| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2005;27:104-110
© 2005 Elsevier Science NL
Department of Cardiac Surgery, Heart Center Leipzig, University of Leipzig, Leipzig, Germany
Received 9 February 2004; received in revised form 16 August 2004; accepted 23 August 2004.
* Corresponding author. Tel.: +49 341 865 1421; fax: +49 341 869 6646. (E-mail: rastan{at}rz.uni-leipzig.de).
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
-actinin) was used to assess cardiomyogenic differentiation. Action potential characteristics were recorded in whole cell patch clamp mode and to investigate intercellular communication a second gap junction permeable fluoroscopic dye was brought into BMCs by microinjection (Lucifer yellow, LY). Results: From day 7 in coculture BMCs beated synchronously with neonatal rat cardiomyocytes. On day 14, 55.9% of BMCs expressed -actinin and 98.3% were positive for gap junction protein connexin 43. BMC action potential duration (APD90) was mean 11.1ms with dV/dtmax of 26.8V/s and similar to atrial cardiac type. However, microinjection of LY revealed only little dye transfer into adjacent rat cardiomyocytes. Conclusions: Cocultured BMCs have the potential for early expression of muscle specific proteins in about 60% after 14 days and for cardiac gap junction proteins. Synchronous beating indicates an effective electromechanical coupling. In this heterologous setting we could prove only weak metabolic coupling.
Key Words: Bone marrow cells • Cell transplantation • Action potential • Patch clamp • Transdifferentiation
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
However, at the present state it is not clearly understood which mechanisms lead to improved left ventricular function after BMC transplantation. Considerations include transdifferentiation of BMCs to contractile cardiomyocytes and cell fusion as well as mechanisms-like improved angiogenesis by modulation of environmental factors. Little is known about the electrophysiological properties of BMCs and their potential for action potential generation. Experimental studies indicated both expression of cardiomyocyte-specific protein -actinin [6] and gap junction protein connexin 43 [7] implicating a functional integration of BMCs into recipient myocardium and formation of cell-to-cell docking. However, a close connexon contact of two neighboring cells to low resistance gap junction channels allowing functional electromechanical and metabolic coupling between BMCs and cardiomyocytes could not be demonstrated yet [8].
Aim of our study was to investigate the functional and electrophysiological integration of BMCs in the cardiac syncytium. For that we investigated morphological and electrophysiological characteristics as well as intercellular coupling of BMC-derived cells by coculturing BMCs with cardiomyocytes in a heterologous in vitro-model.
|
2. Material and methods |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
The in vitro-model is shown in Fig. 1. Neonatal rat ventricular cardiomyocytes were used as feeder cell model to simulate cardiac environment.
|
After two washing steps BMCs were resuspended in 12ml Dulbecco's Modified Eagles Medium (DMEM, Sigma–Aldrich, Steinheim, Germany) by adding 10% fetal bovine serum, 100µg/ml penicillin G, 100µg/ml streptomycin and 10µM 5-azacytidine (Sigma–Aldrich, Steinheim, Germany). 5-Azacytidine is a cytidine analog of Streptoverticillium ladakamus. It is a demethylating agent inhibiting DNA methyltransferase. Doing this an improved BMC transdifferentiation potential by hypomethylating regulatory genes was achieved as described elsewhere [9]. The cells were seeded on gelatin-coated tissue culture dishes. After 24h BMCs were washed and further cultured in DMEM without 5-azacytidine.
To identify the transplanted cells in the feeder myocardium BMCs were incubated for 4h with 1µg/ml 1,1'-dioctadecyl-3,3,3',3',-tetramethylindocarbocyanine perchlorate (DIL in DMEM and 0.1% dimethyl sulfoxide) after 6 days of cultivation. DIL (MW 933.9, Sigma–Aldrich, Steinheim, Germany) is a red fluoroscopic lipophilic dye accumulating in the membrane of rough endoplasmic reticulum (rER) [10]. It is a non-translocating dye, which allows detection of the cells even after 5 weeks (data not shown). Because of the high molecular weight it is considered to be non-gap junction permeable [8].
