Ultrastructural evidence of long-lasting cerebellar degeneration after early exposure to phenobarbital in mice

Previous studies in this laboratory demonstrated a 20 to 30% reduction in cerebellar Purkinje and granule cells after exposure to phenobarbital (PhB) early in life. Therefore, neurons in the cerebellar cortex were examined for signs of cytologic degeneration using transmission electron microscopy (TEM) after exposure to PhB pre- and postnally. Pregnant mice were given the acid form of PhB in their milled food (3 g/kg, gestation days 9 to 18) and water, ad libitum. Neonates were injected s.c. with an aqueous solution of sodium PhB (50 mg/kg body weight), days 2 to 21 after delivery. Controls were fed regular food or injected with the vehicle. The offspring were anesthetized on day 14 or 50 by an acute overdose of PhB and immediately perfused with a formaldehyde-paraformaldehyde or glutaraldehyde solution. The pyramis vermis of the cerebellar cortex was excised and processed routinely for TEM. The three layers of the cortex were examined. A short-term effect (at day 14) was found. More significantly, the treatment appeared to establish or trigger a degenerative process, the results of which were still apparent at day 50, more than 30 days after the termination of PhB treatment. Using double-blind evaluation for the presence and frequency of abnormalities, the cerebellar neurons of treated animals had 155 to 300% more abnormalities compared with control animals. Abnormalities included (i) Mitochondrial degeneration, ranging from swelling, collapse of cristae, vacuolization, to total granularization; (ii) lamellar bodies distributed throughout the cytoplasm and in cell processes; and (iii) myelin sheath degeneration, including periodic swelling and collapse, twisting of the coat, and scattered, unevenly stained areas. Damage was usually focal. Affected cells were found adjacent to normal cells in all areas of the cortex. PhB may cause the neural damage through a possible hormonal role.     

Neuronal deficits in mice following prenatal exposure to phenobarbital.

Pregnant mice (heterogeneous stock) were fed milled food containing 3 g/kg phenobarbital in the acid form (PhB) and water as their only nutritional source from gestation day 9 to 18, with smaller doses on gestation days 6 to 9 and 18 to parturition (B group). Control (C) females received milled food and water. Blood PhB concentrations of females and fetuses were between 40 and 200μg/ml blood. The brains of C and B male offspring were removed at age 50 days, fixed, sectioned, and stained with hematoxylin and eosin. Matching sagittal sections were selected for the study of the cerebellum, hippocampus, and the cerebral cortex. The brain weights of B offspring were 8% smaller than C offspring (P < 0.001). The area of the saggital section of the cerebellar and the hippcampal layers did not differ between groups. There were 30% fewer Purkinje cells in B offspring than in C offspring (P < 0.001). Consequently, their number per square millimeter was 26% smaller (P < 0.01). The number of the hippocamal pyramidal cells was 15% fewer in the B offspring (P < 0.02). There was no difference in the number of granule cells in both the cerebellum and hippocampus. The cerebral cortex was not affected by prenatal phenobarbital administration. The results suggest that the prenatally forming large neurons were affected by phenobarbital administration. However, there was no effect on the subsequent formation of the postnatally developing granule cells.     

Autoradiographic study of phenobarbital's effect on development of the central nervous system

Pregnant Sabra mice received 3 g phenobarbital (PhB)/kg milled food from gestation days 9 to 18 as their only food source; control female mice received milled food. All animals were divided into three groups and injected i.p. with a single dose of [H³]thymidine on gestation days (GD) 13, 15, or 17. The offspring survived until 50 days of age when they were anesthetized and perfused with 10% neurtral Formalin and the brains were removed and prepared for autoradiography. Matching sagittal sections of different offspring were selected for the study of the cerebellum, hippocampus, or cortex. The number of labeled Purkinje cells of the cerebellum was reduced 75% (P < 0.001) on GD 13 in experimental female, but not male offspring. The number of labeled cerebellar granule cells per section did not differ between groups. The hippocampus showed a 67% reduction in female offspring (P < 0.01) and a 51% reduction in male offspring. (P < 0.01) of labeled pyramidal cells per section on GD 13 after prenatal treatment of phenobarbital. As in the cerebellum, no deficits occurred in the hippocampal granule cells. In the cortex the drug decreased cell labeling 52% in female (P < 0.05) and 30% in male offspring (P < 0.05) per square millimeter on GD 13. Control groups demonstrated a gender difference in the prenatal labeling of cortical cells: male offspring had 46% more labeled cells per section (P < 0.01) on GD 13. A breakdown of the cortex into different regions showed that the greatest gender difference occurred in the deep strata (close to the corpus callosum) on GD 13. A 58% reduction in treated female (P < 0.001) and 43% reduction in treated male offspring (P < 0.001) was noted. The results suggest that phenobarbital induced alterations in the times of origin of neurons or in necrosis of recently formed neurons.     

