Novel aspects of neuroblastoma : to hypoxia and beyond

Neuroblastoma (NB) is a pediatric cancer arising from the neural crest cells forming the sympathetic nervous system. Just as other types of pediatric cancer the driving mutations of neuroblastoma are few and the tumors are instead categorized according to genetic abbreviations such as amplification, loss of heterozygosity, gains and translocations. Neuroblastoma continues to be therapeutically challenging and flabbergasts scientists and doctors with it vast heterogeneity both clinically and biologically. The tumors range from aggressive, fast growing, lethal cancer to metastatic tumors that will spontaneously regress and disappear without any clinical interventions. The survivors of high-risk neuroblastoma struggle with life-long side effects due to the aggressive therapeutic treatment used in these young children. In this thesis I have focused on discovering alternative mechanisms that can contribute to the development and malignancy of neuroblastoma.

Paper I. High-risk neuroblastoma has been shown to have high level of DNA methylation of putative tumor suppressors. We designed a therapeutic strategy where we exploited the reversibility of DNA methylation and combined the DNA-demethylating drug 5-Aza-deoxycytidine (AZA) with the differentiation-promoting activity of retinoic acid (RA) as an alternative strategy to treat high-risk neuroblastoma. In this paper we showed that treatment with AZA restores high-risk neuroblastomas sensitivity to RA. Additionally, the combined systemic distribution of AZA and RA impedes tumor growth and prolongs survival in vivo. Genomewide analysis of treated tumors revealed that the combined treatment induced a HIF2α-associated hypoxia-like transcriptional response followed by an increase in neuronal gene expression and a decrease in cell-cycle gene expression. We performed a loss-of-function experiment using a small-molecule inhibitor of HIF2α which resulted in diminished tumor response to AZA+RA treatment. Our study indicated that the increase in HIF2α levels is a key component in the tumor response to AZA+RA and that high levels of HIF2α, but not HIF1α, significantly correlate with expression of neuronal differentiation genes and better prognosis, but negatively correlate with key features of high-risk tumors. Contrary to previous studies, our findings indicate an unanticipated tumor suppressive role for HIF2α in neuroblastoma.

Paper II. We explored the role of hypoxia inducible transcription factor EPAS1/HIF2α in neuroblastoma. We analyzed several neuroblastoma tumor expression datasets which showed that EPAS1 expression is associated with better patient outcome and characteristics of low-risk tumors and did not support an oncogenic role as previously shown in other studies. We continued by treating xenografted mice with HIF2α inhibitors, which did not block in vitro neuroblastoma cell proliferation nor tumor growth. To illuminate the role of EPAS1 during embryonic development we analyzed single cell data sets from the developing mouse sympathoadrenal lineage, wherein expression of Epas1 was a strong predictor of the most differentiated cells and negatively correlated with key progenitor characteristics. Additionally, the genes co-expressed with Epas1 in the sympathoadrenal lineage were associated favorable patient outcome and features of low-risk neuroblastoma.

Paper III. Due to the lack of recurrent mutations in neuroblastoma it has become more important to focus on alternative mechanisms that can influence the development of pediatric cancers. We have studied the role of fusion transcripts and proteins in neuroblastoma to illuminate possible oncogenic properties. Utilizing sequenced neuroblastoma datasets we have been able to identify neuroblastoma specific fusion transcripts. We could identify enrichments of fusions in regions frequently gained or lost in high-risk neuroblastoma as well as fusions of well-known drivers of neuroblastoma. To explore potential oncogenic properties of fusion proteins we focused on the ZNF451-BAG2 fusion which generates a truncated BAG2 protein, that we call ΔBAG2. When overexpressing ΔBAG2 in neuroblastoma cell lines we see impaired retinoic acid-induced differentiation. Indicating that ΔBAG2 could play a role in neuronal maturation, potentially leading to a less differentiated and more aggressive tumor. Our findings reveal an overlooked mechanism capable of generating altered gene products, which are relevant for neuroblastoma pathogenesis and presents new potential drug targets.

List of scientific papers

I. Combined epigenetic and differentiation-based treatment inhibits neuroblastoma tumor growth and links HIF2α to tumor suppression. Isabelle Westerlund, Yao Shi, Konstantinos Toskas, Stuart M. Fell, Shuijie Li, Olga Surova, Erik Södersten, Per Kogner, Ulrika Nyman, Susanne Schlisio and Johan Holmberg. Proc Natl Acad Sci U S A. 2017 Jul 25;114(30):E6137-E6146.
https://doi.org/10.1073/pnas.1700655114

II. EPAS1/HIF2α correlates with features of low-risk neuroblastoma and with differentiation during sympathoadrenal development. Isabelle Westerlund, Yao Shi and Johan Holmberg. Biochem Biophys Res Commun. 2019 Jan 22;508(4):1233-1239.
https://doi.org/10.1016/j.bbrc.2018.12.076

III. Recurrent fusion transcripts associated with key tumor characteristics occur at high frequency in neuroblastoma. Yao Shi, Vilma Rraklli, Eva Maxymovitz, Shuijie Li, Isabelle Westerlund, Christofer Juhlin, Adam Stenman, Catharina Larsson, Per Kogner, Maureen J. O’Sullivan, Susanne Schlisio and Johan Holmberg. [Manuscript]