Involvement of nitric oxide in the deregulation of cytosolic calcium in cerebellar neurons during combined glucose-oxygen deprivation.

Nitric oxide (NO) has been proposed as a neuronal messenger molecule in hypoxic/ischemic cell injury (Nowicki et al., 1991; Trifiletti, 1992). We conducted studies in a model of combined glucose-oxygen deprivation using cultured rat cerebellar granule cells. Experiments were designed to test the hypothesis that sustained elevation of cytosolic calcium ([Ca2+]i) and NO generation act in concert to trigger neuronal injury after anoxic insult. A hypoxic state was achieved by perfusing the cells with medium pre-equilibrated with argon gas. [Ca2+]i was monitored using digital-imaging fluorescence microscopy in cells loaded with fura-2 AM. Under short-term hypoxic conditions, cells displayed a progressive and sustained, moderate increase of [Ca2+]i, which returned to near basal levels on restoration of O2-containing medium. Prolonged hypoxic conditions (> 60 min) caused irreversible elevation of [Ca2+]i followed by disruption of cell membrane integrity, as indicated by severe swelling, loss of regular cell shape and processes, leakage of dye fura-2, and propidium iodide uptake ("point of no return"). Pretreatment with NG-nitro-L-arginine methyl ester (L-NAME, 100 microM), a specific NO synthase inhibitor, markedly delayed the onset of intensity of the rise of [Ca2+]i. The hypoxia-induced elevation of [Ca2+]i was also greatly attenuated if L-NAME (100 microM) was added to the argon-perfused medium before the cells demonstrated signs of irreversible injury. Prolonged or repeated hypoxic conditions, however, caused a rapid and intense increase of [Ca2+]i, which could not be blocked by inhibition of NO synthase (NOS). In addition, reoxygenation after the "point of no return," as characterized above, greatly potentiated [Ca2+]i overload and facilitated the process of cell injury. The potentiation and facilitation of cell damage, as demonstrated by rapid massive increase of [Ca2+]i and subsequent cell death, was not blocked by NOS inhibitor, L-NAME.

Tityustoxin-K alpha, from scorpion venom, blocks voltage-gated, non-inactivating potassium current in cultured central neurons.

Whole-cell voltage-clamp was used to examine the effects of tityustoxin-K alpha (TsTX-K alpha), from the venom of the scorpion Tityus serrulatis, on voltage-gated K+ currents in cultured hippocampal neurons and cerebellar granule cells. Slowly activating, noninactivating outward currents (IK) were generated by depolarizing steps from a holding potential of -60 mV to potentials positive to -40 mV. TsTX-K alpha produced a dose-dependent block of the sustained outward current. The fraction of total current blocked ranged from 10 to 60% over a concentration range of 2.5-120 nM in both cerebellar and hippocampal neurons. A hyperpolarizing prepulse to -100 mV was used to generate a rapidly inactivating current with properties like those of IA. When IA was isolated pharmacologically (with 5-10 mM TEA to block IK) or by subtracting IK from total outward current, TsTX-K alpha had no effect on the IA in either cell type. TsTX-K alpha also had no apparent effect on the leak conductance or on the inward rectifier current in these cells. The data indicate that TsTX-K alpha in cultured mammalian neurons is a potent and selective blocker of a voltage-gated, non-inactivating K+ current with properties like those of a delayed rectifier.

Potentiation of quinolinate-induced hippocampal lesions by inhibition of NO synthesis.

Low doses of quinolinic acid (QUIN) administered intracerebroventricularly (ICV) to rats produced either no damage or mild to moderate damage in the pyramidal cell layer of the hippocampus and resulted in mild, limbic seizures in the majority of animals treated. The same dose of QUIN following ICV pretreatment with the nitric oxide synthase inhibitor N-nitro-L-arginine (NARG), produced extensive hippocampal lesions with complete loss of the pyramidal layer in 50% of the animals, and moderate damage with total neuronal loss in areas CA1 and CA3 in the remainder of the group. Animals treated with both NARG and QUIN also exhibited a greater incidence of severe convulsive behavior (9/11) and 3 deaths. Pretreatment with the nitric oxide-generating drug molsidomine attenuated the enhanced toxicity observed with combined NARG-QUIN treatment, resulting primarily in no detectable hippocampal damages and mild seizures resembling those produced by QUIN alone. Administration of NARG alone produced neither seizure activity nor histological evidence of neurotoxicity. We conclude that inhibition of nitric oxide production with NARG potentiates the neurotoxicity of quinolinic acid in the rat hippocampus.

Biphasic action of dextrorphan on penicillin induced bursting in rat hippocampal slice.

Effects of dextrorphan (DX), a metabolite of the over-the-counter antitussive, dextromethorphan, were investigated in rat hippocampal slices exposed to the epileptogenic agent penicillin. At 50 microM and 100 microM concentrations dextrorphan suppressed late components of the epileptiform CA1 field potential elicited by afferent electrical stimulation, and partially suppressed the intracellularly recorded paroxysmal depolarization shift. These effects were not due to non-specific changes in cell excitability, since resting cell membrane potential, input resistance, and the ability of cells to fire action potentials in response to direct depolarizing current were unaffected. The depressant effect of 100 microM dextrorphan was probably due to actions at the NMDA receptor, since pretreatment with the competitive NMDA antagonist D-APV prevented any further depressant effects of dextrorphan in this model. In contrast, at a 10 microM concentration DX enhanced the amplitude of evoked epileptiform field potentials and intracellularly recorded EPSPs. These findings support a role for dextrorphan and similar agents as anticonvulsants at high concentrations, but raise a caution regarding possible excitatory actions of dextrorphan at low concentrations.

