The RTR Complex Partner RMI2 and the DNA Helicase RTEL1 Are Both Independently Involved in Preserving the Stability of 45S rDNA Repeats in Arabidopsis thaliana
Sarah Röhrig, Susan Schröpfer, Alexander Knoll, Holger Puchta
The stability of repetitive sequences in complex eukaryotic genomes is safeguarded by factors
suppressing homologues recombination. Prominent in this is the role of the RTR complex.
In plants, it consists of the RecQ helicase RECQ4A, the topoisomerase TOP3α and
RMI1. Like mammals, but not yeast, plants harbor an additional complex partner, RMI2.
Here, we demonstrate that, in Arabidopsis thaliana, RMI2 is involved in the repair of aberrant
replication intermediates in root meristems as well as in intrastrand crosslink repair. In
both instances, RMI2 is involved independently of the DNA helicase RTEL1. Surprisingly,
simultaneous loss of RMI2 and RTEL1 leads to loss of male fertility. As both the RTR complex
and RTEL1 are involved in suppression of homologous recombination (HR), we tested
the efficiency of HR in the double mutant rmi2-2 rtel1-1 and found a synergistic enhancement
(80-fold). Searching for natural target sequences we found that RTEL1 is required for
stabilizing 45S rDNA repeats. In the double mutant with rmi2-2 the number of 45S rDNA
repeats is further decreased sustaining independent roles of both factors in this process.
Thus, loss of suppression of HR does not only lead to a destabilization of rDNA repeats but
might be especially deleterious for tissues undergoing multiple cell divisions such as the
Repair of adjacent single-strand breaks is often accompanied by the formation of tandem sequence duplications in plant genomes
Simon Schiml, Friedrich Fauser and Holger Puchta
Duplication of existing sequences is a major mechanism of genome evolution. It has been previously shown that duplications can occur by replication slippage, unequal sister chromatid exchange, homologous recombination, and aberrant double-strand break-induced synthesis-dependent strand annealing reactions. In a recent study, the abundant presence of short direct repeats was documented by comparative bioinformatics analysis of different rice genomes, and the hypothesis was put forward that such duplications might arise due to the concerted repair of adjacent single-strand breaks (SSBs). Applying the CRISPR/Cas9 technology, we were able to test this hypothesis experimentally in the model plant Arabidopsis thaliana. Using a Cas9 nickase to induce adjacent genomic SSBs in different regions of the genome (genic, intergenic, and heterochromatic) and at different distances (∼20, 50, and 100 bps), we analyzed the repair outcomes by deep sequencing. In addition to deletions, we regularly detected the formation of direct repeats close to the break sites, independent of the genomic context. The formation of these duplications as well as deletions may be associated with the presence of microhomologies. Most interestingly, we found that even the induction of two SSBs on the same DNA strand can cause genome alterations, albeit at a much lower level. Because such a scenario reflects a natural step during nucleotide excision repair, and given that the germline is set aside only late during development in plants, the repair of adjacent SSBs indeed seems to have an important influence on the shaping of plant genomes during evolution.
AtRAD5A is a DNA translocase harboring a HIRAN domain which confers binding to branched DNA structures and is required for DNA repair in vivo
Daniela Kobbe, Andy Kahles, Maria Walter, Tobias Klemm, Anja Mannuss, Alexander Knoll, Manfred Focke and Holger Puchta
DNA lesions such as crosslinks represent obstacles for the replication machinery. Nonetheless, replication can proceed via the DNA damage tolerance pathway also known as postreplicative repair pathway. SNF2 ATPase Rad5 homologs, such as RAD5A of the model plant Arabidopsis thaliana, are important for the error-free mode of this pathway. We able to demonstrate before, that RAD5A is a key factor in the repair of DNA crosslinks in Arabidopsis. Here, we show by in vitro analysis that AtRAD5A protein is a DNA translocase able to catalyse fork regression. Interestingly, replication forks with a gap in the leading strand are processed best, in line with its suggested function. Furthermore AtRAD5A catalyses branch migration of a Holliday junction and is furthermore not impaired by the DNA binding of a model protein, which is indicative of its ability to displace other proteins. Rad5 homologs possess HIRAN (Hip116, Rad5; N-terminal) domains. By biochemical analysis we were able to demonstrate that the HIRAN domain variant from Arabidopsis RAD5A mediates structure selective DNA binding without the necessity for a free 3′OH group as has been shown to be required for binding of HIRAN domains in a mammalian RAD5 homolog. The biological importance of the HIRAN domain in AtRAD5A is demonstrated by our result that it is required for its function in DNA crosslink repair in vivo.