Tumor necrosis factor-α-mediated threonine 435 phosphorylation of p65 nuclear factor-κB subunit in endothelial cells induces vasogenic edema and neutrophil infiltration in the rat piriform cortex following status epilepticus

Abstract

Background: Status epilepticus (SE) induces severe vasogenic edema in the piriform cortex (PC) accompanied by neuronal and astroglial damages. To elucidate the mechanism of SE-induced vasogenic edema, we investigated the roles of tumor necrosis factor (TNF)-α in blood-brain barrier (BBB) disruption during vasogenic edema and its related events in rat epilepsy models provoked by pilocarpine-induced SE.

Methods: SE was induced by pilocarpine in rats that were intracerebroventricularly infused with saline-, and soluble TNF p55 receptor (sTNFp55R) prior to SE induction. Thereafter, we performed Fluoro-Jade B staining and immunohistochemical studies for TNF-α and NF-κB subunits.

Results: Following SE, most activated microglia showed strong TNF-α immunoreactivity. In addition, TNF p75 receptor expression was detected in endothelial cells as well as astrocytes. In addition, only p65-Thr435 phosphorylation was increased in endothelial cells accompanied by SMI-71 expression (an endothelial barrier antigen). Neutralization of TNF-α by soluble TNF p55 receptor (sTNFp55R) infusion attenuated SE-induced vasogenic edema and neuronal damages via inhibition of p65-Thr435 phosphorylation in endothelial cells. Furthermore, sTNFp55R infusion reduced SE-induced neutrophil infiltration in the PC.

Conclusion: These findings suggest that impairments of endothelial cell functions via TNF-α-mediated p65-Thr 485 NF-κB phosphorylation may be involved in SE-induced vasogenic edema. Subsequently, vasogenic edema results in extensive neutrophil infiltration and neuronal-astroglial loss.

Figure 1

Expression of TNF-α and TNF receptor in the PC 12 hr-post SE. (A) TNF-α immunoreactivity in Ox-42-positive microglia (arrows). (B) TNFp55R expression in astrocytes (arrows). (C-D) TNFp75R expression in astrocytes as well as endothelial cells (arrows). Bar = 25 (A-D) μm.

Figure 2

Effect of sTNFp55R infusion on SE-induced serum-protein extravasation and neuronal damage. (A-D) Serum-protein extravasation and FJB-positive neuronal damages in the PC 3 days after SE. Compared to saline-infused animals, serum-protein extravasation and FJB-positive neuronal damage is markedly ameliorated in sTNFp55R-infused animals. Bars = 400 (A and C) and 50 (B and D) μm. (E) Quantitative analyses of serum-protein extravasation and FJB-positive neuronal damage in the PC 3 days after SE (mean ± S.E.M). Significant differences from saline-treated animals, *p < 0.05.

Figure 3

Effect of sTNFp55R infusion on NF-κB phosphorylation in astrocytes 12 hr after SE. In 12 hr-post SE animals of the saline-infused group (A, C, E and G), astrocytes show p65-Ser276 (A), p65-Ser311 (C), p65-Ser529 (E), and p65-Ser536 (G) phosphorylation (arrows). sTNFp55R infusion (B, D, F and H) effectively reduces p65-Ser276 (B) and p65-Ser311 (D) phosphorylation, while it does not affect p65-Ser529 (F) and p65-Ser536 (H) phosphorylation (arrows). Bar = 12.5 μm.

Figure 4

Quantitative analyses of the effect of sTNFp55R infusion on NF-κB phosphorylation and SMI-71 expression. (A) Quantitative analysis of NF-κB phosphorylation 12 hr after SE (mean ± S.E.M). Significant differences from saline-infused animals, *p < 0.05. (B) Quantitative analysis of SMI-71 expression 1 day after SE (mean ± S.E.M). Significant differences from saline-infused animals, *p < 0.05. (C) Linear regression analysis between p65-Thr435 phosphorylation and SMI-71 in the PC following SE.

Figure 5

Effect of sTNFp55R infusion on p65-Thr435 phosphorylation in endothelial cells following SE. (A-B) Inhibition of p65-Thr435 phosphorylation by sTNFp55R infusion 12 hr after SE. p65-Thr435 phosphorylation is rarely observed in astrocytes (arrows). (C) Endothelial p65-Thr435 phosphorylation in non-SE animals. (D-E) Endothelial p65-Thr435 phosphorylation in saline-infused animals 1 day after SE. p65-Thr435 phosphorylation is enhanced, while SMI-71 expression is reduced in GLUT-1-positive endothelial cells (arrows). (F-G) Endothelial p65-Thr435 phosphorylation in sTNFp55R-infused animal 1 day after SE. sTNFp55R infusion effectively reduces p65-Thr435 and preserves SMI-71 expression in GLUT-1-positive endothelial cells (arrows). Bars = 12.5 (A-D) and 25 (E-G) μm.

Figure 6

Effect of sTNFp55R infusion on neutrophil infiltration and MIP-2 expression following SE. (A) Neutrophil infiltration in vasogenic edema lesion 1 day after SE. (B) MIP-2 expression in the PC 1 day after SE. (C) Astroglial expression of MIP-2 (arrows). Bars = 12.5 (A and C) and 150 (B) μm. (D) Quantitative analysis of neutrophil infiltration 1 day after SE (mean ± S.E.M). Significant differences from saline-infused animals, *p < 0.05. (E) Linear regression analysis between the number of infiltrated neutrophils/area in the vasogenic edema region and the area of vasogenic edema in the PC. (F) Quantitative analysis of the number of MIP-2 positive cells per the unit area of vasogenic edema 1 day after SE (mean ± S.E.M). There is no difference in the number of MIP-2-positive cells per unit area of vasogenic edema between sTNFp55R-infused animals and saline-infused animals. (G) Linear regression analysis between the number of MIP-2 positive cells per unit area in vasogenic edema region and the area of vasogenic edema in the PC.