We report a novel mechanism of gene regulation in skeletal muscle fibers. Within an individual myofiber nucleus, not all muscle loci are transcriptionally active at a given time and loci are regulated independently. This phenomenon is particularly remarkable because the nuclei within a myofiber share a common cytoplasm. Both endogenous muscle-specific and housekeeping genes and transgenes are regulated in this manner. Therefore, despite the uniform protein composition of the contractile apparatus along the length of the fiber, the loci that encode this structure are not transcribed continuously. The total number of active loci for a particular gene is dynamic, changing during fetal development, regeneration, and in the adult, and potentially reflects the growth status of the fiber. The data reveal that transcription in particular stages of muscle fiber maturation occurs in pulses and is defined by a stochastic mechanism.

The anthracycline antibiotic doxorubicin produces a characteristic myopathy in cardiac muscle that limits its use in cancer therapy. We have shown in cultured neonatal rat cardiac muscle cells that doxorubicin treatment resulted in a rapid, selective decrease in the expression of muscle-specific genes, which preceded other changes characteristic of doxorubicin cardiomyopathy. Doxorubicin selectively and dramatically decreased the levels of mRNA for the sarcomeric genes, a-actin, troponin I, and myosin light chain 2, as well as the muscle-specific, but nonsarcomeric M isoform of creatine kinase. However, doxorubicin did not affect nonmuscle gene transcripts (pyruvate kinase, ferritin heavy chain, and 3-actin). Actinomycin D, an inhibitor of DNA-dependent RNA polymerase, did not show a similar selective decrease of muscle-specific mRNAs but, rather, produced a nonspecific, dose-dependent decrease of muscle and nonmuscle transcripts. The doxorubicin effect on muscle gene expression was limited to cardiac muscle; cultured skeletal myocytes were resistant to the effects of doxorubicin at 100-fold greater doses than those causing changes in mRNA levels in cardiac muscle cells. These effects of doxorubicin were reproduced in vivo; rats ih'ected with doxorubicin showed a dose-dependent decrease in the levels of mRNAs for a-actin, troponin I, myosin light chain 2, and M isoform of creatine kinase in cardiac but not skeletal muscle. These selective changes in gene expression in cardiocyte cultures and cardiac muscle precede classical ultrastructural changes and may explain the myofibrillar loss that characterizes doxorubicin cardiac injury.

Recent reports have demonstrated the presence of two isoforms of troponin I in the human fetal heart, namely, cardiac troponin I and slow skeletal muscle troponin I. Structural and physiological considerations indicate that these isoforms would confer differing contractile properties on the myocardium, particularly on the phosphorylation-mediated regulation of contractility by adrenergic agonists. We have investigated the developmental expression of these isoforms in the human heart from 9 weeks of gestation to 9 months of postnatal life, using Western blots revealed with troponin I antibodies to detect troponin protein isoforms and Northern blots to detect the corresponding mRNAs. The results show the following: 1) Slow skeletal muscle troponin I is the predominant isoform throughout fetal life. 2) After birth, the slow skeletal isoform is lost, with cardiac troponin I being the only isoform detectable by 9 months of postnatal development. 3) The protein isoforms and their corresponding mRNAs follow the same pattern of accumulation, suggesting that the transition in troponin expression is regulated at the level of gene transcription. The developmental transition in troponin I isoform content has implications for contractility of the fetal and postnatal myocardium. We further analyzed right and left ventricular muscle samples from 17 hearts in end-stage heart failure resulting from pulmonary hypertension, ischemic heart disease, or dilated cardiomyopathy. Cardiac troponin I mRNA remained abundant in each case, and slow skeletal muscle troponin I mRNA was not detectable in any of sample. We conclude that alterations in troponin I isoform content do not therefore contribute to the altered contractile characteristics of the adult failing ventricle.

Several reports have documented that thapsigargin is a potent inhibitor of the SR Ca2+ ATPase isolated from cardiac or skeletal muscle. We have characterized the specificity of this agent in intact rat cardiac myocytes using cells maintained in the whole cell voltage clamp configuration. We have shown that thapsigargin decreases the magnitude of the Ca2+ transient and the twitch by about 80% while it slows the decay rate for these responses. These changes were not accompanied by any alterations in sarcolemmal currents or in the trigger Ca2+ generated by the inward calcium current. Taken together these results reveal that the action of thapsigargin is restricted to the SR Ca2+ ATPase in intact cardiac myocytes. Furthermore, it is demonstrated unambiguously that SR intracellular Ca2+ stores are an absolute requirement for the development of contractile tension in rat heart myocytes. It is shown that thapsigargin is a valuable probe to examine the importance of SR pools of Ca2+ and the role of the Ca2+ ATPase in intact myocytes as well as in genetically altered heart cells.