Neurodevelopmental ATM deficiency affects glutamatergic transmission through KCC2-KARs expression in the hippocampus but not in the cortex

Interest in the function of ataxia-telangiectasia mutated protein (ATM) is extensively growing as evidenced by preclinical studies which continuously associate ATM loss with new intracellular pathways. To further investigate ATM involvement in neurotransmission, we initially exploited ATM+/- and ATM-/- mice. ATM defective animals display cognitive defects rescued by the delivery of the antidepressant Fluoxetine (Fluox). Prenatal Fluox increases levels of the chloride intruder NKCC1, previously described lowered in ATM defective mice. Surprisingly, this result occurs exclusively at hippocampal level suggesting an ATM context-specificity. Indeed, a deeper investigation of excitatory synapses revealed increased Gluk1 and Gluk5 subunit-containing kainate receptors (KARs) levels in the hippocampus, but not cortex, of ATM+/- and ATM-/- mice. The comparable composition between wt and ATM+/- post-synapses (TIF purification) suggests a presynaptic KARs localization and in vitro and in ex vivo electrophysiology that KARs are fully active. Since KAR subunits functionally interact with the chloride cotransporter KCC2, which is highly expressed in ATM deficient mice, we blocked KCC2 activity during the development of ATM+/- cultures and found reduced KARs levels. Conversely, the transient and acute blockade of ATM kinase activity achieved by KU55933 in adults does not move similar changes as indicated by a completely different panel of deregulated genes identified in RNA-seq investigation performed on hippocampi explanted from KU mice. Our data, indicating that such alterations are developmental regulated and linked to ATM kinase activity through the KCC2 regulation, add new findings useful for a deeper understanding of ATM function in central neurons. Importantly, controlling the KCC2/NKCC1 levels, ATM affects both the GABAergic and glutamatergic system in the hippocampus. Overall design: Hippocampal RNAs from mice intranasally treated with the ATM kinase blocker KU55933 (n=4) or vehicle (4% DMSO in saline; n = 4). Direct comparison was performed