PROJECT SUMMARY
Glioblastoma multiforme (GBM) is the most common primary brain tumor. Cohesin is a chromatin-bound ring
complex involved in 3D genome organization, sister chromatid cohesion, gene expression, and DNA repair.
Mutational inactivation of genes encoding components of the cohesin complex is common in GBM, and
mutations of the STAG2 subunit account for >50% of all cohesin mutations. However, the mechanism(s) of
STAG2 tumor suppression remain unknown. Recent ground-breaking studies in the basic biology of cohesin
have shown that cohesin plays a critical role in generating and maintaining the chromatin loops that underly
much of 3D genome organization and that link enhancers to the promoters the regulate. However, the
relationship of these functions of cohesin to GBM pathogenesis is undefined. This grant will test the hypothesis
that tumor-derived STAG2 mutations result in alterations to 3D genome organization and enhanced Polycomb
Group (PcG)-mediated transcriptional repression to drive neoplastic transformation in GBM. This hypothesis is
based on experiments performed in the Waldman (multi-PI) and Jin (multi-PI) labs that utilized gene editing to
correct the endogenous mutant allele of STAG2 in GBM cells, with matched corrected and uncorrected cells
analyzed by Hi-C and RNA-seq. The data were analyzed using a new bioinformatics pipeline Jin developed
called “HiCorr” and “DeepLoop” that makes it possible to clearly identify chromatin loops from sub-billion read-
pair sequencing depth with the highest possible sensitivity. These experiments showed that whereas STAG2
was dispensable for maintenance of Topologically Associating Domains (TADs), STAG2 was essential for
regulating the size and strength of individual CTCF and H3K27me3-anchored chromatin loops, leading to
alterations in the expression of adjacent genes. The preliminary studies also showed that STAG2-mutant GBM
cells have dramatically increased levels of chromatin-bound H3K27me3, enhanced repression of Polycomb
Group (PcG)-regulated genes, and sensitivity to inhibitors of PcG signaling in vitro. Based on these data, two
aims are proposed. In Aim #1 we will examine the relationship between STAG2-regulated chromatin loops and
gene expression in GBM cells and tumors. In Aim #2 we will define the role of STAG2 in PcG-mediated chromatin
looping and transcriptional repression in GBM. Completion of the research proposed in this grant will define the
role of tumor-derived STAG2 gene mutations in 3D genome organization and PcG-mediated transcriptional
repression in GBM cells and tumors. These findings will provide a long-sought molecular mechanism for cohesin-
mediated tumor suppression, providing important clues for how cohesin mutations can be targeted for
therapeutic purposes in GBM.