Genome-Wide CRISPR Screen Reveals Dependency of PPM1D-mutant Cells on DNA Repair

PPM1D (Protein Phosphatase Mg2+/Mn2+ Dependent 1D) is a protein phosphatase frequently mutated in clonal hematopoiesis (CH), therapy-related myeloid neoplasms (t-MN), and solid tumors. PPM1D is best known for its role as a negative regulator of the DNA damage response. Mutations in the terminal exon of PPM1D are gain-of-function and lead to a truncated, dominant form of PPM1D with preserved phosphatase activity. This stabilized, mutant protein leads to persistent dephosphorylation of PPM1D-targets including p53 and other key DNA damage response proteins. Our lab previously demonstrated that mutant Ppm1d hematopoietic stem and progenitor cells (HSPCs) outcompete their wild type (WT) counterparts in vivo following bone marrow transplantation and repeated cisplatin exposure. Given the prevalence of PPM1D mutations in CH and cancer, PPM1D is an attractive therapeutic target and ongoing efforts are underway to develop a small molecular inhibitor.

We have sought to identify genetic vulnerabilities of PPM1D mutant cells by performing a whole-genome CRISPR/Cas9 dropout screen in PPM1D WT and mutant leukemia cell lines. Analysis of gene candidates that were essential for mutant cell survival, but not WT, showed an enrichment for genes involved in DNA repair including the Fanconi Anemia (FA) pathway genes. Given the role of PPM1D in the inhibition of p53 and the DNA damage response, we hypothesized that mutant PPM1D cells have increased genomic instability and therefore, rely on the non-canonical FA DNA repair pathway for survival.

Indeed, we have found that PPM1D mutant cells harbored more single- and double-stranded DNA breaks as well as chromosomal aberrations both without external stressors and after cisplatin treatment compared to their WT counterparts. We have also found that PPM1D mutant cells exhibit significantly more FANCD2 foci at baseline compared to WT cells, suggesting increased activation of the FA pathway. In addition to interstrand crosslink repair, the FA pathway also plays a critical role in replication fork protection during replication stress to prevent the accumulation of toxic double-stranded breaks. To assess replication stress and kinetics in WT and PPM1D mutant cells, we performed DNA fiber assays to visualize nascent DNA synthesis at the single-molecule resolution. Surprisingly, in comparison to WT cells, PPM1D mutant cells had increased replication fork speed. These results demonstrate that PPM1D mutant cells have more genome instability and activate the FA pathway for DNA repair and replication progression.

Increased DNA damage often results in increased mutagenesis and accumulation of single-base substitutions. To validate our cell line findings and determine if Ppm1d-mutant HSPCs harbor more mutations, we performed whole-genome sequencing on WT and Ppm1d HSPCs from single-cell derived methocult colonies. Surprisingly, we did not find a significant increase in the overall mutation burden in the Ppm1d-mutant colonies compared to WT. This suggests that while Ppm1d-mutant cells have more genomic instability, the fidelity of repair is maintained. This may be due to a delay in the recruitment of DNA repair factors. Overall, our work identifies genetic candidates that give rise to opportunities for synthetic lethal approaches to target PPM1D-mutant clonal expansion in patients with PPM1D-mutated malignancies and those at high-risk for developing t-MN.

Takahashi:GSK: Consultancy; Symbio Pharmaceuticals: Consultancy; Ostuka Pharmaceuticals: Honoraria; Agios: Consultancy; Celgene/BMS: Consultancy; Novartis: Consultancy; Mission Bio: Honoraria; Illumina: Honoraria. Vassiliou:AstraZeneca: Research Funding; STRM.BIO: Consultancy.