Epilepsy-induced Changes in Signaling Systems of Human and Rat Postsynaptic Densities


Address correspondence and reprint requests to Dr. U. Wyneken at Facultad de Medicina, Universidad de los Andes, San Carlos de Apoquindo 2200, Las Condes, Santiago 6782468, Chile. E-mail: uwyneken@uandes.cl


Summary:  Purpose: To study seizure-induced changes in signaling proteins present in postsynaptic densities (PSDs) isolated from human epileptic neocortex and from rat cortex in which seizures were induced by injection of kainic acid.

Methods: We performed Western blot analysis of signaling proteins in PSDs isolated from cortical tissue.

Results: Seizures induce a strong upregulation of TrkB, the receptor for brain-derived neurotrophic factor (BDNF), whereas components of the N-methyl-d-aspartate (NMDA)-receptor complex are downregulated in both human and rat PSDs.

Conclusions: These data show that long-term changes in PSD composition occur as a consequence of epileptic seizure activity.

Epilepsy is a brain disorder with a variable age-adjusted prevalence ranging from 0.4 to 0.8%(1). About 20% of the epilepsy patients have seizures that are not adequately controlled by antiepileptic drugs (AEDs). It is commonly assumed that an imbalance between excitation and inhibition initiates seizure activity (2,3), although cellular and molecular mechanisms underlying epileptogenesis and maintenance of neuronal hyperexcitability during epileptic seizures are poorly understood. Moreover, changes in synaptic physiology induced by seizures may represent either adaptive or epileptogenic mechanisms. Understanding the mechanisms involved in generating and maintaining synaptic changes could provide novel therapeutic approaches for the management of the disease and its consequences.

The postsynaptic density (PSD) is a specialization of the postsynaptic membrane of excitatory synapses that contains the glutamate receptor subtypes and numerous other signaling molecules involved in the generation and regulation of the synaptic response to the neurotransmitter l-glutamate (4). It has been suggested that PSDs are dynamic structures (5), and it has been shown that the thickness and composition changes during synaptic activity (6). This activity-induced synaptic plasticity may cause long-lasting changes in the excitatory response to glutamate.

In a previous study we showed, by using an animal model in which seizure activity was induced by systemic application of the convulsant kainate, that extensive rearrangement in PSDs of forebrain chemical synapses occurs within 4 h after epilepsy-like seizure activity (7).

To study synaptic signaling systems known to be involved in excitatory synaptic plasticity phenomena, we focused our interest on two synaptic signaling systems that are integral components of the PSD: (a) N-methyl-d-aspartate (NMDA) receptors (NMDARs), which are anchored to the PSD by the scaffolding protein SAP90/PSD-95 (8) and play a crucial role in synaptic plasticity and excitotoxicity (9); and (b) TrkB, the high-affinity receptor for brain-derived neurotrophic factor (BDNF).

Contradictory roles have been assigned to BDNF: although it is proposed that BDNF upregulation by seizure activity constitutes a neuroprotective function (10), it also has been proposed to be epileptogenic (11). We analyzed changes in TrkB and in constituents of the NMDAR complex (i.e., NMDAR subunits NR1 and NR2A or B and SAP90/PSD-95), in PSDs of epilepsy patients and compared them with control PSDs isolated from normal cortical tissue that is inevitably resected during tumor surgery. To verify that these changes are a consequence of seizure activity, we studied the time course of TrkB and NMDAR-complex changes in rat PSDs after kainate-induced seizures.


Ten human neocortical specimens used for PSD isolation were obtained at the Asenjo Neurosurgery Institute, either from the focus of epilepsy patients whose seizures were refractory to pharmacotherapy (cases 6 to 10), that responded to therapy after one convulsive episode (cases 4 and 5), or from control patients (cases 1 to 3) who had never had seizures. In control cases, tissue was obtained from the tumor margins during brain tumor resection. General preoperative patient evaluation included detailed clinical histories, neurologic examination, scalp electroencephalogram (EEG), cerebral computed tomography (CT), and magnetic resonance imaging (MRI). The presurgical study for patients with epilepsy refractory to AEDs included extensive neuropsychological tests, interictal and ictal scalp–sphenoidal EEG, video-EEG, and intraoperative cortical EEG. All surgical procedures applied were those currently approved by the Asenjo Neurosurgery Institute's Ethical Committee. For the use of part of the biopsy for additional studies, informed consent was obtained from the patients or their relatives. The characteristics of the patients and the cortical areas taken for the PSD preparation are shown in Table 1.

Table 1.  Clinical characteristics and changes of protein immunoreactivity in PSDs
 Controls*Epileptic patients
Case number12345678910
  • *

    : control values obtained from four independent immunoblots were set as 1.0; NS, no seizures; GTC: generalized tonic–clonic seizures; AT, atonic seizures; P, partial seizures; ABS, absences; PHT, phenytoin; PB, phenobarbital; CBZ: carbamazepine; VPA, valproic acid; TU tumor; CD, cortical displasia; ♂, male; ♀, female; F, frontal; T, temporal; P, parietal; O, occipital.