Neuronal deficits after neonatal exposure to phenobarbital

Heterogeneous Sabra mice pups were injected daily with 50 mg/kg (B50) phenobarbital or 40 mg/kg (B40) from age 2 to 21 days. Control litter mates (C) received vehicle injection. Brain phenobarbital concentrations attained 50 μg/g. The brains of C, B40, and B50 mice were removed at age 50 days, fixed, sectioned, and stained with hematoxylin and eosin. Matching sagittal sections were selected for the study of the cerebellum, hippocampus, and the cerebral cortex. Brain weight was reduced 8 to 12% compared with control among the phenobarbital-treated animals. There were reductions of about 20% in the area of the cerebellar layers in B50-treated mice but no significant differences in B40-treated mice. In both groups Purkinje and granule cell numbers were reduced by about one-third, the packing density of the cerebellar cells was also reduced (18% to 20%), and the area of the hippocampal layers was reduced by about 15 to 20%. There were one-third fewer pyramidal cells in B50- and B40-treated mice than in controls and 22 to 25% fewer granule cells. Therefore, the ratio granule:pyramidal cells was significantly different in the hippocampus. The cortical area was smaller than control in the B50 group (14%) and the B40 group (20%). The neuronal deficits in the cortex were less extensive than in the other brain parts studied (B50, 9%; B40, 19%). The resuls suggest that unlike many neonatal insults including undernutrition, neonatal administration of relatively small doses of phenobarbital destroys even brain cells which are already formed.     

Alterations in mice dopamine receptor characteristics after early exposure to phenobarbital

The effects of microinjection of 5-10 micrograms of morphine into the midbrain periaqueductal gray (PAG) on the activity of neurons in the rostral ventral medulla (RVM) were studied in lightly anesthetized rats. Based on the relationship between changes in neuronal activity and the occurrence of the tail-flick reflex (TF), RVM neurons were divided into 3 groups: off-cells, on-cells and neutral cells. The off-cells exhibited an abrupt pause and the on-cells an acceleration beginning just prior to the occurrence of the TF. Neutral cell firing did not change at the time of the TF. Microinjections of morphine into the PAG which inhibited the TF had differential effects on the spontaneous activity of the 3 groups of neurons in RVM. Off-cells showed an increase and on-cells a decrease in spontaneous activity which preceded the inhibition of the TF. These microinjections also reduced the TF-related responses of off- and on-cells. The effects on cell activity were reversed by systemically administered naloxone and were not seen following microinjections which failed to block the TF. Neutral cell activity was unchanged following microinjection of morphine into the PAG. These results support the hypothesis that off- and on-cells in the RVM mediate the effects of microinjection of morphine into the PAG on spinal nociceptive reflexes.     

Normal zinc and iron concentrations in mice after early exposure to phenobarbital

Zinc and iron levels were studied in mice with early (pre/neonatal) exposure to phenobarbital, as the levels of these trace metals are known to be correlated with specific behaviors shown in our previous and present experiment to be affected by early phenobarbital administration. Mice were exposed to phenobarbital prenatally or neonatally. At adulthood they showed marked reduction from control in all parameters of eight-arm maze performance (P < 0.001). Since zinc is known to be correlated with this behavior, it was subsequently studied in barbiturate exposed animals. The differences between barbiturate exposed and control offspring for zinc levels in plasma, brain and hippocampus did not reach statistical significance. Our previous studies have shown that the number of dopamine receptors and the resulting apomorphine-induced climbing behavior is altered after early exposure to phenobarbital. The effect of iron level on dopamine receptors is now well established. Subsequently, a group of mice were tested for iron levels in their brain and liver. No significant differences were found.

