- Top of page
Summary: Purpose: Maternal deprivation is stressful for the neonate. The aim of this study was to investigate the short- and long-term effects of maternal separation on recurrent seizures in the developing brain.
Methods: Rats were divided into four groups according to whether the rat pups were treated with maternal deprivation from postnatal day 2 (P2) to P9 or neonatal seizures induced by intraperitoneal (i.p.) injection of pentylenetetrazol (PTZ) from P10 to P14. Rats in the control group received saline i.p. injection from P10 to P14; rats in the isolation group underwent daily separation from their dams from P2 to P9; rats in the PTZ-treated group were subjected to PTZ-induced recurrent seizures from P10 to P14; rats in the isolation plus PTZ–treated group were subjected to maternal deprivation from P2 to P7 followed by serial seizures from P10 to P14. In addition, subsets of rats at P15 were killed and the brains assessed for acute neuronal degeneration. Visual–spatial memory test using the Morris water maze task was performed at P80. After testing, the hippocampus was evaluated for histologic lesions and cyclic adenosine monophosphate (cAMP)-responsive element-binding protein phosphorylation at serine-133 (pCREBSer-133), an important transcription factor underlying learning and memory.
Results: All rats given PTZ developed recurrent seizures. After PTZ administration, rats with a history of maternal deprivation had more intense impairment than did rats with maternal deprivation and neonatal seizures than those without deprivation. Neuronal degeneration was most prominent in the rats exposed to maternal deprivation plus recurrent seizures. Rats receiving maternal deprivation or PTZ-induced recurrent seizures exhibited only spatial deficits, but no morphologic changes in the hippocampus. However, rats with maternal deprivation plus PTZ-induced recurrent seizures exhibited worse visual–spatial learning compared with rats with either isolation or PTZ-induced recurrent seizures alone. The levels of pCREBSer-133 may play a role in the decrease in the hippocampus from the rats subjected to maternal deprivation and/or PTZ-induced recurrent seizures, as compared with rats exposed to vehicle-control saline. These results indicate that repeated maternal deprivation can exacerbate long-term cognitive deficits resulting from neonatal seizures. In addition, impaired phosphorylation of CREBSer-133.
Conclusions: Repeated maternal deprivation stress has synergistic effects with recurrent seizures in inducing neurologic damage in the developing brain.
Seizures occur more frequently in the neonatal period and early childhood than at any other time in life. In rodent studies, recurrent neonatal seizures result in long-term cognitive deficits (1–3), reduced dentate granule cell neurogenesis (4), and synaptic reorganization in the terminal field of the mossy fiber pathway (1–3).
Neonatal physiology and development are regulated to a great extent by mother–child interactions. In animal studies, environmental manipulation during the early postnatal period induces a decrease in anxiety-like behavior in adulthood (5). Maternal deprivation (neonatal isolation) appears to be stressful to the pup. Isolation of the rat pups from their mother, for even a brief period, evokes vocalizations (6) and activates the hypothalamic– pituitary–adrenal (HPA) axis (7), demonstrating that the separation experience is stressful for the neonate. If the isolation experience is repeated, enduring effects including behavioral abnormalities (7,8), neurochemical and endocrinologic changes (8,9), alterations of hippocampal neuroplasticity (10), and infralimbic cortex synaptic connections (11) can be observed in adulthood.
Little information is available in the literature regarding the molecular and cellular mechanisms that might contribute to memory deficits after seizures in the developing brain. Understanding the cellular mechanisms underlying memory deficits after seizures may open new therapeutic avenues for these transcriptional factor, is critically required in synaptic plasticity, learning and memory (12–14). CREB is a member of a large family [CREB/ATF (activating transcription factor)] of structurally related proteins that bind to the CRE promoter. CREB can be activated seconds or minutes after external stimulation through an intracellular increase in cyclic adenosine monophosphate (cAMP) or calcium. cAMP then activates the catalytic subunit of protein kinase A (PKA), by releasing its binding to the regulatory subunit. The catalytic subunit passively translocates to the nucleus where it phosphorylates one or more CREB-related transcription factors that activate the transcription of genes that lead to synaptic plasticity (15). Ser-133 phosphorylation is considered to be a critical event that mediates the initiation of transcription.), an important patients. A growing body of evidence suggests that phosphorylation of cAMP response element-binding protein at serine-133 (pCREBSer-133).
Support for the role of CREB in memory comes from models varying from the fruit fly to rodent (15–17). CREB activation occurs in the formation of new memories in Drosophila(18–20), Aplysia(21–23), and mice (12,18,24). Long-term memory and long-term potentiation is defective in CREB mutant mice (25). Spaced electrical tetani, which are sufficient to induce long-term potentiation in hippocampal slices from mice, also induce CREB-mediated gene expression (26). Antisense oligonucleotides specific for CREB transcripts injected into the hippocampus have been shown to impair water maze performance (27). Similarly, Lamprecht et al. (28) disrupted long-term memory of conditioned taste aversion by injecting antisense oligonucleotides into the amygdala. Thus it is possible that disrupted pCREBSer-133 function may be instrumental in the memory deficit after seizures during certain environmental challenges, such as maternal deprivation, during the neonatal period.
In most published studies, seizures have been induced in the developing animals under normal housing conditions. However, in clinical situations, neonates with seizures may also be in a stressful environment, such as the neonatal intensive care unit where the infants are separated from the mother for prolonged periods. In this study, we (a) examined whether maternal deprivation aggravated long-term cognitive effects after pentylenetetrazol (PTZ)-induced recurrent seizures, a conventional model for inducing seizures (2,3), in immature animals; and (b) explored the possible mechanisms underlying the memory deficits after recurrent seizures in the presence or absence of isolation stress in the developing brains. We found that maternal deprivation accentuated not only the immediate neuronal injury but also long-term seizure-induced cognitive deficits.