Protein content in PSD:
 Fold change from control
Trk B1.
NR2 A/B0.
PSD 95/SAP-900.
Age (yr)/Sex41/♀15/♂10/♂24/♂5/♂4/♂11/♀0.6/♂7/♂10/♀
Seizure frequencyNSNSNS1 event1 event40/day1/wk2/day20/day2/wk
Localization of resectionFFFTTTPOTT

Sprague–Dawley rats were injected intraperitoneally with saline (C group) or kainate at 13 mg/kg body weight (S group). Cortical PSDs were isolated from rats that developed status epilepticus, or stages 5–6 according to Zhang et al. (12), after 2, 6, or 24 h after injection (n = 4 for each time point). In contrast, in rats killed 1 h after injection, full seizures did not develop; they reached only stages 3–4 (12). Human and rat PSD isolation was performed as described previously (7). Equal amounts of PSD proteins were subjected to Western blot analysis with antibodies against NR1 (Pharmingen), NR2A/B (Chemicon), TrkB, and PSD-95 (Transduction Laboratories). Rats were treated by following the recommendations of the Chilean National Research Agency Fondecyt, and the experimental protocols were approved by the Ethics Committee of Universidad de los Andes.

Statistical analysis of densitometric quantification of the Western blots was performed (one-sample Student's t test) by using InStat3 obtained from GraphPad (San Diego, CA, U.S.A.). For statistical comparisons of fractional changes ranging from 0 to 1, the arc sine transformation was applied.


Quantitative densitometry of the immunoblots showed that the TrkB content in epilepsy patients increased 2.6 ± 0.26 (fold change ± SEM), whereas NR1, NR2A/B, and PSD-95 were downregulated to 0.7 ± 0.01, 0.7 ± 0.06, and 0.6 ± 0.07 of controls, respectively (Fig. 1). In rats, TrkB in PSDs increased by a factor of 3.3 ± 0.70 at 2 h after kainate injection (i.e., ∼30 min after the beginning of seizures), and to 6.2 ± 0.20 at 6 h (Fig. 2). This large increase was maintained for ≥24 h. In contrast, NR1, NR2A/B, and PSD-95 levels in PSDs decreased to 0.6 ± 0.07, 0.4 ± 0.05, and 0.6 ± 0.06, respectively, when measured 24 h after kainate injection. For PSD-95, a significant decrease to 0.8 ± 0.01 of control levels was already seen at 6 h after kainate injection.

Figure 1.

Upper panel: Examples of immunoblots of TrkB, NR1, NR2A/B and PSD95 in human PSDs of case 3 (no seizures), and two epileptic cases. Lower panel: Mean fold change ± S.E.M. (scale bars) as compared to controls (n=3) of indicated immunoreactivities in PSDs obtained from 7 epileptic patients. (* = p < 0.05, ** = p < 0.01).

Figure 2.

Time course of changes in TrkB, NR1, NR2A/B and PSD95 in rat control (C) and kainate-injected (S) PSDs obtained 1, 2, 6 and 24 hours after injection. Upper panel: Representative immunoblots in C- and S-PSDs. Lower panel: Mean fold change ± S.E.M. (scale bars) from controls of the same immunoreactivities in S-PSDs. (n=4; *= p<0.05, **= p<0.01).


These results show that significant and selective changes in postsynaptic signaling structures occur in response to seizure activity. These changes occurred in the human neocortex and were independent of the age of the patient, as well as in the adult rat telencephalon (neocortex plus hippocampus) after kainate-induced status epilepticus. The fact that in the rat model the increase of TrkB is quite rapid favors the hypothesis that BDNF triggers long-lasting synaptic modifications (13). To date, the function of specific anchoring of TrkB receptors to postsynaptic structures is unknown. It remains to be established whether these changes favor epileptogenesis or neuroprotection, allowing the design of therapeutic approaches to strengthen or inhibit such changes. We suggest that the observed changes in PSD composition could represent neuronal adaptive mechanisms in response to a severe brain insult, as decreases in the NMDAR complex may help to prevent the excitotoxicity related to glutamatergic overstimulation, whereas TrkB upregulation may favor neuronal survival. This idea is supported by the fact that, in patients with chronic epilepsy and in the rat model, only 24 h after seizures, the same patterns of protein changes were observed.

Acknowledgment: We thank neurosurgeons Drs. P. Aros, F. Ayach, and A. Zuleta for their help in specimen collection. This work was supported by grants from the Volkswagen Foundation, Fonds der Chemischen Industrie, Fondecyt 1020257, and Universidad de los Andes.