It is suggested that deficits in the hippocampal behaviors, mainly eight-arm maze, after early exposure to phenobarbital are not related to changes in zinc levels. Similarly, early phenobarbital-induced alternation in dopamine receptors and the resulting dopaminergic behaviors are not related to changes in iron levels.     

Stereologic study of the cerebellar Purkinje cells submitted to alcoholic intoxication in Wistar rats

BACKGROUND: to analyze the effect of the alcohol on the cells of Purkinje. METHOD: Wistar rats received alcoholic solutions orally in different concentrations 4%, 12% and 24%. The animals were sacrificed with 4, 8 and 12 weeks and the cerebella were randomly cut and embedded in paraffin. Sections of 6 micrometer (H&E) were stereologically analyzed. RESULTS: The differences among the density for area and density of surface of the cells of Purkinje of all of the experimental groups (E) and the respective controls (C) were significant. With 12 weeks the cell of Purkinje volume density decreased among the groups C and E in the concentrations of 4% and 12%, but not for the concentration of 24%, probably due to smaller liquid ingestion by the animals. CONCLUSION: The alcohol has toxic effect on the Purkinje cellular body in the three studied concentrations from 4 weeks.     

Long-term reduction in spontaneous alternations after early exposure to phenobarbital

Spontaneous alternation behavior is related to the integrity of the hippocampus. Our earlier studies demonstrated hippocampal deficits after early phenobarbital (PhB) exposure. In the present study, we examined spontaneous alternation of mice who had been exposed to PhB prenatally or neonatally. Prenatal PhB was administered transplacentally: pregnant females were fed 3 g PhB/kg milled food on gestation days 9–18. Neonates were treated directly with daily injections of 50 mg PhB/kg on postnatal days 2–22. The animals were tested for spontaneous alternation in a T maze at the ages of 22, 28, 35 and 42 days. The test was conducted at each age for two consecutive days. A maximum of four alternations were allowed on the first day, and one alternation on the second day. Animals treated neonatally had reductions in alternation from the control group for every age group. Looking at the mean of the four trials on the first day there was a reduction of 35% at age 22 (P < 0.001), 8% at age 28, 21% at age 35 (P < 0.05) and 36% at age 42 (P < 0.02). On the second day the respective reductions were 32, 19, 24 and 36% (P < 0.05). The differences in alternation between animals treated with PhB prenatally and the control group were too small to reach statistical significance. Subsequently a more sensitive test, delayed spontaneous alternation (30 s), was applied to an additional group of animals at age 42 which had been prenatally exposed to PhB: 31% reduction from the control group was found on day 1 (P < 0.001), and 34% on day 2 (P < 0.02). The greater differences after neonatal as opposed to prenatal administration could be related to the more extensive hippocampal damage that was found in adults after neonatal treatment.     

Neural and behavioral alterations after early exposure to phenobarbital

Mice who were exposed to phenobarbital prenatally (B mice) had at adulthood deficits in the hippocampal eight-arm maze, spontaneous alternations, and water maze behaviors. Morphological studies revealed neuronal losses in the hippocampus. The surviving neurons had reductions from control in the number of dendritic branches, area and spine density, but wider fission angle than control. Neurochemical studies on the hippocampus revealed the following alterations: (a) decrease in NE level and the number of the NE cell bodies (b) no change in the serotonergic system (c) an increase in muscarinic receptors Bmax in the hippocampus; (d) no changes in GABA and benzodiazepine receptors. However, neonatal phenobarbital exposure caused an increase in the Bmax of GABA and benzodiazepine receptors. Transplantation of fetal septal cholinergic neurons into the hippocampus of B mice reversed most of the deficits in eight-arm maze behavior, while transplantation of noradrenergic cells did not affect the performance of B mice. In further studies on cholinergic mechanisms, the dopaminergic innervations in the septum (originating from A10), which are known to indirectly inhibit the activity of the septohippocampal cholinergic pathways, were destroyed by 6-OHDA. B mice treated with 6-OHDA had an increase in hippocampal ChAT activity and improved their eight-arm maze performance. Thus, understanding of the mechanism of a particular behavioral deficit enables one to correct it despite the nonspecific action of the neuroteratogen.     

