Introduction

BRCA2 is a tumor suppressor gene that functions in the homologous recombination (HR) pathway of DNA repair. Individuals with a germline loss-of-function (LOF) variant in BRCA2 are predisposed to several adult-onset cancers. Although not originally thought to have an impact on pediatric cancer, a large-scale study from our laboratory in pediatric rhabdomyosarcoma (RMS), as well as other recent studies have reported an enrichment of germline variants in BRCA2, and to a lesser extent, PALB2, amongst pediatric cancer patients. Limited data to date suggest that unlike adult tumors, tumors from these pediatric patients often retain a functional copy of BRCA2 or PALB2. Based on these findings, we have initiated a study to assess the impact of germline heterozygous BRCA2 and PALB2 LOF variants and their potential involvement in the development of pediatric RMS.

Methods

In vitro studies first used CRISPR to create LOF BRCA2 variants in an immortalized line of mesenchymal stem cells (iMSCs), the cell type of origin for RMS. We now are using Prime Editing to precisely create BRCA2 variants of interest. We will perform a battery of genome stability and DNA repair assays with engineered iMSCs to ascertain whether any of BRCA2’s typical functions are impacted by the induced mutations. We will also assess whether engineered iMSCs are susceptible to therapetuics including PARP or POLQ inhibitors. We have also generated a genetically engineered mouse model (GEMM) with a heterozygous LOF allele in Brca2, to cross into a mouse model of RMS to assess changes in the pace of tumorigenesis in the presence of heterozygous alleles.

Results

An effort from our laboratory revealed germline BRCA2 findings are statistically enriched in a large (n=615) pediatric RMS cohort (Li et al., JNCI, 2020). Subsequently, 609 paired tumor-normal RMS patient samples have undergone whole-genome and RNA sequencing, including 228 never-before sequenced samples, through the Gabriella Miller Kids First program. Germline data will identify whether there are new patients with BRCA2 or PALB2 germline variants. Somatic data will determine whether BRCA2 is expressed in the tumors of patients with heterozygous variants, and whether their tumor genomes possess hallmark signatures of HR deficiency.

            Initial efforts to engineer iMSCs with traditional CRISPR methods yielded multiple cell lines with both compound heterozygous and homozygous LOF mutations in BRCA2. Characterization of these cell lines demonstrated variability in both mRNA and protein expression, thus they may not be optimal for analysis. Thus, prime editing to create patient BRCA2 variants (p.D156Ter, p.E1035Ter, p.C1200Ter, p.V1283fs, p.S2378Ter, p.W2619Ter) is underway.

 Generation of a GEMM with a heterozygous removal of exon 11 in Brca2 has been successful. LOF of Brca2 is embryonic lethal in mice (Sharan, et al. Nature, 1997). Consistently, no homozygous mice were obtained following 7 Brca2 crosses (50 pups), a departure from the expected mendelian ratio of 25% (p=0.0001). This allele is now being crossed into two different mouse models of RMS: one with a conditional loss of p53 using a myoblast-specific (MyoD1-Cre) perturbation of TP53, and another with a gain-of-function oncogenic mutant p53, LSL-p53^R172H.

Conclusion

Ongoing experiments analyzing cell models will identify whether heterozygous LOF of BRCA2 and PALB2 confer a subtle, but detectable, effect on genomic instability. Further, our analyses will also determine whether these cells are susceptible to targeted therapies. GEMM testing will identify whether heterozygous variants in Brca2 affect RMS tumorigenesis or metastasis. Finally, analysis of the recently completed sequencing data will identify additional relevant variants, measure BRCA2 expression in tumors of interest, and determine whether BRCA2 mutant tumor genomes exhibit diminished HR function. These efforts will further elucidate what role these HR genes may play in increasing risk for development of pediatric cancers.