In this study, we isolated cardiac resident stem cells from L2G85 transgenic mice that constitutively expressed Fluc-eGFP, enabling us to track stem cell by both noninvasive imaging and invasive histopathology. Upon culturing, we demonstrated that these phase-bright cells upregulated Sca-1 and expressed surface markers resembling mesenchymal stem cells. We also present evidence that molecular imaging can be used to track CSCs survival in a murine model of myocardial infarction. When injected directly into the infarcted heart, these cells initially demonstrated viability by bioluminescence imaging. But after a period of eight weeks, the majority of the imaging signals were no longer present, suggesting elimination from the myocardium. Consequently, we found no significant physiologic benefit from CSC therapy as determined from multiple parameters, including fractional shortening by echocardiogram, myocardial glucose uptake by [18F]-FDG PET scan, ejection fraction by MRI, invasive PV loop analysis, and infarct size by histopathologic assessment.
Recent works have documented the presence of a reservoir of stem and progenitor cells in the myocardium (8
). These cells have been successfully isolated and expanded ex vivo,
and have been differentiated into cardiomyocytes, smooth muscle cells, and endothelial cells (10
). Previous works using CSCs have also demonstrated significant improvement in cardiac function after CSC injection, which was in part attributed to the persistent engraftment of the injected CSCs into the infarct zone (9
). Higher percentage of viable myocardium within the infarct zone and improved ejection fraction were demonstrated up to five weeks after cell transplantation (9
). In contrast, bioluminescence imaging data from our study suggests that by week 8, <0.5% of the transplanted CSCS are still alive. By echocardiography and PET imaging, no significant changes were observed at day 28 and day 56 in cardiac contractility and viability.
At present, we can only speculate that the discrepancy in findings may be attributed to differences
in isolation techniques (cardiosphere, side population), stem cell marker expression (c-Kit, Sca-1), donor cell sources (human, mice), particular strains of mice used, or host animal models (non-reperfused LAD ligation, ischemia-reperfusion) (34
). From the procedural standpoint, we can not exclude that the possibility that our mice may have been oversedated (2% isoflurane) given the lower baseline fractional shortening values (35–40%) observed compared to previous studies (35
). This could have undermined the chances of detecting a small but significant change between the two groups. From the transgenic model standpoint, it is possible that Fluc reporter gene silencing may have occurred and limited our ability to detect surviving CSCs at late time points. However, a previous study using hematopoietic stem cells isolated from these mice was able to demonstrate complete reconstitution of the irradiated bone marrow at late time points as detected by BLI, which argues against significant transgene silencing (17
). From the technical standpoint, BLI is limited to small animal imaging as the low energy photons (2–3 eV) can become attenuated within deeper tissues (e.g., heart) compared with more superficial locations (e.g., skeletal muscles). At present, a conservative estimate of minimal detectable cell number by BLI is ~1,000 in the heart vs. ~100 in the leg (14
). Finally, from the instrumentation standpoint, it is plausible that the image resolution (~1.5 mm3
) and detection sensitivity (10−9
molar) of the small animal microPET scanner is still insufficient to distinguish subtle but significant differences in [18
F]-FDG uptake between the two groups. Thus, additional studies will be needed in the future to further examine each of these variables separately.
Just as the mechanism of stem cells exerting benefit on myocardial function remains to be fully elucidated, the mechanism of their elimination over time also remains poorly understood. The pattern of acute donor cell death seen in our CSC transplantation is also consistent with previous studies showing poor donor cell survival after transplantation of neonatal cardiomyocytes (36
), mesenchymal stem cells (37
), bone marrow mononuclear cells (14
), and human embryonic stem cell-derived cardiomyocytes (38
). A number of culprits may be involved, including inflammation, ischemia, apoptosis, anoikis, and autophagy may all play contributory roles. Furthermore, the myocardial milieu in vivo
is a stark contrast to the rich nutrients and oxygen concentration that have been optimized for their cell culturing, growth, and expansions in vitro
. It is interesting to note historically that myoblast transplantation as a potential cell-based therapy for patients with Duchenne muscular dystrophy (DMD) met with disappointing failures in the 1990s (39
). One of the main reasons was acute donor cell death following transplantation. Future clinical studies aimed at transplanting CSCs in patients will likely need to first understand the mechanism of acute donor cell death.
Despite the limitation of acute donor stem cell death, both animal and clinical studies have suggested beneficial effects following transplantation of various stem cell types (40
), perhaps acting through paracrine pathways (41
). Thus, strategies aimed at prolonging cell survival (and hence prolonging paracrine activation) may lead to even more favorable results. To that extent, the application of bioengineering methods (42
) or pro-survival cocktails (43
) or genetic modification (44
), rather than direct stem cell injection, may prove to be a more viable approach for achieving long-term engraftment in the future. Encouragingly, recent study by Tillmanns et al.
showed that CSCs activated with insulin-like growth factor 1 and hepatocyte growth factor before their injection into infarcted sites had significantly improved cell survival rate and was able to form conductive arterioles and capillaries in the host myocardium (45
In summary, our study demonstrates that imaging can be used to monitor CSC fate noninvasively in living animals and that CSC transplantation provides no significant benefits to cardiac function in a mouse myocardial infarction model. As the field of cardiac stem cell therapy continues to mature, it is critical to better understand the underlying mechanism of the treatment modality, and to identify the potential pitfalls of the therapy prior to its incorporation into the clinical realm (1
). Although a stem cell source from the heart is of tremendous potential, cautious optimism is necessary prior to full clinical use of these cells.