Abstract
Abstract
In addition to the known cytogenetic heterogeneity of MDS, systematic application of new generation sequencing technologies (NGS) and SNP-arrays have further unraveled the complexity of MDS, revealing previously unknown somatic mutational patterns and chromosomal abnormalities. While many of the mutational events are secondary and acquired during disease progression, some may be ancestral in nature. Discovery of novel somatic defects may contribute to the understanding of molecular pathogenesis of MDS and lead to the introduction of new prognostic biomarkers or new therapeutic targets.
When we performed analysis of whole exome NGS in patients with MDS and other myeloid neoplasms, including 205 from our own MDS cohort as well as 201 primary AML cases from TCGA, we noted that the BRCC3 complex gene was recurrently mutated in 9 patients (2%). These somatic mutations were confirmed by both Sanger and targeted deep sequencing. UIMC1, FAM175A, BABAM1 and FAM175B gene mutations were each found in separate single patients. The most commonly affected gene was BRCC3, found in 5 patients (1%): 1 with primary AML, 2 with CMML and 2 with MDS. There were 3 canonical mutations at exon 4 (p.R81X) and 2 in exon 1 (p.Q7X). SNP-array analysis (N=682) showed deletion of the BRCC3 complex gene in 45 patients (7%). Deletion involving the UIMC1 locus (5q35.2) and the BRCC3 locus (Xq28) were found in 31 (4.5%) and 8 (1%) patients, respectively. Evaluation of deep sequencing results demonstrates that the variant allelic frequencies of BRCC3 mutations were more than 48% in the early MDS stage, suggesting that BRCC3 mutations are initial events.
The BRCC3 complex is located in the nucleus and participates in DNA double-strand break (DSB) repair. BRCC3 is a member of the JAMM/MPN+ family of zinc metalloproteases and specifically cleaves Lys-63 linked polyubiquitin chains. BRCC3 is a component of two complexes, the BRCA1-A complex and the BRISC complex. The BRCA1-A complex consists of UIMC1, FAM175A, BABAM1, BRE, BARD1, BRCC3 and BRCA1. DSB defects may be deleterious to genomic stability and instigate tumorigenesis.
Clinical analysis revealed that mutations and deletions of BRCC3 complex genes were more common in MDS than in pAML (P<.01). When 684 patients genotyped for these defects were analyzed, the presence of BRCC3 complex defects was associated with shorter survival (HR=2.44 95%CI; 1.73-3.34; P<.001). The presence of BRCC3 complex gene defects affected survival in all subgroups of MDS, MDS/MPN, sAML, and pAML when analyzed separately. Overall, analysis of the additional mutational spectrum of patients with BRCC3 complex defects showed that the most frequently mutated genes were KRAS/NRAS, TET2 and U2AF1, raising the possibility of synergistic leukemogenic effects of multiple mutations. However, multivariate analysis identifies a BRCC3 complex gene defect as an independent adverse prognostic factor (HR=2.3 95%CI; 1.63-3.14; P<.001).
For functional studies, we first generated a BRCC3 shRNA lentivirus vector and used it in in vitro immortalization assays utilizing a serial replating principle. BRCC3 knockdown resulted in decreased colony formation and lacked any immortalization properties. To further elucidate the functional consequences of BRCC3 lesions in the pathogenesis of MDS we performed silencing studies targeting BRCC3 in K562 cells and normal human CD34+ cells. While depletion of BRCC3 alone was not lethal, it led to enhanced etoposide-induced apoptosis. Consistent with these results, overall cell viability was substantially lower in shRNA-BRCC3–treated cells following etoposide when compared with control cells (46% vs. 55%). RPA2 expression, a single-strand DNA-binding protein used as a marker, was higher in BRCC3 knockdown cells than control cells and was further enhanced by etoposide treatment, thus indicating an increased end resection of single strand breaks. We concluded that defects in BRCC3 decrease non-homologous end joining and increased homologous recombination.
In sum, our study demonstrates for the first time detection of BRCC3 complex defects, leading to impairment of DSBs repair in patients with myeloid neoplasms. Furthermore, BRCC3 defects are associated with an aggressive phenotype and shorter survival.
Disclosures:
Makishima: AA & MDS international foundation: Research Funding; Scott Hamilton CARES grant: Research Funding. Polprasert:MDS foundation: Research Funding. Maciejewski:NIH: Research Funding; AA/MDS foundation: Research Funding.
