The Effects And Consequences Of Recurrent Neonatal Seizures On The Neurogenesis

BackgroundSeizures occur more frequently in the neonatal period than at any other time in life. A controversy which has been debated for the recent years is whether recurrent neonatal seizures can lead to long-term adverse consequences or are simply a reflection of underlying brain dysfunction and are not intrinsically harmful. Despite numerous clinical observation that seizures in the developing brain may be detrimental, the pathological mechanism has not yet been completely understood.Now, it is generally known that ongoing neurogenesis occurs in certain regions of the brain, such as the subgranular zone of hippocampal dentate gyrus and subventicular zone of the lateral ventricles, throughout the life in human and most mammalian animals. Among those, the neurogenesis in the subgranular zone of hippocampal dentate gyrus is more paid attention to by the scholars who are studying the mechanism of epileptogenesis. Animal studies have demonstrated numerous pathological conditions including epilepsy can enhance neurogenesis. For example, a single episode of status epilepticus induced by pilocarpine or kainic acid, kindling epileptic model and repeated electroconvulsive shock stimulation and so on, can increase the proliferation of granule cells of the dentate gyrus. In immature younger rodents, parts of foreign research and our previous study have also manifested that seizures can increase the neurogenesis of granule cells of the dentate gyrus.However, in the neonatal period of rats during which the neurogenesis is the most robust, the effects of neonatal seizure on the rodent neurogenesis within differentages have not been clear. Previous studies have shown that electroconvulsive seizures, recurrent flurothyl seizures and bicuculline-induced status epilepticus irreversibly decrease brain DNA content, cell number, and total brain weight. In addition, neurogenesis of the granule cells in the dentate gyrus is age-dependent, while corticosterone levels and N-methyl-D-aspartate (NMDA) receptor activation appear to regulate the rate of this proliferation.The mossy fibers, granule cell axons, terminate normally to stratum lucidum of hippocampal CA3 area passing through hilus of the dentate gyrus. But mossy fiber sprouting has been observed in a number of adult epileptic models including prolonged seizures induced by kainic acid or pilocarpine, electrical or pentylenetetrazole (PTZ) kindling, and pathological specimen removed from patients with tempoal lobe epilepsy and so on. Mossy fiber sprouting means that sprouted axon collaterals of granule cells project from the hilus of dentate gyrus into the inner molecular layer or stratum oriens in CA3 region. Moreover, inreasing evidence have proved that these sprouting are involved in new functional synaptic connections, i.e. synaptic reorganization. And it is assumed that some synaptic reorganization may form a kind of excitatory circuit, then subsequently result in chronic spontaneous epileptic seizure. In addition, in the mature animal status epilepticus causes neuronal loss in hippocampal fields CA1, CA3, and the dentate hilus, leads to long-term deficits in learning, memory, and behavior.Although clinical practice shows that neonatal seizure is harmful, many animal studies suggest that the immature brain is less vulnerable to the effects of prolonged seizures than the mature brain. For instance, a number of researches have demonstrated that a single prolonged seizure in an immature animal results in less cell loss and sprouting, and fewer deficits in learning, memory, and behavior than similar seizures in adults.In a word, immature brain differs from adult in the susceptibility, features and prognosis of seizures, probaly due to the differences between their epileptogenesis. Therefore, the pathogenetic mechanism of seizures in immature animal, especially the neonatal, are still to be explored further.ObjectiveTo establish a kind of rat model with recurrent neonatal seizures, and evaluate its feasibility and characteristics. Afterwards we intended to investigate the effects of recurrent neonatal seizures on neurogenesis at different ages, the differences of neurogenesis between dorsal and ventral hippocampi, and the relationship among the neurogenesis and neuronal loss, glucocorticosteroids levels. And then, to study the long-term consequences of recurrent neonatal seizures, in another word, to observe the behavior changes including learning, memory and seizure susceptibility, and the pathological morphological changes including hippocampal cell loss and mossy fiber sprouting , when the rats become adult. Methods1.Neonatal Wistar rats (n=10) were subjected to three times of pilocarpine injections intraperitonealy at postnatal day 1, 4 and 7. The doses of pilocarpine were adjusted accordiong to the dosage usually used in adult rat until the seizures were induced. Before pilocarpine injection, scopolamine was administered in order to prevent its cholinergic reaction. If status epilepticus lasts for a too long time, chloralhydrate was given to stop seizure properly. Measure the mean time to the onset of induced seizure, and observe the manifestations of seizure. Neonatal rats (n=6) in the control group were given saline injection intraperitonealy other than piloparpine at the same time as the studied group.2.After establishing the rat model with recurrent neonatal seizures, both the experimental rats and the control group were administered bromodeoxyuridine (BrdU) once intraperitonealy separately at the four time points: immediately after the third seizure (P7), the fourth day after the seizure (PI 1), the fourteen day (P21) and the fourty fifth day (P52). Eight rats with seizure and six seizure-free rats were used at each time point as above. All rat pups were killed 36h after BrdU injection, and blood was collected for the measurement of serum corticosterone level. Brain tissue sections were prepared and carried out by Nissl staining for neuronal loss, by BrdU labeling for cell proliferation and by BrdU-NF200 double labeling for the identification of the new formed cells.3.After establishing the