Project Summary:
Posterior fossa group A (PFA) ependymoma is a type of central nervous system tumor primarily found
in the brain or spinal cord of infants and toddlers. PFA ependymoma is thought to be an epigenetically driven
tumor characterized by a global loss of H3K27me3 levels, which promotes cancer cell proliferation. Polycomb
repressive complex 2 (PRC2) is the sole enzyme responsible for histone H3K27 methylation in mammalian
cells. Low H3K27me3 in PFA ependymoma is dictated by at least two mechanisms, including PRC2 inhibition
and metabolic rewiring. EZH1 and EZH2 are the catalytic subunits of PRC2. The EZH1/2 inhibitory protein
EZHIP (a.k.a. CXorf67) is normally expressed in gonads. EZHIP is abnormally expressed in PFA ependymoma,
where it inhibits PRC2 enzymatic activity using a protein sequence mimicking the H3K27M oncohistone. The
H3K27M-like sequence occupies the lysine-binding channel at the active site of PRC2 and thereby precludes
histone substrate binding. FPA ependymoma cells are addicted to H3K27 hypomethylation, which underlies
enhanced expression of glycolysis and TCA cycle metabolism genes that help fuel cancer cells. PRC2
inhibition by EZHIP ultimately depends on their physical interaction. Using a hybrid approach combining
AlphaFold structural modeling, biochemistry, structural biology, and crosslinking mass spectrometry, we will
clarify an important missing link between EZHIP and PRC2, which we found critical for PRC2 inhibition in the
preliminary study. Specifically, in Aim 1, we will build an atomic model of a previously uncharacterized EZHIP
motif bound to PRC2. We will also show that the newly identified binding interface between EZHIP and PRC2
can be specifically mutated or targeted to relieve PRC2 inhibition. Finally, we will generate a structural model
of full-length EZHIP engaged with a PRC2 holo complex by integrating spatial constraints from crosslinking
mass spectrometry into AlphaFold structural modeling. In Aim 2, we will be focused on the direct competition
between EZHIP and EPOP, a naturally occurring accessory subunit of PRC2, in PRC2 binding. This is an
exciting mechanistic model strongly supported by our preliminary data showing that EPOP competes off EZHIP
from PRC2 and restores H3K27me3 levels in EZHIP-expressing cells. Importantly, we will study the molecular
mechanism of how EPOP may impact the epigenetic and metabolic pathways by disrupting the EZHIP–PRC2
complex in PFA ependymoma cell models. In addition, EPOP-based protein inhibitors of the EZHIP–PRC2
interaction will be optimized and tested both alone and in synergy with metabolism-targeting drugs in the
treatment of PFA ependymoma.