OUP user menu

Predominant fusion of bone marrow-derived cardiomyocytes

Jacob Andrade, Jason T. Lam, Monica Zamora, Chengqun Huang, Diego Franco, Noemi Sevilla, Peter J. Gruber, Jonathan T. Lu, Pilar Ruiz-Lozano
DOI: http://dx.doi.org/10.1016/j.cardiores.2005.09.016 387-393 First published online: 1 December 2005


Objectives: Here we address the capacity of bone marrow-derived cells (BMDCs) to trans-differentiate into mature myocytes under the physiological stimulus of exercise training.

Methods: For this purpose, we have transplanted bone marrow from mice ubiquitously expressing enhanced green fluorescence protein (eGFP) into host mice that have been subjected to a prolonged program of exercise.

Results: In all successful bone marrow reconstitutions (greater than 80%), we observed rare but consistent events of bone marrow-derived cardiomyocytes, the frequency of which was unchanged upon exercise training. We have further determined whether these recruited myocytes are a product of trans-differentiation or fusion by the use of a genetic system that distinguishes cell fusion from trans-differentiation in a single-cell assay.

Conclusions: We concluded that both in the unchallenged mouse and in the trained specimens, fusion is the most prominent mechanism by which bone marrow-derived cells are observed in the myocyte compartment.

  • Bone marrow-derived cell
  • Heart failure
  • Cardiomyocytes

1. Introduction

Heart failure is a leading cause of lethality in western countries. Recent advances in regenerative medicine have pointed to the possibility of repairing damaged hearts [1–5] and there are several candidate sources of cells with potential to generate cardiomyocytes, including the adult heart [1,6,7] (reviewed in [8,9]), adult bone marrow mesenchymal stem cells (EPC) [2,10,11], multipotent adult progenitor cells (MAPC) from bone marrow [12], cardiac side population cells [13] and embryonic stem cells [5].

Several lines of evidence support the notion of bone marrow-derived cells as a source of cardiac myocytes in humans. Female allograft hearts transplanted into male patients show minute repopulation of myocytes by extra-cardiac precursors [14] and gender-mismatched bone marrow transplant patients suggest the possibility of bone marrow-derived cardiac myocytes in the human heart [15]. Major efforts have been devoted to the determination of the type of marrow cells responsible for myocardiogenesis, with some divergence in results from different groups. Isolation of the bone marrow hematopoietic lineage fails to yield cells to the myocyte compartment, neither to brain, kidney, gut, liver or muscle [16]. Independent studies have shown that cells negative for blood lineage (Lin) and positive for hematopoietic stem cell (ckit+) can be found in the adult heart, can be purified and may support cardiac regeneration [1]. The mechanisms by which bone marrow cells derive into diverse type of non-hematopoietic cells, including cardiomyocytes and epithelia, are not clear but cell fusion seems to be prevalent for cardiomyocytes [17] while trans-differentiation is prevalent in epithelial cells [18].

Aside from some yet unresolved questions regarding the molecular and cellular mechanisms that drive bone marrow-derived cells into the myocardial lineage, bone marrow-derived cardiomyocytes are becoming a therapeutic tool. Bone marrow-derived cells are currently being used in several clinical trials around the world to repair damage associated with heart failure and occluded coronary arteries [19]. One critical aspect in regenerative medicine is the analysis of stem cell niches and their ability to expand. Here we investigate the possibility as to whether physiological stimulus can enhance the potential of bone marrow-derived cells to become cardiac myocytes and the mechanism by which bone marrow-derived cells appear as myocytes under physiological conditions.

2 Materials and methods

2.1 Bone marrow cell isolation

Crude bone marrow cells for transplantation were isolated from transgenic mice donors between 8 and 24 weeks of age (Table 1). Cells were flushed out of femurs and tibias with PBS using a 27-gauge needle and screened through a 100-μm mesh. The flushed solution was centrifuged at 1200 rpm for 8 min to collect the cells. The cells were then incubated in 1 ml of 0.85% ammonium chloride for 3 min (eliminates RBCs), washed with PBS, centrifuged and resuspended to a final concentration of 1 × 107 cells/ml in sterile PBS. All procedures were done under sterile conditions.

