Mesenchymal stem cells rescue cardiomyoblasts from cell death in an in vitro ischemia model via direct cell-to-cell connections

Abstract

Background: Bone marrow derived mesenchymal stem cells (MSCs) are promising candidates for cell based therapies in myocardial infarction. However, the exact underlying cellular mechanisms are still not fully understood. Our aim was to explore the possible role of direct cell-to-cell interaction between ischemic H9c2 cardiomyoblasts and normal MSCs. Using an in vitro ischemia model of 150 minutes of oxygen glucose deprivation we investigated cell viability and cell interactions with confocal microscopy and flow cytometry.

Results: Our model revealed that adding normal MSCs to the ischemic cell population significantly decreased the ratio of dead H9c2 cells (H9c2 only: 0.85 +/- 0.086 vs. H9c2+MSCs: 0.16 +/- 0.035). This effect was dependent on direct cell-to-cell contact since co-cultivation with MSCs cultured in cell inserts did not exert the same beneficial effect (ratio of dead H9c2 cells: 0.90 +/- 0.055). Confocal microscopy revealed that cardiomyoblasts and MSCs frequently formed 200-500 nm wide intercellular connections and cell fusion rarely occurred between these cells.

Conclusion: Based on these results we hypothesize that mesenchymal stem cells may reduce the number of dead cardiomyoblasts after ischemic damage via direct cell-to-cell interactions and intercellular tubular connections may play an important role in these processes.

Figure 1

Ischemia model on cardiomyoblasts. (A) Follow up of OGD on cardiomyoblasts. Cells were stained with calcein-AM (ex/em 494/517 nm) for live cells (green) and ethidium homodimer (ex/em 528/617 nm) for dead cells (red). (B) Flow cytometry analysis of control and ischemic cardiomyoblasts labeled with ethidium homodimer after OGD. The green curve represents the control cardiomyoblasts and the red shows the ischemic cardiomyoblasts. The complete rightward shift of the red curve based on these representative data indicates that OGD increased the number of dead cells nearly maximally.

Figure 2

Morphology and viability of H9c2 cells 24 hours after OGD. (A) DiO-labeled H9c2 cells without MSCs observed 24 hours after OGD were predominantly rounded up and stained with ethidium homodimer indicating cell death in progress. (B) Flow cytometry analysis of control and ischemic H9c2 cells cultured for 24 hours after OGD labeled with ethidium homodimer (ex/em 528/617 nm) showed that the number of dead H9c2 cells was elevated compared to the control group (median fluorescence from 19 to 65 units). (C) Co-cultivation of DiO-labeled H9c2 (ex/em 488/501 nm) cells and DiD-labeled MSCs (ex/em 633/665 nm) for 24 hours after OGD showed that the morphology of ischemically damaged cells were normal after 24 hours. (D) Flow cytometry analysis revealed that after co-cultivation of cells the number of dead H9c2 cells remained on the control level (median fluorescence 24 versus 23 units).

Figure 3

Co-cultivation of H9c2 cells with MSCs decreased cell death. (A) Experimental layouts after in vitro ischemia. (B) The ratio of dead H9c2 cells was significantly smaller when MSCs were added after OGD (0.85 ± 0.086 vs. 0.16 ± 0.035, n = 5), but MSCs added in cell culture inserts did not decrease significantly the ratio of dead H9c2 cells (0.90 ± 0.055, n = 5). Data represent mean ± SEM. *p < 0.05 C+MSC vs. C and C+MSC vs. C+MSC ins. (C: H9c2 cells only; C+MSC: H9c2 cells and MSCs; C+MSC ins: H9c2 cells and MSCs in cell culture inserts) (C) Absolute number of live H9c2 cells before and after OGD shows that before OGD the H9c2 cells were close to confluence (63,120 ± 7,694) and there was little increase in cell numbers during the next 24 hours if the cells were left to grow without OGD (76,116 ± 3,396). 24 hours after OGD the number of viable cells was very low when cultured alone or with MSCs in cell insert (1,757 ± 1,081 and 990 ± 608 respectively), which was significantly increased (15,174 ± 3,975) if MSCs added directly. (D) MSCs labeled with Vybrant DiD were growing on cell culture inserts in the same manner as under normal culture conditions after 24 hours of cultivation. Scale bar represents 100 μm.

Figure 4

Formation of intercellular connections after OGD. (A) Nanotubular network formation was observed among DiO-labeled cardiomyoblasts (green) and DiD-labeled MSCs (red) after 24 hours of co-culture. (B) MitoTracker staining (red) revealed active mitochondria in the nanotubular network (yellow arrows). (C) Time lapse pictures of the formation of a nanotube between a DiO-labeled cardiomyoblast (green) and DiD-labeled stem cell (red).

Figure 5

Cell fusion of a H9c2 cell and a mesenchymal stem cell. The time lapse pictures demonstrate steps of the fusion of a H9c2 cell (green) and a mesenchymal stem cell (red). Fused cells with double nuclei exhibit a combined yellow staining.

Figure 6

Co-culture of H9c2 and stem cells in normal conditions. (A) Cardiomyoblasts (green) and MSCs (red) after one day co-cultivation. (B) A representative double labeled cell with double nuclei (nuclei were stained with Hoechst). (C) DiO-labeled cardiomyoblasts (green) and DiD-labeled MSCs (red) analyzed with flow cytometry after one day co-cultivation. We found three different cell populations: 59.42% of DiO-labeled H9c2 cells, 30.1% of DiD-labeled MSCs and 8.14% of double labeled cells. (D) The distribution according to the forward scatter plot demonstrates that several double labeled yellow cells are mostly the same size as the green H9c2 cells or the red MSCs disapproving complete cell fusion.