- Email: [email protected]
Effect of seizures on cerebral hypoxic–ischemic lesions in immature rats
Developmental Brain Research 113 Ž1999. 83–95
Research report
Effect of seizures on cerebral hypoxic–ischemic lesions in immature rats Javad Towfighi
a,)
, Cathy Housman a , David Mauger b, Robert C. Vannucci
c
a
b
Department of Pathology (Anatomic Pathology), The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, P.O. Box 850, Hershey, PA 17033-0850, USA Department of Health EÕaluation Sciences, The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, Hershey, PA 17033-0850, USA c Department of Pediatrics, The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, Hershey, PA 17033-0850, USA Accepted 22 December 1998
Abstract The present investigation was designed to study the effect of chemically induced seizures on cerebral hypoxic–ischemic ŽHI. damage in immature animals. Accordingly, cerebral HI was produced in 7-day postnatal Žp7. rats and p13 rats by combined unilateral common carotid artery ligation and hypoxia with 8% oxygen. Seizures were induced chemically by the subcutaneous injection of kainic acid ŽKA. or inhalation of flurothyl vapor. Three types of experiments were conducted in each age group and for each convulsant. In some animals Žgroup 1., seizures were produced at 24 h and again at 6 h prior to HI. In groups 2 and 3, seizures were induced 2 h or 24 h post HI, respectively. The results indicate that in group 1 animals, the first seizure significantly reduced duration of the second seizure challenge 18 h later at both p7 and p13 Ž p s 0.001.. Histologic examination of brains of animals in group 1 subjected to seizures prior to HI and their HI-only controls showed that seizures prior to HI conferred protection against cerebral damage. This effect was significant for flurothyl seizures in p13 rats for all cerebral regions, especially hippocampal CA1 Ž p s 0.0004., and in p7 rats for hippocampus Ž p s 0.04. and particularly cerebral cortex Ž p s 0.007.. For KA seizures, the protective effect was only significant in p13 rats and was limited to hippocampal CA regions and subiculum Ž p s 0.0009.. Histologic assessment of cerebral lesions of p7 and p13 rats in the other two groups showed no significant difference between the animals subjected to seizures 2 h or 24 h post HI and their HI-only controls Ž p ) 0.05.. In conclusion, the results of the present study provide no evidence that seizures in early postnatal development aggravate pre-existing cerebral HI damage. They do suggest that seizures prior to HI or prior to a second seizure confer tolerance to both conditions. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Seizure; Hypoxia–ischemia; Kainic acid; Flurothyl
1. Introduction Clinical observations suggest that hypoxic–ischemic ŽHI. insults to brain during fetal and early postnatal life are an important risk factor for the occurrence of epilepsy in later years w12,19x. These observations have been supported by studies in experimental animals w14x. It is not yet clear whether seizures that complicate perinatal HI brain damage result in further cerebral lesions which might lead to cerebral palsy or mental retardation. Results of a prior pathologic study in immature rats indicate that chemically
) Corresponding [email protected]
author.
Fax:
q 1-717-531-5021;
E-mail:
induced seizures following cerebral HI do not cause any additional brain damage w5x. In that study, HI brain damage was first produced in 7 day postnatal Žp7. rats and then seizures were induced by bicuculline, a drug which blocks the postsynaptic inhibitory action of gamma-aminobutyric acid ŽGABA.. In the present communication, seizures were produced in a similar model either prior to or following cerebral HI in p7 and p13 rats by using two different convulsant agents; specifically, kainic acid ŽKA. and flurothyl, both of which have modes of action different from bicuculline. Our rationale for the conduct of these experiments related to the different brain damaging potential of specific chemical convulsants in adult animals Žrats. w2x and to any potential protective influence of a prior metabolic stress Že.g., hypoxia. on subsequent HI brain damage w11,40x.
0165-3806r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 3 8 0 6 Ž 9 9 . 0 0 0 0 4 - 8
84
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
2. Materials and methods 2.1. Animals Dated, pregnant Wistar rats were purchased ŽCharles River Laboratories, Wilmington, MA. and housed in individual cages. The newborn rats were reduced to 10 per litter and reared with their dams until the time of experiment, at ages p6–p7 or p12–p13. The rationale for using these two age groups was related to the differences in the susceptibility of brain structures to HI in these two age groups w36x. Moreover, the development of rat brain at p6–p7 and p12–p13 relative to the adult rat is roughly comparable to that of human neonate at gestational ages of 32–34 weeks and 38–40 weeks, respectively w24,25x. All procedures were approved by the Animal Care and Use Committee of The Pennsylvania State University College of Medicine. 2.2. Induction of cerebral HI Cerebral HI was produced by ligation of right common carotid artery ŽCCA. and hypoxia w24,37x. Briefly, rats were lightly anesthetized with halothane during which a ligature was applied across the artery. Following 10–12 min of recovery from anesthesia, the animals were individually placed in jars of 500 ml partially submerged in a 378C water bath. The rats were left in jars with lids partially open to room air for 2 to 2.5 h. Then the lids were tightened and a gas mixture of 8% O 2 –92% N2 was circulated via inlet and outlet portals. The rats were exposed to this gas mixture for either 90 min Žp7 rats. or 60 min Žp13 rats.. These two different intervals of HI were used because they produce lesions that are comparable in their severity in the two age groups w37x.
it vaporized. The infusion pump flow rate was the lowest rate of flow needed to produce and maintain seizures. The p6 and p12 rats pre-HI, and p7 and p13 rats post-HI were placed in groups of 4 or 5 in clear plastic chambers of 2.5 l Žp6–p7 rats. or 5 l Žp12–p13 rats. size. They were allowed to sustain seizures for 60 min. Then they were returned to their dam following a 5 min observation period. 2.4. Preliminary studies These studies were performed to determine the optimal dose of KA that produced severe seizures for 60 min or longer Žstatus epilepticus. in the majority of animals but with a low mortality rate. Accordingly, rats of p7 Ž n s 30. and p13 Ž n s 30. received a single s.c. KA injection of 4 mgrkg, 3 mgrkg, or 2 mgrkg. The first two doses resulted in severe seizures for 90–180 min in all animals of both age groups with a mortality rate approaching 40%. Single injections of 2 mgrkg KA resulted in severe seizures of 60 min or longer in the majority of animals with a mortality of less than 5%. Therefore, this dose was selected for the remainder of the study. For the flurothyl experiment, the vapor was adjusted to the lowest concentration of the drug that caused severe seizures for 60 min in all animals. Among rats of both age groups subjected to flurothyl seizures, the mortality rate was close to 50% Ž10 of 25 p7 rats and 17 of 34 p13 rats died.. Histologic examination of brains of the surviving animals at 24 h or 72 h post seizures failed to show any abnormalities. Since seizures induced by single injection of KA or inhalation of flurothyl did not produce neuronal death, a second seizure was induced in additional groups of animals 18 h after the first episode of seizure. These groups consisted of KA seizures in p6 Ž n s 5. and p12 rats Ž n s 5. or of flurothyl seizures in p6 Ž n s 5. and p12 rats Ž n s 5.. No animals died in KA group and 3 out of 10 rats died in the flurothyl group. No histologic abnormality was seen in brains of the surviving animals.
2.3. Induction of seizures 2.5. Experiments Kainic acid ŽKA. Ž2 carboxyP 3-carboxymethyl 4-isopropenylpyrrolidine. was purchased from Sigma ŽSt. Louis, MO. and dissolved in phosphate buffer at a concentration of 0.5 mgrml. A 2 mgrkg dose of freshly made solution of KA was given by subcutaneous Žsc. injection to p6 and p12 rats pre-HI, and p7 and p13 rats post-HI Žsee below for determination of dosage.. Following the KA injection the animals were placed in a padded chamber open to the room air. At the termination of seizures, rats were returned to their dams. Flurothyl, bis-Ž2,2,2 trifluorethyl ether., was purchased from Aldrich Chemical ŽMilwaukee, WI.. The liquid flurothyl was administered by an infusion pump ŽHarvard Apparatus, Pump 22. at a constant rate through the ceiling of a padded, air-tight chamber onto a filter pad from which
Three types of experiments were conducted at each age group and for each convulsant. In some animals ŽGroup 1., seizures were produced at 24 h and again at 6 h prior to HI. This was done to determine the effect of repeated seizures on the subsequent HI brain damage. In the second group ŽGroup 2., seizures were produced at 2 h post-HI. In group 3, seizures were induced 24 h post-HI. In each group, for every seizure plus HI animal an HI only littermate was randomly selected as an HI control and was kept under the same temperature conditions as the convulsing animal. In the KA experiments HI controls were injected s.c. with vehicle. When an animal died during a seizure or HI, the littermate control was also sacrificed. The surviving animals were returned to their dams between the
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
experiments and until the time of sacrifice. In p7 rats that had repeated seizures 24 h and 6 h prior to HI or had seizures 2 h post-HI, examination of their brains were
85
performed 24 h post-HI. The brains of p7 rats that had seizures 24 h post-HI were examined 48 h post-HI. In p13 rats, brains of all animals were examined 72 h post-HI.
Fig. 1. Comparison of light microscopic structural preservation between brains of p8 rats fixed by immersion and fixed by transcardiac perfusion. ŽA. Immersion fixed cortex of a rat that underwent cerebral HI at age p7 and died during seizure 24 h later. ŽB. Perfusion fixed cortex of a rat that underwent cerebral HI at age p7 and examined 24 h later. Dead neurons Ž). with pyknotic or karyorrhectic nuclei are readily distinguished from normal neurons. H and E = 260.