Absence of behavioral effects of intrauterine phenobarbital exposure in rats

Phenobarbital (Ph) exposure in experimental animals has been associated with impaired brain growth and maturation. In order to look for behavioral correlates of these structural changes, rat pups were tested after intrauterine exposure to Ph. Five Sprague-Dawley rat dams were fed powdered chow containing Ph (0.75 mg/g of chow) from day 8 through the remainder of pregnancy, resulting in serum Ph levels of 19 to 36 micrograms/ml. Control (C) dams were fed plain chow. All pups were cross-fostered to control dams at birth. During their first 3 weeks of age, the acquisition of signs of physical maturation and neurologic development (such as incisor eruption, free-fall righting, rope climbing and descending, etc.) and the development of exploratory behavior in an open chamber were assessed. On days 39 to 48, 24 pups were tested in a water-T maze. Litter size and pup size were smaller in Ph-exposed dams than in controls, but acquisition of all signs of development except eye opening, activity in the exploratory chamber, and time and error scores on water-T maze testing did not differ between the two groups. The reasons that behavioral effects were not found in this study compared with previous studies may include the young age of the animals tested, the differences among behavioral parameters examined, the fact that controls were matched by weight with the experimental animals in this study, and different susceptibility to teratogenic effects among different species.     

Long term reduction in eight arm maze performance after early exposure to phenobarbital

Performance in the hippocampal eight arm maze was studied in mice after early exposure to phenobarbital (PhB). since previous studies suggested that these animals suffered neural deficits in the hippocampus. For prenatal exposure pregnant mothers were fed 3 g PhB/kg milled food on gestation days 9–18. Neonates were injected daily with 50 mg PhB/kg. on postnatal days 2–21. After a week of water deprivation, the animals were tested at age 50 days for 5 days preceded by 1 day of habituation. Deficits in eight arm maze performance were demonstrated in early treated mice on every testing day. For example, on day 5 of testing the number of correct entries during the first eight attempts in the prenatally treated group were 12% below control level (P<0.01), the respective reduction in the neonatal group was 10% (P< 0.001). The number of trials needed to enter all arms on day 5 was 27% above control level among prenatally treated mice (P< 0.001), and 13% in neonatally treated mice (P< 0.05). It took prenatal PhB animals twice the time to reach criterion than their controls (P< 0.001) and four times as long for neonatally treated mice (P< 0.001).     

Effect of Purkinje cell loss on cerebellar gangliosides in nervous mutant mice

The distribution of cerebellar gangliosides was studied in adult (73 ± 2 days) nervous (nr/nr) mutant mice which lose 50–90% of their Purkinje cells. This neuronal loss is associated with significant reductions in cerebellar weight and ganglioside concentration. The cerebellar dry weights (mg) and the ganglioside concentrations (μg N-acetylneuraminic acid per 100 mg dry weight) in nr/nr mice and age-matched normal littermates (+/?) are 7.4 ± 0.3 mg and 13.2 ± 0.4 mg; and 411.7 ± 4.8 μg and 438.5 ± 2.1 μg, respectively. Abnormalities were also observed for the concentration of certain ganglioside species. Most notably, GT1a is significantly reduced by 42%, and GD3 is significantly increased by 29% in the nr/nr mice compared to the +/? mice. The nr/nr mice also express a slight but significant reduction in GT1b. No ganglioside abnormalities were observed between the nr/nr and +/? mice in cerebral cortex. We previously found reduced cerebellar GT1a content in other mutants that also lose Purkinje cells, i.e., sg/sg, pcd/pcd, and Lc/+. GT1a is not reduced, however, in wv/wv mice that lose mostly granule cells. The findings in nr/nr mice are therefore consistent with our hypothesis that GT1a is enriched in Purkinje cells. GD1a, which is enriched in mature granule cells, is not reduced in the nr/nr mice. Since we previously found that GD3 is a good marker for reactive glia in neurological disease, the elevated GD3 concentration in the nr/nr mice indicates a mild gliosis. Our findings with nr/nr and the other neurological mutants indicate that gangliosides can be useful as cell-surface markers for monitoring changes in the cytoarchitecture of the mouse cerebellum.     

Hippocampal cholinergic alterations and related behavioral deficits after early exposure to phenobarbital.

