Maintaining genome stability is essential for cellular homeostasis. Genetic instability leads to cancer predisposition, accelerated aging and inflammatory and neurodegenerative pathologies. DNA double-strand breaks (DSB) are one of the most toxic lesions and can be generated by exogenous sources (exposure to radiation, chemical molecules, etc.) or endogenous (free radicals, disruption of genome replication, etc.). DSBs must necessarily be repaired to avoid cell death. Several repair mechanisms exist, some of which preserve genetic information and others generate modifications of the genetic heritage. The choice of the appropriate mechanism depending on the nature of the lesion, its location, or the state of the cell is decisive for the stability of the genome.
This selection is organized in two stages: 1- A first choice between the protection of DNA ends to promote their direct ligation by NHEJ (Non Homologous End Joining) and the resection, i.e. the partial degradation ends to generate single-stranded DNA (ssDNA) regions; 2- This ssDNA is a necessary intermediate for several repair systems: for mechanisms that are necessarily mutagenic such as the pairing of complementary single-stranded zones (Single Strand Annealing, SSA) or the ligation of DNA ends after resection (Alternative End Joining, A-EJ) , but also for Homologous Recombination (HR), which is the only post-resection mechanism conservative of the genetic heritage. The key HR protein, RAD51 (conserved in all living species) binds to ssDNA and scans the intact DNA in the genome for a sequence that will be used as a template to repair the DSB, thus maintaining the stability of the genome.
In this work, we investigated the mechanisms by which RAD51 prevents post-resection mutagenic repairs alternative to HR. For this we worked with reporter systems introduced in the genome of cultured human cells to measure repair by each of the existing pathways, NHEJ, SSA, A-EJ and RH. DSBs are introduced in a targeted manner into the genome by the expression of a meganuclease.
We observed that the depletion of RAD51 (or BRCA2, which allows the binding of RAD51 to DNA; BRCA2 is frequently mutated in hereditary breast or ovarian cancers) stimulates both SSA and A-EJ , but not NHEJ, validating the two-step model. We then used three different dominant negative forms of RAD51 (DN-RAD51) which repress HR and stimulate SSA and A-EJ. In living cells, these three DN-RAD51 do not bind efficiently to damaged chromatin and prevent the binding of endogenous RAD51 protein. These three DN-RAD51 are deficient for ATP hydrolysis, which led to the conclusion that the hydrolysis of ATP by RAD51 is necessary for its binding to damaged DNA, in vivo.
In parallel, we used a fourth DN-RAD51 which also inhibits HR but does not stimulate SSA and A-EJ. Unlike the first three DN-RAD51s, it binds efficiently to damaged chromatin. Therefore, the promotion of HR is separable from the inhibition of SSA and A-EJ mutagenic repair mechanisms and this inhibition is only due to the binding of RAD51 to DNA. Finally, we show that the binding of RAD51 to DNA does not regulate the efficiency of DSB resection, but prevents the pairing of complementary single-stranded DNA regions, which is a necessary step of SSA and of the A-EJ.
Figure legend: The roles of RAD51 in the protection of DNA double strand breaks against mutagenic repair . 1. NHEJ competes with resection on DSBs. 2. After resection, BRCA2 loads RAD51 onto ssDNA and this is dependant of ATP hydrolysis by RAD51. The occupancy of ssDNA by RAD51 triggers Homologous Recombination (in a catalytic manner) and suppresses the annealing step (independently of the promotion of Homologous recombination) of nonconservative SSA and A-EJ, which can be RAD52-dependent. The blockage of annealing by RAD51 does not require its strand exchange activity. Copyright © 2022, Oxford University Press
Our work revealed a new role of RAD51 in the selection of the DSB repair pathway. RAD51 preserves the integrity of the genome by protecting breaks from non-conservative repair mechanisms, this by binding to the ssDNA and independently of HR promotion.
This work was made possible thanks to the financial support of the Ligue Nationale contre le Cancer ("équipe labellisée"), INCa and ANR.
Learn more
PMID: 35137208