NEW YORK (GenomeWeb) – Enhancer networks in central nervous system tumors called ependymomas appear to offer clues to potential targets for the disease, according to an international team led by investigators in the US, Germany, and Canada.
The researchers used a combination of exome sequencing, whole-genome sequencing, copy number analyses, RNA sequencing, chromatin immunoprecipitation sequencing, and/or array-based DNA methylation profiling to look at mutation, expression, regulatory, and epigenetic patterns in samples from 42 individuals with ependymomas in Heidelberg or Toronto. Focusing on regions of interest in both cohorts, they identified 1,682 ependymoma super enhancers, which offered a glimpse at some of the genes and pathways contributing to the tumors.
With the help of small molecule or RNA interference (RNAi), the team targeted a handful of these enhancers in patient-derived neurosphere cell culture samples or in mouse models of the disease. Results from those follow-up analyses hint that it may be possible to slow tumor growth by staunching super-enhancer activity in ependymomas, which typically do not respond to chemotherapy. The study appeared online today in Nature.
"[W]e were able to identify almost 1,700 super enhancers, and assign them to specific molecular groups of ependymoma," co-corresponding author Marcel Kool, a pediatric neuro-oncology researcher affiliated with the German Cancer Research Center and the Hopp Children's Cancer Center at the NCT Heidelberg, said in a statement. "We subsequently demonstrated that many of these super enhancers influence the activity of genes that are implicated in the development of cancer."
The researchers began by using Illumina HiSeq 2000 instruments to do ChIP-seq — targeting the active chromatin mark histone 3 lysine 27 acetylation (H3K27ac) — on 42 fresh-frozen primary ependymoma tumor samples. After folding in mutation, copy number, expression, and methylation data, they flagged almost 2,200 apparent super enhancers in 24 tumors from the Heidelberg cohort and nearly 3,200 super enhancers in the 18 Toronto cohort ependymoma tumors.
From the 1,682 overlapping candidate sites, the team selected 15 of the most highly ranked potential super enhancers for short hairpin RNA knockdown experiments in patient-derived cell cultures. There, RNAi aimed at suspected super enhancers led to impaired ependymoma cell growth over a week.
The researchers searched for super enhancers that were specific to a given ependymoma subtype. For example, in a relatively aggressive, somatic mutation-poor form of ependymoma that appears in the posterior fossa region of the skull — known as the posterior fossa ependymoma group A (PF-EPN-A) — they narrowed in on a super enhancer that regulates IGF2BP1. And knocking down that gene appeared to curb cell proliferation in the PF-EPN-A ependymomas.
In other ependymoma subgroups, tumor-specific super-enhancer candidates turned up near genes such as CACNA1H, or appeared to influence the activity of transcription factors such as SOX9 or SOX2. Again, the researchers reported, RNAi targeting such transcription factors appeared to impair ependymoma cells and animal models.
Along with efforts to establish enhancer-regulated networks, the team incorporated available information from a drug-gene interaction database to find small molecules that might staunch ependymoma growth via genes influenced by the newly identified super enhancers, including compounds targeting genes such as FGFR1 or WEE1.
"By integrating our data with drug interaction databases, we identified and validated novel cancer dependencies of ependymoma that are responsive to pharmacologic inhibition," the authors wrote. "Our study further demonstrates that knowledge of enhancer landscapes can be used to dissect the molecular differences between histologically similar tumor entities and to provide unique information that may inform precision therapies."