86
Age Žpost- natal days.
Outcome
p7
p13
a
Kainic acid plus HI Group 1c Žseizures prior to HI.
Group 2 Žseizure 2 h post HI.
Group 3 Žseizure 24 h post HI.
Flurothyl plus HI Group 1c Žseizures Prior to HI.
Group 2 Žseizure 2 h post HI.
Group 3 Žseizure 24 h post HI.
Mild seizurea Severe seizureb Died seizure Died, HI Total
2 11 0 0 13
6 4 0 0 10
4 6 0 0 10
0 10 7 0 17
0 11 17 1 29
0 11 11 1 23
Mild Seizure Severe seizure Died, seizure Died, HI Total
0 10 0 0 10
8 12 3 1 24
11 9 0 0 20
0 10 11 0 21
0 12 7 5 24
0 10 8 0 18
Surviving rats that had either grades 1 and 2 seizures or seizures of higher grade with duration less than 60 min Žsee text for explanation.. Surviving rats that had grades 3, 4 or 5 seizures with duration of 60 min or longer. c These rats had seizure 24 h and 6 h prior to HI. They were categorized based on the severity of their first seizure. b
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Table 1 Number of rats in each group with kainic acid induced seizures plus hypoxia–ischemia ŽHI. and flurothyl seizures plus HI
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
This was done due to the presence of delayed neuronal death in this age group w36x. 2.6. Histologic methods Surviving animals were deeply anesthetized and their brains were perfusion fixed intracardially with a solution of 1:1:8 formaldehyde:glacial acetic acid:methanol ŽFAM. Žfor details, see Ref. w37x.. The brains of p7 rats that died during seizures 24 h following HI and their HI only littermate controls were immediately immersion fixed in FAM for histologic examination. At this age, due to the small size of brain and thin skull, this method of fixation is satisfactory for evaluation of dead neurons at 24 h post HI ŽFig. 1.. This method of fixation was not done in p7 rats that died 2 h post-HI due to the difficulty in assessing the population of dead neurons due to the early nature of the lesions. For the same reason, immersion fixation was not used in p13 rats that died 24 h post HI during seizure due to the presence of delayed neuronal death at this age group w36x. Moreover, the larger size of the brain and thicker skull increased the risk of suboptimal fixation. In brief, all fixed brains were sectioned coronally into 2 mm slices. Two brain slices were selected for paraffin embedding. One of the slices Žanterior block., contained the anterior commissure, anterior cerebral cortex, white matter, and anterior striatum. The other slice Žposterior block., included the posterior cortex, white matter, dorsal hippocampus, amygdala, thalamus, and the median eminence. Sections of 6 mm were obtained and stained with hematoxylin and eosin ŽH and E. or with acid fuchsin–Cresyl violet ŽAF–CV.. For identification of the anatomic structures, a stereotaxic atlas of developing rat brain was used w31x. Evidence of irreversible neuronal damage included clearing or acidophilia of perikaryon and nuclear changes such as pyknosis or karyorrhexis. The degree of damage in various regions was graded as following: Grade 0 Žno damage.; grade 1 consisted of only a few dead cells roughly not exceeding 5%; grade 5 consisted of mild damage involving greater than 5% but not exceeding onequarter population of cells; grade 10 consisted of moderate damage with the population of damaged neurons greater than one-quarter, but not exceeding one-half of the cell population; grade 15 consisted of moderately severe damage with the population of dead neurons being greater than one-half, but not exceeding the three-quarters of total; grade 20 consisted of severe damage involving greater than three-quarters of the cell population, including infarction. The histologic evaluation of the damage was made under a 20 = objective in a microscope equipped with an ocular grid Žfor details, see Ref. w5x.. Evaluation of each structure will be described separately. In the cerebral cortex, both anterior and posterior cortex were divided into six segments, consisting of medial, dorsal superior, dorsal
87
inferior, lateral superior, lateral inferior, and entorhinal cortex plus olfactory tubercle Žsee Fig. 1 in Ref. w5x.. The dorsal hippocampus was divided into the following regions: subiculum, medial CA1, lateral CA1, CA2rlateral CA3, medial CA3, and CA4. Thalamus, striatum and amygdala were divided into lateral and medial parts. 2.7. Statistical analysis Statistical comparisons between seizure plus HI and HI only control groups were based on the Wilcoxon Rank Sum test. All analyses were stratified by age Žp7 and p13., seizure inducing agents ŽKA and flurothyl. and time of seizures Ž24 h and 6 h prior to HI, 2 h post-HI, and 24 h post-HI.. To assess the impact of animal mortality during seizures, when sufficient data existed we also compared animals that survived seizures with those that did not. In addition, we compared animals that died during flurothyl seizures with their HI only littermate controls when sufficient data were available Žsee Section 3..
Fig. 2. Dot chart comparison of change in seizure susceptibility 18 h after kainic acid ŽKA. seizures in rats of postnatal day 6 Žp6. and postnatal day 12 Žp12.. Points indicate the length of first seizure minus the second seizure. Note a reduction in the duration of the second seizure in both p6 Ž ps 0.001. and p12 Ž ps 0.001. rats.
88
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
3. Results 3.1. BehaÕioral obserÕations and mortality In KA injected animals, seizures followed a latency of 20 to 30 min and were similar in the two age groups Žp6–7 and p12–13.. Seizures began with scratching-like move-
ments of the hind paws, which were followed in most animals by loss of postural control and periodic falling onto one side or on the back plus scratching-like movements. In animals with more severe seizures Žstatus epilepticus., scratching-like movements were soon replaced by clonic or tonic–clonic movements. Based on these observations, the seizures were graded as following: grade 0: no
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
seizures; grade 1: brief scratching-like movements only; grade 2: periodic scratching-like movements and falling; grade 3: animals laying on their side or back with scratching-like movements; grade 4: alternate clonic and tonic– clonic seizures with animals laying on their side or back; grade 5: tonic–clonic seizures often with periods of cyanosis. The duration of seizures ranged from 10 min to 180 min and ended without intervention. In p12 and p13 rats with grades 4 or 5 seizures, ‘wet dog shake’ movements were observed near the end of the seizures. These movements were not seen in p6 and p7 rats. According to the severity and duration of KA induced seizures, rats were arbitrarily divided into two major groups; specifically, grades 3, 4, or 5 seizures of 60 min or longer Žsevere seizures., and grades 3, 4 or 5 seizures of less than 60 min or with lower grades seizures of any length Žmild seizures. ŽTable 1.. Prior HI reduced the severity of subsequent seizures, which was statistically significant in p13 rats at 2 h Ž p s 0.03. as well as at 24 h Ž p s 0.004. of recovery, and in p7 rats at 2 h of recovery Ž p s 0.04. ŽTable 1.. This was not the case in p7 rats at 24 h of recovery Ž p s 0.3.. There was no significant correlation between the animal’s age Ž p ) 0.05. or weight Ž p ) 0.05. and the severity of seizures. Comparison of the effect of the first seizure on the duration of the second seizure in p6 and p12 rats showed a significant reduction in the duration of the second seizure in both age groups Ž p s 0.001. ŽFig. 2.. In the flurothyl experiments, all rats developed tonic– clonic seizures within 2 min with an intensity comparable to grades 3–5 of KA seizures Žsee above.. The rats were allowed to convulse for 60 min at which time the flurothyl flow was discontinued. The behavioral seizures stopped within 5 min of exposure to room air. In p12 and p13 rats, seizures were preceded by a short period of running or circling, which was not seen in p6 and p7 rats. Unlike the KA group, flurothyl rats had a high mortality rate ŽTable 1.. The minimum rate of flurothyl flow required to produce and maintain status epilepticus in rats could not be evaluated for individual rats Žsee Section 2.. 3.2. Neuropathologic alterations The extent of brain damage in various regions will be described separately for the KA group and flurothyl group.
