Sexually Antagonistic Mitonuclear Coevolution in Duplicate Oxidative Phosphorylation Genes

Mitochondrial function is critical in eukaryotes. To maintain an adequate supply of energy, precise interactions must be maintained between nuclear- and mitochondrial-encoded gene products. Such interactions are paramount in chimeric enzymes such as the oxidative phosphorylation (OXPHOS) complexes. Mutualistic coevolution between the two genomes has therefore been suggested to be a critical, ubiquitous feature of eukaryotes that acts to maintain cellular function. However, mitochondrial genomes can also act selfishly and increase their own transmission at the expense of organismal function. For example, male-harming mutations are predisposed to accumulate in mitochondrial genomes due to their maternal inheritance ("mother's curse"). Here, we investigate sexually antagonistic mitonuclear coevolution in nuclear-encoded OXPHOS paralogs from mammals and Drosophila. These duplicate genes are highly divergent but must interact with the same set of mitochondrial-encoded genes. Many such paralogs show testis-specific expression, prompting previous hypotheses suggesting they may have evolved under selection to counteract male-harming mitochondrial mutations. We found increased rates of evolution in OXPHOS paralogs with testis-specific expression in mammals and Drosophila, supporting this hypothesis. However, further analyses suggested such patterns may be due to relaxed, not positive selection, especially in Drosophila. Structural data also suggest that mitonuclear interactions do not play a major role in the evolution of many OXPHOS paralogs in a consistent way. In conclusion, no single OXPHOS paralog met all our criteria for being under selection to counteract male-harming mitochondrial mutations. We discuss alternative explanations for the drastic patterns of evolution in these genes, including mutualistic mitonuclear coevolution, adaptive subfunctionalization after gene duplication, and relaxed selection on OXPHOS in male tissues.