Not only does G a GLP complex appear to play
Not only does G9a/GLP complex appear to play a role (albeit a seemingly conflicting one) in LTP maintenance, it may also be involved in long-term depression (LTD) (). At hippocampal CA1 synapses, LTD can be induced by pharmacological or synaptic stimulation of Group 1 metabotropic glutamate receptors (Gp1 mGluRs) – either by short applications of R,S-dihydroxyphenylglycine (DHPG), an agonist of Gp1 mGluRs, or by prolonged low-frequency stimulation of SC-CA1 afferents (, , ). It has been reported recently that pharmacological inhibition of G9a/GLP complex blocked the maintenance of DHPG-induced mGluR-LTD (). Bath application of G9a/GLP inhibitor BIX or UNC hampered the maintenance of mGluR-LTD without negating initial mGluR-LTD induction by DHPG bath-application – synaptic strength decreased 10 min after DHPG application but returned to baseline within 20 min in the presence of BIX or UNC. This was presumably through protein kinase Mζ (PKMζ)- and N-ethylmaleimide-sensitive factor (NSF)-mediated reduction in internalisation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic rifadin receptor (AMPA) receptors. In addition, measured H3K9me2 levels increased upon DHPG application but not upon co-application of DHPG and BIX/UNC, suggesting that mGluR activation-dependent changes in the epigenome were mediated by G9a/GLP. Notably, the reversion of mGluR-LTD by G9a/GLP inhibition was only observable in intact acute hippocampal slices but not when the cell body layer of CA1 neurons was severed. This indicates the existence of activity-dependent, G9a/GLP-regulated signals and changes that originate from the nucleus but are able to override local LTD-supporting machinery at the synapses. Furthermore, showed that not only did G9a/GLP inhibition prevent mGluR-LTD, it also reinforced weak tetanization-induced early-LTP into late-LTP in the same group of SC-CA1 synapses, even when the G9a/GLP inhibitors had been washed out for 30 min. This implies that G9a/GLP inhibition exerted a metaplastic effect and drove the CA1 neurons towards a potentiation-prone state – G9a/GLP repression did not only reverse LTD but also reinforced LTP given the correct stimulus. Taken together, recent studies show that G9a/GLP complex plays a part in both LTP and LTD maintenance, and hence suggest that this histone lysine methyltransferase acts as a bidirectional epigenetic regulator of transcription underlying activity-dependent synaptic plasticity. What are the molecular underpinnings of the above electrophysiological observations of the role of G9a/GLP in synaptic plasticity? G9a/GLP complex has been implicated in the regulation of multiple canonical plasticity-related genes and proteins – including brain-derived neurotrophic factor (BDNF), PKMζ and activity-regulated cytoskeletal (Arc) gene. Basal expression of immediate early genes (IEG, e.g. and ), which are usually quickly activated and transcribed upon stimulation, were altered in G9a-deficient NAc MSNs in a cell type-specific manner (). Changes in IEG may also be activity-dependent. For instance, G9a and H3K9me2 occupancy at some of these IEGs in the NAc have been shown to respond dynamically to acute and chronic cocaine stimulation (). With regards to synaptic plasticity, G9a/GLP inhibition repressed the DHPG-induced upregulation of mRNA and protein levels, implying that normal G9a/GLP function is required for the essential expression in mGluR-LTD (). Also, mice also exhibited misregulated mRNA expression in response to dark rearing, which is associated with the animals’ inability to exhibit homeostatic synaptic scaling up after sensory deprivation (). Moreover, studies suggest that G9a and H3K9me2 binding at promoters and correlated BDNF expression also respond dynamically to various environmental stimuli in various brain regions, including the NAc, the amygdala, the cortex, and the hippocampus (, , , , ). Notably, different substances alter H3K9me2 occupancy at different promoters in distinct brain regions. Cocaine stimulation downregulated G9a and H3K9me2 occupancy at promoter VI in the NAc () while nicotine downregulated H3K9me2 occupancy at promoter I, IV and IXa () in the cortex. Functionally, G9a/GLP repression-mediated reinforcement of LTP in the hippocampal area CA1 required BDNF (). Also, in primary rat cortical neurons, knockdown and pharmacological inhibition of G9a/GLP prevented the repression of mRNA that is usually observed in TTX-mediated synaptic scaling up. And , mice did not display the typical increase in H3K9me2 occupancy at promoters and the corresponding decrease in mRNA expression that is associated with sensory deprivation in WT animals ().