Nuclear sensing of breaks in mitochondrial DNA enhances immune surveillance.

Mitochondrial DNA double-strand breaks (mtDSBs) are toxic lesions that compromise the integrity of mitochondrial DNA (mtDNA) and alter mitochondrial function1. Communication between mitochondria and the nucleus is essential to maintain cellular homeostasis; however, the nuclear response to mtDSBs remains unknown2. Here, using mitochondrial-targeted transcription activator-like effector nucleases (TALENs)1,3,4, we show that mtDSBs activate a type-I interferon response that involves the phosphorylation of STAT1 and activation of interferon-stimulated genes. After the formation of breaks in the mtDNA, herniation5 mediated by BAX and BAK releases mitochondrial RNA into the cytoplasm and triggers a RIG-I-MAVS-dependent immune response. We further investigated the effect of mtDSBs on interferon signalling after treatment with ionizing radiation and found a reduction in the activation of interferon-stimulated genes when cells that lack mtDNA are exposed to gamma irradiation. We also show that mtDNA breaks synergize with nuclear DNA damage to mount a robust cellular immune response. Taken together, we conclude that cytoplasmic accumulation of mitochondrial RNA is an intrinsic immune surveillance mechanism for cells to cope with mtDSBs, including breaks produced by genotoxic agents.

Transcription is a major driving force for plastid genome instability in Arabidopsis.

Though it is an essential process, transcription can be a source of genomic instability. For instance, it may generate RNA:DNA hybrids as the nascent transcript hybridizes with the complementary DNA template. These hybrids, called R-loops, act as a major cause of replication fork stalling and DNA breaks. In this study, we show that lowering transcription and R-loop levels in plastids of Arabidopsis thaliana reduces DNA rearrangements and mitigates plastid genome instability phenotypes. This effect can be observed on a genome-wide scale, as the loss of the plastid sigma transcription factor SIG6 prevents DNA rearrangements by favoring conservative repair in the presence of ciprofloxacin-induced DNA damage or in the absence of plastid genome maintenance actors such as WHY1/WHY3, RECA1 and POLIB. Additionally, resolving R-loops by the expression of a plastid-targeted exogenous RNAse H1 produces similar results. We also show that highly-transcribed genes are more susceptible to DNA rearrangements, as increased transcription of the psbD operon by SIG5 correlates with more locus-specific rearrangements. The effect of transcription is not specific to Sigma factors, as decreased global transcription levels by mutation of heat-stress-induced factor HSP21, mutation of nuclear-encoded polymerase RPOTp, or treatment with transcription-inhibitor rifampicin all prevent the formation of plastid genome rearrangements, especially under induced DNA damage conditions.

Cell cycle-dependent spatial segregation of telomerase from sites of DNA damage.

Telomerase can generate a novel telomere at DNA double-strand breaks (DSBs), an event called de novo telomere addition. How this activity is suppressed remains unclear. Combining single-molecule imaging and deep sequencing, we show that the budding yeast telomerase RNA (TLC1 RNA) is spatially segregated to the nucleolus and excluded from sites of DNA repair in a cell cycle-dependent manner. Although TLC1 RNA accumulates in the nucleoplasm in G1/S, Pif1 activity promotes TLC1 RNA localization in the nucleolus in G2/M. In the presence of DSBs, TLC1 RNA remains nucleolar in most G2/M cells but accumulates in the nucleoplasm and colocalizes with DSBs in rad52Δ cells, leading to de novo telomere additions. Nucleoplasmic accumulation of TLC1 RNA depends on Cdc13 localization at DSBs and on the SUMO ligase Siz1, which is required for de novo telomere addition in rad52Δ cells. This study reveals novel roles for Pif1, Rad52, and Siz1-dependent sumoylation in the spatial exclusion of telomerase from sites of DNA repair.

Short-range inversions: rethinking organelle genome stability: template switching events during DNA replication destabilize organelle genomes.

In the organelles of plants and mammals, recent evidence suggests that genomic instability stems in large part from template switching events taking place during DNA replication. Although more than one mechanism may be responsible for this, some similarities exist between the different proposed models. These can be separated into two main categories, depending on whether they involve a single-strand-switching or a reciprocal-strand-switching event. Single-strand-switching events lead to intermediates containing Y junctions, whereas reciprocal-strand-switching creates Holliday junctions. Common features in all the described models include replication stress, fork stalling and the presence of inverted repeats, but no single element appears to be required in all cases. We review the field, and examine the ideas that several mechanisms may take place in any given genome, and that the presence of palindromes or inverted repeats in certain regions may favor specific rearrangements.

Organelle DNA rearrangement mapping reveals U-turn-like inversions as a major source of genomic instability in Arabidopsis and humans.

Failure to maintain organelle genome stability has been linked to numerous phenotypes, including variegation and cytosolic male sterility (CMS) in plants, as well as cancer and neurodegenerative diseases in mammals. Here we describe a next-generation sequencing approach that precisely maps and characterizes organelle DNA rearrangements in a single genome-wide experiment. In addition to displaying global portraits of genomic instability, it surprisingly unveiled an abundance of short-range rearrangements in Arabidopsis thaliana and human organelles. Among these, short-range U-turn-like inversions reach 25% of total rearrangements in wild-type Arabidopsis plastids and 60% in human mitochondria. Furthermore, we show that replication stress correlates with the accumulation of this type of rearrangement, suggesting that U-turn-like rearrangements could be the outcome of a replication-dependent mechanism. We also show that U-turn-like rearrangements are mostly generated using microhomologies and are repressed in plastids by Whirly proteins WHY1 and WHY3. A synergistic interaction is also observed between the genes for the plastid DNA recombinase RECA1 and those encoding plastid Whirly proteins, and the triple mutant why1why3reca1 accumulates almost 60 times the WT levels of U-turn-like rearrangements. We thus propose that the process leading to U-turn-like rearrangements may constitute a RecA-independent mechanism to restart stalled forks. Our results reveal that short-range rearrangements, and especially U-turn-like rearrangements, are a major factor of genomic instability in organelles, and this raises the question of whether they could have been underestimated in diseases associated with mitochondrial dysfunction.

Structural and functional characterization of the phosphorylation-dependent interaction between PML and SUMO1.

PML and several other proteins localizing in PML-nuclear bodies (PML-NB) contain phosphoSIMs (SUMO-interacting motifs), and phosphorylation of this motif plays a key role in their interaction with SUMO family proteins. We examined the role that phosphorylation plays in the binding of the phosphoSIMs of PML and Daxx to SUMO1 at the atomic level. The crystal structures of SUMO1 bound to unphosphorylated and tetraphosphorylated PML-SIM peptides indicate that three phosphoserines directly contact specific positively charged residues of SUMO1. Surprisingly, the crystal structure of SUMO1 bound to a diphosphorylated Daxx-SIM peptide indicate that the hydrophobic residues of the phosphoSIM bind in a manner similar to that seen with PML, but important differences are observed when comparing the phosphorylated residues. Together, the results provide an atomic level description of how specific acetylation patterns within different SUMO family proteins can work together with phosphorylation of phosphoSIM's regions of target proteins to regulate binding specificity.