Poster Presentation 27th Lorne Cancer Conference 2015

A comprehensive and systematic analysis of miRNA function in neuroblastoma (#152)

Eoin Dodson 1 2 , Iva Nikolic 1 2 , Daniel Thomas 3 , Kaylene Simpson 3 , Alexander Swarbrick 1 2
  1. The Kinghorn Cancer Centre & Cancer Research Division, Garvan Institute of Medical Research, Darlinghurst, NSW, Australia
  2. St Vincents Clinical School, Faculty of Medicine, UNSW, Sydney, NSW, Australia
  3. Victorian Centre for Functional Genomics , Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia

Background:

Neuroblastoma (NB) is a cancer of the sympathetic nervous system found in children aged 0-5. It accounts for ~15% of total childhood cancer mortality as high-risk NB patients have poor 5-year survival outcomes below 50% despite aggressive multi-modal therapy. MYCN oncogene amplification occurs in ~22% of patients with higher degrees of amplification conferring much poorer outcomes, even in otherwise favourable patient groups.

MicroRNAs (miRNAs) are a class of non-coding RNAs which repress multiple target genes via base-pairing at a seed sequence. Recently several miRNAs have been shown to be oncogenic or tumour suppressive in NB, but a comprehensive genome-wide analysis of miRNA function in NB has not been published.

Methods:

A genome-wide functional screen of >1200 miRNAs using both miRNA mimics (overexpression) and antisense inhibitors was carried out across a panel of 3 cancer types: breast, prostate and NB. This included both a highly MYCN amplified cell line (Kelly) and a non-MYCN amplified NB cell line (Shep). Cell viability was quantified in the presence of vincristine, doxorubicin and without chemotherapy.

In my project I aim to pursue two classes of miRNAs arising from this screen: (1) those exhibiting a synthetic lethal interaction with chemotherapy, which have no impact on cell viability alone but synergise with chemotherapy to kill cells. (2) miRNAs that are lethal to NB independent of chemotherapy. The synthetic lethals and most effective lethals will first be validated in a secondary screen which will also include two additional MYCN-amplified NB cell lines. Validated hits from the secondary screen will be examined for mechanisms of action including direct target genes of the miRNA. We also intend to assess these miRNAs in vivo as potential therapeutics. Alternatively, there is potential for synthetic lethal miRNAs to predict patient response to chemotherapy based on their expression in patient tissue. In addition we can also search for over-represented target genes and pathways.

Results:

In the genome-wide functional screen we found 351 lethals (viability fold change <0.5) for the MYCN amplified cell line and 265 in the non-MYCN amplified cell line. There is a large overlap between the NB cell lines, but not with the other 2 cancer types of the functional screen. Furthermore, 22 miRNAs were synthetic lethal (media >0.7; chemotherapy <0.5 of media viability) with either vincristine or doxorubicin, strikingly all of these occurred only in the MYCN-amplified cell line.