Statins, which reduce LDL-cholesterol by inhibition of 3-hydroxy-3-methylglutaryl-coenzyme A reductase, areamongthemostwidelyprescribeddrugs.Skeletalmyopathyis aknown statin-inducedadverseeffectassociatedwith mitochondrial changes. We hypothesized that similar effects would occur in cardiac myocytes in a lipophilicitydependent manner between 2 common statins: atorvastatin (lipophilic) and pravastatin (hydrophilic). Neonatal cardiac ventricular myocytes were treated with atorvastatin and pravastatin for 48 h. Both statins induced endoplasmic reticular (ER) stress, but only atorvastatin inhibited ERK1/2T202/Y204, AktSer473, and mammalian target of rapamycin signaling; reduced protein abundance of caveolin-1, dystrophin, epidermal growth factor receptor, and insulin receptor-b; decreased Ras homolog gene family memberA activation; and induced apoptosis. In cardiomyocyte-equivalent HL-1 cells, atorvastatin, but not pravastatin, reduced mitochondrial oxygen consumption. When male mice underwent atorvastatin and pravastatin administration per os for up to 7 mo, only long-term atorvastatin, but not pravastatin, induced elevated serum creatine kinase; swollen, misaligned, size-variable, and disconnected cardiac mitochondria; alteration of ER structure; repression of mitochondria-and endoplasmic reticulum-related genes; and a 21% increase in mortality in cardiac-specificvinculin-knockoutmiceduringthe first 2monthsof administration.Toourknowledge,we arethefirstto demonstrate in vivo that long-term atorvastatin administration alters cardiac ultrastructure, a finding with important clinical implications.
The authors thank Drs. Marilyn Farquhar (University of California, San Diego), Velia Fowler, and Malcom Wood (both from Scripps Research, La Jolla, CA, USA) for helpful discussions regarding the TEM analysis; Ying Jones (University of California, San Diego) for excellent technical assistance in preparation of tissues for electron microscopic analysis; Drs. Joan Heller Brown and Sunny Y. Xiang (both from the University of California, San Diego) for providing the GST plasmid; the Veterans Affairs/Veterans Medical Research Foundation (VA/VMRF) Microarray and Next Generation Sequencing (NGS) Core (San Diego, CA, USA) for microarray profiling; and Dr. Benedict Lucchesi (University of Michigan, Ann Arbor, MI, USA) for valuable scientific discussions and encouragement to pursue this project. The authors declare no conflicts of interest. This work was funded by the U.S. National Institutes of Health HL107200, HL091071 (National Heart, Lung, and Blood Institute), and AG052722 (National Institute on Aging) (to H.H.P.), and by American Heart Association Beginning Grant-in-Aid (BGIA) 2260359 (to A.E.Z.-H.).