Abstract
Early-Life Seizures Result in Deficits in Social Behavior and Learning.
Lugo JN, Swann JW, Anderson AE. Exp Neurol 2014;256:74–80.
Children with epilepsy show a high co-morbidity with psychiatric disorders and autism. One of the critical determinants of a child's behavioral outcome with autism and cognitive dysfunction is the age of onset of seizures. In order to examine whether seizures during postnatal days 7–11 result in learning and memory deficits and behavioral features of autism we administered the inhalant flurothyl to induce seizures in C57BL/6J mice. Mice received three seizures per day for five days starting on postnatal day 7. Parallel control groups consisted of similarly handled animals that were not exposed to flurothyl and naïve mice. Subjects were then processed through a battery of behavioral tests in adulthood: elevated-plus maze, nose-poke assay, marble burying, social partition, social chamber, fear conditioning, and Morris water maze. Mice with early-life seizures had learning and memory deficits in the training portion of the Morris water maze (p < 0.05) and probe trial (p < 0.01). Mice with seizures showed no differences in marble burying, the nose-poke assay, or elevated plus-maze testing compared to controls. However, they showed a significant difference in the social chamber and social partition tests. Mice with seizures during postnatal days 7–11 showed a significant decrease in social interaction in the social chamber test and had a significant impairment in social behavior in the social partition test. Together, these results indicate that early life seizures result in deficits in hippocampal-dependent memory tasks and produce long-term disruptions in social behavior.
Commentary
There have been a number of large-scale epidemiologic studies that have associated childhood epilepsy with a variety of neurobehavioral comorbidities including cognitive delay, depression, anxiety, attention-deficit/hyperactivity disorder as well as autism spectrum disorder (e.g., [1, 2]). While some headway has been made into the neurobiological substrates that may contribute to seizure-related cognitive impairments, there have been relatively few studies that have attempted to address additional comorbidities. In addition, little is known to what extent seizure events that occur very early in life might impact cognition later in life.
A recent study by Lugo et al. suggests that early-life seizures can produce fairly specific deficits in spatial learning, and alterations in the way that adult mice interact socially. In this study, seizures were induced repeatedly in young mice over the course of 5 days using the volatile convulsant flurothyl. Seizures were induced beginning at 7 days of age to emulate seizure activity in children less than 1 year old. Mice were subsequently tested as young adults (60 days old) in a variety of behavioral tasks designed to assess cognitive capacity and autistic-like behaviors. Consistent with previous work (3), Lugo et al. found that early-life seizures impair performance in the Morris water maze in which mice are trained to find an escape platform that is submerged just below the surface of the pool. Mice that had experienced early-life seizures took significantly longer to find the hidden platform during training and exhibited a less-selective search strategy 30 minutes following training during a probe trail in which the platform was removed. Of importance, mice in the early-life seizure group were able to learn the location of the platform when it was clearly marked, demonstrating that the deficits in the hidden platform version were not due to alterations in motor performance or visual acuity required to perform the task. In contrast, early-life seizures did not produce deficits in associative learning as assessed by Pavlovian fear conditioning. Mice in the early-life seizure group were able to learn that an auditory conditioned stimulus (CS) was predictive of a foot-shock. Unlike the Morris water maze, this form of learning, commonly referred to as cued conditioning, can be accomplished even when the functional integrity of the hippocampus has been compromised. Although the authors did not assess the impact of early-life seizures on the hippocampal-dependent form of fear conditioning, where context is used as the CS, the water maze data and cued conditioning data taken collectively suggest that the early-life seizures selectively impact hippocampal-dependent memory formation.
In addition to selective deficits in spatial memory, mice in the early-life seizure group also exhibited deficits in social exploration. Under normal conditions, mice are very social animals and will typically choose mice over inanimate objects when given the choice to explore either. Reduced social interactions in mice are often thought to reflect autistic-like behavior (4). To investigate the possibility that early-life seizures might result in autistic-like behavior in mice, the authors examined mice in two tests of social interaction. In the first test—termed the social partition test—two mice (of the same age and weight, one being the experimental subject) were housed in the same cage but on opposite sides of a porous barrier. After 24 hours of cohousing, the mice were observed, and the number of approaches and the time spent at the partition by the experimental mouse were recorded. Next, the nonexperimental, familiar mouse which was previously co-housed with the experimental mouse was replaced with a novel mouse, and the same measurements were made for 5 minutes. In the final phase, the previously novel mouse was removed and replaced again with the overnight partner. Mice in the control group exhibited a clear increase in the amount of time near the partition when the novel mouse was introduced compared with the time spent at the partition during either exposure to the overnight partner. In contrast, the mice that had experienced early-life seizures did not exhibit any preference for the novel mouse, spending equal amounts of time at the partition, regardless of which mouse was on the other side. It should be noted that only male mice were examined in this test (sham female mice do not exhibit a preference).
In a separate series of experiments, the authors examined the impact of early-life seizures on sociability in the social approach apparatus. The social approach apparatus consists of a three chambered arena in which the two outer chambers contain small wire cages. The center chamber, which is connected to the outer chambers by sliding plastic doors, remains empty. Prior to introduction of the experimental mouse, a demonstrator mouse was placed in one of the wire cages and an inanimate object was placed in the alternate wire cage. The experimental mouse was released into the center chamber and allowed to explore. Typically, mice will traverse all three chambers but will spend more time investigating the mouse compared with the inanimate object. Similar to the results obtained in the social partition experiments, the mice exposed to early-life seizures spent significantly less time engaged with mice that they had not been previously exposed to when compared with sham-treated mice. Like the spatial-learning deficit, the deficits in social behavior appear to be fairly specific. The mice in the early-life seizure group exhibit similar levels of exploratory behavior and basal levels of anxiety when compared with control mice as assessed in the elevated-plus maze and nose-poke assay. Finally, the mice in the early-life seizure group were not significantly different from controls in the marble-burying task, which is used to assess chronic repetitive behaviors. Thus, results from the social partition and social approach experiments suggest that a relatively brief bout of repeated seizures early in life while the brain is developing can produce long-lasting changes in social behavior.
According to The Diagnostic and Statistical Manual of Mental Disorders (5), the hallmarks of Autism Spectrum Disorder (ASD) include deficits in reciprocal social interactions, stereotyped restricted repetitive patterns of behavior, and deficits in communication (both verbal and nonverbal). The experiments presented by Lugo suggest that early-life seizures produce deficits in social interactions in the absence of chronic repetitive patterned behavior. It remains unknown to what extent early-life seizures might alter communication in adult mice, but these experiments could be accomplished by analyzing ultrasonic vocalization (6). Given the selective impairments in hippocampal-dependent learning, it is tempting to speculate that the social behavioral deficits may also be the providence of the hippocampus, which if true, might suggest an important anatomic segregation of the major symptoms of ASD. This work also may have broader implications for other neurobehavioral comorbidities associated with early-life seizures.