
activity is negligible in undifferentiated myoblasts, the activity
increases during the course of differentiation (Figure 1). Therefore
The Journal of Thoracic and Cardiovascular Surgery • Volume 122, Number 4 763
TX
formed to assess GJIC8,14 with the aid of an Eppendorff
Micromanipulator 5171 and Transjector 5246. Single cells in a
subconfluent monolayer were microinjected with 0.5% 6-carboxyfluorescein (6-CF, Sigma) with a pressure of 250 psi under phasecontrast microscopy. The extent of intercellular dye transfer was
determined by means of continuous recording with the confocal
microscope (Odyssey LSCM) equipped with Intervision software
(Noran Instruments).
ET
CSP
D
EDITORIAL
Cardiopulmonary Support and Physiology
CHD
CK activity is used as a common marker of myogenic differentiation in skeletal myoblasts. CK activity in the homogenates was
measured by using the spectrophotometric method.15 Specific
activity is expressed in units of CK activity per gram of protein.
For the morphologic study on differentiation, the cultures were
stained with hematoxylin and eosin.
Cells (1 × 104) were grown in 24-well plates (Nunc) in growth
medium to examine cellular proliferation capacity. The number of
cells per plate was determined from counts obtained with a hemocytometer8 at the selected time period (n = 6 in each point).
Statistical Analysis
CPS
All values are expressed as means ± SEM. Statistical comparison
of the data was performed by using 1-way repeated-measures
analysis of variance. If a significant F ratio was obtained, further
comparisons were determined with the Bonferroni/Dunn post hoc
test. Analyses were performed with the StatView version 4.0 statistical package (SAS Institute).
Results
ACD
CSP
ET
TX
Detection of Cx43-Overexpressing L6 Skeletal
Myoblast Cell Lines
Six surviving cloned lines were derived from gene transfection
with a Cx43 cDNA cloned into the expression vector pCI-neo
and from selection with G418. These lines, named Cx43(8),
Cx43(11), Cx43(15), Cx43(16), Cx43(19), and Cx43(21),
were subjected to Western blotting for Cx43, demonstrating
that all of them expressed a higher level of Cx43 than control
L6 cells that had been transfected by pCI-neo without the gene
encoding Cx43 (Figure 2, A). Cx43 protein appeared as a doublet, suggesting phosphorylation of Cx43. Cx43(15) and
Cx43(19), whose levels of Cx43 expression were the highest
(14.5 ± 1.3-fold and 17.3 ± 1.88-fold from Figure 2, B, data at
day 1, respectively), as well as the control cells, were cultured
in growth medium for 2 days, followed by feeding with differentiation medium. Cx43 expression of the control cells,
which was clear in the early stage, was downregulated to a
minimum level in the late phase (Figure 2, B). As regards
Cx43(15) and Cx43(19), the extended level of expression continued throughout the period studied, although the expression
was decreased during differentiation at a similar rate.
Nevertheless, Cx43 overexpression still remained 5 to 8 times
higher in day 7 differentiated cells compared with the level in
control-transfected L6 cells.
Immunocytochemical analysis with a monoclonal antibody specific for Cx43 demonstrated low Cx43 expression
in cultures of control cells. In contrast, Cx43(19) myoblasts
displayed obvious immunofluorescent labeling, both in the
cytoplasm and on the cell surfaces where intercellular gap
junctions should exist (Figure 3).
Enhanced Intercellular Communication
Scrape loading with Lucifer yellow dye was performed on
confluent monolayers to evaluate the extent of GJIC. Dye
Suzuki et al
spread more widely throughout the monolayer up to 7 or 9
cell diameters in Cx43(19) cells (Figure 4, D), in contrast to
the more limited spread (up to 3 cell diameters) seen in control-transfected L6 cells (Figure 4, C). This finding was further confirmed by means of visualization of the transfer of
another dye (6-CF) microinjected into single cells.
