Protective role of S-nitrosoglutathione (GSNO) against cognitive impairment in rat model of chronic cerebral hypoperfusion

Abstract

Chronic cerebral hypoperfusion (CCH), featuring in most of the Alzheimer's disease spectrum, plays a detrimental role in brain amyloid-β (Aβ) homeostasis, cerebrovascular morbidity, and cognitive decline; therefore, early management of cerebrovascular pathology is considered to be important for intervention in the impending cognitive decline. S-nitrosoglutathione (GSNO) is an endogenous nitric oxide carrier modulating endothelial function, inflammation, and neurotransmission. Therefore, the effect of GSNO treatment on CCH-associated neurocognitive pathologies was determined in vivo by using rats with permanent bilateral common carotid artery occlusion (BCCAO), a rat model of chronic cerebral hypoperfusion. We observed that rats subjected to permanent BCCAO showed a significant decrease in learning/memory performance and increases in brain levels of Aβ and vascular inflammatory markers. GSNO treatment (50 μg/kg/day for 2 months) significantly improved learning and memory performance of BCCAO rats and reduced the Aβ levels and ICAM-1/VCAM-1 expression in the brain. Further, in in vitro cell culture studies, GSNO treatment also decreased the cytokine-induced proinflammatory responses, such as activations of NFκB and STAT3 and expression of ICAM-1 and VCAM-1 in endothelial cells. In addition, GSNO treatment increased the endothelial and microglial Aβ uptake. Additionally, GSNO treatment inhibited the β-secretase activity in primary rat neuron cell culture, thus reducing secretion of Aβ, suggesting GSNO mediated mechanisms in anti-inflammatory and anti-amyloidogenic activities. Taken together, these data document that systemic GSNO treatment is beneficial for improvement of cognitive decline under the conditions of chronic cerebral hypoperfusion and suggests a potential therapeutic use of GSNO for cerebral hypoperfusion associated mild cognitive impairment in Alzheimer's disease.

Fig. 1

Effect of GSNO treatment on spatial learning and memory deficits and Aβ accumulation in rat model with permanent bilateral common carotid artery occlusion (BCCAO). A) Time needed to reach the hidden platform (latency in seconds) in the Morris maze by vehicle or S-nitrosoglutathione (GSNO; 50 μg/kg, i.p. for 2 months)-treated rats (n = 6) (i). Data were also further analyzed with respect to acquisition (averages for day 1–3 of training) (ii), consolidation (averages for day 4–6 of training) (iii), and combined (overall) performance (average distance and latency for days 1–6) (iv). Following the training phase, the hidden platform in the target quadrant was removed and the time spent in the target quadrant (permanence time) was measured for analysis of spatial memory function (v). B) Rats were dissected for the extraction of brain cortices from above experimental and control animals for analysis of Aβ₄₀ (i) and Aβ₄₂ (ii) levels by sandwich ELISA assay. The data were analyzed for mean values and standard error for all experiments; *p < 0.05, **p < 0.01, and ***p < 0.001 for comparison to sham control group; #p < 0.05 for comparison to vehicle treated BCCAO group; NS for no-significant changes compared to vehicle treated BCCAO group.

Fig. 2

Anti-inflammatory activity of GSNO. A) The effect of GSNO treatment on the expression of vasoinflammatory markers (ICAM-1 and VCAM-1) in the brains of rats treated with permanent BCCAO was examined by real time quantitative PCR and by immunofluorescence staining of paraffin sections as described in method section. In the quantitative PCR analysis, Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA was quantified as an internal control. B) Further, to study the mechanism of GSNO in endothelial inflammatory signaling events, the cultured mouse brain endothelial cells (bEND3) were pretreated with GSNO (250 μM) for 3 h and then treated with cytokine mix (Cyto; TNFα, IL1β, and IFNγ; 25 ng/ml each). Following incubation for 0.5 h with cytokine mix, the activities of nuclear factor κ-B (NFκB) were analyzed by western analysis for nuclear p65 and cytoplasmic IκB and by gel-shift assay of NFκB DNA binding activity (i). The activities of signal transducer and activator of transcription 1 (STAT1) and STAT3 were analyzed by western analysis of phospho-(Tyr701)-STAT1 and phospho-(Tyr705)-STAT3 (ii). C) Following GSNO cytokine treatment, the mRNA (i) and protein (ii) levels of ICAM-1 and VCAM-1 were analyzed. In these experiments, GSNO was pretreated for 3 h before treatment with cytokine mix. The mRNA and proteins were extracted 6 h and 12 h after cytokine treatment. The data was analyzed for mean values and standard error for all experiments; *p < 0.05 and **p < 0.01 for comparison to vehicle (VHC) treated control (CTL) group; ⁺p <0.05 and ⁺⁺p <0.01 for comparison to cytokine (Cyto) treated group.

