Impact of stirred suspension bioreactor culture on the differentiation of murine embryonic stem cells into cardiomyocytes

Abstract

Background: Embryonic stem cells (ESCs) can proliferate endlessly and are able to differentiate into all cell lineages that make up the adult organism. Under particular in vitro culture conditions, ESCs can be expanded and induced to differentiate into cardiomyocytes in stirred suspension bioreactors (SSBs). However, in using these systems we must be cognizant of the mechanical forces acting upon the cells. The effect of mechanical forces and shear stress on ESC pluripotency and differentiation has yet to be clarified. The purpose of this study was to investigate the impact of the suspension culture environment on ESC pluripotency during cardiomyocyte differentiation.

Results: Murine D3-MHC-neo(r) ESCs formed embyroid bodies (EBs) and differentiated into cardiomyocytes over 25 days in static culture and suspension bioreactors. G418 (Geneticin) was used in both systems from day 10 to enrich for cardiomyocytes by eliminating non-resistant, undifferentiated cells. Treatment of EBs with 1 mM ascorbic acid and 0.5% dimethyl sulfoxide from day 3 markedly increased the number of beating EBs, which displayed spontaneous and cadenced contractile beating on day 11 in the bioreactor. Our results showed that the bioreactor differentiated cells displayed the characteristics of fully functional cardiomyocytes. Remarkably, however, our results demonstrated that the bioreactor differentiated ESCs retained their ability to express pluripotency markers, to form ESC-like colonies, and to generate teratomas upon transplantation, whereas the cells differentiated in adherent culture lost these characteristics.

Conclusions: This study demonstrates that although cardiomyocyte differentiation can be achieved in stirred suspension bioreactors, the addition of medium enhancers is not adequate to force complete differentiation as fluid shear forces appear to maintain a subpopulation of cells in a transient pluripotent state. The development of successful ESC differentiation protocols within suspension bioreactors demands a more complete understanding of the impacts of shear forces on the regulation of pluripotency and differentiation in pluripotent stem cells.

Figure 1

Pluripotency of bioreactor differentiated cells as shown by tumor formation in SCID mice. A) The 125-mL NDS stirred-suspension bioreactor. B) Sections of teratoma generated from non-drug selected pluripotent mESCs in bioreactor showing cells from all three germ layers: (Star) Mesoderm; (Circle) Ectoderm; (Triangle) Endoderm (H&E staining). C) Teratoma derived from mESCs after differentiation under static culture system. The majority of teratoma contains unorganized tissue. Only some gut-like epithelial cells were recognizable (Square).

Figure 2

Expression of pluripotency and cardiac markers during differentiation in suspension culture. A) RT-PCR showed the expression of pluripotency as well as cardiac lineage genes during differentiation in the suspension bioreactor. The G418 used from day 10 of differentiation to select cardiomyocytes: w: with drug; w/o: without drug, Neg: Negative control. The numbers show days past differentiation. Cardiac markers were examined in non-drug selected bioreactors. B) RT-PCR showed the absence of pluripotency markers gene expression during static differentiation. The expression of the cardiac lineage gene α-MHC increased during differentiation in static culture in non-drug selected cultures. C) Quantitative RT-PCR results in non-drug selected cultures revealed the fold increase of gene expression due to shear forces in suspension bioreactor compared to static culture during the time course of cardic differentiation. Values represent means±SD of 3 independent experiments. p < 0.05.

Figure 3

Expression of pluripotency and cardiac markers after 25 days in suspension culture. A) Fluorescence-activated cell sorting (FACS) analysis of Oct4 and MF-20. The percentage of cells expressing Oct4 increased from 20% to 54%. The maximum percentage of cells expressing MF-20 was on day 20 and declined afterward. A population of cells was expressing both markers at the same time during differentiation. Statistical analysis (ANOVA) was performed using GraphPad Prism4 (GraphPad Software) and significance was set at p < 0.05. B) Confocal microscopy confirmed the expression of Oct4 and Nanog in a sub-population of cells 20 days after differentiation in bioreactor. C) Distribution of cell populations in suspension culture system as revealed by FACS analysis of Oct4 and MF-20 genes.

Figure 4

Pluripotency of bioreactor-differentiated embryoid bodies. A) Differentiated ESCs retained their ability to form ESC-like colonies upon plating on gelatin-coated culture plates. The morphology of the colonies as well as expression of pluripotency markers showed the pluripotency of differentiated cells after 25 days in suspension bioreactor without drug selection. Immunostaining was performed with antibodies against Nanog and Oct4. B) Confocal microscopy of day 25 EBs showed the simultaneous expression of α-MHC and Oct4 within the bioreactor-derived aggregates. Scale bar: 100 μm.

Figure 5

Mouse ESC^neo differentiation in suspension culture. A) Murine ES cell aggregates. The size of EBs was growing over time period of differentiation. B) Beating cardiac body (CB) 21 days after differentiation. The CBs were beating rhythmically with various rates. C) Percentage of beating EBs. The maximum percentage of beating was 50% on day 14 of EBs. D) Dissociated cardiomyocytes stained with specific cardiac antibody Myosin sarcomere (MF-20) green and DAPI-blue. E) Beating cardiac bodies showed specific subcellular ultrstructure of cardiomyocytes using electron microscopy. The presence the cardiomyocytes specific structure including nucleus and myofibrils, mitochondria, Endoplasmic reticulum, gap junctions and desmosomes was shown. Hb: H-band; Ib: I-Band; Ab: A-band; Zb: Z-band; Dj: Desmosome Junction; GJ: Gap junction; Mi: Mitochondria; RE: Reticulum Endoplasmic; N: Nucleus. Scale bars: 100 μM.

Figure 6

Chronotropical responses of bioreactor-derived cardiomyocytes. A) Variation in the number of beatings after drug application. The highest change was due to application of Bay K8664. B) Changes in beating rate. The frequency of changes in beating clearly showed the development of mature ion channels in the membrane of cardiomyocytes, which enabled them to respond to drugs (P < 0.05).