- Top of page
The present study shows that PTZ-induced recurrent seizures in the developing brain lead to long-term cognitive deficit and support the growing evidence that recurrent seizures in the developing brain are harmful (35,36). We demonstrated that both the abilities of spatial memory and pCREBSer-133 were reduced in rats subjected to PTZ-induced recurrent seizures during the neonatal period. Furthermore, this study demonstrates for the first time that repeated neonatal isolation stress exhibits synergism with neonatal seizures in inducing both short- and long-term detrimental effects.
In rodent studies, both prenatal stress (37) and postnatal stress (38) result in decreased neurogenesis in the granule cells of the dentate gyrus. The dentate gyrus has an important role in learning (39) and undergoes the majority of its development during the first 2 weeks of life (40). Repeated neonatal isolation stress caused abnormal development of the functional pathway, in particular in the limbic system, and these changes may persist into adulthood (41). Kehoe and Bronzino (42) demonstrated that repeated neonatal isolation stress produced an alteration of long-term potentiation (LTP) in the adult animals. As shown in this study, intense silver staining of granule cells of the dentate gyrus was detected immediately in rats subjected to maternal isolation plus recurrent seizures. Our results suggest that the dentate gyrus is particularly vulnerable to early environmental manipulations and contributes to long-term cognitive deficits after seizures and isolation. Conversely, cresyl violet staining of adult rat hippocampus showed no neuronal loss in any of the groups. These results suggest that silver staining is a marker of acute neuronal degeneration (30), but not a definite marker, per se, for permanent cell loss (31).
Activation and expression of CRH is developmentally regulated and is associated with hippocampal excitability and seizure generation in the developing brain (43). Repeated isolation results in potentiation of the hypothalamus–pituitary–adrenal (HPA) axis in adulthood (44). Maternal deprivation enhances the HPA response after exposure to a novel environment (45). Furthermore, after repeated neonatal isolation, neurosteroids are altered, as indicated by a reduction of dihydroprogesterone and an increase of allopregnanolone (46). Exposure to high levels of glucocorticoids during this critical period in development has been shown to lead to numerous detrimental effects on the developing central nervous system (47). Prolonged elevations of glucocorticoid levels are toxic to hippocampal neurons by increasing their vulnerability to a variety of insults (48–50). Furthermore, glucocorticoids exacerbate neuronal damage caused by pathophysiologic challenges, including hypoxia–ischemia, traumatic brain injury, and repeated seizures (51,52). We suggest that neonatal stress, through potentiation of the HPA axis and an alteration of neurosteroids, can exacerbate immediate neuronal insults, possibly by alterations of neuron connectivity. Neonatal seizures and stress appear to be synergistic, leading to enduring detrimental effects in cognitive functions.
The possible mechanisms underlying long-term cognitive deficits resulting from various pathophysiologic environments, such as recurrent seizures in the presence of maternal deprivation during the neonatal period, remain unclear. Although there is now a relative agreement of molecular-signaling pathways for certain forms of learning and memory in invertebrates such as Aplysia and Drosophila, the mechanisms that are responsible for learning and memory in mammals remain unclear. Studies of several forms of learning and memory (e.g., behavioral sensitization, Morris water maze task, and classic conditioning in vertebrates) indicate that both behavioral long-term memory and its neural representation require gene expression triggered by pCREBSer-133 that consequently leads to the growth of new synaptic connections (15,18,53,54). Despite the correlation of pCREBSer-133 with performance on the Morris water maze, the failure to show a difference between the isolation, the PTZ-treated, and the combined isolation plus PTZ–treated groups indicates that the absolute deficits in pCREBSer-133 are not the sole factor determining the severity of spatial memory performance. PCREBSer-133-independent mechanisms must be involved in memory impairment after the seizures.
In surgically treated patients with temporal lobe epilepsy, a history of prolonged febrile seizures in childhood can usually be found (55). Conversely, epidemiologic studies showed that the risk of hippocampal sclerosis and temporal lobe epilepsy after an initially provoked or unprovoked seizure was low (56). Mathern et al. (57) and Katzir et al. (58) proposed that secondary physiologic decompensations, such as loss of cerebral autoregulation during seizure attacks, might contribute to the seizure-induced neuronal damage and hence individual variability in long-term outcomes. Here we showed that stress also could exert synergistically detrimental effects on recurrent seizures in the developing brain. Taken together, our results support the concept that, under certain circumstances, the brain is more vulnerable to seizure-induced damage that causes permanent molecular signaling, morphologic, and behavioral changes in the hippocampus.
The demonstration that isolation stress has immediate and enduring effects in association with recurrent seizures in the developing brain may have clinical implications. The environment immediately after birth, which provides important socioemotional experience during the earliest phases of postnatal brain maturation, is the mother–child interaction. Neonates with seizures are usually separated from their mothers, and therapeutic interventions, such as noise, light, endotracheal suctioning, and vein punctures are both painful and stressful. Our findings suggest that psychological and pharmacologic interventions to prevent stress responses in the neonates are needed to reduce aggravating effects of stress on seizure-induced brain damage.
Acknowledgment: The study was supported in part by grants CMRPG8013 from Chang Gung Memorial Hospital, New Century Health Care Promotion Foundation, and NSC 91-2314-B-182A-050 (L-T.H.), 1261 from Chang Gung Memorial Hospital and NHRI-EX91-8909BP (S.N.Y.), a Mental Retardation Research Center grant from NIH (2P30HD18655), and a grant from the NINDS (NS27984) (G.L.H.).