After another 24h DIL-labeled BMCs were detached with PBS containing 0.6mM EDTA and 500 units trypsin/ml over a 7-min period at 37°C (1 unit trypsin is defined as 
A253 of 0.001min–1 at pH 7.6 at 25°C using -benzoyl-L-arginine ethyl ester (BAEE) as substrate, Sigma–Aldrich, Steinheim, Germany). The detached cells were centrifuged at 170g for 5min, washed twice, resuspended in culture medium M199 and cocultured with 4 day old neonatal cardiomyocytes.
Neonatal rat ventricular myocytes were prepared from 0 to 1 day old rats, as described elsewhere [11] with some modifications. In brief, neonatal rats were decapitated and ventricles were dissociated fermentally at 37°C by a repetitive protocol using collagenase type CLS II (Biochrom AG, Berlin, Germany) with an enzyme activity of 270U/ml. The cardiomyocytes were washed and resuspended in M199 medium with Earls salts and HEPES (Gibco, Paisley, UK) by adding 100µg/ml penicillin, 100µg/ml streptomycin, 5% fetal bovine serum and 10% horse serum. The cells were seeded in gelatin-coated cover slips.
All cell cultures were maintained in a humidified incubator with an atmosphere of 5% CO2 and 95% air at 37°C. Unsettled cells were washed out after 24h, and the growing medium was replaced. Medium was replaced on alternating days.
2.2. Immunhistological staining
Cocultures were washed with PBS, fixed with methanol at –20°C and incubated with 1% Triton-X 100 in PBS (Sigma, Steinheim, Germany) for 30min. After incubation with PBS–BSA (bovine serum albumin, 0.1%) for 20min the cells were incubated with monoclonal antibodies against cardiomyocyte-specific -actinin (A-7811 monoclonal mouse antibody, Sigma, Steinheim, Germany, 1:100) and connexin 43 (MAB-3068 monoclonal mouse antibody, Chemicon, Temecula, USA, 1:500) for 30min at room temperature. Afterwards the cells were washed with PBS–BSA and incubated with the secondary antibody (F-8771, Fab specific anti-mouse IgG, Sigma, Steinheim, Germany, 1:200), conjugated with fluorescein isothiocyanate (FITC). The samples were washed again and mounted with Pertex (medite Medizintechnik, Burgdorf, Germany). Final evaluation was performed with a fluorescence microscope (Axioplan 2, Zeiss, Jena, Germany) at 400x and 1000x magnification.
2.3. Electrophysiological examination
Electrophysiological studies were performed 4, 7 and 14 days of coculture. They were transferred to a 1ml perfusion chamber of a patch clamp inverted microscope (Axiovert 25, Zeiss, Jena, Germany) and were superfused (2ml/min) with an extracellular tyrode solution containing (in mM) NaCl 135, KCl 4, CaCl2 2, MgCl2 1, NaH2PO4 0.33, HEPES 10, glucose 10, pH 7.4) at room temperature.
Electrophysiological experiments were performed using a discontinuous single electrode voltage clamp amplifier (dSEVC, SEC 05, switching frequency 35kHz, NPI electronic, Tamm, Germany). Data were recorded with a sampling rate of 10kHz and low-pass-filtered at 2kHz (Bessel filter) using cellworks software package (NPI electronic). Measurements were performed in the whole cell patch clamp configuration employing micropipettes of 4–5M
resistance made with a vertical puller (PIP5, Heka, Lambrecht, Germany) in a two step procedure from 1.5mm borosilicate capillary tubes (GB150F-8P, Science Products, Hofheim, Germany) and filled with an intracellular solution of (in mM) KCl 140, MgCl2 1, ethylenglycol-bis(ß-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) 5, HEPES 10, NaCl 5, Mg–ATP 3, pH 7.2).
Action potential (AP) measurements were performed if seal resistance exceeded 3G. Recording was started 5min after break-in allowing for equilibration of the patch pipette solution with the intracellular milieu.
Non-spontaneously beating BMCs were stimulated at 1Hz with two-fold threshold pulses of 1ms duration in current clamp mode. Action potentials were recorded and evaluated with regard to resting membrane potential (RMP), maximal upstroke velocity (dV/dtmax), overshoot potential (OVP), AP amplitude and action potential duration at 90% depolarization (APD90).