89
3.2.1. KA group No abnormalities were present in brains of surviving rats that sustained seizures only Žsee Section 2.4.. The majority of surviving animals that had cerebral HI with or without KA induced seizures showed lesions that ranged from selective neuronal death to infarcts. Comparison of the histologic grade of damage in various brain regions between the surviving animals that had HI plus seizure Žseizure prior to or following HI. and their HI controls showed the following findings. In p7 rats, no differences were observed in the severity of lesions between the animals that had seizures prior to or following HI and their HI controls Ž p ) 0.05. ŽFig. 3.. In contrast to p7 rats, p13 rats that had seizures prior to HI showed less extensive cerebral lesions than the HI controls ŽFig. 4, top.. This difference was significant for hippocampal CA1 Ž p s 0.03. and CA2rCA3 Ž p s 0.0003. and all CA regions combined plus subiculum Ž p s 0.0009.. The difference was not significant for cerebral cortex Ž p s 0.60., striatum Ž p s 0.60., or thalamus and amygdala combined Ž p s 0.20.. No significant differences were observed in the severity of lesions between the animals that had seizures 2 h or 24 h post-HI and their HI controls ŽFig. 4, bottom.. Due to the variation in seizure intensity among the KA animals and the possibility that the effect of status epilepticus might have been obscured by including the animals with milder seizures, comparisons were also made between the animals with severe seizures Žgrades 3–5 G 60 min. plus HI and their HI controls. The results were still not significant. 3.2.2. Flurothyl group No abnormalities were seen in brains of surviving rats that had seizures only Žsee Section 2.4.. Most surviving animals that had either cerebral HI plus flurothyl induced seizures or HI only, developed lesions that ranged from selective neuronal death to infarcts. Comparison of the histologic lesions in different regions of brain between the rats that had HI plus seizure Žseizure prior to or following HI. and their HI controls showed the following findings. In p7 rats, the group that had seizures prior to HI showed less extensive CA1 Ž p s 0.04. and cortical Ž p s 0.007. damage than the HI controls ŽFig. 5, top.. The differences were not significant for other regions Ž p ) 0.05.. There were no significant differences in the severity of cerebral lesions between the animals that had seizure at 2 h or 24 h post-HI
Fig. 3. Box plot comparison of severity of brain damage in p7 rats subjected to KA seizures 24 h and 6 h prior to hypoxia–ischemia ŽHI. and their HI controls ŽTop., and in p7 rats subjected to KA seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. No significant differences exist for any brain region between the groups that had seizures plus HI and HI controls Ž p ) 0.05.. wThe box plot provides a graphical summary of data by drawing attention to several key factors. The ends of the box Žshaded area. represent the 25th and 75th percentiles of the data, i.e., the box contains the middle 50% of the data points. The heavy horizontal bar inside the box represents the median of the data. Horizontal bars connected to the box by dashed lines represent extreme data points that are within a fixed distance from the box, 1.5 times the interquartile range Žheight of the box.. Horizontal bars not connected to the box by dashed lines represent data points that are not within this fixed distance from the box and may therefore be potential outliers. The box plot may have an unusual appearance when many data points take on the same value. For example, if the 25th and 75th percentiles are equal to each other, then the box will have zero height Žthe median will be equal to the 25th and 75th percentiles by definition..x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
90
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Fig. 4. Box plot comparison of severity of brain damage in p13 rats subjected to KA seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p13 rats subjected to KA seizures 2 h or 24 h post HI and their HI controls ŽBottom.. There is significantly less damage in all hippocampal regions of animals with seizures prior to HI compared to HI controls Ž) p s 0.03 y 0.0003., but not for any brain region between the groups that had seizures post-HI and HI controls Ž p ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
and their HI controls ŽFig. 5, bottom.. In p13 rats, the group that had seizures prior to HI showed significantly
less extensive cerebral lesions in all regions than their HI controls Žfor CA1, p s 0.0004; CA2rCA3, p s 0.01; all
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
91
Fig. 5. Box plot comparison of severity of brain damage in p7 rats subjected to flurothyl seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p7 rats subjected to flurothyl seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. There is significantly less damage in hippocampal CA1 and cerebral cortex of animals with seizures prior to HI compared to HI controls Ž) p s 0.04 y 0.007., but not for any brain region between the groups that had seizures post-HI and HI controls Ž p ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala
CA regions combined, p s 0.002; cerebral cortex, p s 0.001; striatum, p s 0.002; thalamus and amygdala com-
bined, p s 0.02. ŽFig. 6, top.. No significant differences were observed in the severity of cerebral lesions between
92
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Fig. 6. Box plot comparison of severity of brain damage in p13 rats subjected to flurothyl seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p13 rats subjected to flurothyl seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. There is significantly less damage in all brain regions of animals with seizures prior to HI compared to HI controls Ž) p s 0.02 y 0.0004., but not for any brain region between the groups that had seizures post-HI and HI controls Žp ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
the animals that had seizures 2 h or 24 h post-HI and HI controls Ž p ) 0.05. ŽFig. 6, bottom..
To assess the impact of animal mortality during seizures, we compared the extent of brain damage in p7 rats be-
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
tween those that had HI 24 h prior to seizures and died and the rats that survived for additional 24 h. No significant difference was seen in the extent of brain damage between the two groups Ž p ) 0.05.. Comparison was also made between the cerebral lesions of p7 rats that died during flurothyl seizure at 24 h post-HI and their paired HI littermates that were sacrificed at the same time. No significant difference was seen in the extent of brain damage between the two groups Ž p ) 0.05.. Such comparison was not done in p7 rats that died during seizure 2 h post HI and p13 rats that died 2 h or 24 h post HI Žsee Section 2.6..
4. Discussion It is well known that status epilepticus leads to neuronal death and even infarction of specific regions of brain in adult animals including rats and primates w20,21,23,32,35x. Investigations in immature animals of many species, including the present study in p7 and p13 rats, have failed to produce seizure related neuronal death during the early postnatal period w1,5,6,10,22,30,33,41 x. An exception to the finding in immature rats is the observation of neuronal death in the brains of immature rabbits subjected to status epilepticus by the systemic administration of KA or lithium-pilocarpine w9,34x. The reason for the selective vulnerability of the immature brain in rabbit to seizures and its relative resistance in other species remains unexplained. Among the various species, immature rats have been used extensively to study seizures as well as cerebral HI during early development w 1–4,13,22,24,26,29,30,36, 38,39x. In a recent study of immature rats, Cataltepe et al. w5x showed that chemically induced seizures following cerebral HI do not result in additional cerebral lesions. In that study, cerebral HI was first produced in p7 rats, and then single or repeated seizures were induced by the systemic administration of bicuculline, a drug that blocks the postsynaptic inhibitory action of GABA. The present experiments in p7 and p13 rats extend these observations by using chemical convulsants with modes of action different from bicuculline. Kainic acid acts by binding to a specific subtype of the glutamate receptor, while flurothyl opens membrane sodium channels, thereby inducing a hyperexcitable state w8x. The results of the present study provide no evidence that seizures accentuate any pre-existing HI brain damage during early postnatal development. Although unlikely, such a possibility exists if the study is repeated in a larger group of animals. The most significant finding of the present investigation was the protective effect of seizures induced prior to cerebral HI on the severity of the ultimate brain damage. This effect was significant for flurothyl seizures in all brain regions of p13 rats and cortex and hippocampus of p7 rats. For KA seizures, the protective effect was signifi-
93
cant only in p13 rats and was limited to the hippocampus. To our knowledge, such protection against HI conferred by prior seizures has not been demonstrated previously, even in adult animals. It remains to be investigated whether this protective effect of seizures prior to HI is related to the seizure activity per se or to other factors, e.g., hyperthermia or hypoxia induced by the seizures. This issue can be answered only by studies in intubated and paralyzed animals, in which various systemic variables can be monitored and controlled. It is noteworthy that a prior episode of transient sublethal global ischemia or hypoxia in both adult w15,16,18,28x and developing animals w11x as well as pre-ischemic hyperthermia w7,17x confer a protective effect against ischemic or HI brain damage. Terms such as ‘ischemic tolerance’ and ‘hypoxic preconditioning’ have been used to describe these adaptive influences on ischemic or HI insults w11,17x. The results of the present study also suggest that an initial seizure confers partial protection against a subsequent seizure. This ameliorating effect of the first seizure on the second seizure occurred in both p6 and p12 rats, and has not been shown previously in immature animals. A similar observation has been made by Sasahira et al. w27x in bicuculline induced seizures in adult rats. These investigators noted a marked reduction in the severity of a second seizure when the animals were re-challenged with bicuculline 1d or 3d after the first seizure. The reduction in the severity of the second seizure and consequent epileptic brain damage was attributed to the use of diazepam to terminate the first seizure on the prior day. In the present investigation, such resistance to a second seizure was present 18 h after the first seizure despite the absence of a seizure-suppressing drug. The protective effect of a first seizure against the brain damaging potential of a second seizure in adult rats lasts only for few days w27x. The underlying cellular mechanismŽs. of this transient tolerance is not understood. The resistance might be mediated through interference with excitotoxic injury including intracellular calcium entry or by blocking other ionic and metabolic purturbations. Among the cellular mechanisms, induction of stress proteins such as heat shock proteins ŽHSPs., particularly HSP-72, have been implicated in both cellular injury and tolerance. An early transient induction of HSP-72 in the hippocampus of adult rats subjected to seizures suggests that this protein might play a role in the adaptive tolerance to a second seizure challenge w27x. However, it must be emphasized that the stress induced tolerance might not be type specific. The tolerance induced by a prior seizure or hypoxia might not be directed only against a subsequent seizure and ischemic insult, respectively. The findings of the present study suggest that seizures not only induce tolerance against a subsequent seizure but also confer tolerance to cerebral HI. In conclusion, the findings of the present study in immature rats provide no evidence that seizures accentuate
94
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
pre-existing HI brain damage, but they do suggest that seizures induced prior to HI or prior to a second seizure confer tolerance for both conditions. Further studies are required to clarify whether this tolerance is related to seizure activity per se or is due to systemic perturbations which arise from the seizure activity.
w15x
w16x
w17x
Acknowledgements This research was supported by grant number P01 HD30704 from The National Institute of Child Health and Human Development. The authors also thank Tina Eberly for secretarial assistance.
