NEW YORK (GenomeWeb) – An international team of researchers has identified epigenetic differences between subtypes of medulloblastoma, identifying the potential cells of origin for the most common subtype, known as Group 4.
The team — which included researchers from the Dana Farber Cancer Institute, the European Molecular Biology Laboratory, the German Cancer Research Center, St. Jude's Children Research Hospital, and elsewhere — performed ChIP-seq on 28 primary tumor samples and three well-characterized cell lines in order to characterize the regulatory enhancers in the tumors, as they reported this week in Nature.
Prior to the study, the enhancer landscape of only one medulloblastoma cell line had been characterized, and previous research had found that the epigenome of cell lines is often quite different from the epigenome seen in clinical tumor samples. So the researchers sought out to gain a better understanding of the regulatory landscape of medulloblastoma, which is a rare, but highly malignant brain tumor most often found in children.
Medulloblastoma can be grouped into four different subtypes — WNT, SHH, Group 3, and Group 4 — and the 28 clinical samples represented all four subtypes. By performing ChIP-seq on all samples, the researchers identified more than 78,000 potential enhancers, 26 percent of which were differentially active among the tumor subtypes. The other 74 percent of enhancers showed varied activity "suggesting either ubiquitous activity, for example, 'housekeeping' genes, or a general role in medulloblastoma or cerebellar identity," the authors wrote.
By analyzing the enhancers, the researchers identified six distinct classes, including one for each subtype as well as one class that was shared between Group 3 and 4 and a class that was shared between the WNT and SHH subtypes.
Because Group 3 and Group 4 medulloblastoma have some "transcriptional similarity," the finding that they also shared a set of enhancers was not surprising, the authors noted. However, the subset of shared enhancers between WNT and SHH was "unexpected."
Next, the team wanted to see if they could match the enhancers to genes using RNA-seq data from the same samples. They were able to assign 8,775 enhancers to at least one gene with several enhancers predicted to converge on the same gene.
Some of their data confirmed previous findings. For instance, the team identified a 450-kb focal amplification in one Group 3 sample that encompassed the ACVR2A gene and the upstream enhancer regions, including the TGF-beta pathway. "These data, combined with our prior observations that TGF-beta receptor genes are recurrently amplified in Group 3, further suggest TGF-beta signaling as a putative oncogenic driver in this subgroup," the authors wrote.
Looking at so-called super enhancers, which drive the oncogenes required for cells to maintain their tumor identity and genes associated with cell-type specific functions, the team found around 3,000 super enhancer loci and found that super enhancers tended to cluster by subgroup. In addition, they were able to validate some of their findings in zebrafish, for instance, identifying some super enhancers that appear to regulate the oncogene MYC.
Among some of the subgroup-specific super enhancer target genes, the researchers identified transcription factors involved in neuronal development.
The origins of Group 3 and Group 4 subtypes are "unknown, yet essential to define, as these tumors account for approximately 60 percent of all diagnoses, lack targeted therapies, and are frequently associated with poor clinical outcomes," the authors wrote.
By looking at the target genes of super enhancers and performing a circuitry analysis, the researchers were able to trace back the origin of Group 4 to neurons originating from early progenitors of the upper rhombic lip — a portion of the brain that begins developing in the embryo.
Understanding the cellular origins of medullablastoma has important treatment implications. Other cancers have been treated effectively by targeting the cells in its lineage, for example targeting B cells in leukemia, the authors wrote.
"As medulloblastoma is believed to originate from cell populations that normally exist ephemerally during development, targeting the aberrant persistence of tumor cells from these lineages may represent a novel therapeutic strategy with minimal effect on the normal tissue compartment," they added.