For the dye transfer experiments Lucifer Yellow dilithium salt (Sigma–Aldrich, Steinheim, Germany) was added in a concentration of 1% to the pipette solution. After break-in to the cell Lucifer Yellow distributes immediately to the cell by dialysis. Because of a molecular weight of 457.2Da it is considered to be gap junction permeable. In contrast to DIL Lucifer Yellow is a green fluorochrome (Emax 528nm). Therefore, LY transfer to adjacent cells would demonstrate a functional metabolic coupling.
2.4. Statistical analysis
For electrophysiological measurements a total of 10 neonatal rat ventricular myocytes and 10 BMCs were examined each. All values are expressed as mean±SE. Data were compared by two-tailed Student's t-test at a level of significance of P<0.05.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
The DIL labeling of the BMCs on day 6 was highly efficient and mild, more than 95% of the cells were alive after labeling. BMCs incorporated DIL by accumulating it in the rough endoplasmic reticulum membrane. Comparable to native BMCs the DIL labeling did not lead to any morphological alteration.
DIL-labeled BMCs could easily be identified in coculture by red fluorescence (Fig. 2a and b). After 7 days 143 of 256 investigated vital cells (55.9%) were positive for the cardiomyocyte-specific -actinin forming cell clusters of 2–10 cells with no increase after 14 days (Fig. 2d). From day 7 onwards more than 98% of these cells were also positive for gap junction protein connexin 43 indicating possible intercellular coupling. Gap junction proteins were perceptible both between the DIL-labeled bone marrow-derived cardiomyocytes and between the BMCs and rat cardiomyocytes (Fig. 2c).
|
3.3. Action potential characteristics
For the AP measurements non-beating BMCs with cardiomyocyte phenotype were identified at the end of the coculture period. Action potential properties of 10 BMCs were significantly different from those of neonatal rat ventricular cardiomyocytes (Fig. 3a and b). AP characteristics are described in Table 1. The resting membrane potential of the BMCs was mean –66.9mV with a wide dispersion (range –50.3 to –88.4mV). In all cells action potentials similar to atrial cardiac type was recorded. Mean maximum upstroke velocity was 26.8V/s (range 17.5–36.5). The action potential duration was extremely short. APs showed heterogeneity regarding maximum diastolic potential, AP upstroke velocity and overshoot potential. However, various forms of AP phenotypes based on RMP, AP upstroke, and AP duration allowing a division in atrial, ventricular or sinus node-like type were not recognized. Electrophysiological properties were not associated with cellular morphology.
|
|
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
-actinin. However, action potential characteristics displayed an atrial-like cardiac type with reduced RMP and very short APD90. In several animal models a positive postischemic effect on left ventricular function was shown for different cell types including cultured heart cells, smooth muscle cells, fetal cardiomyocytes and different subpopulations of progenitor cells [1–3,6,7,13]. Recently, in small series progenitor cells were successfully applied also in humans, using a transcoronary or epimyocardial approach [4,5,14]. Subpopulations with potential benefits were characterized as CD31+, CD34+/CD45+, CD34–/AC133+ and murine c-kit+/sca-1+ cells separated from bone marrow cells [1,2,4,14,15]. However, the best subpopulation is not identified yet. Therefore, in our study we isolated mononuclear cells from porcine bone marrow and cocultured this unselected pool of BMCs into ventricular cardiomyocytes. This is in accordance to Assmus and coworkers [4], who found a left ventricular improvement by transplanting an unselected BMC pool in humans. Similar to them we found 2.5±1.2% CD45/CD34-positive cells.
