Accurate estimation of 5-methylcytosine in mammalian mitochondrial DNA

Abstract

Abstract Whilst 5-methylcytosine (5mC) is a major epigenetic mark in the nuclear DNA in mammals, whether or not mitochondrial DNA (mtDNA) receives 5mC modification remains controversial. Herein, we exhaustively analysed mouse mtDNA using three methods that are based upon different principles for detecting 5mC. Next-generation bisulfite sequencing did not give any significant signatures of methylation in mtDNAs of liver, brain and embryonic stem cells (ESCs). Also, treatment with methylated cytosine-sensitive endonuclease McrBC resulted in no substantial decrease of mtDNA band intensities in Southern hybridisation. Furthermore, mass spectrometric nucleoside analyses of highly purified liver mtDNA preparations did not detect 5-methyldeoxycytidine at the levels found in the nuclear DNA but at a range of only 0.3–0.5% of deoxycytidine. Taken together, we propose that 5mC is not present at any specific region(s) of mtDNA and that levels of the methylated cytosine are fairly low, provided the modification occurs. It is thus unlikely that 5mC plays a universal role in mtDNA gene expression or mitochondrial metabolism.