The role of ATRX in neuroblastoma: Uncovering novel targets for medicinal interventions

Abstract

Cancer is worldwide the number one cause of death among young children. A frequently occurring type of paediatric cancer is neuroblastoma, which arises in parasympathetic ganglia. A specific subgroup of neuroblastoma patients with high-risk disease is characterised by genetic aberrations in the chromatin remodeller ATRX. ATRX is responsible for the incorporation of the histone variant H3.3 at repetitive DNA. It is assumed that this function is important for genome stability. In neuroblastoma distinct mutations occur within the ATRX gene. However, an overview of all occurring mutations, their frequencies and their associations with patient and tumour characteristics was lacking. Therefore, I first focussed on generating an overview of the ATRX mutational landscape in neuroblastoma and all other types of paediatric cancer. To accomplish this, we determined the frequency of ATRX nonsense mutations, missense mutations, and multi-exon deletions (MEDs). This led us to the discovery that the latter occurred almost exclusively in neuroblastoma and that 70% of the ATRX aberrations in neuroblastoma are MEDs. Furthermore, we predicted that 75% of these MEDs are likely in-frame and result in the production of shorter protein products. Additionally, we discovered a slightly better overall survival for patients with ATRX missense mutations compared to the other mutation types. We also found that 11q deletions, which are independently associated with poor survival, co-occur more frequently with ATRX MEDs compared to the other ATRX mutation types or wild-type. Altogether, this suggests the existence of different ATRX sub-types within neuroblastoma. To investigate whether such different ATRX subtypes exist we generated multiple isogenic cell line models with distinct ATRX aberrations, namely knock-out, exon 2-10 MED and exon 2-13 MED models. On these models we conducted total RNA-sequencing and for our analysis we also included the RNA-sequencing data of eight neuroblastoma cell lines, including three with an ATRX MED, and data of nine neuroblastoma tumours, including two with an exon 2-10 MED. By comparing the gene expression against the ATRX wild-type cells we discovered in all ATRX mutant changes in the expression of genes involved in ribosome biogenesis and metabolism. We found reduced gene expression related to these processes in the ATRX exon 2-10 MEDs models, but in sharp contrast we observed increased expression in ATRX KO and ATRX exon 2-13 models. In this manner we confirmed the existence of distinct ATRX sub-types. Subsequently, we conducted drug screens and genome-wide CRISPR-Cas9 synthetic lethality screens on ATRX aberrant cells to discover fast implementable drugs or identify novel drug targets. By performing drug screens we discovered three drugs that showed effectivity against all neuroblastoma cell lines tested independent of the ATRX mutational status. Additionally, we discovered 541 and 376 potential synthetic lethal interactions for ATRX KO and ATRX MED, respectively, compared to ATRX-wildtype cells. These synthetic lethal interactions give hope for the future as they could potentially be used as novel therapeutic targets. Finally, I discuss the implications of our findings in relation to one another and to the current knowledge regarding ATRX, additionally I mention potential directions for future research.