w18x
w19x
w20x
References w21x w1x B.J. Albala, S.L. Moshe, ´ R. Okada, Kainic-acid-induced seizures: a developmental study, Dev. Brain Res. 13 Ž1984. 139–148. w2x Y. Ben-Ari, E. Tremblay, D. Riche, G. Ghilini, R. Naquet, Electrographic, clinical and pathological alterations following systemic administration of kainic acid, bicuculline or pentetrazole: metabolic mapping using the deoxyglucose method with special reference to the pathology of epilepsy, Neuroscience 6 Ž1981. 1361–1391. w3x M.L. Berger, E. Tremblay, L. Nitecka, Y. Ben-Ari, Maturation of kainic acid seizure-brain damage syndrome in the rat. III: Postnatal development of kainic acid binding sites in the limbic system, Neuroscience 13 Ž1984. 1095–1104. w4x O. Cataltepe, T.F. Barron, D.F. Heitjan, R.C. Vannucci, J. Towfighi, Effect of hypoxiarischemia on bicuculline-induced seizures in immature rats: behavioral and electrocortical phenomena, Epilepsia 36 Ž1995. 396–403. w5x O. Cataltepe, R.C. Vannucci, D.F. Heitjan, J. Towfighi, Effect of status epilepticus on hypoxic–ischemic brain damage in the immature rat, Pediatr. Res. 38 Ž1995. 251–257. w6x E.A. Cavalheiro, D.F. Silva, W.A. Turski, L.S. Calderazzo-Filho, Z.A. Bortolotto, L. Turski, The susceptibility of rats to pilocarpineinduced seizures is age-dependent, Dev. Brain Res. 37 Ž1987. 43–58. w7x M. Chopp, H. Chen, K.-L. Ho, M.O. Dereski, E. Brown, F.W. Hetzel, K.M.A. Welch, Transient hyperthermia protects against subsequent forebrain ischemic cell damage in the rat, Neurology 39 Ž1989. 1396–1398. w8x S. Cohen, A. Goldschmid, G. Shtacher, S. Srebrenik, S. Gitter, The inhalation convulsants: a pharmacodynamic approach, Mol. Pharmacol. 11 Ž1975. 379–385. w9x J.E. Franck, P.A. Schwartzkroin, Immature rabbit hippocampus is damaged by systemic but not intraventricular kainic acid, Dev. Brain Res. 13 Ž1984. 219–227. w10x D.G. Fujikawa, B. Soderfeldt, C.G. Wasterlain, Neuropathological ¨ changes during generalized seizures in newborn monkeys, Epilepsy Res. 12 Ž1992. 243–251. w11x J.M. Gidday, J.C. Fitzgibbons, A.R. Shah, T.S. Park, Neuroprotection from ischemic brain injury by hypoxic preconditioning in the neonatal rat, Neurosci. Lett. 168 Ž1994. 221–224. w12x A. Hill, J.J. Volpe, Hypoxic–ischemic encephalopathy of the newborn, in: K.F. Swaiman ŽEd.., Pediatric Neurology: Principles and Practice, CV Mosby, St. Louis, 1989, pp. 373–392. w13x G.L. Holmes, The long-term effects of seizures on the developing brain: clinical and laboratory issues, Brain Dev. 13 Ž1991. 393–409. w14x F.E. Jensen, C.D. Applegate, D. Holtzman, T.R. Belin, J.L. Burch-
w22x
w23x
w24x
w25x
w26x
w27x
w28x w29x w30x
w31x w32x
w33x
w34x w35x
fiel, Epileptogenic effect of hypoxia in the immature rodent brain, Ann. Neurol. 29 Ž1991. 629–637. T. Kirino, Y. Tsujita, A. Tamura, Induced tolerance to ischemia in gerbil hippocampal neurons, J. Cereb. Blood Flow Metab. 11 Ž1991. 299–307. K. Kitagawa, M. Matsumoto, M. Tagaya, R. Hata, H. Ueda, M. Niinobe, N. Handa, R. Fukunaga, K. Kimura, K. Mikoshiba, T. Kamada, ‘Ischemic tolerance’ phenomenon found in the brain, Brain Res. 528 Ž1990. 21–24. K. Kitagawa, M. Matsumoto, M. Tagaya, K. Kuwabara, R. Hata, N. Handa, R. Fukunaga, K. Kimura, T. Kamada, Hyperthermia-induced neuronal protection against ischemic injury in gerbils, J. Cereb. Blood Flow Metab. 11 Ž1991. 449–452. Y. Liu, H. Kato, N. Nakata, K. Kogure, Protection of rat hippocampus against ischemic neuronal damage by pretreatment with sublethal ischemia, Brain Res. 586 Ž1992. 121–124. C.T. Lombroso, J.L. Burchfiel, Etiologic and preventive aspects of epilepsy in the child—bridging the gap between laboratory and clinic, Epilepsia 28 Ž1987. S25–35. B.S. Meldrum, J.B. Brierley, Prolonged epileptic seizures in primates: ischemic cell change and its relation to ictal physiological events, Arch. Neurol. 28 Ž1973. 10–17. G. Nevander, M. Ingvar, R. Auer, B.K. Siesjo, ¨ Status epilepticus in well-oxygenated rats causes neuronal necrosis, Ann. Neurol. 18 Ž1985. 281–290. L. Nitecka, E. Tremblay, G. Charton, J.P. Bouillot, M.L. Berger, Y. Ben-Ari, Maturation of kainic acid seizure-brain damage syndrome in the rat. II: Histopathological sequeale, Neuroscience 13 Ž1984. 1073–1094. B.K. O’Connell, J. Towfighi, W.A. Kofke, R.A. Hawkins, Neuronal lesions in mercaptopropionic acid-induced status epilepticus, Acta Neuropathol. 77 Ž1988. 47–54. J.E. Rice, R.C. Vannucci, J.B. Brierley, The influence of immaturity on hypoxic–ischemic brain damage in the rat, Ann. Neurol. 9 Ž1981. 131–141. H.J. Romijn, M.A. Hoffman, A. Gramsbergen, At what age is the developing cerebral cortex of the rat comparable to that of the full-term newborn human baby?, Early Hum. Dev. 26 Ž1991. 61–67. H.J. Romijn, A.W.J.W. Janszen, M.J.D. van Voorst, R.M. Buijs, R. Balazs, ´ D.F. Swaab, Perinatal hypoxic ischemic encephalopathy affects the proportion of GABA-immunoreactive neurons in the cerebral cortex of the rat, Brain Res. 592 Ž1992. 17–28. M. Sasahira, T. Lowry, R.P. Simon, D.A. Greenberg, Epileptic tolerance: prior seizures protect against seizure-induced neuronal injury, Neurosci. Lett. 185 Ž1995. 95–98. R.P. Simon, M. Niiro, R. Gwinn, Prior ischemic stress protects against experimental stroke, Neurosci. Lett. 163 Ž1993. 135–137. E.F. Sperber, S.L. Moshe, ´ Age related differences in seizure susceptibility to flurothyl, Dev. Brain Res. 39 Ž1988. 295–297. E.F. Sperber, P.K. Stanton, K.Z. Haas, S. Shinnar, S.L. Moshe, ´ Resistance of the immature brain to hippocampal damage following flurothyl-induced status epilepticus, Ann. Neurol. 30 Ž1991. 495. N.M. Sherwood, P.A. Timuras, A Stereotaxic Atlas of the Developing Rat Brain, University of California Press, Berkley, 1970. B. Soderfeldt, H. Kalimo, Y. Olsson, B.K. Siesjo, ¨ ¨ Pathogenesis of brain lesions caused by experimental epilepsy: light and electron microscopic changes in the rat cerebral cortex following bicuculline induced status epilepticus, Acta Neuropathol. 54 Ž1981. 219–231. C.E. Stafstrom, J.L. Thompson, G.L. Holmes, Kainic acid seizures in the developing brain: status epilepticus and spontaneous recurrent seizures, Dev. Brain Res. 65 Ž1992. 227–236. K. Thompson, C. Wasterlain, Lithium-pilocarpine status epilepticus in the immature rabbit, Dev. Brain Res. 100 Ž1997. 1–4. J. Towfighi, W.A. Kofke, B.K. O’Connell, C. Housman, J.M. Graybeal, Substantia nigra lesions in mercaptopropionic acid induced status epilepticus: a light and electron microscopic study, Acta Neuropathol. 77 Ž1989. 612–620.
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95 w36x J. Towfighi, D. Mauger, Temporal evolution of neuronal changes in cerebral hypoxia–ischemia in developing rats: a quantitative light microscopic study, Dev. Brain Res. 109 Ž1998. 169–177. w37x J. Towfighi, D. Mauger, R.C. Vannucci, S.J. Vannucci, Influence of age on the cerebral lesions in an immature rat model of cerebral hypoxia–ischemia: a light microcopic study, Dev. Brain Res. 100 Ž1997. 149–160. w38x J. Towfighi, N. Zec, C. Housman, R.C. Vannucci, Temporal evolution of neuropathologic changes in an immature rat model of cere-
95
bral hypoxia: a light microscopic study, Acta Neuropathol. 90 Ž1995. 375–386. w39x R.C. Vannucci, Experimental biology of cerebral hypoxia–ischemia: relation to perinatal brain damage, Pediatr. Res. 27 Ž1990. 317–326. w40x R.C. Vannucci, J. Towfighi, S.J. Vannucci, Hypoxic preconditioning and hypoxic–ischemic brain damage in the immature rat: pathologic and metabolic correlates, J. Neurochem. 71 Ž1998. 1215–1220. w41x C.G. Wasterlain, F. Plum, Vulnerability of developing rat brain to electroconvulsive seizures, Arch. Neurol. 29 Ž1973. 38–45.
