Perirhinal cortex hyperexcitability in pilocarpine-treated epileptic rats

Abstract

The perirhinal cortex (PC), which is heavily connected with several epileptogenic regions of the limbic system such as the entorhinal cortex and amygdala, is involved in the generation and spread of seizures. However, the functional alterations occurring within an epileptic PC network are unknown. Here, we analyzed this issue by using in vitro electrophysiology and immunohistochemistry in brain tissue obtained from pilocarpine-treated epileptic rats and age-matched, nonepileptic controls (NECs). Neurons recorded intracellularly from the PC deep layers in the two experimental groups had similar intrinsic and firing properties and generated spontaneous depolarizing and hyperpolarizing postsynaptic potentials with comparable duration and amplitude. However, spontaneous and stimulus-induced epileptiform discharges were seen with field potential recordings in over one-fifth of pilocarpine-treated slices but never in NEC tissue. These network events were reduced in duration by antagonizing NMDA receptors and abolished by NMDA + non-NMDA glutamatergic receptor antagonists. Pharmacologically isolated isolated inhibitory postsynaptic potentials had reversal potentials for the early GABA(A) receptor-mediated component that were significantly more depolarized in pilocarpine-treated cells. Experiments with a potassium-chloride cotransporter 2 antibody identified, in pilocarpine-treated PC, a significant immunostaining decrease that could not be explained by neuronal loss. However, interneurons expressing parvalbumin and neuropeptide Y were found to be decreased throughout the PC, whereas cholecystokinin-positive cells were diminished in superficial layers. These findings demonstrate synaptic hyperexcitability that is contributed by attenuated inhibition in the PC of pilocarpine-treated epileptic rats and underscore the role of PC networks in temporal lobe epilepsy.

FIGURE 1

Scheme of the main afferent/efferent connections of the (PC) (area 35). The drawing corresponds to a section taken at −7.6 mm from the bregma according to the Paxinos and Watson (2007) atlas. Major afferent projections are from olfactory regions and lateral amygdala (LA), insular and piriform cortices, and claustrum; the most conspicuous efferent projection of PC is directed to the entorhinal cortex (EC), but insular, frontal, and parietal cortices are also targeted. Subcortical efferents terminate in a variety of regions, including basal ganglia and basomedial nuclei of the amygdala (cf. Furtak et al., 2007).

FIGURE 2

Photomicrographs of NeuN distribution in the perirhinal cortex (PC) of NEC (A) and pilocarpine-treated (B) rats. Note that no clear cut difference is noticeable in PC of the two different sections. C: Analysis of neuronal cell densities did not reveal significant differences. The analyzed areas are indicated by the broken lines in (A) and (B). Abbreviations: ab, angular bundle; EC, entorhinal cortex; rs, rhinal sulcus; Sub, subiculum. Scale bar, 150 μm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIGURE 3

Changes in interneuron subpopulations in the pilocarpine-treated, epileptic PC. Photomicrographs of interneurons identified with antibodies against parvalbumin (PV, panels A and B), neuropeptide Y (NPY, C and D), and cholecystokinin (CCK, E and F) in the PC of NEC (A, C, and E) and pilocarpine-treated epileptic (B, D, and F) rats. Note that all the different classes of interneurons are decreased in pilocarpine-treated epileptic rats, as shown in the plot histogram in G. Statistical analysis of neuronal counts was performed by using the Mann–Whitney test: **P < 0.01, *P < 0.05 versus NECs. Abbreviations: ab, angular bundle; EC, entorhinal cortex. Scale bar, 100 μm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIGURE 4

