| BACKGROUNDFear memory is defined as the ability to associate new environmental cues with physical danger.Extensive research on fear memory revealed it is extremely robust and enduring.Fear memory is usually investigated by classical Pavlovian fear conditioning experimental model.Fear memory extinction refers to the weaken of previously acquired responses when a Pavlovian signal is no longer paired with a reinforcer.Extinction is used as a major component of exposure therapies,a primary treatment of a variety of psychiatric disorders,such as posttraumatic stress disorder(PTSD),anxiety disorders,and stress-related disorders.Extinction involves the formation of new learning rather than erasure or attenuation of the initial learning.Previous studies have shown that extinction learning is highly context dependent,which requires extensive training and results in a relatively fragile extinction memory trace.Some patients respond insufficiently after exposure therapies due to the fragile extinction memory.Epigenetics refers to the heritable changes in the expression or regulation of gene without any alteration in the primary DNA sequence.Epigenetic mechanisms have critical regulatory roles in gene expression which may produce long-lasting changes in cellular function.Targeting the epigenetic modifications may help to reduce the initial fear memory and contributing to exposure therapies.Recent discoveries in support of epigenetic regulation of fear memory formation,consolidation,extinction and renewal.N6-methyladenosine(m6A)modification is the most abundant inner RNA epigenetic modification in eukaryotes.Currently,it has been identified that m6A marks groups of m RNAs and involved in hippocampus-dependent learning and memory,formation and consolidation of long-term memory,and stress response.However,it remains to reveal whether and how m6A modifications may regulate fear memory extinction.Understand the underlying m6A effects on fear memory extinction processes will help to points to potential targets for treating a series of psychiatric disorders related to dysfunction of fear memory extinction.PURPOSESIn the current study,we examined the changes of m6A modifications levels during fear memory extinction.Through deletion of the methyltransferase Mettl3 in glutamate neurons,GABAergic neurons or astrocytes,we identified whether m6A modifications in different cell types may influence fear memory extinction.By next-generation sequencing,Me RIP-q PCR and rescue-experiment,we seek a better understanding of m6A and its regulatory roles in fear memory extinction.METHODS AND RESULTS1.We established an animal model of fear memory extinction.The levels of m6A on m RNAs were significantly increased in the hippocampus of mice 0.5 h after fear extinction training.2.We crossed Mettl3flox/flox mice with genetically modified Cre-expressing mouse lines to achieve tamoxifen-inducible Mettl3 depletion specifically in glutamatergic neurons(Mettl3-Vglut1-Cre),GABAergic neurons(Mettl3-Gad2-Cre)and astrocytes(Mettl3-Gfap-Cre).Immunofluorescence staining validated the loss of METTL3 in specific categories of neurons and astrocytes in genetically modified mice.We then detected decreased Mettl3 expression along with decreased m6A levels in the hippocampus of these mice with conditional knockout of the Mettl3 gene using qRT-PCR and Epi Quick RNA Methylation Quantification Kit(Epigentek).Loss of METTL3 did not alter the motor activities or generate anxiety-like behaviour of the mice in the open-field test and elevated-plus maze test.We demonstrated that METTL3 expressed in hippocampal astrocytes,but not in glutamatergic or GABAergic neurons is specifically involved in fear memory extinction,although Mettl3 depletion failed to affect the capability of developing or retrieval fear memory.Fear extinction and reversal learning is a component of cognitive flexibility.Reversal learning were assessed in the Morris water maze(MWM)spatial learning task,spatial reversal task performance of Mettl3-Gfap-Cre mice were impaired,although initial acquisition learning was not affected.Loss of METTL3 in astrocytes did not alter the fear discriminative abilities,spatial learning and memory,motor coordination or short-term memory abilities of the mice.3.Intrahippocampal injection of Cre-recombination-based adeno-associated viruses to Mettl3flox/flox mice,we validated AAV efficiently and widely transduce astrocytes in the mouse hippocampus after 3 weeks of intrahippocampal injection.Immunofluorescence staining validated the loss of METTL3 in astrocytes in the hippocampus of the mice compared to control mice.Loss of METTL3 in hippocampus astrocytes did not alter the motor activities or generate anxiety-like behaviour of the mice in the open-field test and elevated-plus maze test.Loss of METTL3 in hippocampus astrocytes did not alter the fear discriminative abilities of the mice.Loss of METTL3 in hippocampus astrocytes impaired extinction learning and extinction of fear memory of the mice.4.Biological triplicates of next-generation sequencing using hippocampal transcriptome of Control and Mettl3-Gfap-Cre mice,the results indicated Grin2b was one of the most important differentially expressed genes.5.We next used qRT-PCR and found Grin2b levels were increased in the hippocampus24 h after conditioning training(right before extinction training).Antagonizing Glu N2B-containing NMDA receptors both by intraperitoneal injection or intrahippocampal injection of Ro 25-6981 significantly reverses the knockout phenotype of Mettl3-Gfap-Cre mice.The Me RIP-q PCR results indicated that m6A modification abundance on Grin2b in hippocampus samples of Mettl3-Gfap-Cre mice were downregulated.CONCLUSIONThe m6A tagging on hippocampal astrocytes facilitates fear memory extinction through finetuning the activity of Glu N2B-containing NMDA receptors. |