Recurrent neonatal seizures result in long-term increases in neuronal network excitability in the rat neocortex

Abstract

Neonatal seizures are associated with a high likelihood of adverse neurological outcomes, including mental retardation, behavioral disorders, and epilepsy. Early seizures typically involve the neocortex, and post-neonatal epilepsy is often of neocortical origin. However, our understanding of the consequences of neonatal seizures for neocortical function is limited. In the present study, we show that neonatal seizures induced by flurothyl result in markedly enhanced susceptibility of the neocortex to seizure-like activity. This change occurs in young rats studied weeks after the last induced seizure and in adult rats studied months after the initial seizures. Neonatal seizures resulted in reductions in the amplitude of spontaneous inhibitory postsynaptic currents and the frequency of miniature inhibitory postsynaptic currents, and significant increases in the amplitude and frequency of spontaneous excitatory postsynaptic currents (sEPSCs) and in the frequency of miniature excitatory postsynaptic currents (mEPSCs) in pyramidal cells of layer 2/3 of the somatosensory cortex. The selective N-methyl-D-aspartate (NMDA) receptor antagonist D-2-amino-5-phosphonovalerate eliminated the differences in amplitude and frequency of sEPSCs and mEPSCs in the control and flurothyl groups, suggesting that NMDA receptors contribute significantly to the enhanced excitability seen in slices from rats that experienced recurrent neonatal seizures. Taken together, our results suggest that recurrent seizures in infancy result in a persistent enhancement of neocortical excitability.

Figure 1

Different seizure susceptibility of somatosensory cortex in controls and rats with a prior history of neonatal seizures. Extracellular field potentials were recorded from L2/3 of somatosensory cortex in slices from P20–30 and P60–80 control and flurothyl-treated rats. A. Example of interictal (a) and ictal-like (b) activity evoked by 10 μM gabazine in a slice from a flurothyl-treated rat at P71. B. Summary graphs show probability of SLA induction by gabazine in L2/3 of somatosensory cortex in slices from control (light grey) and flurothyl-treated (dark grey) rats at P20–30 (left) and P60–80 (right). Number of slices used for analysis shown in parenthesis. Note that in both age groups there was a significant increase in SLA probability in flurothyl-treated rats.

Figure 2

Propagation of gabazine-induced seizures through layers of somatosensory cortex at P60–80. Examples of oscillations that occurred simultaneously in L2/3 and L5/6 (A) and oscillations initiated in L5/6 and propagated to L2/3 with a delay (B) in control group. Corresponding probability histograms of SLA delay in L2/3 vs L5/6 are shown below.

Figure 3

Participation of interneurons and pyramidal neurons in initiation and maintenance of gabazine-induced SLA in L2/3 of somatosensory cortex. Dual field potential recording and cell-attached recording from interneuron (A) or pyramidal cell (B) in slice from flurothyl-treated rat at P65. During the initial phase of gabazine-induced SLA pyramidal cell synchronously fired with field events while the interneurons remained silent. 1–2 min after the first oscillation occurs interneurons began to show action potential activity highly synchronized with field oscillations.

Figure 4

Long-term effect of neonatal flurothyl-induced seizures on spontaneous inhibitory and excitatory synaptic transmission in L2/3 pyramidal cell of somatosensory cortex at P60–80. Aa. Example of sIPSCs recorded at holding potential 0 mV. Cumulative probability plots of amplitude (Ab) and interevent intervals (Ac) of IPSCs and corresponding bar graphs (mean ± SE) from control (light grey) and flurothyl-treated (dark grey) groups show decrease in amplitude of IPSCs in flurothyl-treated group, but no change in the interevent interval of sIPSC. Ba. Example of sEPSCs recorded at holding potential −80 mV in the presence of 10 μM gabazine. Representative cumulative probability plots and bar graphs (mean ± SE) show increasing in frequency (Bc) and amplitude (Bb) of sEPSCs recorded in flurothyl-treated group (dark grey) vs control group (light grey). Number of cells used for analysis shown in parenthesis above each column. Ca. Example of sEPSCs recording at holding potential −80 mV in the presence of 10 μM gabazine and 50 μM D-APV. Representative cumulative probability plots and bar graphs (mean ± SE) show no difference in interevent interval (Cc) and amplitude (Cb) of sEPSCs in flurothyl-treated group (dark grey) vs control group (light grey) recorded with NMDA receptor blocker.

Figure 5

Long-term effect of recurrent neonatal seizures on miniature inhibitory and excitatory synaptic transmission in L2/3 pyramidal cell of somatosensory cortex at P60–80. (Aa) Example of mIPSC recording (top and middle traces). Addition of gabazine to the extracellular solution completely blocks mIPSCs (bottom trace) (Ab and Ac) Cumulative probability of amplitude and interevent intervals and corresponding bar graphs of mIPSCs in flurothyl-treated group (dark grey) vs control group (light grey). (Ba) Recordings of mEPSC in control (ACSF) and during perfusion with D-APV and DNQX in a representative neuron. (Bb and Bc) Cumulative mEPSC amplitude and interevent distributions reveals a significant decrease in mEPSC interevent interval in flurothyl treated group. (Ca) Example of mEPSC recording in the presence of D-APV (top and middle traces) and D-APV and DNQX (bottom trace). Representative cumulative probability plots and bar graphs (mean ± SE) show no difference in interevent interval (Cc) and amplitude (Cb) of mEPSCs in flurothyl-treated group (dark grey) vs control group (light grey) recorded with NMDA receptor blocker. All values are means ± SEM.