- Top of page
Summary: Purpose: We investigated the activation of microglia and astrocytes, induction of cytokines, and hippocampal neuronal damage, 4 and 24 h after kainic acid–induced status epilepticus (SE) in postnatal day (PN) 9, 15, and 21 rats.
Methods: Limbic seizures were induced by systemic injection of kainic acid. Glia activation and neuronal cell loss were studied by using immunocytochemistry and Western blot. Cytokine expression was analyzed by reverse transcriptase–polymerase chain reaction (RT-PCR) followed by Southern blot quantification.
Results: After SE onset, hippocampal glia activation, cytokine expression, and neuronal damage are all age-dependent phenomena. In the hippocampus, neuronal injury occurs only when cytokines are induced in glia, and cytokine synthesis precedes the appearance of degenerating neurons. Neuronal injury is more pronounced when interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) are produced in addition to IL-1β.
Conclusions: This study shows that cytokine induction in rat brain after sustained seizures is age dependent, and it is associated with the appearance of cell injury.
Proinflammatory cytokines and related inflammatory and antiinflammatory molecules are rapidly overexpressed by glia in adult rodent hippocampus in various models of limbic seizures (1–3). In the adult brain, cytokines are expressed to a larger extent when seizures are associated with neuronal damage, suggesting a link between cytokine production and the occurrence of neuronal injury (1,4). Thus it has been shown that in mature rat brain, proinflammatory cytokines, and in particular interleukin (IL)-1β, act as modulators of various forms of neurodegeneration (5–7).
In humans and in experimental models of epilepsy, seizure susceptibility and the associated neuronal damage are age-dependent phenomena, changing dramatically during postnatal development. In the first 2 postnatal weeks, the brain is more prone to seizure activity, but it is relatively resistant to irreversibile seizure-induced damage as compared with adult brain (8–12).
The factors implicated in the occurrence of age-dependent seizure-related injury are still unclear. We tested the hypothesis that glia activation and the subsequent cytokine production may be involved in this process. We used rats at distinct postnatal (PN) ages (PN9, PN15, PN21), known to have age-specific seizure susceptibility and neuronal injury after kainic acid–induced seizures.
- Top of page
These data show an age-dependent activation of microglia and astrocytes and induction of cytokines in the rat hippocampus after SE. In PN9 rats, little activation of both glia populations was found, and we did not observe any cytokine neosynthesis. In PN15 and PN21 rats, glia were markedly activated. IL-1β is the only cytokine induced in PN15 rats, whereas in PN21 rats, all cytokines were produced, similar to adult rat brain (3). The limited activation of microglia and astrocytes and the lack of cytokine expression after SE in PN9 rats are not due to the inability of glia to react to injurious stimuli at this age. Thus after hypoxic or ischemic events in PN9 rats, microglia, astrocytes, and cytokines are activated (17–19).
The analysis of neuronal cell loss showed that in immature brain, neuronal injury occurs only when cytokines (and in particular IL-1β) are induced, and their synthesis precedes the appearance of neuronal damage. In PN15 and PN21 rats, IL-1β mRNA was enhanced 4 h after SE onset, thus before neurodegenerating neurons appear (i.e., 24 h after seizures). Neuronal injury was larger in PN21 rats, when all proinflammatory cytokines were produced. These data suggest that IL-1β must act synergistically with other cytokines to induce significant neuronal injury. Accordingly, in neuronal cell cultures, although IL-1β and TNF-α alone were not toxic, their combination caused pronounced neuronal injury (20). If the induction of inflammatory processes by seizures plays a role in age-dependent neuronal death, at least two pathways related to cytokine-activated signal transduction may be involved: the induction of iNOS and cyclooxygenase (COX)-2. Thus these pathways are developmentally regulated, as IL-1β induction by seizures. After kainic acid-induced seizures, no COX-2 mRNA induction was found before PN14 (21). Despite prolonged seizures, immature rats do not show increase in reactive oxygen species as in adults (22).
The mechanism by which IL-1β may contribute to neuronal injury also may be related to the functional interaction between IL-1β–receptor type I and N-methyl-d-aspartate (NMDA) receptors. We recently found that the exposure of hippocampal neurons to IL-1β enhances the phosphorylation of the NR2A/2B subunit of the NMDA receptor (23). This receptor modification upregulates NMDA activity, thus increasing its channel-gating properties (24). The age-dependent induction of IL-1β during seizures may contribute to the occurrence of neuronal damage by inducing posttranslational modification in NR2 subunit.
Experiments are in progress to validate these hypotheses. Further insights into this field will be of value for better understanding molecular pathways critically involved in seizure-associated neuronal damage.