Fig. 3

Promotion of Aβ uptake by GSNO in endothelial and microglial cells. A) To assess the effect of GSNO on dynamin-2-mediated endothelial Aβ uptake, the mammalian expression vector expressing human dynamin-2 was transfected to bEND3 murine brain endothelial cells, and then the cellular levels of dynamin-2 and the endothelial uptake of fluorescence-labeled Aβ (FAM-Aβ) were analyzed (i). To examine whether GSNO is able to S-nitrosylate dynamin-2, the bEND3 cells were treated with GSNO (250 μM/3 h) and subjected to biotin-switch assay. The resulting S-biotinylated (S-nitrosylated) proteins in whole cell lysates were detected by western analysis using antibody specific to biotin (ii). To assess the involvement of S-nitrosylation mechanism on GSNO-induced Aβ uptake, the wild type, C86A mutant, or C607A mutant dynamin-2 was transfected to bEND3 cells and biotin-switch assay and Aβ uptake assay were performed following the GSNO treatment (250 μM/3 h). Following the pull-down of S-biotinylated proteins with avidin-conjugated agarose beads, the levels of co-precipitated dynamin-2 (DNM-2) were analyzed by western blot. The levels of dynamin-2 in whole cell lysates were analyzed to examine the total dynamin-2 levels. The levels of β-actin in whole cell lysates were used for internal loading control. B) To examine the effect of GSNO on Aβ uptake, FAM-Aβ uptake assay was performed in bEND3 and BV2 microglial cells. Before the uptake assay, FAM-Aβ peptides were preincubated with basal media (fresh DMEM) or astroglial conditioned media (CM), which included lipidated ApoE (see Supplementary Data 1). To examine whether CM-mediated Aβ uptake is mediated through LRP1, the Aβ uptake assay was performed following the treatment of cells with LRP1 inhibitor, receptor-associated protein (Rap) (i). To examine whether GSNO enhances basal media (BM) or CM-mediated Aβ uptake, the cells were pretreated with GSNO (250 μM/3 hrs) prior to performance of Aβ uptake assay (ii). To examine whether GSNO treatment enhances the endothelial uptake of Aβ₄₀ as well as Aβ₄₂, the bEND3 cells were incubated with GSNO and then subjected to Aβ uptake assay using FAM-Aβ₄₀ and FAM-Aβ₄₂ peptides which were pre-incubated with CM (iii). Aβ uptake was also visualized under a fluorescent microscope (iv). C) In addition to bEND3 cells, the Aβ uptake assay was also performed on BV2 microglia under the same experimental conditions. The data were analyzed for mean values and standard error for all experiments; *p < 0.05, **p < 0.01, and ***p < 0.001 for comparison to mock control (empty), BM, or vehicle (VHC) treated group; ⁺p < 0.05 for comparison to CM treated group.

Fig. 4

Inhibitory effect of GSNO on neuronal Aβ production. A) To examine the effect of GSNO treatment on neuronal Aβ production, primary cultured rat cortical neurons were treated with increasing doses of GSNO for 24 h, and the levels of Aβ₄₀ (i) and Aβ₄₂ (ii) secreted into the media were determined by ELISA. In addition, the effects of glutathione ethylester (GSHEE), a cell permeable GSH analogue, and S-nitroso-N-acetylpenicillamine (SNAP), a S-nitroso donor, on Aβ₄₂ secretion (iii) and neuronal survival (iv and v) were compared with the equal concentrations of GSNO (200 μM). For the assay of neuronal survival, lactate dehydrogenase (LDH) release assay (iv) and MTT assay (v) were performed. B) To examine the effect of GSNO treatment on the regulation of α- (i) or β-secretase (ii) activity, the neuron cells were treated with increasing doses of GSNO and α- or β-secretase activity assay was performed by method as described in materials and methods. Further, the effect of equimolar concentrations of GSNO, decomposed GSNO (dsGSNO), SNAP, GSHEE, and nitrite (200 μM each) on neuronal β-secretase activity was assayed (iii). C) To examine whether GSNO is able to S-nitrosylate neuronal BACE1, the neuron cells treated with increasing concentrations of GSNO for 3 h were subjected to biotin-switch assay. The resulting S-biotinylated (S-nitrosylated) proteins in whole cell lysates were detected by western analysis using antibody specific to biotin. Following the pull-down of S-biotinylated proteins with avidin-conjugated agarose beads, the levels of co-precipitated BACE1 were analyzed by western blot. The levels of BACE1 in whole cell lysates were analyzed to examine the total BACE1 levels. The levels of β-actin in whole cell lysates were used for internal loading control. The data were analyzed for mean values and standard error for all experiments; *p < 0.05, **p < 0.01, and ***p < 0.001 for comparison to vehicle (VHC) treated control (CON) group.