Both Reelin (RELN) and glutamate decarboxylase 67 (GAD1) have been
implicated in the pathophysiology of Autism Spectrum Disorders (ASD). We have
previously shown that both mRNAs are reduced in the cerebella (CB) of ASD
subjects through a mechanism that involves increases in the amounts of MECP2
binding to the corresponding promoters. In the current study, we examined the
expression of RELN, GAD1, GAD2, and several other mRNAs implicated in this
disorder in the frontal cortices (FC) of ASD and CON subjects. We also focused
on the role that epigenetic processes play in the regulation of these genes in
ASD brain. Our goal is to better understand the molecular basis for the
down-regulation of genes expressed in GABAergic neurons in ASD brains.
We measured mRNA levels corresponding to selected GABAergic genes using
qRT-PCR in RNA isolated from both ASD and CON groups. We determined the extent
of binding of MECP2 and DNMT1 repressor proteins by chromatin
immunoprecipitation (ChIP) assays. The amount of 5-methylcytosine (5mC) and
5-hydroxymethylcytosine (5hmC) present in the promoters of the target genes was
quantified by methyl DNA immunoprecipitation (MeDIP) and hydroxyl MeDIP
(hMeDIP).
We detected significant reductions in the mRNAs associated with RELN and
GAD1 and significant increases in mRNAs encoding the Ten-eleven Translocation
(TET) enzymes 1, 2, and 3. We also detected increased MECP2 and DNMT1 binding to
the corresponding promoter regions of GAD1, RELN, and GAD2. Interestingly, there
was decreased amounts of 5mC at both promoters and little change in 5hmC content
in these same DNA fragments.
Our data demonstrate that RELN, GAD1, and several other genes selectively
expressed in GABAergic neurons, are down-regulated in post-mortem ASD FC. In
addition, we observed increased DNMT1 and MECP2 binding at the corresponding
promoters of these genes. The finding of increased MECP2 binding to the RELN,
GAD1 and GAD2 promoters, with reduced amounts of 5mC and unchanged amounts of
5hmC present in these regions, suggests the possibility that DNMT1 interacts
with and alters MECP2 binding properties to selected promoters. Comparisons
between data obtained from the FC with CB studies showed some common themes
between brain regions which are discussed.