Cardiosphere-derived cells (CDCs), a naturally heterogeneous mixture of cell sub-populations, were grown from percutaneous endomyocardial adult human biopsy specimens (n=6). c-Kit
+
and CD90
+
CDCs were selected using magnetic-activated cell sorting with excellent purity as determined by flow cytometry. Immunostaining revealed that ~30% of c-Kit
+
CDCs expressed Nkx2.5, ~100% of CD90
+
CDCs expressed procollagen type I, and ~100% of both sub-populations expressed CD105. When placed in co-culture with neonatal myocytes and fibroblasts, c-Kit
+
CDCs expressed cardiac troponin I, while CD90
+
CDCs expressed vimentin. In order to assess the therapeutic potential of purified CDCs, acute myocardial infarcts (MIs) were created in immunodeficient mice and c-Kit
+
(n=16), CD90
+
(n=14), or CD105
+
(n=3) CDCs were injected into the border zone. Echocardiograms were performed 3 weeks post-MI to measure left ventricular ejection fraction (LVEF). CD105-injected mice were comparable to an historical control group of mixed CDC-injected mice (LVEF = 41.3±2.9% CD105 vs. 42.8±10.4% CDC [n=11], p=0.60), indicating that the sorting process did not itself impair the therapeutic potential of CDCs. c-Kit- and CD90-injected mice were indistinguishable from one another (LVEF=31.7±8.2% c-Kit vs. 32.1±11.8% CD90, p=0.92), and both groups were significantly outperformed by the CD105-injected mice (p=0.01 and p=0.03, respectively). All groups were then compared to two other historical control groups, mice treated with normal human dermal fibroblasts (NHDFs [n=7]) and mice treated with phosphate-buffered saline (PBS [n=11]). c-Kit-injected mice did significantly outperform both NHDF- (p=0.04) and PBS-injected mice (p=0.03), while more variability in the CD90-injected group resulted in nearly significant comparisons with the NHDF (p=0.08) and PBS groups (p=0.08). While the therapeutic mechanisms of action of these two distinct sub-populations are undoubtedly different, both offer similar global functional benefits in the setting of acute MI. We conclude that the spontaneously-emerging unselected CDC population serves as a therapeutic cell cocktail, and that no functional advantage is conferred by the extra step of sorting for c-Kit
+
or CD90
+
sub-populations.