View this table:
Table 1

Summary of bone marrow transplantation (BMT) experiments performed

ExperimentGenotype of marrow donorGenotype of recipientTime of cell isolation (post-transplant)Method of analysisNumber of cells examinedNumber of animalsNumber of positive cardiocytes
eGFP BMTGFPC57Bl/61 weekHistology15,000,00050
2 weeks50
3 weeks50
4 weeks51–2
16 weeks58
28 weeks512
36 weeks516
16 weeksCell isolation2,000,00039
Nebulette BMTNebuletteC57Bl/616 weeksHistology15,000,00033
Cell isolation3,000,00030
BMT TACGFPC57Bl/63.5 weeksHistology6,000,00021
BMT Sham21
BMT TrainedGFPC57Bl/620 weeksCell isolation9,000,000310
BMT Untrained20 weeks39
FusionZEGMLC2v28 weeks1,500,00035 (GFP+) 0 (β-Gal+)
  • Chart indicates the number of animals analyzed, number of cells scored, methodology used, type of donor and recipient, and the time when the samples were analyzed after transplantation.

2.2 Bone marrow transplantation

Recipient mice of either C57BL/6 (The Jackson Laboratory) or transgenic mice for MLC2v-Cre were irradiated with 1000 rad in the UCSD School of Medicine Irradiation Facility and maintained in a clean environment with autoclaved food, water and housing for a minimum of 2 weeks post-irradiation. 1 × 106 bone marrow-derived cells were injected into the tail veins of the recipient mice 24 h after irradiation. Animals were then sacrificed by cervical dislocation at several time points after transplantation and tissue was prepared for histological analysis or cell isolation studies. All animals were kept under normal stress-free conditions except for those set aside for pressure overload and treadmill exercise experiments. All animal procedures were approved by the UCSD Animal Subjects Committee.

2.3 Transplantation of bone marrow cells labeled with muscle-restricted reporters

Bone marrow was isolated from nebulette-LacZ reporter mice. Nebulette is a sarcomeric protein exclusively expressed in the myocardium [20]. A 3.5-kb promoter-driven β-galactosidase expression fully recapitulates the endogenous nebulette gene expression (Supplementary Fig. 1).

2.4 Fusion vs. trans-differentiation

Mice expressing Cre recombinase under the control of the ventricular myosin light chain 2, (MLC2v) promoter [21] were lethally irradiated and transplanted with bone marrow from a double reporter mouse line, Z/EG (The Jackson Laboratory). This mouse ubiquitously expresses LacZ under the control of the CMV enhanced chicken beta-actin promoter. Once this line is crossed with a mouse expressing Cre recombinase, β-galactosidase expression is replaced with enhanced green fluorescence protein (eGFP). The hearts were removed from these transplanted animals at least 6 months after transplantation and prepared for cell isolation studies or histological analysis. Cardiomyocytes positive for LacZ expression indicate trans-differentiation of hematopoetic cells, whereas cardiomyocytes found expressing eGFP indicate fusion of a hematopoetic cell with a cardiac myocyte.

2.5 Cardiomyocyte isolation

Cardiomyocytes were isolated as previously described method [22]. Briefly, hearts were excised from heparinized animals and rinsed in ice-cold Tyrode Solution containing (in mM) 140 NaCl, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 10 HEPES, 10 glucose (pH 7.4). The hearts were cannulated and mounted on a Langendorff perfusion apparatus via the aorta. They were then perfused with Ca2+-free Tyrode at 37 °C for 3 min. Perfusion was then switched to the same solution containing collagenase type II at 1 mg/ml (Gibco BRL) and continued until the hearts became flaccid (15–20 min). The heart was rinsed with Ca2+-free Tyrode for 5 min. Left ventricular tissue was minced and dissociated with pipette. The myocytes solution was filtered through nylon filter. Myocytes were plated onto laminin-coated dish for examination.

2.6 Exercise training

Physiological hypertrophy was induced in irradiated adult C57BL/6 using a long-term exercise training program previously described [23] on 3-month-old mice after receiving bone marrow transplantation from ubiquitously expressing GFP mice. The hearts were removed after the 8-week training program and prepared for cell isolation analysis or histological analysis.

2.6.1 Ethical statement

This investigation conforms with the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publication NO. 85-23, revised 1996).

3 Results

3.1 Bone marrow-derived myocytes in the unchallenged mouse

To determine the ability of bone marrow-derived cells to contribute to cardiac tissue, we transplanted crude bone marrow cells–extracted from ubiquitously expressing eGFP mice–into lethally irradiated wild-type mice. Recipient animals were sacrificed and screened at 1 week, 2 weeks, 3 weeks, 1 month, 2 months, 4 months, 7 months and 9 months for positive expression of eGFP in the marrow. The hearts of those animals with transplanted eGFP-positive marrow and reconstituted at least 80% (Fig. 1B) were subsequently screened for detectable levels of eGFP-positive myocytes. We observed eGFP expression in a limited number of mature cardiomyocytes, identified by morphology, striation, and α-actinin staining (Fig. 1A). There was no particular preference for these cells to be found in the left or right ventricle. However, none were found in the atrial tissue. No detectable levels of eGFP-expressing cardiomyocytes were found before 1 month post-transplantation. However, numbers of eGFP-positive cardiomyocytes increased with time after transplantation (Fig. 1D). Isolated eGFP-positive cardiac cells from host eGFP-reconstituted bone marrow (Fig. 1B) displayed the normal cardiomyocyte rod-shaped morphology and were functionally contracting (Fig. 1B).

