Mitochondrial Russian doll genes may explain some discrepancies in links between mtDNA mutations and mitochondrial diseases
Sophie Breton
Abstract
Mitochondrial DNA mutations are responsible for a variety of human disorders due to defects in oxidative energy metabolism, and they are also involved in aging, aging-related diseases and multiple types of cancer.[1, 2] The variety of pathologies caused by mtDNA mutations is remarkable, but the relationship between genotype and disease phenotype is not always straightforward, most probably due to the complexity of mitochondrial biology. Indeed, because mtDNA is present in multiple copies within each cell, when mutated and wild-type mtDNA molecules coexist in a given tissue (heteroplasmic condition), the mutated mtDNA has to reach a certain copy number to cause mitochondrial dysfunction (threshold effect). The threshold effect varies among mutation types (e.g., in protein-coding genes or mt-tRNAs) and also among tissues because of their different dependence on oxidative metabolism. The heterogeneity of phenotypes can be explained by differences in mutational load reaching the pathogenic threshold in some tissues but not in others.[1, 2] Nevertheless, several discrepancies in links between mtDNA mutations and disease phenotypes remain, such as mutations within genes that cannot be directly linked to effects on oxidative energy metabolism or mitochondrial protein synthesis. In this issue of BioEssays, Pozzi and Dowling[3] propose a very elegant hypothesis to explain some of these discrepancies between theoretical expectation and experimental observation when studying mitochondrial diseases linked to mutations in the mtDNA. The “mitochondrial interference” hypothesis stipulates that certain mitochondrial disease phenotypes could be due to mutations in small RNAs encoded within known mitochondrial genes that would disturb the dialogue between mitochondria and nucleus, thus leading to disease. There is growing evidence for the widespread presence of functional small mitochondrial RNAs in human and other species, but their involvement in mitochondrial disorders and the underlying mechanisms have not yet been studied. The authors provide strong support for their hypothesis, which is congruent with previous literature that has reported tissue-specific pathological effects of mtDNA mutations. Indeed, the authors explain that small mtRNAs could affect nuclear mRNA targets that have tissue-specific expression and thus could be pathogenic only in specific tissues. They propose that mutations in small mtRNAs could not only reduce their affinity with their typical targets but also enhance it for other new targets, thus providing an additional mechanism to explain the complexity of biochemical effects and clinical symptoms linked to mtDNA mutations.[3] It is true that current studies of mitochondrial diseases linked to mtDNA mutations focus mostly on effects of sequence mutations on oxidative metabolism, and that mitochondrial interference provides a new way to study mitochondrial diseases in a context independent of OXPHOS functionality. It is also true that molecular screening of mitochondrial disorders has been usually restricted to common harmful mutations and deletions of mtDNA, that is, without taking into account rare mutations, and perhaps never synonymous mutations in protein-coding genes.[4] A situation that is not discussed by Pozzi and Dowling is that small mtRNAs lying within mitochondrial protein-coding genes could also explain why some mitochondrial disorders and mitochondria-related cancer are associated with synonymous mutations,[1] which are habitually considered functionally neutral. This also raises a question: Why would such a convenient “mitochondrial Russian doll genes” hypothesis be limited to small mtRNAs? Small open reading frames (smORFs) of functional importance encoded within typical human mitochondrial genes have also been recently reported.[5, 6] These smORFs encode for micropeptides that have key roles in the retrograde signaling network and cellular metabolism, and some of them could possibly be found within protein-coding genes in alternative reading frames.[7] Together with the mitochondrial interference mechanism, the existence of an alternative mitochondrial proteome could well explain some of the discrepancies discussed here and by Pozzi and Dowling. These recent discoveries in mitochondrial research clearly deserve attention. The author declares no conflict of interest. Data sharing not applicable – no new data generated.