Research report
Effect of seizures on cerebral hypoxic–ischemic lesions in immature rats Javad Towfighi
a,)
, Cathy Housman a , David Mauger b, Robert C. Vannucci
c
a
b
Department of Pathology (Anatomic Pathology), The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, P.O. Box 850, Hershey, PA 17033-0850, USA Department of Health EÕaluation Sciences, The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, Hershey, PA 17033-0850, USA c Department of Pediatrics, The Milton S. Hershey Medical Center, The PennsylÕania State UniÕersity College of Medicine, Hershey, PA 17033-0850, USA Accepted 22 December 1998
Abstract The present investigation was designed to study the effect of chemically induced seizures on cerebral hypoxic–ischemic ŽHI. damage in immature animals. Accordingly, cerebral HI was produced in 7-day postnatal Žp7. rats and p13 rats by combined unilateral common carotid artery ligation and hypoxia with 8% oxygen. Seizures were induced chemically by the subcutaneous injection of kainic acid ŽKA. or inhalation of flurothyl vapor. Three types of experiments were conducted in each age group and for each convulsant. In some animals Žgroup 1., seizures were produced at 24 h and again at 6 h prior to HI. In groups 2 and 3, seizures were induced 2 h or 24 h post HI, respectively. The results indicate that in group 1 animals, the first seizure significantly reduced duration of the second seizure challenge 18 h later at both p7 and p13 Ž p s 0.001.. Histologic examination of brains of animals in group 1 subjected to seizures prior to HI and their HI-only controls showed that seizures prior to HI conferred protection against cerebral damage. This effect was significant for flurothyl seizures in p13 rats for all cerebral regions, especially hippocampal CA1 Ž p s 0.0004., and in p7 rats for hippocampus Ž p s 0.04. and particularly cerebral cortex Ž p s 0.007.. For KA seizures, the protective effect was only significant in p13 rats and was limited to hippocampal CA regions and subiculum Ž p s 0.0009.. Histologic assessment of cerebral lesions of p7 and p13 rats in the other two groups showed no significant difference between the animals subjected to seizures 2 h or 24 h post HI and their HI-only controls Ž p ) 0.05.. In conclusion, the results of the present study provide no evidence that seizures in early postnatal development aggravate pre-existing cerebral HI damage. They do suggest that seizures prior to HI or prior to a second seizure confer tolerance to both conditions. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Seizure; Hypoxia–ischemia; Kainic acid; Flurothyl
1. Introduction Clinical observations suggest that hypoxic–ischemic ŽHI. insults to brain during fetal and early postnatal life are an important risk factor for the occurrence of epilepsy in later years w12,19x. These observations have been supported by studies in experimental animals w14x. It is not yet clear whether seizures that complicate perinatal HI brain damage result in further cerebral lesions which might lead to cerebral palsy or mental retardation. Results of a prior pathologic study in immature rats indicate that chemically
) Corresponding [email protected]
author.
Fax:
q 1-717-531-5021;
E-mail:
induced seizures following cerebral HI do not cause any additional brain damage w5x. In that study, HI brain damage was first produced in 7 day postnatal Žp7. rats and then seizures were induced by bicuculline, a drug which blocks the postsynaptic inhibitory action of gamma-aminobutyric acid ŽGABA.. In the present communication, seizures were produced in a similar model either prior to or following cerebral HI in p7 and p13 rats by using two different convulsant agents; specifically, kainic acid ŽKA. and flurothyl, both of which have modes of action different from bicuculline. Our rationale for the conduct of these experiments related to the different brain damaging potential of specific chemical convulsants in adult animals Žrats. w2x and to any potential protective influence of a prior metabolic stress Že.g., hypoxia. on subsequent HI brain damage w11,40x.
0165-3806r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 3 8 0 6 Ž 9 9 . 0 0 0 0 4 - 8
84
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
2. Materials and methods 2.1. Animals Dated, pregnant Wistar rats were purchased ŽCharles River Laboratories, Wilmington, MA. and housed in individual cages. The newborn rats were reduced to 10 per litter and reared with their dams until the time of experiment, at ages p6–p7 or p12–p13. The rationale for using these two age groups was related to the differences in the susceptibility of brain structures to HI in these two age groups w36x. Moreover, the development of rat brain at p6–p7 and p12–p13 relative to the adult rat is roughly comparable to that of human neonate at gestational ages of 32–34 weeks and 38–40 weeks, respectively w24,25x. All procedures were approved by the Animal Care and Use Committee of The Pennsylvania State University College of Medicine. 2.2. Induction of cerebral HI Cerebral HI was produced by ligation of right common carotid artery ŽCCA. and hypoxia w24,37x. Briefly, rats were lightly anesthetized with halothane during which a ligature was applied across the artery. Following 10–12 min of recovery from anesthesia, the animals were individually placed in jars of 500 ml partially submerged in a 378C water bath. The rats were left in jars with lids partially open to room air for 2 to 2.5 h. Then the lids were tightened and a gas mixture of 8% O 2 –92% N2 was circulated via inlet and outlet portals. The rats were exposed to this gas mixture for either 90 min Žp7 rats. or 60 min Žp13 rats.. These two different intervals of HI were used because they produce lesions that are comparable in their severity in the two age groups w37x.
it vaporized. The infusion pump flow rate was the lowest rate of flow needed to produce and maintain seizures. The p6 and p12 rats pre-HI, and p7 and p13 rats post-HI were placed in groups of 4 or 5 in clear plastic chambers of 2.5 l Žp6–p7 rats. or 5 l Žp12–p13 rats. size. They were allowed to sustain seizures for 60 min. Then they were returned to their dam following a 5 min observation period. 2.4. Preliminary studies These studies were performed to determine the optimal dose of KA that produced severe seizures for 60 min or longer Žstatus epilepticus. in the majority of animals but with a low mortality rate. Accordingly, rats of p7 Ž n s 30. and p13 Ž n s 30. received a single s.c. KA injection of 4 mgrkg, 3 mgrkg, or 2 mgrkg. The first two doses resulted in severe seizures for 90–180 min in all animals of both age groups with a mortality rate approaching 40%. Single injections of 2 mgrkg KA resulted in severe seizures of 60 min or longer in the majority of animals with a mortality of less than 5%. Therefore, this dose was selected for the remainder of the study. For the flurothyl experiment, the vapor was adjusted to the lowest concentration of the drug that caused severe seizures for 60 min in all animals. Among rats of both age groups subjected to flurothyl seizures, the mortality rate was close to 50% Ž10 of 25 p7 rats and 17 of 34 p13 rats died.. Histologic examination of brains of the surviving animals at 24 h or 72 h post seizures failed to show any abnormalities. Since seizures induced by single injection of KA or inhalation of flurothyl did not produce neuronal death, a second seizure was induced in additional groups of animals 18 h after the first episode of seizure. These groups consisted of KA seizures in p6 Ž n s 5. and p12 rats Ž n s 5. or of flurothyl seizures in p6 Ž n s 5. and p12 rats Ž n s 5.. No animals died in KA group and 3 out of 10 rats died in the flurothyl group. No histologic abnormality was seen in brains of the surviving animals.
2.3. Induction of seizures 2.5. Experiments Kainic acid ŽKA. Ž2 carboxyP 3-carboxymethyl 4-isopropenylpyrrolidine. was purchased from Sigma ŽSt. Louis, MO. and dissolved in phosphate buffer at a concentration of 0.5 mgrml. A 2 mgrkg dose of freshly made solution of KA was given by subcutaneous Žsc. injection to p6 and p12 rats pre-HI, and p7 and p13 rats post-HI Žsee below for determination of dosage.. Following the KA injection the animals were placed in a padded chamber open to the room air. At the termination of seizures, rats were returned to their dams. Flurothyl, bis-Ž2,2,2 trifluorethyl ether., was purchased from Aldrich Chemical ŽMilwaukee, WI.. The liquid flurothyl was administered by an infusion pump ŽHarvard Apparatus, Pump 22. at a constant rate through the ceiling of a padded, air-tight chamber onto a filter pad from which
Three types of experiments were conducted at each age group and for each convulsant. In some animals ŽGroup 1., seizures were produced at 24 h and again at 6 h prior to HI. This was done to determine the effect of repeated seizures on the subsequent HI brain damage. In the second group ŽGroup 2., seizures were produced at 2 h post-HI. In group 3, seizures were induced 24 h post-HI. In each group, for every seizure plus HI animal an HI only littermate was randomly selected as an HI control and was kept under the same temperature conditions as the convulsing animal. In the KA experiments HI controls were injected s.c. with vehicle. When an animal died during a seizure or HI, the littermate control was also sacrificed. The surviving animals were returned to their dams between the
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
experiments and until the time of sacrifice. In p7 rats that had repeated seizures 24 h and 6 h prior to HI or had seizures 2 h post-HI, examination of their brains were
85
performed 24 h post-HI. The brains of p7 rats that had seizures 24 h post-HI were examined 48 h post-HI. In p13 rats, brains of all animals were examined 72 h post-HI.
Fig. 1. Comparison of light microscopic structural preservation between brains of p8 rats fixed by immersion and fixed by transcardiac perfusion. ŽA. Immersion fixed cortex of a rat that underwent cerebral HI at age p7 and died during seizure 24 h later. ŽB. Perfusion fixed cortex of a rat that underwent cerebral HI at age p7 and examined 24 h later. Dead neurons Ž). with pyknotic or karyorrhectic nuclei are readily distinguished from normal neurons. H and E = 260.
86
Age Žpost- natal days.
