Increased glial glutamate transporter EAAT2 expression reduces epileptogenic processes following pilocarpine-induced status epilepticus

Abstract

Several lines of evidence indicate that glutamate plays a crucial role in the initiation of seizures and their propagation; abnormal glutamate release causes synchronous firing of large populations of neurons, leading to seizures. In the present study, we investigated whether enhanced glutamate uptake by increased glial glutamate transporter EAAT2, the major glutamate transporter, could prevent seizure activity and reduce epileptogenic processes. EAAT2 transgenic mice, which have a 1.5-2 fold increase in EAAT2 protein levels as compared to their non-transgenic counterparts, were tested in a pilocarpine-induced status epilepticus (SE) model. Several striking phenomena were observed in EAAT2 transgenic mice compared with their non-transgenic littermates. First, the post-SE mortality rate and chronic seizure frequency were significantly decreased. Second, neuronal degeneration in hippocampal subfields after SE were significantly reduced. Third, the SE-induced neurogenesis and mossy fiber sprouting were significantly decreased. The severity of cell loss in epileptic mice was positively correlated with that of mossy fiber sprouting and chronic seizure frequency. Our results suggest that increased EAAT2 expression can protect mice against SE-induced death, neuropathological changes, and chronic seizure development. This study suggests that enhancing EAAT2 protein expression is a potential therapeutic approach.

Figure 1

Evaluation of EAAT2 transgenic mice at the acute epileptic stage. After injections of pilocarpine (290 mg/kg, i.p.), the first seizure activity at each stage that mouse reached was noted with the corresponding time. The scale to evaluate seizure stage is described in the Methods section (Racine, 1972). A total of 240 EAAT2 mice and 268 wild-type littermates (WT) were tested in 28 independent groups. (A) The maximal seizure activity of each animal is presented. In 28 independent groups, stage II and below: 6.4±1.5% (WT) vs. 10.3±2.2% (EAAT2); stage III: 2.5±1.1% (WT) vs. 7.2±1.9% (EAAT2); stage IV: 7.2±2.4% (WT) vs. 8.2±2.3% (EAAT2); stage V: 5.2±1.6% (WT) vs. 4.2±1.3% (EAAT2); and SE: 78.7±2.8% (WT) vs. 70.0±3.4% (EAAT2). (B) Latency was defined as the time interval between the injection and the indicated stage. Neither the latency period of stage III seizure nor the latency up until SE (p=0.60 and 0.07, respectively; Mann-Whitney rank sum test) was significantly affected. (C) A reduction of acute mortality rate, defined by mean death rate within 7 days after injection, was detected in EAAT2 mice (n=28 groups, 44.0±3.8% (WT) vs. 25.2±3.6% (EAAT2)).

Figure 2

Reduced chronic seizure frequency and total mortality rate in EAAT2 mice. Recording of spontaneous seizures started at 4 weeks after SE for 2 weeks (8 hr each day). Stage V seizure frequencies (i.e., average stage V counts per 8 hr according to Racine’s scale) were determined in each mouse and compared between 22 wild-type (WT) and 20 EAAT2 mice. (A) Percentages of mice that developed fewer than 1, 1~2, 2~3, or over 3 stage V seizures per 8 hr were: 27.3% (WT) vs. 40% (EAAT2), 27.3% (WT) vs. 40% (EAAT2), 22.7% (WT) vs. 10% (EAAT2), and 13.6% (WT) vs. 5% (EAAT2), respectively. 9.1% of WT and 5% of EAAT2 mice died during the recording weeks. (B) SE-induced chronic seizures were less severe in EAAT2 mice (1.15±0.19, n=23; *p<0.05, t test) when compared with wild-type littermates (2.01±0.26, n=23). (C) Stage V seizure frequencies during each day of the recording. (D) A reduction of total mortality rate, defined by mean death rate in 17 independent groups within 8 weeks after injection, was detected in EAAT2 mice.

Figure 3

Reduced acute neurodegeneration in EAAT2 mice after SE. Coronal sections from eight sets of brains at 3 days after SE were stained with Fluoro-Jade C. Positively labeled degenerating cells were counted under the microscope in several hippocampal subfields: CA1, CA3, dentate gyrus (DG) and hilus. (A) Representative images of hippocampal neurodegeneration. Scale bar, 50 μm. (B) FJC-positive cells were significantly less abundant in EAAT2 mice. *p<0.05, Turkey test.

Figure 4

Hippocampal damage was significantly reduced in epileptic EAAT2 mice. Coronal sections from eight sets of brains at 8 weeks after SE were stained with cresyl violet for neuronal loss. Live neurons in the hilus were counted under the microscope. (A) Representative hippocampal damage were presented. Arrow heads indicate severe neuronal loss. Scale bar, 50 μm. (B) Compared to control mice, both wild-type SE and EAAT2 SE mice had significant neuronal loss in hilus (*p<0.05, Turkey test). More importantly, increased EAAT2 significantly attenuated the damages (*p<0.05, Turkey test). (C) A strong positive correlation was detected between hilar cell loss and chronic seizure frequencies. Hilar cell loss was defined as (cell number in control hilus – live cell number in SE hilus)/cell number in control hilus×100. Correlation coefficient=0.69, p<0.01, Spearman rank order correlation test.

Figure 5

Reduced hippocamal neurogenesis in epileptic EAAT2 mice. Coronal sections from six sets of brains harvested at 12 days after SE were immunostained with a doublecortin (DCX) antibody for newly born granule cells. (A-C) Representative images of DCX-stained dentate gyrus. DCX-positive immature neurons appeared at the border between the subgranular zone and the granular layer. (D-F) Representative images of DCX-positive cells with DCX-labeled hilar basal dendrites (arrow heads). (G) There was a significant reduction of newly born DCX-positive neurons in hippocampal dentate gyrus in EAAT2 mice (*p<0.05, Turkey test). (H) The mean percentage of DCX-labeled cells with hilar basal dendrites was significantly increased in both wild-type mice (56.0±3.5%, ***p<0.001, Holm-Sidak method) and EAAT2 littermates (33.7±3.1%, **p<0.01, Holm-Sidak method) compared to control (19.8±2.0%), but the increase was significantly lower in EAAT2 mice compared to WT littermates (***p<0.001, Holm-Sidak method). Scale bar, 25 μm.

Figure 6

Aberrant mossy fibers sprouting decreased in the dentate gyrus of epileptic EAAT2 mice. Timm staining was performed to detect mossy fiber spouting with mice harvested at 8 weeks after SE. Sections were scored by 2 investigators independently on a scale of 0~5 (Fig. S1). (A-C) Representative images of Timm-stained dentate gyrus. scale bar, 50 μm. (D) EAAT2 mice exhibited significantly lower Timm scores compared to wild-type littermates (n=16, *p<0.05, Dunn’s method). (E) A moderate but not significant positive correlation (correlation coefficient=0.45, p=0.052, Spearman rank order correlation test) was detected between the severity of aberrant mossy fiber sprouting and chronic seizure frequency. (F) A strong positive correlation was detected between the hilar cell loss and the Timm score (correlation coefficient=0.59, p<0.001, Spearman rank order correlation test).