Spontaneous synaptic activity in NEC and pilocarpine-treated tissue. A: (a) Simultaneous field [deep perirhinal cortex (PC_f ) and lateral amygdala nucleus (LA_f )] and intracellular recording (−68 mV, PC) in NEC tissue (i) reveals depolarizing PSPs indicated by arrows in insert and (ii) robust spontaneous hyperpolarizing IPSPs indicated by asterix. Expansion of these events is depicted in the right lower insert. (b) Response of PC neuron to intracellular current injection. B: Simultaneous field (PC_f, LA_f ) and intracellular activity (−68 mV, PC) recorded in the majority of pilocarpine-treated tissue. Note the absence of field activity (PC_f, LA_f ). PSPs are indicated by arrows in the intracellular trace and in the expansion in the right lower insert. (b) Response of PC neuron to intracellular current injection. C: (a) Simultaneous field (PC_f, LA_f ) and intracellular activity (−65 mV) recorded in a sub-set of pilocarpine-treated slices reveals robust network activity (PC_f, LA_f ). Expansion of an event demonstrates initiation in PC (arrow) and spread to LA (right lower insert). (b) Response of PC neuron to intracellular current injection.

FIGURE 5

Responses of PC neurons to local single-shock stimulation. A: Sub- and suprathreshold responses of PC neurons in (a) NEC, (b) pilocarpine (no field activity), and (c) pilocarpine (with field activity) recorded in “Control” conditions (i.e., ACSF only). Note the absence of an inhibitory response (insert) and the evoked epileptiform-like activity in (c) Pilocarpine (with field activity). B: Responses of PC neurons in (a) NEC, (b) pilocarpine (no field activity), and (c) pilocarpine (with field activity) in the presence of glutamatergic antagonists (+CPP + CNQX). Note the biphasic IPSP responses specifically at more depolarized membrane potentials (inserts). ▲ represent stimulation artifact.

FIGURE 6

Responses of PC neurons to single-shock stimulation of LA networks. Comparison of the intracellular responses evoked by single-shock stimulation of LA networks in PC neurons recorded from NEC (A), pilocarpine (no field activity) (B), and pilocarpine (with field activity) (C) brain slices. In each cell, the responses were recorded at membrane potentials set to different levels by intracellular current injection. Note that in NEC, LA stimulation induces a “pure” IPSP sequence while EPSP–IPSP responses can be recorded in the pilocarpine (no field activity) tissue; also note that LA stimulation in pilocarpine (with field activity) results in robust discharges in PC cells which at more depolarized membrane levels (i.e., −54 mV) is preceded by a hyperpolarizing component (asterisk). ▲ represent stimulation artifact.

FIGURE 7

GABA_A-mediated component of evoked IPSP in the PC of pilocarpine-treated tissue exhibits a more depolarized reversal potential. A: Intracellular recordings of the “monosynaptic” IPSPs evoked in NEC and pilocarpine-treated tissue in the presence of CPP + CNQX; membrane potential was set to different levels by intracellular current injection. Note that the reversal potential of the IPSP early component is more depolarized in NEC (reversal potential = −74.7 mV) versus pilocarpine-treated tissue (reversal potential = −69.5 mV). B: Histogram of the distribution of the IPSP early component reversal potential values in NEC and pilocarpine-treated PC neurons. The statistical analysis is illustrated in Table 3. Skewness, that quantifies the asymmetry of the histogram distribution, suggests a tail to the right in pilocarpine-treated rats and a tail to the left in NECs. Kurtosis, which identifies the tendency to group values in the center, shows that pilocarpine-treated rats have more values in the center, whereas NECs are close to a Guassian distribution but have more values in the tail. In any case, the Kolgomorov–Smirnov test was significant (P < 0.01), suggesting a non-normal distribution of values when the two groups are considered all together.

FIGURE 8

Potassium-chloride cotransporter 2 (KCC2) immunoreactivity in the PC of nonepileptic control (NEC) rats (A) and pilocarpine-treated rats (B). Note that immunoreactivity is mainly localized on nerve fibers and cell surface, as easily appreciated in the CA1 hippocampal region [arrows in panel A; cf. de Guzman et al., 2006; Fig. 9C, for magnification of the cellular KCC2 staining]. The perirhinal cortex (PC) appears to be more intensely stained in the NEC than in the pilocarpine-treated rat. The densitometric analysis (histogram) demonstrated a significant (**P < 0.01, Mann–Whitney test) decrease in KCC2 optical densities in the PC of epileptic rats. Abbreviations: ab, angular bundle; EC, entorhinal cortex; rs, rhinal sulcus; Sub, subiculum. Scale bar, 200 μm.