Mice were exposed to phenobarbital (PhB) prenatally and neonatally. Prenatal exposure was accomplished by feeding the mother PhB (3 g/kg milled food) on gestation days 9-18. Neonatal exposure was accomplished by daily injections of 50 mg/kg sodium PhB directly to the pups on days 2-21. Long-term biochemical alterations in the pre- and postsynaptic septohippocampal system, as well as related behavioral deficits, were assessed in the treated animals. Significant increase in B(max) values for binding of [3H]QNB to muscarinic cholinergic receptors was obtained on both ages 22 and 50 in prenatally (40-90%, respectively, p less than 0.001) and neonatally exposed (58-89%, p less than 0.001) mice whereas Kd remained normal. Similarly, a significant increase of inositol phosphate (IP) formation in response to carbachol was found after both prenatal and neonatal exposure to PhB (p less than 0.05). No alterations in choline acetyltransferase (ChAT) activity were observed in the prenatally or neonatally treated animals. The early exposed mice showed deficits in the performance in Morris water maze, a behavior related to the septohippocampal pathway. The results suggest that early exposure to PhB induces alterations in postsynaptic components of the hippocampal cholinergic system and concomitantly to impairment in hippocampus-related behavior.     

Altered inhibitory input to Purkinje cells of dystrophin-deficient mice

We investigated evoked EPSPs and spontaneous IPSPs in cerebellar slices from dystrophin-deficient mdx mice. In the presence of the GABA(A) antagonist bicuculline the increase in EPSP amplitude was less in mdx Purkinje cells compared to control, and the amplitude of miniature IPSCs in mdx cells was also significantly less than in controls. This reduced inhibitory input is most likely due to the reported reduction in the size of GABA(A) channel clusters.     

The role of cerebral serotonin in the development of tolerance to centrally administered phenobarbital

Chronic intracerebroventricular injection of phenobarbital results in the development of tolerance to the depressant effects of the drug. The neurotransmitters involved and the manner in which cerebral neurons adapt to this depressant effect are at present unknown. This study examines whether brain serotonin containing neurons participate in the attenuation of the hypnotic response caused by chronic barbiturate administration. Depletion of serotonin with p-chlorophenylalanine, p-chloroamphetamine and 5,7-dihydroxytryptamine did not affect the initial dose-response curve to the centrally injected barbiturate, but all treatments resulted in significant delays in tolerance development. A negative correlation between the extent of whole brain serotonin remaining and the duration of loss of righting reflex on the last day of the chronic phenobarbital regimen was obtained after pretreatment with p-chlorophenylalanine, p-chloroamphetamine and saline. The sleep times of animals pretreated with 5,7-dihydroxytryptamine did not fit this linear relationship. Treatment with p-chloroamphetamine after cessation of the chronic phenobarbital regimen did not influence the rate of tolerance reversal. Steady state levels of serotonin and the concentration of its metabolite, 5-hydroxyindoleacetic acid, in different brain areas were comparable in controls and tolerant rats when examined at various stages intolerance development. However, tolerant and non-tolerant rats sacrificed at time points immmediately after central phenobarbital injection had different temporal patterns of 5-hydroxyindoleacetic acid elevation in the striatum, hypothalamus and midbrain which were observable only during the loss of righting reflex. The data indicate that cerebral serotonin neurons participate in the attenuation of hypnosis following chronic phenobarbital injections.     

Ganglioside GD3 of cerebral neurons and Purkinje cells in aged human brains

Immunocytochemical staining for ganglioside GD3 (II^3α (NeuAcα2–8NeuAc)-LacCer, GD3) in neuronal cells of the cerebral cortices (cerebral neurons), and cerebellar dentate nucleus (dentate neurons) and Purkinje cells, in human autopsy cases of progressive supranuclear palsy, senile dementia of the Alzheimer type, Pick's disease, amyotrophic lateral sclerosis and olivopontocerebellar atrophy (OPCA) was undertaken using mouse IgM anti-GD3 monoclonal antibody. Cerebral neurons and dentate neurons were constantly GD3-immunoreactive and immunoreactivity was observed in the cytoplasm. The peroxidase reaction product for GD3 (RP) in cerebral and dentate neurons was granular in appearance. It appeared that RP was associated with lipofuscin granules. However, immunoreactivity of Purkinje cells varied among cases, and the RP was slightly granular even when they were positive. This study suggests that lipofuscin granules contributed to the neuronal immunoreactivity of GD3 in aged human brains.     

