CRISPR-dependent base editing screens identify separation of function mutants of RADX with altered RAD51 regulatory activity

Background

The recombinase RAD51 is an essential factor in maintaining genome stability by orchestrating the repair of DNA double-strand breaks (DSBs) by homologous recombination (HR) (Chakraborty et al., 2023; Kowalczykowski, 2015). Furthermore, RAD51 plays an essential role in maintaining the genome during replication and replicative stress. It promotes replication fork reversal at stalled replication forks and protects the fork ends from degradation (Bhat & Cortez, 2018). Moreover, RAD51 has been found to regulate replication fork restart, making it a central element in the cellular response to replicative stress (Mason et al., 2019).

As such, factors that regulate the binding of RAD51 to damaged replication forks or influence the activity of RAD51 are hypothesized to play a crucial role in RAD51-mediated genome integrity in response to replicative stress. For example, Breast Cancer Type 2 Susceptibility Protein (BRCA2) has been deemed an imported regulator that stimulates RAD51 nucleofilament formation at stalled replication forks, fork reversal and fork end protection (Chakraborty et al., 2023). In addition, Cortez and colleagues reported that the DNA single-strand binding protein RADX is a major player in the regulation of RAD51 at replication forks. Indeed, RADX depletion causes an aberrant increase in RAD51 activity, which reduces replication processivity and leads to aberrant DSB formation (Dungrawala et al., 2017). Conversely, RADX overexpression leads to nuclease-dependent fork degradation. It was discovered that RADX competes with RAD51 and limits the level of RAD51 at damaged replication forks. Ultimately, RADX regulates fork reversal and genome maintenance during replication and in response to replicative stress (Bhat et al., 2018).

In this preprint, which follows up on a previous study showing that RADX binds to RAD51 and destabilized its nucleofilament form (Adolph et al., 2021), the Cortez Lab uses a CRISPR-dependent base editing approach to introduce mutations throughout the RADX gene and, thereby, characterize the role of RADX in genome stability (Madison B. Adolph et al., 2023). They identified motifs necessary for RADX function and uncovered that specific RADX mutants are still able to bind RAD51 but hinder its ATP hydrolysis activity.

Key findings

Cellular context dictates the impact of RADX inactivation

The authors were first of all interested in determining whether RADX inactivation results in a cell growth disadvantage. They implemented a two-color growth competition assay using an RPE-1 (Human Retinal Pigment Epithelial-1) immortalized cell line. RPE-1 wild-type (WT) cells either expressed GFP or mCherry transfected with siRNA targeting RADX gene (siRADX) or a non-targeting control siRNA (siNT). The results showed that while RADX depletion does not affect cell growth in the absence of P53, it is strongly affected in the presence of P53.

Previous studies demonstrated that RADX regulates the accumulation of RAD51 at replication forks, although the mechanisms behind how RADX promotes or inhibits RAD51 at these sites is highly dependent on the cellular context and remains unclear. Remarkably, depletion of RAD51 resulted in hypersensitivity to replication stress induced by hydroxyurea (HU). The inactivation of RADX, combined with RAD51 depletion (siRAD51) or its inhibition (B02), suppressed the hypersensitivity to replication stress induced by HU.

Deleterious RADX mutants identified using a CRISPR base editor screen

Madison Adolph and colleagues used a CRISPR base editor screen to introduce mutations via single-guide RNAs (sgRNAs) throughout the RADX gene and identified mutants conferring either a loss- or gain-of function. This experiment was performed using both RPE-1 cells proficient (WT) and deficient (p53-/-) for p53 as well as cells treated with HU and B02. The results were as follows: (1) sgRNAs were depleted in WT RPE-1 cells and those same sgRNAs were enriched in p53-/- RPE cells, (2) certain sgRNAs were enriched in both conditions at the RADX N-terminus and (3) sgRNAs targeting the amino acid residues 678-750 were depleted in both conditions.

The authors chose to focus on the mutations introduced in residues 678-750 because they were deleterious for cell growth. Using Alphafold, these mutations were predicted to be on the protein’s surface between the third and fourth oligosaccharide/oligonucleotide binding domains (OB), a previously uncharacterized RADX region (Fig 2A). The use of predictive tools to check if the mutations may be deleterious identified that variants S705N and E713K may be disruptive. Additionally, the variants E713K, D741N and K743Q have been implicated in TCGA (The Cancer Genome Atlas) studies.

The residues identified in the oligosaccharide/oligonucleotide binding domains (OB) are of physiological importance

To assess the physiological importance of the RADX residues 678-740, the authors determined the impact of their mutations – referred to as OB3-4 interdomain mutants or 3,4-ID mutants – on replication and in response to DNA damage. While these mutations did not affect the localization of RADX to replication forks, they considerably attenuated the elongation of replication forks and generated DNA damage during S phase. Furthermore, the expression of 3,4-ID RADX mutants in RADX deficient cells resulted in increased hypersensitivity to DNA damaging agents and replication stress.