BMCs were pretreated with 5-aza-2'deoxycytidine, which is the bioactive in vivo metabolite of 5-azacytidine. 5-Azacytidine is a hypomethylating agent, which may lead to a reprogramming of progenitor cells by hypomethylating regulatory genes and improve their transdifferentiation potential. On the other hand this might be the reason for the missing proliferation of bone marrow cells during the first 7 days of monoculture [9]. Although the exact mechanisms are not fully understood yet, it was reported that 5-azacytidine pretreatment leads to cardiomyogenic transdifferentiation of the implanted cells [16]. It has also been shown that the proportion of cells differentiating into a cardiac lineage was greater in the 5-azacytidine group [9]. Creating optimal conditions for the BMCs to analyze the basic potential for cardiomyogenic differentiation was the reason to use 5-azacytidine in our experiment despite the fact that this could limit the conclusions of our data for the clinical practice. However, recently Fukuhara et al. [17] could show that direct cell-to-cell interaction is the key for untreated BMCs to go into cardiac lineage. That findings implicate that contractile environment plays the major role for cell differentiation, which might have been achieved in our study by the presence of beating cardiomyocytes. We used the active metabolite 5-aza-2'deoxycytidine in a concentration of 10µM, which was in fact 33-fold higher than in the report of Bittira, who employed 5-aza-2'deoxycytidine in a concentration of 0.3µM. In comparison, in vitro concentrations of 5-azacytidine were applied between 3µM [16] and 10mM [3].
Most of the published in vivo studies focussed on the improvement of cardiac left ventricular function after recently suffered myocardial infarction. Positive effects of stem cell or progenitor cell transplantation were derived from improvement in left ventricular ejection fraction, reduction in infarction size and left ventricular wall thinning [3–5,18,19,20]. Less is known about the processes taking place on the cellular level. Tomita and coworkers [3] found some troponin I positive bone marrow-derived cells 4 weeks after cell transplantation of marrow stromal cells in a porcine myocardial infarction model. Shake et al. [19] could demonstrate an expression of the muscle-specific proteins -actinin, troponin T and myosin heavy chain of mesenchymal stem cells 2 weeks after transplantation in ischemic porcine myocardium and Kawamoto and colleagues [15] found a higher capillary density after implantation of CD34+ endothelial progenitor cells in the infarction border zone in pigs with posterolateral ischemia induced by ameroid constrictor. Using different cell preparations these results are not directly comparable yet. Our model allowed the identification of bone marrow-derived cells and a phenotypic and functional characterization under cardiac in vitro conditions.
Synchronous beating with neonatal rat cardiomyocytes was observed in most of the beating BMCs. In areas of monolayer cell structure a ‘systolic’ reduction in BMC size was observed. That makes a passive ‘contraction’ by neighboring cardiomyocytes unlikely. Besides electromechanical coupling via gap junction channels another potential explanation for synchronously beating BMCs is due to mechanosensitive stimulation by mechanical stretching. This item has to be investigated more deeply in future studies.
Gap junctions typically establish cell-to-cell communication. The immunfluoroscopic evidence of connexin 43 at the cardiomyocyte-BMC cell border demonstrated gap-junction elements. This is in accordance with Fukuhara et al. [17] who found connexin 43 between murine BMCs and rat cardiomyocytes. However, bringing BMCs into myocardial scar tissue Tomita [3] could not prove gap junction expression. This underlines the demand of vital neighboring cardiomyocytes to allow BMCs for cardiac differentiation.
Injecting Lucifer Yellow in a synchronous beating DIL-positive cell enabled us to examine transjunctional metabolic coupling to adjacent rat ventricular cardiomyocytes. The finding of weak cell-to-cell communication is in agreement with other groups [20,21]. To elucidate the metabolic coupling in detail further experiments have to focus on coculture interval and on other dyes, which can pass gap junctions more easily. However, compared to microinjection into monocultured ventricular cardiomyocytes (Fig. 4f) we found that dye transfer from BMCs into cardiomyocytes was extremely reduced (Fig. 4c and d). Reasons for that finding include the use of two distinct cell types and the short investigation period as well as an insufficient potential for gap-junction expression. Another reason for reduced coupling findings may reflect the species difference between porcine bone marrow cells and rat ventricular cardiomyocytes. However, using neonatal rat ventricular cardiomyocytes is a well-established feeder cell model. In contrast to adult rat cardiomyocytes they do not dissimilate their gap junctions in vitro. Because extracellular loops, membrane spanning domains and connexin distributions are highly conserved among species we decided for the heterologous setting [8].
