Hyper-excitability and epilepsy generated by chronic early-life stress

Abstract

Epilepsy is more prevalent in populations with high measures of stress, but the neurobiological mechanisms are unclear. Stress is a common precipitant of seizures in individuals with epilepsy, and may provoke seizures by several mechanisms including changes in neurotransmitter and hormone levels within the brain. Importantly, stress during sensitive periods early in life contributes to 'brain programming', influencing neuronal function and brain networks. However, it is unclear if early-life stress influences limbic excitability and promotes epilepsy. Here we used an established, naturalistic model of chronic early-life stress (CES), and employed chronic cortical and limbic video-EEGs combined with molecular and cellular techniques to probe the contributions of stress to age-specific epilepsies and network hyperexcitability and identify the underlying mechanisms. In control male rats, EEGs obtained throughout development were normal and no seizures were observed. EEGs demonstrated epileptic spikes and spike series in the majority of rats experiencing CES, and 57% of CES rats developed seizures: Behavioral events resembling the human age-specific epilepsy infantile spasms occurred in 11/23 (48%), accompanied by EEG spikes and/or electrodecrements, and two additional rats (9%) developed limbic seizures that involved the amygdala. Probing for stress-dependent, endogenous convulsant molecules within amygdala, we examined the expression of the pro-convulsant neuropeptide corticotropin-releasing hormone (CRH), and found a significant increase of amygdalar--but not cortical--CRH expression in adolescent CES rats. In conclusion, CES of limited duration has long-lasting effects on brain excitability and may promote age-specific seizures and epilepsy. Whereas the mechanisms involved require further study, these findings provide important insights into environmental contributions to early-life seizures.

Keywords: amygdala; corticotropin releasing hormone; epilepsy; infantile spasms; seizures; stress.

Fig. 1

Chronic early-life stress (CES) enhances excitability in amygdala, manifest as the presence of epileptiform spike series. A and B are sample EEGs from two individual rats recorded around two weeks each. Rats were implanted with bipolar electrodes in the right amygdala and with two cortical electrodes, one each over right and left frontoparietal cortices, as described in the methods. The montage used for these animals consisted of: Channel 1: from within the right amygdala (bipolar); Channel 2: Activity between the two cortical electrodes. These epileptiform discharges occurred throughout the 16 and 17 day recordings in the two rats, and their typical contour is shown in an expanded time-scale (of the boxed segments) in A1, B1. Such spike-trains were never observed in the 20 video-EEG recorded control rats.

Fig. 2

Chronic early-life stress (CES) leads to limbic spontaneous seizures in a minority of rats. Shown are EEGs traces of spontaneous limbic seizures from a CES rat. Rats were implanted bilaterally with bipolar electrodes in amygdala as well as with an electrode over the right frontoparietal cortex. The montage consisted of two channels: Channel 1: intra-left amygdala (bipolar); Channel 2: between one of the twisted electrodes in the right amygdala and the right cortical electrode. A, B show the onset and progression of two spontaneous seizures. The expanded time-scale view (A1,B1) allows observation of the temporal sequence of the onset of the seizures (denoted by the blue arrows). The vertical line aids in discerning that the first voltage deflection associated with the seizure originates in the amygdala channels earlier than in the cortex. The second expanded time-scale views (right) suggest that during the course of the seizure, the amygdala spikes ‘lead’ cortical sharp waves and waves. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

CES provokes flexion events resembling human infantile spasms in 11/23 rats. (A) Serial still images from a video recording of a CES rat cortical electrodes. These images depict the onset and progression of two consecutive spasm-like behaviors: In the event in the top row, note the body flexion of the rat. The sustained, slower phase of this flexion is more apparent in the spasm-like seizure shown in the bottom row. (B) Correlation of spasm-like behaviors and EEG in a different rat, with electrodes within the amygdala. The blue arrow points to the concurrent onset of behavioral flexion and epileptiform discharges. The behavioral event ends at the time noted by the red arrows. (C–D) Temporal and quantitative distribution of spasm-like events among rats. In C, each animal is represented by a symbol and color; the X axis is the age of the rat, in days, and the Y axis is the number of spasms for each days. Triangles denote rats who had multiple spasms over multiple days. These were the majority (6/11), as is depicted in (D). Circles denote rats (n = 3) that had several spasms during a single recording day, and squares denote two rats in whom only a single spasm was captured. Note that rats were recorded for only 1–2 h per day prior to weaning, and thus the actual number of spasms was likely much higher. (D). The majority of rats had flexion, spasm-like events multiple times over several days.

Fig. 4

Corticotropin releasing hormone (CRH) expression is augmented in the amygdala of CES rats. (A) Representative photomicrographs of amygdalae after immunohistochemistry using an antiserum directed against CRH. CRH immunoreactivity (ir) was enhanced in the amygdala of CES rats compared with controls at P45 as shown in C: Semi-quantitative analysis of CRH expression levels revealed a significant increase of the optical density of CRH-ir in rats that experienced CES, (p = 0.048). (B) In frontoparietal-cortex, CRH-ir cell number and density did not differ significantly in the same control and CES rats distinguished by amygdala expression. Representative micrographs show the typical distribution of CRH-expressing cortical interneurons. (D) A graph depicting the number of cells expressing CRH above detection levels in an area of 2 square mm (controls: 36.3 ± 2.1; CES: 35.8 ± 13.2; p = 0.97, t-test with Welch correction for unequal variance). Cell numbers were not used in amygdala sections because, as apparent in the photos in A, most CRH in this nucleus is found in fibers. (E) At the age of onset of spasm-like events (pre-weaning or infancy, P19), CRH mRNA expression levels were borderline higher in CES rats compared to controls (CES: 0.12 ± 0.01; controls: 0.08 ± 0.002, n = 3–4 per group; p = 0.04, Mann–Whitney test). Scale: 250 μm.