The mutations identified lead to aberrant and toxic levels of RAD51 at the replication forks

Previous results from the Cortez lab and the analyses described above led the authors to investigate whether the phenotypes described earlier could result from a defect in the regulation of RAD51 during replication. While mutations did not alter the ability of RADX to bind single-stranded DNA, oligomerize and interact with RAD51, the authors observed that the effect of RADX mutations (3,4-ID) on the reduction of replication fork progression is mediated by RAD51. These results strongly suggest that the RADX mutants no longer regulate RAD51 levels at replication forks. Indeed, the authors observed aberrant levels of RAD51 at replication forks in cells expressing these mutants in an extent similar to those observed in RADX-deficient cells.

There is a defect in RAD51 ATPase activity in cells expressing 3,4-ID RADX mutants

In addition, and in a remarkable way, the authors found that the ATPase activity of RAD51 is impaired in cells expressing 3,4-ID RADX mutants (Fig 2B). Thereby, the authors revealed that these RADX mutants fail to stimulate RAD51 ATPse activity, which is crucial in the regulation of RAD51 at replication forks, for fork progression and in response to replicative stress.

Why this work is important?

Genomic instability is at the forefront of cancer initiation and progression. It has become paramount in cancer research to understand the mechanisms leading to defects in genome maintenance. RADX, the key protein in this study, modulates RAD51, a major player in DNA repair whose deregulation may pave the way towards tumorigenicity. Nevertheless, the mechanisms by which RADX promotes or inhibits RAD51 at fork sites remains unknown making this study by the Cortez Lab all the more important. The use of a CRISPR base editor screen to introduce mutations throughout the RADX gene is innovative and has allowed the researchers to pinpoint mutations in previously uncharted RADX regions. Further experiments using these mutants demonstrated that RADX stimulates RAD51 ATP hydrolysis, crucial for DNA replication and replicative stress responses. This surely adds another piece to the puzzle!

Questions for the authors

1: Do you expect the role of the OB3-4 interdomain to be conserved across species?

2: Do you envisage that the mutations you have identified in this study could be used as markers to predict the response of certain patients to the genotoxic agents used in cancer therapy?

3: RAD51 expression levels are altered in certain cancers. Do you think that expression levels of RADX (and by extension BRCA2) may also be a crucial factor in the regulation of RAD51 at replication forks?

Sources

[1] Adolph, M. B., Mohamed, T. M., Balakrishnan, S., Xue, C., Morati, F., Modesti, M., Greene, E. C., Chazin, W. J., & Cortez, D. (2021). RADX controls RAD51 filament dynamics to regulate replication fork stability. Molecular Cell, 81(5), 1074-1083.e5. https://doi.org/10.1016/j.molcel.2020.12.036

[2] Bhat, K. P., & Cortez, D. (2018). RPA and RAD51: fork reversal, fork protection, and genome stability. Nature Structural & Molecular Biology, 25(6), 446–453. https://doi.org/10.1038/s41594-018-0075-z

[3] Bhat, K. P., Krishnamoorthy, A., Dungrawala, H., Garcin, E. B., Modesti, M., & Cortez, D. (2018). RADX Modulates RAD51 Activity to Control Replication Fork Protection. Cell Reports, 24(3), 538–545. https://doi.org/10.1016/j.celrep.2018.06.061

[4] Chakraborty, S., Schirmeisen, K., & Lambert, S. A. (2023). The multifaceted functions of homologous recombination in dealing with replication-associated DNA damages. DNA Repair, 129, 103548. https://doi.org/10.1016/j.dnarep.2023.103548

[5] Dungrawala, H., Bhat, K. P., Le Meur, R., Chazin, W. J., Ding, X., Sharan, S. K., Wessel, S. R., Sathe, A. A., Zhao, R., & Cortez, D. (2017). RADX Promotes Genome Stability and Modulates Chemosensitivity by Regulating RAD51 at Replication Forks. Molecular Cell, 67(3), 374-386.e5. https://doi.org/10.1016/j.molcel.2017.06.023

[6] Kowalczykowski, S. C. (2015). An Overview of the Molecular Mechanisms of Recombinational DNA Repair. Cold Spring Harbor Perspectives in Biology, 7(11), a016410. https://doi.org/10.1101/cshperspect.a016410

[7] Madison B. Adolph, Atharv S. Garje, Swati Balakrishnan, Florian Morati, Mauro Modesti, Walter J. Chazin, & David Cortez. (2023). CRISPR-dependent base editing screens identify separation of function mutants of RADX with altered RAD51 regulatory activity. BioRxiv.

[8] Mason, J. M., Chan, Y.-L., Weichselbaum, R. W., & Bishop, D. K. (2019). Non-enzymatic roles of human RAD51 at stalled replication forks. Nature Communications, 10(1), 4410. https://doi.org/10.1038/s41467-019-12297-0

 

doi: https://doi.org/10.1242/prelights.35502

Read preprint (1 votes)