BackgroundGlutamate metabolism disorder is an important mechanism of sepsis-associated encephalopathy (SAE). Astrocytes regulate glutamate metabolism. In septic mice, α2A adrenoceptor (α2A-AR) activation in the central nervous system provides neuroprotection. α2A-ARs are expressed abundantly in hippocampal astrocytes. This study was performed to determine whether hippocampal astrocytic α2A-AR activation confers neuroprotection against SAE and whether this protective effect is astrocyte specific and achieved by the modulation of glutamate metabolism. MethodsMale C57BL/6 mice with and without α2A-AR knockdown were subjected to cecal ligation and puncture (CLP). They were treated with intrahippocampal guanfacine (an α2A-AR agonist) or intraperitoneal dexmedetomidine in the presence or absence of dihydrokainic acid [DHK; a glutamate transporter 1 (GLT-1) antagonist] and/or UCPH-101 [a glutamate/aspartate transporter (GLAST) antagonist]. Hippocampal tissue was collected for the measurement of astrocyte reactivity, GLT-1 and GLAST expression, and glutamate receptor subunit 2B (GluN2B) phosphorylation. In vivo real-time extracellular glutamate concentrations in the hippocampus were measured by ultra-performance liquid chromatography tandem mass spectrometry combined with microdialysis, and in vivo real-time hippocampal glutamatergic neuron excitability was assessed by calcium imaging. The mice were subjected to the Barnes maze and fear conditioning tests to assess their learning and memory. Golgi staining was performed to assess changes in the hippocampal synaptic structure. In vitro, primary astrocytes with and without α2A-AR knockdown were stimulated with lipopolysaccharide (LPS) and treated with guanfacine or dexmedetomidine in the presence or absence of 8-bromo- cyclic adenosine monophosphate (8-Br-cAMP, a cAMP analog). LPS-treated primary and BV2 microglia were also treated with guanfacine or dexmedetomidine. Astrocyte reactivity, PKA catalytic subunit, GLT-1 an GLAST expression were determined in primary astrocytes. Interleukin-1β, interleukin-6 and tumor necrosis factor-alpha in the medium of microglia culture were measured. ResultsCLP induced synaptic injury, impaired neurocognitive function, increased astrocyte reactivity and reduced GLT-1 and GLAST expression in the hippocampus of mice. The extracellular glutamate concentration, phosphorylation of GluN2B at Tyr-1472 and glutamatergic neuron excitability in the hippocampus were increased in the hippocampus of septic mice. Intraperitoneal dexmedetomidine or intrahippocampal guanfacine administration attenuated these effects. Hippocampal astrocytes expressed abundant α2A-ARs; expression was also detected in neurons but not microglia. Specific knockdown of α2A-ARs in hippocampal astrocytes and simultaneous intrahippocampal DHK and UCPH-101 administration blocked the neuroprotective effects of dexmedetomidine and guanfacine. Intrahippocampal administration of DHK or UCPH-101 alone had no such effect. In vitro, guanfacine or dexmedetomidine inhibited astrocyte reactivity, reduced PKA catalytic subunit expression, and increased GLT-1 and GLAST expression in primary astrocytes but not in primary astrocytes that received α2A-AR knockdown or were treated with 8-Br-cAMP. Guanfacine or dexmedetomidine inhibited microglial reactivity in BV2 but not primary microglia. ConclusionsOur results suggest that neurocognitive protection against SAE after hippocampal α2A-AR activation is astrocyte specific. This protection may involve the inhibition of astrocyte reactivity and alleviation of glutamate neurotoxicity, thereby reducing synaptic injury. The cAMP/protein kinase A (PKA) signaling pathway is a potential cellular mechanism by which activating α2A-AR modulates astrocytic function.