Prenatal diclofenac sodium administration increases the number of Purkinje cells in female rats: a stereological study

Diclofenac sodium (DS) may affect the number of Purkinje cells in the developing cerebellum since DS can easily be transported from the maternal to the fetal physiological system during the pregnancy. In the present study, the effects of prenatal exposure to DS on the number of Purkinje cells in the cerebellum of 4-week-old (4W-old) and 20-week-old (20W-old) female rats were investigated. There were two main groups: the drug-treated group (DTG) and the control group (CG). Beginning from the 5th day after mating for a period of 15 days, a daily dose of 1 mg/kg of DS (Voltaren, 75 mg/3 ml ampul, Novartis, Mefar Ilaç Sanayi A.S., Kartal, ?stanbul, Turkey) was intraperitoneally injected in the DTG of pregnant rats. In contrast, a daily dose of 1 ml/kg of isotonic saline was intraperitoneally administered to the CG of pregnant rats during the same period. After spontaneous delivery, female offspring were obtained, and the main groups’ offspring were divided into two subgroups as a 4W-old group and a 20W-old group. Therefore, there were four groups at the end of the experiment: the 4W-old DTG and the CG, and the 20W-old DTG and the CG. At the end of 4W and 20W, offspring were perfused, their brains were dissected, and the number of cells estimated via the optical fractionator technique. Our results showed that while the total number of Purkinje cells in the cerebellum of offspring of DT 20W-old female rats was significantly higher than that of the CG, there was no significant difference between the 4W-old DTG and the control groups. Therefore, it could be suggested that DS administration during the prenatal period increases the number of Purkinje cells in the cerebellum of a developing female rat throughout postnatal 20W.     

早期运用苯巴比妥钠预防新生儿窒息后脑损伤

目的 探讨早期干预对新牛儿室息后脑损伤发病率的影响。方法 对出生时有室息表现的新生儿存在出生后6h内给予笨巴比妥钠15mg／kg，12h后给予5mg／kg，每12小时一次，维持3～5d。结果 预防朋药组较对照组往新生儿窒息后的缺氧缺血性脑病、颅内出血等脑损伤的发病率有明显降低，两者相比较有显著差异（P〈0．05）。结论 对出生时有窒息表现的新生儿可以使用苯巴比妥钠进行早期干预。     

Developmental exposure to methylmercury elicits early cell death in the cerebral cortex and long-term memory deficits in the rat

Experiments were performed to assess the neurotoxic effects induced by prenatal acute treatment with methylmercury on cortical neurons. To this purpose, primary neuronal cultures were obtained from cerebral cortex of neonatal rats born to dams treated with methylmercury (4 and 8 mg/kg by gavage) on gestational day 15, the developmental stage critical for cortical neuron proliferation. Prenatal exposure to methylmercury 8 mg/kg significantly reduced cell viability and caused either apoptotic or necrotic neuronal death. Moreover, this exposure level resulted in abnormal neurite outgrowth and retraction or collapse of some neurites, caused by a dissolution of microtubules. The severe and early cortical neuron damage induced by methylmercury 8 mg/kg treatment correlated with long term memory impairment, since adult rats (90 days of age) born to dams treated with this dose level showed a significant deficit in the retention performance when subjected to a passive avoidance task. Prenatal exposure to methylmercury 4 mg/kg significantly increased the neuronal vulnerability to a neurotoxic insult. This was determined by measuring the increment of chromatin condensation induced by glutamate, at a concentration (30 μM) able to induce an excitotoxic damage. This exposure level eliciting apoptotic death did not result in cognitive dysfunctions. In conclusion, the methylmercury-induced disruption of glutamate pathway during critical windows of brain development may interfere with cell fate and proliferation resulting in a more or less severe cortical lesions associated or not with loss of function later in life, depending on the exposure levels. Therefore, the early biochemical effects and long-term behavioral changes elicited by high methylmercury levels suggest that the developing brain is impaired in its ability to recover following toxic insult, and the initial effects on cortical neurons may lead to permanent cognitive dysfunctions.