Fig. 1

Bone marrow-derived cells yield ventricular myocytes. (A) Histological analysis of transplanted mice with eGPF expressing bone marrow cells. eGFP-positive cells (green) are detected in the host ventricles and co-expressed sarcomeric α-actinin (red). (B) Purification of bone marrow from the host animals to determine level of reconstitution. This was approximately 80%. (C) Purified eGFP-positive myocyte. Image is slightly out of focus due to cell contraction. (D) Time course of eGFP cells in the heart upon transplantation. Subjects were not analyzed after 9 months.

To gain further insight about the nature of these bone marrow-derived cardiomyocytes, we performed a similar set of experiments in which we transplanted irradiated host C57 mice with crude bone marrow extract of a myocyte-specific nebulette-LacZ transgenic mouse. In the recipient hearts, we observed sparse blue signals in histological sections (see Supplementary data, Fig. 2) but, unlike the eGFP bone marrow experiments, we were unable to isolate LacZ-positive myocytes.

3.2 Engraftment of bone marrow-derived cells in the myocardium is not enhanced by physiological stimuli

To determine if bone marrow-derived cells contributed to the heart in pathological and physiological conditions, we induced hypertrophy in transplanted animals with two methods: pressure overload and exercise.

We induced pathological hypertrophy by pressure overload of the hearts in transgenic bone marrow recipient mice using a surgical transverse aortic banding protocol [24]. We observed an increased heart weight-to-body weight ratio in animals receiving surgery compared to those submitted to mock surgery. Despite the visible difference in heart size, we did not observe a significant increase in number of eGFP-expressing cardiomyocytes when compared to those sham operated (Table 1).

As an additional stimulus, we induced physiological hypertrophy in transplanted mice by placing them on an 8-week rigorous treadmill running program [23]. Trained mice were monitored weekly for improvements in oxygen uptake and carbon dioxide production in relation to running speed. VO2 max was determined as the maximum speed an animal can run aerobically. During training, animals were run at intervals of 80–90% VO2 max for 8 min and 50–60% VO2 max for 2 min for a total of 1 h 4 days per week. Monitoring was done on the fifth day that involved a gradual increase in maximum running speed for VO2 max. After 8 weeks, animals were sacrificed and hearts were screened for eGFP expressing cardiomyocytes. We did not observe any significant increase in the number of eGFD-positive cardiomyocytes compared with untrained (sedentary) transplanted animals (Table 1).

3.3 Cell fusion is a predominant mechanism to yield bone marrow-derived myocytes

In order to establish whether bone marrow-derived cardiomyocytes are a product of trans-differentiation of bone marrow cells into cardiomyocytes or fusion of bone marrow cells with pre-existing myocytes, we used a method based on Cre/lox recombination, a technique extensively used to conditionally turn on or off gene expression in specific cardiac cell lineages [25]. We extracted bone marrow from the double conditional Cre reporter mouse line Z/EG [26] and transplanted into lethally irradiated mice expressing Cre recombinase under the control of ventricular cardiomyocyte promoter (MLC-2v-Cre) [21,27]. In the Z/EG line, transgenic mice constitutively express LacZ under the control of the CMV enhancer/chicken β-actin promoter. When Cre-expressing (Cre+) cells fuse with Z/EG, Cre recombinase excises the β-galactosidase reporter, resulting in expression of eGFP in the fused cells. Consequently, fused cells can be detected easily by GFP fluorescence, while unfused cells continue to express LacZ. Therefore, both fused and trans-differentiated cells can be detected in a single experiment.

To control for efficiency of recombination, we analyzed GFP activity in neonatal (not shown) and adult animals (Fig. 2A and B) from the offspring of Z/EG crossed with MLC2v-Cre. Fluorescence analysis of double transgenic mice demonstrated strong and specific Cre-mediated recombination in the ventricular myocardium (Fig. 2B).

