A new improved method that reproducibly measures small perturbations of mitochondrial DNA in populations of cells has been developed. It is based on first obtaining a cell count and then analyzing three aliquots of cells: one for total DNA per cell by fluorometry, one for total protein per cell and one for the amount of mitochondrial DNA per microgram of total cell DNA. To quantitate mitochondrial DNA, 0, 1, 2, and 3 nanograms of mouse mtDNA purified from a plasmid are added as internal standard DNA to four 1.0-microgram samples of purified total cell DNA containing an unknown amount of mitochondrial DNA (a sample set). Three sample sets are electrophoresed in an agarose gel devoid of ethidium bromide. Following Southern transfer to nitrocellulose and hybridization to purified 32P-labeled mouse mitochondrial DNA, an autoradiogram is prepared for use as a template to locate the mitochondrial DNA bands. These bands are cut out of the nitrocellulose filters, and their 32P-content is determined using a liquid scintillation counter. For each sample set, the counts per minute is plotted against the amount of mitochondrial DNA added. The plot is linear and the negative average of the values for the three intercepts on the x-axis yields the amount of nanograms of mitochondrial DNA per microgram of total cell DNA. The method is highly reproducible with a standard deviation of approximately 9 percent. The advantages of using this method over others that have been reported are discussed.
