Chronic brain inflammation causes a reduction in GluN2A and GluN2B subunits of NMDA receptors and an increase in the phosphorylation of mitogen-activated protein kinases in the hippocampus

Abstract

Neuroinflammation plays a key role in the initiation and progression of neurodegeneration in Alzheimer's disease (AD). Chronic neuroinflammation results in diminished synaptic plasticity and loss of GluN1 N-methyl-D-aspartate (NMDA) receptors in the hippocampus, leading to the cognitive deficits that are the most common symptoms of AD. Therefore, it is suggested that chronic inflammation may alter expression levels of GluN2A and GluN2B subunits of NMDA receptors and associated intracellular signalling. Chronic neuroinflammation was induced by chronic infusion of lipopolysaccharide (LPS) into the fourth ventricle in Fischer-344 rats. The status of hippocampus-dependent memory was evaluated in control rats and rats chronically infused with LPS. Microglial activation in the hippocampus was examined using immunohistochemical staining. Western blot analysis was used to measure membrane levels of GluN2A and GluN2B subunits of NMDA receptors and mitogen-activated protein kinase (MAPK) in the hippocampi of these rats, and immunofluorescent double labeling was used to assess the cellular location of MAPK. Microglial activation was observed in the hippocampi of rats that showed memory impairments with chronic LPS infusion. Chronic LPS infusion reduced the levels of GluN2A and GluN2B and increased the levels of phosphorylated MAPKs in the hippocampus. MAPK-positive immunoreactivity was observed mostly in the neurons and also in non-neuronal cells. Reductions in GluN2A and GluN2B subunits of NMDA receptors coupled with altered MAPK signaling, in response to inflammatory stimuli may be related to the cognitive deficits observed in AD.

Figure 1

Performance of aCSF- and LPS-infused rats in the spatial version of a Morris water maze. (A) Search error in finding a hidden platform in a spatial learning task during 5 training trial blocks. The artificial cerebrospinal fluid (aCSF)-infused rats (labeled as aCSF) became proficient at locating the submerged platform during training. The lipopolysaccharide (LPS)-infused rats (labeled as LPS) showed less improvement over the course of training when compared with the aCSF-infused rats (*, p < 0.05). (B) The percentage of time spent in the target annulus which was 5 times of the platform size, during the 30 s probe trial. The aCSF-infused rats spent more time in the target area than the LPS-infused rats, and this was statistically significant at the third probe trial (*, p < 0.05).

Figure 2

Membrane levels of GluN2A and GluN2B subunits of NMDARs in the hippocampi of aCSF- and LPS-infused rats. (A) Representative western blot of GluN2A, GluN2B, and actin. (B, C) Membrane levels of GluN2A and GluN2B were decreased in the hippocampi of the LPS-infused rats than in those of the aCSF-infused rats (*, p < 0.05).

Figure 3

Levels of hippocampal phosphorylated extracellular signal-regulated kinase 1/2 and phosphorylated p38 in the aCSF- and LPS-infused rats. (A) Representative western blot of extracellular signal-regulated kinase 1/2 (Erk1/2), phosphorylated Erk1/2 (p-Erk1/2), and actin. (B) Levels of p-Erk1/2 were higher in the hippocampi of the LPS-infused rats than in that of the aCSF-infused rats. (C) Representative western blot of p38, phosphorylated p38 (p-p38), and actin. (D) Levels of p-p38 were higher in the hippocampi of LPS-infused rats than in that of the aCSF-infused rats. *, p < 0.05. The relative expression levels of all proteins were determined by densitometry and normalized to actin.

Figure 4

Confocal photomicrographs of dual staining for p-Erk1/2 and markers of cellular location in the CA1, CA3, and DG of the aCSF- and LPS-infused rats. Photomicrographs showed the overlap between p-Erk1/2-positive signals and NeuN, a neuronal marker (A), or CD11b, a microglial marker (B), or GFAP, an astrocyte marker (C). The p-Erk1/2-positive cells were observed mostly in neurons and also in microglia or astrocytes. No differences between the LPS- and the aCSF-infused rats were observed in p-Erk1/2 levels expressing in the CD11b-positive microglia of the hippocampus (D). Levels of p-Erk1/2 expressing in the GFAP-positive astrocyte were significantly increased in the hippocampal CA1, CA3, and DG areas of the LPS-infused rats compared with the aCSF-infused rats (E). *, p < 0.01.

Figure 5

Confocal photomicrographs of dual staining for p-p38 and markers of cellular location in the CA1, CA3, and DG of the aCSF- and LPS-infused rats. Photomicrographs showed an overlap between p-p38-positive signals and NeuN, a neuronal marker (A), or CD11b, a microglial marker (B), or GFAP, an astrocyte marker (C). The p-p38-positive cells were observed mostly in neurons and also in glial cells. Compared with the aCSF-infused rats, levels of p-p38 expressing in the CD11b-positive microglia were increased only in the hippocampal CA3 region of the LPS-infused (D; *, p < 0.05) and levels of p-p38 expressing in the GFAP-positive astrocyte were significantly increased only in the hippocampal CA1 region of the LPS-infused rats (E; *, p < 0.01).