Representative confocal microscopic images are shown for
control L6 (Figure 4, E) and Cx43(19) cells (Figure 4, F). In
9 of 15 injections in control cells, intercellular transfer of 6CF to immediately neighboring cells was observed only
after 3 minutes of dye injection. In the remaining 6 cases,
intercellular spread of dye was not noted. In contrast, injected dye spread rapidly within 1 minute to Cx43(19) cells
beyond immediate neighbors, reaching cells up to 7 orders
away in all of 12 injections.
Accelerated Myogenic Differentiation and Decreased
Proliferation Capacity
Cx43(15), Cx43(19), and control cells were cultured in
growth medium until subconfluence and then switched to
differentiation medium to assess the differentiation ability.
An enhancement of fusion and differentiation into multinucleated myotubes was evident in microscopic observations
of Cx43-transfected cells (Figure 5, B) by comparison with
control cells (Figure 5, A). Quantified extent of myogenesis
by measuring CK activity was significantly higher in the 7day incubation period in Cx43(15) and Cx43(19) myoblasts
than in control cells (Figure 5, C). The activity in Cx43overexpressing myoblasts increased rapidly until reaching
plateau levels at day 5, whereas control myoblasts showed a
gradual but constant increase in the level of CK activity
until day 7.
Cell proliferation was monitored up to 8 days after plating of 1 × 104 cells on 24-well chamber plates to study the
effect of increased GJIC on growth regulation. Control cells
proliferated significantly (P < .05) more rapidly than
Cx43(15) and Cx43(19) cells in the growth medium (Figure
5, D).
Discussion
This study has documented the feasibility of overexpression
of Cx43 in skeletal myoblasts and its influence on GJIC,
myogenic differentiation, and growth. Cardiomyocytes
must contract in a coordinated fashion for the heart to work
effectively as a pump. This synchronized harmony is
achieved through GJIC, which enables intercellular passage
of molecules, including inorganic ions and second messengers, thus achieving electrical and metabolic coupling of the
cells.5,6 Therefore it is considered that, in skeletal myoblast
transplantation to the heart, GJIC between grafted
myoblasts and native cardiomyocytes is necessary.
Although some reports have suggested the possible presence of such intercellular gap junctions by using electron
764 The Journal of Thoracic and Cardiovascular Surgery • October 2001
We thank Professor C. C. Naus (Department of Anatomy and
Cell Biology, University of Western Ontario, Canada) for the kind
donation of Cx43 cDNA. We also thank Dr E. Dupont (Cardiac
Medicine, Imperial College School of Medicine, United Kingdom)
for his technical assistance with the scrape-loading technique.
Finally, we thank a consultant statistician, Dr Tri Tat (National
Heart and Lung Institute, Imperial College School of Medicine,
United Kingdom) for his contribution to the statistical analysis.
The Journal of Thoracic and Cardiovascular Surgery • Volume 122, Number 4 765
EDITORIAL
CHD
CPS
ACD
and some connexins are proposed to act as tumor suppressors.19 Direct evidence for this inverse relationship has been
provided by experiments in which connexin genes were
transfected into tumorgenic cell lines.8,9 The effect of Cx43
on noncancer cells, however, has not been reported. In this
article we have demonstrated that Cx43 overexpression
attenuates cell proliferation of L6 skeletal myoblasts, which
is associated with enhanced GJIC and improved differentiation ability.
It is interesting and necessary as a further study to investigate whether these Cx43-overexpressing myoblasts
demonstrate enhanced GJIC with cardiomyocytes in comparison with wild-type or control-transfected myoblasts by
using an in vitro coculture (skeletal myoblasts and cardiomyocytes) system. However, a recent article by Reinecke
and associates4 has shown that the results of such in vitro
experiments do not correlate well with those of in vivo studies: primary skeletal myoblasts developed GJIC with cardiomyocytes in vitro, but grafted ones in the heart in vivo
failed to do so. Thus in vivo grafting models would be needed to clarify whether the Cx43-overexpressing cell lines we
have generated are actually advantageous in making GJIC
with cardiomyocytes in the heart, although there are technical limitations to the quantitative evaluation of GJIC in vivo.