Outcome
p7
p13
a
Kainic acid plus HI Group 1c Žseizures prior to HI.
Group 2 Žseizure 2 h post HI.
Group 3 Žseizure 24 h post HI.
Flurothyl plus HI Group 1c Žseizures Prior to HI.
Group 2 Žseizure 2 h post HI.
Group 3 Žseizure 24 h post HI.
Mild seizurea Severe seizureb Died seizure Died, HI Total
2 11 0 0 13
6 4 0 0 10
4 6 0 0 10
0 10 7 0 17
0 11 17 1 29
0 11 11 1 23
Mild Seizure Severe seizure Died, seizure Died, HI Total
0 10 0 0 10
8 12 3 1 24
11 9 0 0 20
0 10 11 0 21
0 12 7 5 24
0 10 8 0 18
Surviving rats that had either grades 1 and 2 seizures or seizures of higher grade with duration less than 60 min Žsee text for explanation.. Surviving rats that had grades 3, 4 or 5 seizures with duration of 60 min or longer. c These rats had seizure 24 h and 6 h prior to HI. They were categorized based on the severity of their first seizure. b
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Table 1 Number of rats in each group with kainic acid induced seizures plus hypoxia–ischemia ŽHI. and flurothyl seizures plus HI
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
This was done due to the presence of delayed neuronal death in this age group w36x. 2.6. Histologic methods Surviving animals were deeply anesthetized and their brains were perfusion fixed intracardially with a solution of 1:1:8 formaldehyde:glacial acetic acid:methanol ŽFAM. Žfor details, see Ref. w37x.. The brains of p7 rats that died during seizures 24 h following HI and their HI only littermate controls were immediately immersion fixed in FAM for histologic examination. At this age, due to the small size of brain and thin skull, this method of fixation is satisfactory for evaluation of dead neurons at 24 h post HI ŽFig. 1.. This method of fixation was not done in p7 rats that died 2 h post-HI due to the difficulty in assessing the population of dead neurons due to the early nature of the lesions. For the same reason, immersion fixation was not used in p13 rats that died 24 h post HI during seizure due to the presence of delayed neuronal death at this age group w36x. Moreover, the larger size of the brain and thicker skull increased the risk of suboptimal fixation. In brief, all fixed brains were sectioned coronally into 2 mm slices. Two brain slices were selected for paraffin embedding. One of the slices Žanterior block., contained the anterior commissure, anterior cerebral cortex, white matter, and anterior striatum. The other slice Žposterior block., included the posterior cortex, white matter, dorsal hippocampus, amygdala, thalamus, and the median eminence. Sections of 6 mm were obtained and stained with hematoxylin and eosin ŽH and E. or with acid fuchsin–Cresyl violet ŽAF–CV.. For identification of the anatomic structures, a stereotaxic atlas of developing rat brain was used w31x. Evidence of irreversible neuronal damage included clearing or acidophilia of perikaryon and nuclear changes such as pyknosis or karyorrhexis. The degree of damage in various regions was graded as following: Grade 0 Žno damage.; grade 1 consisted of only a few dead cells roughly not exceeding 5%; grade 5 consisted of mild damage involving greater than 5% but not exceeding onequarter population of cells; grade 10 consisted of moderate damage with the population of damaged neurons greater than one-quarter, but not exceeding one-half of the cell population; grade 15 consisted of moderately severe damage with the population of dead neurons being greater than one-half, but not exceeding the three-quarters of total; grade 20 consisted of severe damage involving greater than three-quarters of the cell population, including infarction. The histologic evaluation of the damage was made under a 20 = objective in a microscope equipped with an ocular grid Žfor details, see Ref. w5x.. Evaluation of each structure will be described separately. In the cerebral cortex, both anterior and posterior cortex were divided into six segments, consisting of medial, dorsal superior, dorsal
87
inferior, lateral superior, lateral inferior, and entorhinal cortex plus olfactory tubercle Žsee Fig. 1 in Ref. w5x.. The dorsal hippocampus was divided into the following regions: subiculum, medial CA1, lateral CA1, CA2rlateral CA3, medial CA3, and CA4. Thalamus, striatum and amygdala were divided into lateral and medial parts. 2.7. Statistical analysis Statistical comparisons between seizure plus HI and HI only control groups were based on the Wilcoxon Rank Sum test. All analyses were stratified by age Žp7 and p13., seizure inducing agents ŽKA and flurothyl. and time of seizures Ž24 h and 6 h prior to HI, 2 h post-HI, and 24 h post-HI.. To assess the impact of animal mortality during seizures, when sufficient data existed we also compared animals that survived seizures with those that did not. In addition, we compared animals that died during flurothyl seizures with their HI only littermate controls when sufficient data were available Žsee Section 3..
Fig. 2. Dot chart comparison of change in seizure susceptibility 18 h after kainic acid ŽKA. seizures in rats of postnatal day 6 Žp6. and postnatal day 12 Žp12.. Points indicate the length of first seizure minus the second seizure. Note a reduction in the duration of the second seizure in both p6 Ž ps 0.001. and p12 Ž ps 0.001. rats.
88
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
3. Results 3.1. BehaÕioral obserÕations and mortality In KA injected animals, seizures followed a latency of 20 to 30 min and were similar in the two age groups Žp6–7 and p12–13.. Seizures began with scratching-like move-
ments of the hind paws, which were followed in most animals by loss of postural control and periodic falling onto one side or on the back plus scratching-like movements. In animals with more severe seizures Žstatus epilepticus., scratching-like movements were soon replaced by clonic or tonic–clonic movements. Based on these observations, the seizures were graded as following: grade 0: no
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
seizures; grade 1: brief scratching-like movements only; grade 2: periodic scratching-like movements and falling; grade 3: animals laying on their side or back with scratching-like movements; grade 4: alternate clonic and tonic– clonic seizures with animals laying on their side or back; grade 5: tonic–clonic seizures often with periods of cyanosis. The duration of seizures ranged from 10 min to 180 min and ended without intervention. In p12 and p13 rats with grades 4 or 5 seizures, ‘wet dog shake’ movements were observed near the end of the seizures. These movements were not seen in p6 and p7 rats. According to the severity and duration of KA induced seizures, rats were arbitrarily divided into two major groups; specifically, grades 3, 4, or 5 seizures of 60 min or longer Žsevere seizures., and grades 3, 4 or 5 seizures of less than 60 min or with lower grades seizures of any length Žmild seizures. ŽTable 1.. Prior HI reduced the severity of subsequent seizures, which was statistically significant in p13 rats at 2 h Ž p s 0.03. as well as at 24 h Ž p s 0.004. of recovery, and in p7 rats at 2 h of recovery Ž p s 0.04. ŽTable 1.. This was not the case in p7 rats at 24 h of recovery Ž p s 0.3.. There was no significant correlation between the animal’s age Ž p ) 0.05. or weight Ž p ) 0.05. and the severity of seizures. Comparison of the effect of the first seizure on the duration of the second seizure in p6 and p12 rats showed a significant reduction in the duration of the second seizure in both age groups Ž p s 0.001. ŽFig. 2.. In the flurothyl experiments, all rats developed tonic– clonic seizures within 2 min with an intensity comparable to grades 3–5 of KA seizures Žsee above.. The rats were allowed to convulse for 60 min at which time the flurothyl flow was discontinued. The behavioral seizures stopped within 5 min of exposure to room air. In p12 and p13 rats, seizures were preceded by a short period of running or circling, which was not seen in p6 and p7 rats. Unlike the KA group, flurothyl rats had a high mortality rate ŽTable 1.. The minimum rate of flurothyl flow required to produce and maintain status epilepticus in rats could not be evaluated for individual rats Žsee Section 2.. 3.2. Neuropathologic alterations The extent of brain damage in various regions will be described separately for the KA group and flurothyl group.
