Analysis of mtDNA sequences from various eukaryotes, including single-celled protists that diverged from other eukaryotes early in evolution, provides strong support for the idea that the mitochondrion had a single origin. Mitochondria most likely arose from a bacterial symbiote whose closest contemporary relatives are in the Rickettsiaceae group. Bacteria in this group are obligate intracellular parasites. Thus the ancestor of the mitochondrion probably also had an intracellular lifestyle, which placed it in a good position to evolve into an intracellular symbiote. The mtDNA with the largest number of encoded genes so far found is from the protist species Reclinomonas americana. All other mitochondrial genomes contain a subset of the R. americana genes, which strongly implies that they evolved from a common ancestor shared with R. americana, losing different groups of mitochondrial genes by deletion or transfer to the nucleus, or both, over time.

In organisms whose mtDNA includes only a limited number of genes, the same set of mitochondrial genes is retained (see Figure 12-10, orange proteins), regardless of the phyla that include these organisms. One hypothesis for why these genes were never successfully transferred to the nuclear genome is that their encoded polypeptides are too hydrophobic to cross the outer mitochondrial membrane, and therefore would not be imported back into the mitochondria if they were synthesized in the cytosol. Similarly, the large size of rRNAs may interfere with their transport from the nucleus through the cytosol into mitochondria. Alternatively, these genes may not have been transferred to the nucleus during evolution because regulation of their expression in response to conditions within individual mitochondria may be advantageous. If these genes were located in the nucleus, conditions within each mitochondrion could not influence the expression of proteins found in that mitochondrion.