Further investigation is needed to clarify this issue.
At present, the effect of drug treatment in patients with
end-stage heart failure is limited, and heart transplantation
is the only established definitive treatment, although there
are some serious disadvantages, such as complications related to immunosuppression.20 Cell transplantation is a
promising alternative, and several types of cells have been
examined as a graft for cell transplantation in animal models.3 We consider skeletal myoblasts to be the most promising because they retain the capacity to fuse with surrounding myoblasts or with damaged muscle fibers to regenerate
functional skeletal muscle. In addition, skeletal myoblasts
can be isolated from the patients themselves as autografts in
the clinical setting, avoiding the necessity of immunosuppression.1,3 We have shown that Cx43-overexpressing skeletal myoblasts demonstrate improved GJIC, as well as an
enhanced capacity for differentiation. These properties
should prove to be a considerable advantage when these
cells are grafted into myocardium, suggesting a promising
source of cells for transplantation.
CSP
microscopic observation or staining for Cx43,1,2 others have
demonstrated contradictory results.4 We consider that these
conflicting results may be caused by low-level, short-term
expression of Cx43 in grafted myoblasts. Although gap
junctions have been described during the early stages of
muscle development with concurrent expression of Cx43,
they are absent from adult skeletal muscle, with a coincident
downregulation of Cx43 expression.4,7 We generated several myoblast cell lines that continue to express a higher level
of Cx43 for a longer period, even when differentiated, as a
result of stable gene transfection. These cells showed much
enhanced GJIC compared with non-Cx43 transfectants, as
demonstrated by microinjection of 6-CF, as well as scrape
loading with Lucifer yellow dye. We expect such modification of skeletal myoblasts to be advantageous for their integration with host myocardium subsequent to grafting.
Although we focused on evaluating function of gap junctions in this study, further study to clarify the morphologic
differences in gap junctions between Cx43-overexpressing
and control myoblasts by using electron microscopy would
be useful.
Furthermore, we have demonstrated that Cx43 overexpression in L6 myoblasts enhances differentiation capacity.
Although similar findings that Cx43 overexpression induces
myogenic differentiation have been demonstrated with muscle-derived carcinoma cells,9,10 the present data represent
the first report demonstrating a significant role for Cx43 in
myogenic differentiation of skeletal myoblasts. The differentiation of skeletal myoblasts is characterized by withdrawal from the cell cycle, activation of muscle-specific
gene expression, and fusion of myoblasts to form multinucleated myotubes.l6 These myotubes then undergo further
differentiation to form the mature skeletal fiber that is the
functional contractile subunit of muscle tissue. The fusion
of myoblasts is preceded by a complex series of sequential
events, including cell alignment, adhesion, and intercellular
communication.17 It has been suggested that GJIC with concomitant expression of Cx43 might play an important role in
this process because its expression correlates with the
degree of myoblast fusion.5 Because such myogenic differentiation of grafted myoblasts into multinucleated myotubes
and, subsequently, functional mature muscle fibers may also
be another limiting factor of the effectiveness of cellular
cardiomyoplasty, one can expect that overexpression of
Cx43 would additionally improve the efficiency of cell
transplantation by accelerating myogenic differentiation of
grafted myoblasts.
In addition to ensuring intercellular coordination, gap
junctions have been reported to act as growth regulators by
enabling passage of growth-limiting molecules between
cells.18 An inverse relationship is generally seen between
the extent of GJIC and proliferative growth. Uncontrolled
growth, as in cancer, is characterized by decreased GJIC,
ET
Cardiopulmonary Support and Physiology
TX
Suzuki et al
EDITORIAL
Cardiopulmonary Support and Physiology
Suzuki et al
References
CHD
CPS
ACD
1. Taylor DA, Atkins BZ, Hungspreugs P, Jones TR, Reedy MC,
Hutcheson KA, et al. Regenerating functional myocardium: improved
performance after skeletal myoblast transplantation. Nat Med.
