A metabolomics investigation of selected m.3243A>G mutation phenotypes
Abstract
Mitochondrial disease (MD) is a subgroup of inborn errors of metabolism, which can be caused by a mutation in either the nuclear DNA (nDNA) or mitochondrial DNA (mtDNA). One of the most common mtDNA disease causing mutations is the m.3243A>G point mutation, which affects the incorporation of the amino acid leucine into mitochondrial proteins and thus the oxidative phosphorylation system (OXPHOS) system. This mutation was initially linked to mitochondrial myopathy, encephalopathy lactic acidosis and stroke like episodes (MELAS), but various other phenotypes and symptoms was later linked to this mutation, including progressive external ophthalmoplegia (PEO), maternally inherited diabetes-deafness (MIDD) and myopathy. However, the reason why these patients presents with such a broad spectrum of symptoms, even though they harbor the same mutation, remains unknown. Therefore, the aim of this study was to investigate the urine metabolome of a cohort of m.3243A>G diagnosed patients, presenting with different phenotypes (MELAS, MIDD and myopathy), using a multi-platform metabolomics approach. This multi-platform metabolomics approach consisted of untargeted as well as targeted analytical methods. The untargeted analyses consisted of gas chromatography−mass spectrometry (GCMS), nuclear magnetic resonances (NMR) spectroscopy, and liquid chromatography mass spectrometry with ion mobility (LC-IM-MS), in negative as well as positive ionization mode, while the targeted analyses consisted of liquid chromatography tandem-mass spectrometry (LC-MS/MS). Using this multi-platform metabolomics approach enabled us to analyze a larger portion of the metabolome compared to using a single analytical technique. In the first part of the study, we investigated 9 patients presenting specifically with MELAS and 29 healthy controls. We were able to identify 36 metabolites that were altered in the patient group when compared to healthy controls. When investigating these 36 metabolites further, we were able to link them to redox imbalance as a result of a defective OXPHOS system and stalled fatty acid oxidation. Our investigation also resulted in the first association between MELAS and an intricate web of affected pathways consisting of the one-carbon metabolism, methylation cycle and the transsulfuration pathway. In order to validate the 36 markers identified, a new cohort consisting of two MELAS patients and seven controls were used. We demonstrate complete separation of the MELAS patients and controls using principle component analysis, thus indicating that the 36 markers are not unique to the initial cohort used and thus have potential for diagnosis or treatment monitoring. In the second part of the study, we expanded on the findings by investigating two additional m.3243A>G associated phenotypes, MIDD (n = 30) and myopathy (n = 18). These two phenotypes, together with the MELAs cohort were compared to healthy controls, and to one another, in order to find not only metabolic similarities between the different phenotypes, but also phenotypic specific perturbations. Our novel findings indicate, especially in the MELAS patients, increased de novo fatty acid synthesis (FAS) in the mitochondria. We hypothesize that this increased FAS is probably due to lipoic acid synthesis, an essential cofactor for pyruvate dehydrogenase, 2-ketoglutarate dehydrogenase as well as the glycine cleavage system. Furthermore, we show specific metabolic perturbations in each of the three phenotypes. Investigating the metabolic similarities, we found three metabolites that were perturbed in all three phenotypes, 2-hydroxyglutaric acid, glycolic acid and 4-pentenoic acid. We conclude that these metabolites should be further investigated for diagnostic potential. The strength of our study was the utilization of different analytical platforms to generate the robust metabolomics data reported here. We show that urine may be a useful source for disease-specific metabolomics data. Our study contributed to the mitochondrial disease research field by providing significant insight into metabolic alterations caused by the m.3243A>G mutation. Firstly, results obtained in both parts of this study showcased the valuable information that could be obtained when implementing metabolomics as investigation tool. Secondly our results highlighted the potential for mitochondrial disease biosignatures for disease mechanistic understanding. Finally, we pointed out several important metabolic pathways affected in these patients that could be investigated in future studies. Ultimately, understanding the m.3243A>G mutation could lead to better diagnostic and treatment options, which both doctors and patients would benefit from immensely.