89
3.2.1. KA group No abnormalities were present in brains of surviving rats that sustained seizures only Žsee Section 2.4.. The majority of surviving animals that had cerebral HI with or without KA induced seizures showed lesions that ranged from selective neuronal death to infarcts. Comparison of the histologic grade of damage in various brain regions between the surviving animals that had HI plus seizure Žseizure prior to or following HI. and their HI controls showed the following findings. In p7 rats, no differences were observed in the severity of lesions between the animals that had seizures prior to or following HI and their HI controls Ž p ) 0.05. ŽFig. 3.. In contrast to p7 rats, p13 rats that had seizures prior to HI showed less extensive cerebral lesions than the HI controls ŽFig. 4, top.. This difference was significant for hippocampal CA1 Ž p s 0.03. and CA2rCA3 Ž p s 0.0003. and all CA regions combined plus subiculum Ž p s 0.0009.. The difference was not significant for cerebral cortex Ž p s 0.60., striatum Ž p s 0.60., or thalamus and amygdala combined Ž p s 0.20.. No significant differences were observed in the severity of lesions between the animals that had seizures 2 h or 24 h post-HI and their HI controls ŽFig. 4, bottom.. Due to the variation in seizure intensity among the KA animals and the possibility that the effect of status epilepticus might have been obscured by including the animals with milder seizures, comparisons were also made between the animals with severe seizures Žgrades 3–5 G 60 min. plus HI and their HI controls. The results were still not significant. 3.2.2. Flurothyl group No abnormalities were seen in brains of surviving rats that had seizures only Žsee Section 2.4.. Most surviving animals that had either cerebral HI plus flurothyl induced seizures or HI only, developed lesions that ranged from selective neuronal death to infarcts. Comparison of the histologic lesions in different regions of brain between the rats that had HI plus seizure Žseizure prior to or following HI. and their HI controls showed the following findings. In p7 rats, the group that had seizures prior to HI showed less extensive CA1 Ž p s 0.04. and cortical Ž p s 0.007. damage than the HI controls ŽFig. 5, top.. The differences were not significant for other regions Ž p ) 0.05.. There were no significant differences in the severity of cerebral lesions between the animals that had seizure at 2 h or 24 h post-HI
Fig. 3. Box plot comparison of severity of brain damage in p7 rats subjected to KA seizures 24 h and 6 h prior to hypoxia–ischemia ŽHI. and their HI controls ŽTop., and in p7 rats subjected to KA seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. No significant differences exist for any brain region between the groups that had seizures plus HI and HI controls Ž p ) 0.05.. wThe box plot provides a graphical summary of data by drawing attention to several key factors. The ends of the box Žshaded area. represent the 25th and 75th percentiles of the data, i.e., the box contains the middle 50% of the data points. The heavy horizontal bar inside the box represents the median of the data. Horizontal bars connected to the box by dashed lines represent extreme data points that are within a fixed distance from the box, 1.5 times the interquartile range Žheight of the box.. Horizontal bars not connected to the box by dashed lines represent data points that are not within this fixed distance from the box and may therefore be potential outliers. The box plot may have an unusual appearance when many data points take on the same value. For example, if the 25th and 75th percentiles are equal to each other, then the box will have zero height Žthe median will be equal to the 25th and 75th percentiles by definition..x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
90
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Fig. 4. Box plot comparison of severity of brain damage in p13 rats subjected to KA seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p13 rats subjected to KA seizures 2 h or 24 h post HI and their HI controls ŽBottom.. There is significantly less damage in all hippocampal regions of animals with seizures prior to HI compared to HI controls Ž) p s 0.03 y 0.0003., but not for any brain region between the groups that had seizures post-HI and HI controls Ž p ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
and their HI controls ŽFig. 5, bottom.. In p13 rats, the group that had seizures prior to HI showed significantly
less extensive cerebral lesions in all regions than their HI controls Žfor CA1, p s 0.0004; CA2rCA3, p s 0.01; all
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
91
Fig. 5. Box plot comparison of severity of brain damage in p7 rats subjected to flurothyl seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p7 rats subjected to flurothyl seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. There is significantly less damage in hippocampal CA1 and cerebral cortex of animals with seizures prior to HI compared to HI controls Ž) p s 0.04 y 0.007., but not for any brain region between the groups that had seizures post-HI and HI controls Ž p ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala
CA regions combined, p s 0.002; cerebral cortex, p s 0.001; striatum, p s 0.002; thalamus and amygdala com-
bined, p s 0.02. ŽFig. 6, top.. No significant differences were observed in the severity of cerebral lesions between
92
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
Fig. 6. Box plot comparison of severity of brain damage in p13 rats subjected to flurothyl seizures 24 h and 6 h prior to HI and their HI controls ŽTop., and in p13 rats subjected to flurothyl seizures 2 h or 24 h post-HI and their HI controls ŽBottom.. There is significantly less damage in all brain regions of animals with seizures prior to HI compared to HI controls Ž) p s 0.02 y 0.0004., but not for any brain region between the groups that had seizures post-HI and HI controls Žp ) 0.05.. wFor explanation of box plot, see legend to Fig. 3x Ža. Hippocampus represents all CA regions plus subiculum. Žb. Other represents thalamus and amygdala.
the animals that had seizures 2 h or 24 h post-HI and HI controls Ž p ) 0.05. ŽFig. 6, bottom..
To assess the impact of animal mortality during seizures, we compared the extent of brain damage in p7 rats be-
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
tween those that had HI 24 h prior to seizures and died and the rats that survived for additional 24 h. No significant difference was seen in the extent of brain damage between the two groups Ž p ) 0.05.. Comparison was also made between the cerebral lesions of p7 rats that died during flurothyl seizure at 24 h post-HI and their paired HI littermates that were sacrificed at the same time. No significant difference was seen in the extent of brain damage between the two groups Ž p ) 0.05.. Such comparison was not done in p7 rats that died during seizure 2 h post HI and p13 rats that died 2 h or 24 h post HI Žsee Section 2.6..
4. Discussion It is well known that status epilepticus leads to neuronal death and even infarction of specific regions of brain in adult animals including rats and primates w20,21,23,32,35x. Investigations in immature animals of many species, including the present study in p7 and p13 rats, have failed to produce seizure related neuronal death during the early postnatal period w1,5,6,10,22,30,33,41 x. An exception to the finding in immature rats is the observation of neuronal death in the brains of immature rabbits subjected to status epilepticus by the systemic administration of KA or lithium-pilocarpine w9,34x. The reason for the selective vulnerability of the immature brain in rabbit to seizures and its relative resistance in other species remains unexplained. Among the various species, immature rats have been used extensively to study seizures as well as cerebral HI during early development w 1–4,13,22,24,26,29,30,36, 38,39x. In a recent study of immature rats, Cataltepe et al. w5x showed that chemically induced seizures following cerebral HI do not result in additional cerebral lesions. In that study, cerebral HI was first produced in p7 rats, and then single or repeated seizures were induced by the systemic administration of bicuculline, a drug that blocks the postsynaptic inhibitory action of GABA. The present experiments in p7 and p13 rats extend these observations by using chemical convulsants with modes of action different from bicuculline. Kainic acid acts by binding to a specific subtype of the glutamate receptor, while flurothyl opens membrane sodium channels, thereby inducing a hyperexcitable state w8x. The results of the present study provide no evidence that seizures accentuate any pre-existing HI brain damage during early postnatal development. Although unlikely, such a possibility exists if the study is repeated in a larger group of animals. The most significant finding of the present investigation was the protective effect of seizures induced prior to cerebral HI on the severity of the ultimate brain damage. This effect was significant for flurothyl seizures in all brain regions of p13 rats and cortex and hippocampus of p7 rats. For KA seizures, the protective effect was signifi-
93
cant only in p13 rats and was limited to the hippocampus. To our knowledge, such protection against HI conferred by prior seizures has not been demonstrated previously, even in adult animals. It remains to be investigated whether this protective effect of seizures prior to HI is related to the seizure activity per se or to other factors, e.g., hyperthermia or hypoxia induced by the seizures. This issue can be answered only by studies in intubated and paralyzed animals, in which various systemic variables can be monitored and controlled. It is noteworthy that a prior episode of transient sublethal global ischemia or hypoxia in both adult w15,16,18,28x and developing animals w11x as well as pre-ischemic hyperthermia w7,17x confer a protective effect against ischemic or HI brain damage. Terms such as ‘ischemic tolerance’ and ‘hypoxic preconditioning’ have been used to describe these adaptive influences on ischemic or HI insults w11,17x. The results of the present study also suggest that an initial seizure confers partial protection against a subsequent seizure. This ameliorating effect of the first seizure on the second seizure occurred in both p6 and p12 rats, and has not been shown previously in immature animals. A similar observation has been made by Sasahira et al. w27x in bicuculline induced seizures in adult rats. These investigators noted a marked reduction in the severity of a second seizure when the animals were re-challenged with bicuculline 1d or 3d after the first seizure. The reduction in the severity of the second seizure and consequent epileptic brain damage was attributed to the use of diazepam to terminate the first seizure on the prior day. In the present investigation, such resistance to a second seizure was present 18 h after the first seizure despite the absence of a seizure-suppressing drug. The protective effect of a first seizure against the brain damaging potential of a second seizure in adult rats lasts only for few days w27x. The underlying cellular mechanismŽs. of this transient tolerance is not understood. The resistance might be mediated through interference with excitotoxic injury including intracellular calcium entry or by blocking other ionic and metabolic purturbations. Among the cellular mechanisms, induction of stress proteins such as heat shock proteins ŽHSPs., particularly HSP-72, have been implicated in both cellular injury and tolerance. An early transient induction of HSP-72 in the hippocampus of adult rats subjected to seizures suggests that this protein might play a role in the adaptive tolerance to a second seizure challenge w27x. However, it must be emphasized that the stress induced tolerance might not be type specific. The tolerance induced by a prior seizure or hypoxia might not be directed only against a subsequent seizure and ischemic insult, respectively. The findings of the present study suggest that seizures not only induce tolerance against a subsequent seizure but also confer tolerance to cerebral HI. In conclusion, the findings of the present study in immature rats provide no evidence that seizures accentuate
94
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95
pre-existing HI brain damage, but they do suggest that seizures induced prior to HI or prior to a second seizure confer tolerance for both conditions. Further studies are required to clarify whether this tolerance is related to seizure activity per se or is due to systemic perturbations which arise from the seizure activity.
w15x
w16x
w17x
Acknowledgements This research was supported by grant number P01 HD30704 from The National Institute of Child Health and Human Development. The authors also thank Tina Eberly for secretarial assistance.