rat model with recurrent neonatal seizures, rat pus were divided into 4 groups: seizure group I (n=10) and II (n=10), control group I (n=8) and II (n=8). All rats in seizure group and control group were subjected to Morris water maze test at P52 for visual-spatial learning and memory, and to open field test for their activity level at P60. At P63, rats in seizure group I and control group I were induced by PTZ to test their seizure susceptibility, and rats in seizure group II and control group II were killed to prepare brain sections for Nissl stains and Timm stains evaluating for mossy fiber sprouting. Results1.Neonatal rats in the seizure group were induced to epileptic seizure by pilocarpine separately at PI, P4 and P7. With the increase of seizure number and rat age, the mean time to onset of seizure and the dosage of pilocarpine used decreased, and the manifestations of seizurs became intenser. Seven in the seizure group survived after three times of seizures, with one death at P4 and two death at P7. The death rate was 30%. All death was due to too long or severe seizure. No seizure and death were found in the control group.2.At the beginning of experiment, there were 40 in the seizure group, and after three times of seizure, 7 pups were dead. The death rate was 17.5%. Nissl stains showed that there were no evident neuronal loss in CA3, CA1, the granule cell layer (GCL) and hilus of dentate gyms (DG) between the seizure group and the control group at P7, Pll, P21 and P52.The numbers of BrdU-labeled cells were age-dependent both in the seizure group and the control group, decreasing with the increase of ages, and their morphorlogy and distribution changed. BrdU-labeled cells in GCL and hilus decreased significantly in the seizure group comparing with the matched controls at P7 and Pll (,P<0.05), while at P21 there was no difficences between two groups. On the contrary, BrdU-labeled cells in GCL and hilus increased significantly in the seizure group comparing with the matched controls at P52 (PO.05).BrdU-labeled cells distributed mainly in the dorsal hippocampus in both groups at each time point as mentioned above, otherwise rarely in the ventral hippocampus.9Most of BrdU-labeled cells in GCL coexpressed NF200. Serum corticosterone concentrations were negatively correlated with the numbers of BrdU-labeled cells in GCL and hilus in both groups at each time point.3.Seven were dead after 30 rat pups were subjected to three times of neonatal seizures, with 23.3% death.In six days of Morris water maze test, the average time to the plantform in the seizure group (n=20) was higher than the controls (n=16) on each day, but the difference was most pronounced on the first and the second days (P<0.05). In open field test, rats in the seizure group showed less active than controls (mean blocks crossed in controls: 97.6 + 9.87; seizure-induced, 68.4 ±7.10; P<0.05).PTZ elicited seizures in all seizure group I (n=10) and control group I (n=8) rats. Compared with controls, rats in seizure group I had a significantly shorter duration to both myoclonic jerks (seizure-induced, 890 + 90 seconds; controls, 1270 + lOOseconds; P<0.05) and clonic activity (seizure-induced, 1030 + 40 seconds; controls, 1320 + 80seconds; P<0.0\) than controls.The Timm scores in seizure group II (n=10) were significantly higher in CA3 and supragranular region than control group II (n=8) separately (CA3: seizure-induced, 2.5 + 0.3; controls, 1.1+0.7, PO.001. Supragranular region: seizure-induced, 1.1+0.2; controls, 0.2 + 0.1, PO.05).Cell counts by Nissl staining showed that no differences between seizure group II and control group II in CA3, CA1 and hilus. However, seizure group II had greater numbers of dentate granule cells than control group II (PO.05). Conclusions1 .Three times of neonatal seizures were induced by pilocarpine at PI, P4 and P7. With the increase of seizure numbers and rat ages, the mean time to onset of seizure became shorter, and the intensity of seizure got stronger. After three times of neonatal seizures, the death rate of rat pups was 17.5%~30%, indicating that it could be as a rat model with recurrent neonatal seizures.2.There were no obvious neuronal loss in CA3, CA1, GCL and hilus in rats after recurrent neonatal seizures.Neurogenesis changes in an age-dependent manner, with its numbers decreasing accompanied by ages, and its cell morphology and distribution also varied. The neurogenesis in GCL and hilus showed mainly neuron proliferation, decreasing immediately after the third seizure and the fourth day, with no obvious changes at the fourteen day and increase at the fourty fifth day after the third seizure at P7. Neurogenesis occurred mostly in the dorsal hippocampus. Neurogenesis correlated negatively with serum corticosteroid levels.3.Once mature, rats with recurrent neonatal seizures had impaired learning, decreased activity levels and increased seizure susceptibility. The adult rats with a history of recurrent neonatal seizures had sprouting of mossy fibers in CA3 and the supragranular region, with the increase of dentate granule cell numbers and no obvious neuronal loss in CA3, CA1 and hilus. SignificanceThe study has established a rat model with recurrent neonatal seizures, which could be used for the investigation of mechanism in neonatal seizures in the future. Neurogenesis decrease occurs immediately after recurrent neonatal seizures and lasts for several days, which is opposed to the neurogenesis increase observed in the seizure models of adult and younger rats. Suppression of neurogenesis my seizures may be a source of learning deficits in children with epilepsy due to lack of mature hippocampal circuitry, and whether this condition can be corrected by enriched learning environment and physical execise is meaningful in clinical practice. Whether the mechanism that ACTH can be used as treatment of infantile spasm is related to neurogenesis, mossy fiber sprouting and so on, is deserved to be studied. Recurrent neonatal seizures can lead to a long-term detrimental morphological and behavioral effects, which suggests that clinicians should pay much attention to the neonatal seizures to prevent their poor prognosis.