Epigenetic modifications have drawn significant attention due to their crucial roles in
development and disease. The recent emergence of epigenome editing tools provides an
attractive strategy for manipulating chromatin structure and gene expression. They
have been shown to successfully activate and silence targeting genes in conjunction with
chromatin modifying enzymes. A major gap in knowledge pertains to how epigenome
editing differs in efficiency at distinct DNA contexts. The focus of this thesis was to build
a novel system, chromatin inducible targeting (CIT), to test the effect of sequence
context on the efficiency of chromatin modifying enzymes to control gene expression.
Specifically, I tested the effect of CAG/CTG repeat expansion on the ability of HDAC5,
HDAC3, and DNMT1 targeting to modify chromatin structure.
CIT can be divided into three major components. First, a GFP-based reporter monitors
gene expression. It contains an intron carrying a varying number of CAG/CTG triplets.
Second, I adapted the ParB-INT targeting system such that any protein of interest can be
recruited within 300bp of the expanded CAG/CTG repeat tract. Third, an ABA-based
chemical inducible proximity system that allows for spatiotemporal and reversible
targeting of proteins to chromatin. Notably, CIT is also well suited to ask whether
chromatin modifying enzymes work locally to regulate gene expression or in trans.
I found that HDAC5 targeting silences the GFP reporter while decreasing local histone
acetylation. Interestingly, HDAC5 preferentially silences the reporter with the shorter
repeat tract, probably because of the lower levels of histone acetylation present at the
expanded repeat tract before targeting. HDAC5 is thought to deacetylate histones by
recruiting another HDAC, HDAC3. Surprisingly, however, HDAC3 targeting increased
GFP expression, and its effect is insensitive to the size of the repeat tract. This effect is
controversial to the current models of HDAC3 function that deacetylation of histone tails
by HDAC3 in gene body improves transcriptional output. Moreover, I found that Dnmt1
targeting has a similar effect on gene silencing as HDAC5 targeting: it is more efficient in
the context of shorter CAG/CTG repeats.
CIT provides a novel strategy to optimize the efficiency of epigenome editing in a highly
controlled and flexible manner. Our data uncover novel mechanisms of gene regulation
by these chromatin modifiers and guides their use in manipulating chromatin structure.
CIT is suitable for screening and can be adapted to study the effect of virtually any
sequences on epigenome editing.