Fig. 2

Cell fusion is the prevalent mechanisms to achieve bone marrow-derived cardiomyocytes in the unchallenged state. (A and B) Efficiency of the selectable markers. (A) The ZEG mouse expresses LacZ almost ubiquitously in the heart except for small patches detectable in the right ventricle (RV), left ventricle (LV) and inter-ventricular septum (IVS) (B) LacZ and GFP staining of the intercross of Z/EG with MLC-2v-CRE. Double positives for this cross were analyzed at the adult state. Recombination was detectable all throughout the ventricles. Atrial tissue (where MLC2v is not active) remained LacZ positive. (C) Double immunoshistochemistry GFP/LacZ on cryosections from recipient hearts. (D) Cell isolation, using the Langerdorff system of transplanted MLC2v animals yields exclusively GFP-positive cells.

Subsequently, we purified bone marrow cells from the Z/EG mice and transplanted them into lethally irradiated MLC2v-Cre mice. Four months after transplantation, we dissected the hearts of the host mice and analyzed their LacZ/GFP composition. We observed GFP-positive cells by histological analysis of host hearts frozen sections (Fig. 2 C). Upon purification of the myocyte fraction using the Langendorff system, we plated 1.5 million myocytes per animal and counted the number of GFP- and LacZ-positive cells. We found a small number of distinct isolated myocytes (5 GFP positive/preparation) expressing GFP (Fig. 2D). None of the isolated myocytes or the histology preparations were positive for LacZ staining demonstrating that in the unchallenged wild-type, cell fusion is the predominant mechanism of myocyte generation from bone marrow derivatives.

4 Discussion

The ability of any cell type to repopulate the heart for use as a therapeutic tool at this point remains unproven. One fact is clear: the heart has, if any, minimal regenerative capacity though it may have mechanisms to repair micro-injuries. If a reservoir of cardiac precursor cells exists in the adult animal, where is it located? Currently, experimental data exist that support two locations for this very small but putatively significant reservoir: First, stem cells may reside in the heart in a fashion similar to satellite cells for skeletal muscle and/or second, cells of extracardiac origin, such as bone marrow.

The initial discovery that bone marrow-derived cells could differentiate into cardiomyocytes has led to clinical trials [19,28]. However, the use of hematopoietic cells as a therapeutic tool is now being challenged [29,30]. The question arises whether non-hematopoietic cells of bone marrow origin contribute to the heart, either as a mechanism of injury repair or as a rare event of trans-differentiation. A positive answer for any of these situations would set the groundwork for a source of congenic cells to be used as a therapeutic tool. For this reason, we have used unpurified crude bone marrow extracts genetically labeled to analyze total (hematopoietic and non-hematopoietic) bone marrow potential to generate cardiac myocytes in the unchallenged mouse and under conditions of physiological stimulation.

Our initial finding of eGFP-positive cardiomyocytes demonstrates that bone marrow-derived cells arrive to the heart and yield contractile myocytes whose gross morphology is indistinguishable from the host myocytes. This is in agreement with many recent reports that describe bone-marrow-derived cells as able, with variable frequency, to yield myocytes [2,30,31]. More discouraging were the experiments using bone marrow cells driving muscle-specific promoters fused to LacZ. The sparse blue signals in histological sections did not correspond to purifiable contractile LacZ positive myocytes, as are only indicative of background signal in the histological preparation. This is highly suggestive that, not only the hematopoietic fraction, but also non-hematopoietic bone marrow cells are unable to trans-differentiate into cardiomyocytes in the unchallenged mouse. However, we cannot rule out the possibility of promoter shutdown, as others have previously described in independent studies [31].

The combined use of total bone marrow extracts and transplantation with a double indicator mouse in the context of a ventricular muscle-specific Cre-bearing mouse presents two advantages compared to the previous strategies described. First, these studies are independent of promoter shutdown, due to the permanent labeling upon recombination. Second, it unambiguously distinguishes fusion from trans-differentiation, at a single-cell level, in the same preparation. The finding of only eGFP-positive cells in our preparations strongly suggest that marrow trans-differentiation into ventricular myocytes is, at least, a very rare event in the resting state, at least in the mouse model.

Our results also indicate that bone marrow-derived cells do not respond to either acute pathological stimuli or physiological stimulation upon exercise to the myocyte compartment, as demonstrated by a similar level of recruitment compared to the sham-operated or the sedentary mice. In light of our results, further approaches for cell therapy in the heart will have to be searched in alternative cellular sources outside the bone marrow compartment. New discoveries of resident progenitor cells will probably be the source of choice for future strategies.


Authors are thankful to Dr. Kenneth R. Chien for his support and to Dr. Karl-Ludwig Laugwitz for helpful comments on the manuscript. This work is supported by NIH grants to PR-L.

Appendix A. Supplementary Data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.cardiores.2005.09.016.


  • Time for primary review 10 days


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
View Abstract