RADA-dependent branch migration has a predominant role in plant mitochondria and its defect leads to mtDNA instability and cell cycle arrest

Author summaryIn flowering plants, the mitochondrial genome is very large and dynamic, and its stability influences plant fitness and development. Rearrangements by recombination drive its very rapid evolution and can lead to valuable agronomic traits such as cytoplasmic sterility, used by breeders for the production of hybrid seeds. Here we describe RADA, a DNA helicase essential for the stability of the mitochondrial DNA in Arabidopsis. We demonstrate that RADA has branch migrating activity, accelerating the processing of recombination intermediates. radA mutants are severely affected in development and fertility. They display mitochondrial genome instability that results in uncoordinated replication of subgenomes created by recombination. Furthermore, we found that an important component of the growth defects of radA mutants is apparently a cellular response triggered by the sensing of damages to the mitochondrial genome, resulting in the activation of genes that suppress the progression of the cell cycle. Our results underline the importance of better understanding the plant mitochondrial recombination pathways and their cross-talk with nuclear gene expression. Mitochondria of flowering plants have large genomes whose structure and segregation are modulated by recombination activities. The post-synaptic late steps of mitochondrial DNA (mtDNA) recombination are still poorly characterized. Here we show that RADA, a plant ortholog of bacterial RadA/Sms, is an organellar protein that drives the major branch-migration pathway of plant mitochondria. While RadA/Sms is dispensable in bacteria, RADA-deficient Arabidopsis plants are severely impacted in their development and fertility, correlating with increased mtDNA recombination across intermediate-size repeats and accumulation of recombination-generated mitochondrial subgenomes. The radA mutation is epistatic to recG1 that affects the additional branch migration activity. In contrast, the double mutation radA recA3 is lethal, underlining the importance of an alternative RECA3-dependent pathway. The physical interaction of RADA with RECA2 but not with RECA3 further indicated that RADA is required for the processing of recombination intermediates in the RECA2-depedent recombination pathway of plant mitochondria. Although RADA is dually targeted to mitochondria and chloroplasts we found little to no effects of the radA mutation on the stability of the plastidial genome. Finally, we found that the deficient maintenance of the mtDNA in radA apparently triggers a retrograde signal that activates nuclear genes repressing cell cycle progression.