We investigated whether or not the electrophysiological properties of transplanted BMCs were similar to those of native cardiomyocytes after 14 days of coculture. Finding pacing thresholds in all cells we could draw a clear dividing line between action potentials and electrotonic signals. As expected, we found significantly different AP characteristics between neonatal rat ventricular cardiomyocytes and porcine BMC-derived cardiomyocytes (Table 1). The mean RMP of BMCs was comparable to murine embryonic cardiomyocytes after 14 days of transplantation and murine bone marrow cells [16,20]. In contrast to others, we could not identify distinguishable pattern of action potentials such as sinus-node-like, atrial-like or ventricular-like [16,22]. It remains speculative that several AP types may reflect different developmental stages of cellular transdifferentiation. After 14 days in coculture we merely detected a transdifferentiation to atrial-like cardiomyocytes with significant morphological heterogeneity of RMP, AP amplitude and AP duration. The dV/dtmax was slower compared to other reports [22]. This might be explained by a different transplantation interval and reduced Na+ channel expression in BMCs compared to progenitor cells used by others. However, myocytes with atypical AP properties have the potential to lead to arrhythmia or arrhythmic foci.
The APD90 of embryonic stem cells was 35.8ms as indicated by Roell and associates and 60.6ms recorded by Zhang and coworkers [20,22]. Makino [16] described an action potential duration of 46.6ms for marrow stromal cells, which is also considerably shorter than a ventricular AP exhibiting about 100ms duration. In our experiment the mean APD90 was 11.1ms and comparatively short without significant plateau. This may be explained by the use of adult bone marrow cells and different species. Although we are still waiting for detailed characterization of ionic channels during cardiac transdifferentiation these results indicate basic restrictions in ionic channel development even after 2 weeks of BMCs in coculture.
At the present state a differentiation of BMCs into a final cardiac cell lineage is not definitely certain. For skeletal myoblasts an expression of connexin 43 and -actinin as well as sporadic spontaneous contraction could be shown [23]. Therefore, our findings of BMC's action potential characteristics were potentially consistent with skeletal myoblasts.
In 2 weeks old cocultures we found about 30% of our cells being DIL negative connective tissue cells (data not shown). It is well known that fibroblasts, which also express connexin 43, influence cell-to-cell contact. Up to now it is not well understood, whether connective tissue cells express factors that influence BMC differentiation directly or by modification of local environment. In our study we exclusively examined BMCs docking to neighboring cardiomyocytes that we identified by cross-striation. However, because of the in vitro design of our study a transfer to in vivo proceedings is limited.
Cell fusion as a potential mechanism for our findings could not be excluded. However, this phenomenon is not clearly described for adult BMCs. Action potential pattern of fused cells remain speculative, although the electrical cell properties we found were not consistent with mature cardiomyocytes. Interestingly, in this context Deb and associates [18] showed no evidence of cell fusion by chromosomal ploidy analysis in hearts after gender mismatched bone marrow transplantation.
4.1. Limitations
The results of our in vitro experiments do not allow direct clinical transfer to in vivo conditions. Furthermore, they represent a topical status at definitive moments only and cannot give information about the fate of BMCs after prolonged incubation.
In conclusion we found that bone marrow cells transdifferentiate to cardiomyocyte-like cells in coculture expressing cardiac specific proteins. From the electrophysiological aspect these cells were immature despite synchronous contraction. They represent an atrial-like action potential pattern. In this heterologous setting we found only weak metabolic cell coupling. Further investigation is warranted to elucidate cell fusion phenomenon, electrophysiological properties and cell-to-cell docking of bone marrow-derived cells regarding time of monoculture, transplantation interval and optimal BMC subpopulation.
|
Footnotes |
|---|
Presented at the joint 17th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 11th Annual Meeting of the European Society of Thoracic Surgeons, Vienna, Austria, October 12–15, 2003. |
References |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. DiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Orlandi, F. Pagani, D. Avitabile, G. Bonanno, G. Scambia, E. Vigna, F. Grassi, F. Eusebi, S. Fucile, M. Pesce, et al. Functional properties of cells obtained from human cord blood CD34+ stem cells and mouse cardiac myocytes in coculture Am J Physiol Heart Circ Physiol, April 1, 2008; 294(4): H1541 - H1549. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
