Persistent damage induces mitochondrial DNA degradation

Considerable progress has been made recently toward understanding the processes of mitochondrial DNA (mtDNA) damage and repair. However, a paucity of information still exists regarding the physiological effects of persistent mtDNA damage. This is due, in part, to experimental difficulties associated with targeting mtDNA for damage, while sparing nuclear DNA. Here, we characterize two systems designed for targeted mtDNA damage based on the inducible (Tet-ON) mitochondrial expression of the bacterial enzyme, exonuclease III, and the human enzyme, uracil-N-glyosylase containing the Y147A mutation. In both systems, damage was accompanied by degradation of mtDNA, which was detectable by 6h after induction of mutant uracil-N-glycosylase and by 12h after induction of exoIII. Unexpectedly, increases in the steady-state levels of single-strand lesions, which led to degradation, were small in absolute terms indicating that both abasic sites and single-strand gaps may be poorly tolerated in mtDNA. mtDNA degradation was accompanied by the loss of expression of mtDNA-encoded COX2. After withdrawal of the inducer, recovery from mtDNA depletion occurred faster in the system expressing exonuclease III, but in both systems reduced mtDNA levels persisted longer than 144h after doxycycline withdrawal. mtDNA degradation was followed by reduction and loss of respiration, decreased membrane potential, reduced cell viability, reduced intrinsic reactive oxygen species production, slowed proliferation, and changes in mitochondrial morphology (fragmentation of the mitochondrial network, rounding and "foaming" of the mitochondria). The mutagenic effects of abasic sites in mtDNA were low, which indicates that damaged mtDNA molecules may be degraded if not rapidly repaired. This study establishes, for the first time, that mtDNA degradation can be a direct and immediate consequence of persistent mtDNA damage and that increased ROS production is not an invariant consequence of mtDNA damage.