w18x
w19x
w20x
References w21x w1x B.J. Albala, S.L. Moshe, ´ R. Okada, Kainic-acid-induced seizures: a developmental study, Dev. Brain Res. 13 Ž1984. 139–148. w2x Y. Ben-Ari, E. Tremblay, D. Riche, G. Ghilini, R. Naquet, Electrographic, clinical and pathological alterations following systemic administration of kainic acid, bicuculline or pentetrazole: metabolic mapping using the deoxyglucose method with special reference to the pathology of epilepsy, Neuroscience 6 Ž1981. 1361–1391. w3x M.L. Berger, E. Tremblay, L. Nitecka, Y. Ben-Ari, Maturation of kainic acid seizure-brain damage syndrome in the rat. III: Postnatal development of kainic acid binding sites in the limbic system, Neuroscience 13 Ž1984. 1095–1104. w4x O. Cataltepe, T.F. Barron, D.F. Heitjan, R.C. Vannucci, J. Towfighi, Effect of hypoxiarischemia on bicuculline-induced seizures in immature rats: behavioral and electrocortical phenomena, Epilepsia 36 Ž1995. 396–403. w5x O. Cataltepe, R.C. Vannucci, D.F. Heitjan, J. Towfighi, Effect of status epilepticus on hypoxic–ischemic brain damage in the immature rat, Pediatr. Res. 38 Ž1995. 251–257. w6x E.A. Cavalheiro, D.F. Silva, W.A. Turski, L.S. Calderazzo-Filho, Z.A. Bortolotto, L. Turski, The susceptibility of rats to pilocarpineinduced seizures is age-dependent, Dev. Brain Res. 37 Ž1987. 43–58. w7x M. Chopp, H. Chen, K.-L. Ho, M.O. Dereski, E. Brown, F.W. Hetzel, K.M.A. Welch, Transient hyperthermia protects against subsequent forebrain ischemic cell damage in the rat, Neurology 39 Ž1989. 1396–1398. w8x S. Cohen, A. Goldschmid, G. Shtacher, S. Srebrenik, S. Gitter, The inhalation convulsants: a pharmacodynamic approach, Mol. Pharmacol. 11 Ž1975. 379–385. w9x J.E. Franck, P.A. Schwartzkroin, Immature rabbit hippocampus is damaged by systemic but not intraventricular kainic acid, Dev. Brain Res. 13 Ž1984. 219–227. w10x D.G. Fujikawa, B. Soderfeldt, C.G. Wasterlain, Neuropathological ¨ changes during generalized seizures in newborn monkeys, Epilepsy Res. 12 Ž1992. 243–251. w11x J.M. Gidday, J.C. Fitzgibbons, A.R. Shah, T.S. Park, Neuroprotection from ischemic brain injury by hypoxic preconditioning in the neonatal rat, Neurosci. Lett. 168 Ž1994. 221–224. w12x A. Hill, J.J. Volpe, Hypoxic–ischemic encephalopathy of the newborn, in: K.F. Swaiman ŽEd.., Pediatric Neurology: Principles and Practice, CV Mosby, St. Louis, 1989, pp. 373–392. w13x G.L. Holmes, The long-term effects of seizures on the developing brain: clinical and laboratory issues, Brain Dev. 13 Ž1991. 393–409. w14x F.E. Jensen, C.D. Applegate, D. Holtzman, T.R. Belin, J.L. Burch-
w22x
w23x
w24x
w25x
w26x
w27x
w28x w29x w30x
w31x w32x
w33x
w34x w35x
fiel, Epileptogenic effect of hypoxia in the immature rodent brain, Ann. Neurol. 29 Ž1991. 629–637. T. Kirino, Y. Tsujita, A. Tamura, Induced tolerance to ischemia in gerbil hippocampal neurons, J. Cereb. Blood Flow Metab. 11 Ž1991. 299–307. K. Kitagawa, M. Matsumoto, M. Tagaya, R. Hata, H. Ueda, M. Niinobe, N. Handa, R. Fukunaga, K. Kimura, K. Mikoshiba, T. Kamada, ‘Ischemic tolerance’ phenomenon found in the brain, Brain Res. 528 Ž1990. 21–24. K. Kitagawa, M. Matsumoto, M. Tagaya, K. Kuwabara, R. Hata, N. Handa, R. Fukunaga, K. Kimura, T. Kamada, Hyperthermia-induced neuronal protection against ischemic injury in gerbils, J. Cereb. Blood Flow Metab. 11 Ž1991. 449–452. Y. Liu, H. Kato, N. Nakata, K. Kogure, Protection of rat hippocampus against ischemic neuronal damage by pretreatment with sublethal ischemia, Brain Res. 586 Ž1992. 121–124. C.T. Lombroso, J.L. Burchfiel, Etiologic and preventive aspects of epilepsy in the child—bridging the gap between laboratory and clinic, Epilepsia 28 Ž1987. S25–35. B.S. Meldrum, J.B. Brierley, Prolonged epileptic seizures in primates: ischemic cell change and its relation to ictal physiological events, Arch. Neurol. 28 Ž1973. 10–17. G. Nevander, M. Ingvar, R. Auer, B.K. Siesjo, ¨ Status epilepticus in well-oxygenated rats causes neuronal necrosis, Ann. Neurol. 18 Ž1985. 281–290. L. Nitecka, E. Tremblay, G. Charton, J.P. Bouillot, M.L. Berger, Y. Ben-Ari, Maturation of kainic acid seizure-brain damage syndrome in the rat. II: Histopathological sequeale, Neuroscience 13 Ž1984. 1073–1094. B.K. O’Connell, J. Towfighi, W.A. Kofke, R.A. Hawkins, Neuronal lesions in mercaptopropionic acid-induced status epilepticus, Acta Neuropathol. 77 Ž1988. 47–54. J.E. Rice, R.C. Vannucci, J.B. Brierley, The influence of immaturity on hypoxic–ischemic brain damage in the rat, Ann. Neurol. 9 Ž1981. 131–141. H.J. Romijn, M.A. Hoffman, A. Gramsbergen, At what age is the developing cerebral cortex of the rat comparable to that of the full-term newborn human baby?, Early Hum. Dev. 26 Ž1991. 61–67. H.J. Romijn, A.W.J.W. Janszen, M.J.D. van Voorst, R.M. Buijs, R. Balazs, ´ D.F. Swaab, Perinatal hypoxic ischemic encephalopathy affects the proportion of GABA-immunoreactive neurons in the cerebral cortex of the rat, Brain Res. 592 Ž1992. 17–28. M. Sasahira, T. Lowry, R.P. Simon, D.A. Greenberg, Epileptic tolerance: prior seizures protect against seizure-induced neuronal injury, Neurosci. Lett. 185 Ž1995. 95–98. R.P. Simon, M. Niiro, R. Gwinn, Prior ischemic stress protects against experimental stroke, Neurosci. Lett. 163 Ž1993. 135–137. E.F. Sperber, S.L. Moshe, ´ Age related differences in seizure susceptibility to flurothyl, Dev. Brain Res. 39 Ž1988. 295–297. E.F. Sperber, P.K. Stanton, K.Z. Haas, S. Shinnar, S.L. Moshe, ´ Resistance of the immature brain to hippocampal damage following flurothyl-induced status epilepticus, Ann. Neurol. 30 Ž1991. 495. N.M. Sherwood, P.A. Timuras, A Stereotaxic Atlas of the Developing Rat Brain, University of California Press, Berkley, 1970. B. Soderfeldt, H. Kalimo, Y. Olsson, B.K. Siesjo, ¨ ¨ Pathogenesis of brain lesions caused by experimental epilepsy: light and electron microscopic changes in the rat cerebral cortex following bicuculline induced status epilepticus, Acta Neuropathol. 54 Ž1981. 219–231. C.E. Stafstrom, J.L. Thompson, G.L. Holmes, Kainic acid seizures in the developing brain: status epilepticus and spontaneous recurrent seizures, Dev. Brain Res. 65 Ž1992. 227–236. K. Thompson, C. Wasterlain, Lithium-pilocarpine status epilepticus in the immature rabbit, Dev. Brain Res. 100 Ž1997. 1–4. J. Towfighi, W.A. Kofke, B.K. O’Connell, C. Housman, J.M. Graybeal, Substantia nigra lesions in mercaptopropionic acid induced status epilepticus: a light and electron microscopic study, Acta Neuropathol. 77 Ž1989. 612–620.
J. Towfighi et al.r DeÕelopmental Brain Research 113 (1999) 83–95 w36x J. Towfighi, D. Mauger, Temporal evolution of neuronal changes in cerebral hypoxia–ischemia in developing rats: a quantitative light microscopic study, Dev. Brain Res. 109 Ž1998. 169–177. w37x J. Towfighi, D. Mauger, R.C. Vannucci, S.J. Vannucci, Influence of age on the cerebral lesions in an immature rat model of cerebral hypoxia–ischemia: a light microcopic study, Dev. Brain Res. 100 Ž1997. 149–160. w38x J. Towfighi, N. Zec, C. Housman, R.C. Vannucci, Temporal evolution of neuropathologic changes in an immature rat model of cere-
95
bral hypoxia: a light microscopic study, Acta Neuropathol. 90 Ž1995. 375–386. w39x R.C. Vannucci, Experimental biology of cerebral hypoxia–ischemia: relation to perinatal brain damage, Pediatr. Res. 27 Ž1990. 317–326. w40x R.C. Vannucci, J. Towfighi, S.J. Vannucci, Hypoxic preconditioning and hypoxic–ischemic brain damage in the immature rat: pathologic and metabolic correlates, J. Neurochem. 71 Ž1998. 1215–1220. w41x C.G. Wasterlain, F. Plum, Vulnerability of developing rat brain to electroconvulsive seizures, Arch. Neurol. 29 Ž1973. 38–45.