Skeletal myoblasts form stable grafts in the heart and may improve myocardial performance after infarction. The current study compared the ability of different immortalized myoblast lines, or primary myoblast isolates, to form grafts in the normal or cryoinjured heart. With a constant dose of 6 x 10(6) cells, primary rat myoblasts and mouse C2C12 myoblasts formed similarly large skeletal muscle grafts in recipient rat hearts. Grafts often caused transmural replacement of the myocardium in normal hearts and almost completely replaced the damaged region in cryoinjured hearts. To test for possible artifacts due to immunosuppression and xenografting (mouse C2C12 cells into rat hearts), we grafted 1 x 10(6) mouse MM14 or C2C12 myoblasts into normal hearts of nude mice. Again, the C2C12 grafts caused transmural replacement of the left ventricular wall and distorted the epi- and endocardial contours. The MM14 cells, however, formed relatively small grafts. The C2C12 grafts showed substantially higher BrdU incorporation rates at day 4 compared with MM14 cells, suggesting that ongoing proliferation was responsible for the increased graft size. None of the three skeletal muscle cell types expressed detectable amounts of the gap junction protein connexin43 after grafting. The intercellular adhesion protein N-cadherin was not expressed in primary skeletal muscle grafts, but was spotty or abundant in C2C12- and MM14-derived grafts, respectively. The absence of connexin43 precluded electrical coupling between graft and host muscle cells. Thus, when sufficient amounts of proliferation occur after grafting, skeletal muscle cells can effectively replace the volume of lost myocardium. Excess proliferation, however, can cause grafts to expand the ventricular wall and possibly impair pump function. Optimal cardiac repair strategies may need to incorporate methods to control graft cell proliferation.