Mitochondrial gene defects in Arabidopsis can broadly affect mitochondrial gene expression through copy number.

How mitochondria regulate the expression of their genes is poorly understood, partly because methods have not been developed for stably transforming mitochondrial genomes. In recent years, the disruption of mitochondrial genes has been achieved in several plant species using mitochondria-localized TALEN (mitoTALEN). In this study, we attempted to disrupt the NADH dehydrogenase subunit7 (NAD7) gene, a subunit of respiratory chain complex I, in Arabidopsis (Arabidopsis thaliana) using the mitoTALEN method. In some of the transformants, disruption of NAD7 was accompanied by severe growth inhibition and lethality, suggesting that NAD7 has an essential function in Arabidopsis. In addition, the mitochondrial genome copy number and overall expression of genes encoding mitochondrial proteins were generally increased by nad7 knockout. Similar increases were also observed in mutants with decreased NAD7 transcripts and with dysfunctions of other mitochondrial respiratory complexes. In these mutants, the expression of nuclear genes involved in mitochondrial translation or protein transport was induced in sync with mitochondrial genes. Mitochondrial genome copy number was also partly regulated by the nuclear stress-responsive factors NAC domain containing protein 17 and Radical cell death 1. These findings suggest the existence of overall gene-expression control through mitochondrial genome copy number in Arabidopsis and that disruption of single mitochondrial genes can have additional broad consequences in both the nuclear and mitochondrial genomes.

Targeted gene disruption of ATP synthases 6-1 and 6-2 in the mitochondrial genome of Arabidopsis thaliana by mitoTALENs.

We recently achieved targeted disruptions of cytoplasmic male sterility (CMS)-associated genes in the mitochondrial genomes of rice and rapeseed by using mitochondria-targeted transcription activator-like effector nucleases (mitoTALENs). It was the first report of stable and heritable targeted gene modification of plant mitochondrial genomes. Here, we attempted to use mitoTALENs to disrupt two mitochondrial genes in the model plant Arabidopsis thaliana(Arabidopsis) using three different promoters and two types of TALENs. The targets were the two isoforms of the ATP synthase subunit 6 gene, atp6-1 and atp6-2. Each of these genes was successfully deleted and the mitochondrial genomes were recovered in a homoplasmic state. The nuclear genome also has a copy of atp6-1, and we were able to confirm that it was the mitochondrial gene and not the nuclear pseudogene that was knocked out. Among the three mitoTALEN promoters tried, the RPS5A promoter was the most effective. Conventional mitoTALENs were more effective than single-molecule mito-compactTALENs. Targeted mitochondrial gene deletion was achieved by crossing as well as by floral-dip transformation to introduce the mitoTALEN constructs into the nucleus. The gene disruptions were caused by large (kb-size) deletions. The ends of the remaining sequences were connected to distant loci, mostly by illegitimate homologous recombinations between repeats.

FMT, a protein that affects mitochondrial distribution, interacts with translation-related proteins in Arabidopsis thaliana.

KEY MESSAGE: Two translation-related proteins are identified as FMT-interacting proteins. However, FMT, unlike mutants of other CLU genes in fly and human, has no clear impact on the accumulation of mitochondrial proteins. Organelle distribution is critical for effective metabolism and stress response and is controlled by various environmental factors. Clustered mitochondria (CLU) superfamily genes affect mitochondrial distribution and their disruptions cause mitochondria to cluster within a cell in various species including yeast, fly, mammals and Arabidopsis. In Arabidopsis thaliana, Friendly mitochondria (FMT) is a CLU gene that is required for normal mitochondrial distribution, but its molecular function is unclear. Here, we demonstrate that FMT interacts with some translation-related proteins (translation initiation factor eIFiso4G1 and glutamyl-tRNA synthetase OVA9), as well as itself. We also show FMT forms dynamic particles in the cytosol that sometimes move with mitochondria, and their movements are mainly controlled by actin filaments but also by microtubules. Similar results have been reported for animal CLU orthologs. However, an fmt mutant, unlike animal clu mutants, did not show any clear decrease of nuclear-encoded mitochondrial protein levels. This difference may reflect a functional divergence of FMT from other CLU superfamily genes.