1998;4:929-33.
2. Suzuki K, Brand NJ, Smolenski RT, Jayakumar J, Murtuza B, Yacoub
MH. Development of a novel method for cell transplantation through
the coronary artery. Circulation. 2000;102(Suppl):III359-64.
3. Kao RL, Chiu RCJ. Cellular cardiomyoplasty: myocardial repair with
cell implantation. Austin (TX): Medical intelligence unit, Landes
Bioscience; 1997.
4. Reinecke H, MacDonald GH, Hauschka SD, Murry CE.
Electromechanical coupling between skeletal and cardiac muscle:
implications for infarct repair. J Cell Biol. 2000;149:731-40.
5. Verheule S, van Kempen MJ, Welscher PH, Kwak BR, Jongsma HJ.
Characterisation of gap junction channels in adult rabbit atrial and
ventricular myocardium. Circ Res. 1997;80:673-81.
6. Goodenough DA, Goliger JA, Paul DL. Connexins, connexons, and
intercellular communication. Annu Rev Biochem. 1996;65:457-502.
7. Balogh S, Naus CC, Merrifield PA. Expression of gap junctions in cultured rat L6 cells during myogenesis. Dev Biol. 1993;155:351-60.
8. Zhu D, Caveney S, Kidder GM, Naus CC. Transfection of C6 glioma
cells with connexin 43 cDNA: analysis of expression, intercellular
coupling, and cell proliferation. Proc Natl Acad Sci U S A.
1991;88:1883-7.
9. Proulx AA, Lin ZX, Naus CC. Transfection of rhabdomyosarcoma
cells with connexin 43 induces myogenic differentiation. Cell Growth
Differ. 1997;8:553-40.
10. Hirschi KK, Xu CE, Tsukamoto T, Sager R. Gap junction genes Cx26
and Cx43 individually suppress the cancer phenotype of human mam-
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
mary carcinoma cells and restore differentiation potential. Cell
Growth Differ. 1996;7:861-70.
Goldberg GS, Martyn KD, Lau AF. A connexin 43 antisense vector
reduces the ability of normal cells to inhibit the foci formation of
transformed cells. Mol Carcinog. 1994;11:106-14.
Okazaki S, Kawai H, Arii Y, Yamaguchi H, Saito S. Effects of calcitonin gene-related peptide and interleukin 6 on myoblast differentiation. Cell Prolif. 1996;29:173-82.
McKarns SC, Doolittle DJ. Limitations of the scrape-loading/dye
transfer technique to quantify inhibition of gap junctional intercellular
communication. Cell Biol Toxicol. 1992;8:89-103.
Koval M, Geist ST, Westphale EM, Kemendy AE, Civitelli R, Beyer
EC, et al. Transfected connexin 45 alters gap junction permeability in
cells expressing endogenous connexin 43. J Cell Biol. 1995;130:98795.
Szasz G, Gruber W, Bernt E. Creatine kinase in serum: I.
Determination of optimum reaction conditions. Clin Chem.
1976;22:650-6.
Richler C, Yaffe D. The in vitro cultivation and differentiation capacities of myogenic cell lines. Dev Biol. 1970;23:1-22.
Mega RM, Goudou D, Giaume C, Nicolet M, Rieger F. Is intercellular
communication via gap junction required for myoblast fusion? Cell
Adhes Commun. 1994;2:329-43.
Maldonado PE, Rose B, Loewenstein WR. Growth factors modulate
junctional cell-to-cell communication. J Membr Biol. 1988;106:20310.
Yamasaki H, Naus CCG. Role of connexin genes in growth control.
Carcinogenesis. 1996;17:1199-213.
Hunt SA. Current status of cardiac transplantation. JAMA.
1998;280:1692-8.
CSP
ET
TX
766 The Journal of Thoracic and Cardiovascular Surgery • October 2001