Intrahippocampal injection of pertussis toxin blocks adenosine suppression of synaptic responses.

Adenosine exerts prominent inhibitory effects on synaptic transmission via a presynaptic action. Using the hippocampal slice preparation, we have found in electrophysiological experiments that this action of adenosine is blocked by intrahippocampal injections of pertussis toxin. In biochemical studies, we have confirmed that this treatment affects the GTP-binding proteins, Gi and Go, in this preparation. These results indicate that both pre- and postsynaptic actions of adenosine involve pertussis toxin-sensitive GTP-binding proteins.

Selective depression of N-methyl-D-aspartate-mediated responses by dextrorphan in the hippocampal slice in rat.

The effects of dextrorphan (DX) and dextromethorphan (DM) on responses to excitatory amino acids in the CA1 region of the hippocampus of the rat were studied using extracellular and intracellular recording in in vitro slices of brain. Dextrorphan selectively and non-competitively blocked depolarizations evoked by focally-applied N-methyl-D,L-aspartate (NMA), recorded by both extracellular and intracellular techniques. Quisqualate (QUIS) responses and evoked field potentials were not affected by DX. Epileptiform activity elicited in Mg2+-free solution was suppressed by DX. Dextrorphan had no effect on resting membrane potential or input resistance. The antagonism of NMA by DX was dose-dependent with an EC50 of 0.65 microM; DM was also effective but considerably less potent. In the paradigm used in the present study, DX did not produce the clear use-dependent block observed in the presence of MK-801. These data suggest that DX, the metabolite of the widely used antitussive DM, is a potent NMDA antagonist with a potential role as an anticonvulsant and neuroprotective agent.

EEG correlates of neurotoxicity.

Electroencephalographic (EEG) techniques provide an interface between behavioral studies and more mechanistically directed neurophysiological studies, thus constituting a potentially important in vivo method of neurobehavioral assessment. The amplitude and frequency of the EEG signal depends on the behavioral state of the subject and the location of the recording electrode. Systematic assessment of EEG recordings may reveal alterations in integrated neuronal function induced by drug or neurotoxicant exposure. Fourier analysis allows more precise analysis of the frequency content of the recorded signal and yields data that may be further reduced and statistically analyzed. EEG data from neuropharmacologic and neurotoxicologic studies are discussed to illustrate application of power spectral analysis techniques in experimental laboratory settings.

Methylazoxymethanol as a developmental model of neurotoxicity.

Behavioral consequences of exposure to toxic chemicals during development often depend on critical periods of exposure. The present study employed the antimitotic agent methylazoxymethanol (MAM) to interfere with neuron proliferation during specific periods of brain development. Behavioral profiles of neonatal rats were obtained after MAM administration on gestational days 15, 19 (G15 or G19) or postnatal day 1 (PN 1). MAM administration on G15 produced neonates with decreased electroconvulsive shock thresholds (TT50) on postnatal day 8 and increased locomotor activity levels from postnatal day 18 through 29. Treatment on G19 resulted in an increased TT50 and a delay in the developmental locomotor activity curve. Pups treated on PN1 displayed decreased levels of activity on postnatal days 12 through 16 and impaired performance on a rotorod at 29 days of age. MAM treatment on G15 produced decreases in forebrain and cortical mass to 51 and 40% of respective control values. MAM administration on PN1 resulted in small but significant decreases in forebrain, cortex and hindbrain and a 35% decrease in cerebellar size. No changes in regional brain weights were measurable in the G19 treatment group. The results indicate that neonatal behavior can be altered in a relatively specific manner by perturbation of neuronal development at critical time periods and that the functional tests employed clearly distinguish among treatment groups. The findings of this study are interpreted in light of previously described neurochemical and morphological effects of antimitotic treatment.

2,4-Dichlorophenoxyacetic acid (2,4-D) reduces acetylcholinesterase activity in rat muscle.

A single dose (200 mg/kg body weight, i.p.) of 2,4-dichlorophenoxyacetic acid (2,4-D), commonly used as a herbicide, caused significant decreases in acetylcholinesterase (AChE) activity in diaphragm and other muscles of the rat. The 4S, 10S, and 16S forms of AChE were affected. The effect was maximal 15 to 24 h after injection. Choline acetyltransferase (CAT) activity was not affected. Neither AChE nor CAT activities changed in sciatic nerve from 2,4-D-treated animals. Spontaneous locomotor activity decreased dramatically 4 h after 2,4-D treatment. Myotonia that was present 1.5 h after 2,4-D injection became maximal at 2 to 6 h. Twenty-four hours after drug injection, when animals were recovering from myotonia, spontaneous locomotor activity was still depressed to 50% of control values. Prolonged distal motor latencies were observed 15 to 24 h after drug administration. AChE activity and spontaneous locomotor activity returned to control values at 48 h. Thus, 2,4-D causes a decrement of end-plate AChE, as well as behavioral and electrophysiologic changes. Decreased activity of AChE may be an early step in development of the myopathy that occurs after large dose 2,4-D.

Spectral analysis of kindled hippocampal afterdischarges in lead-treated rats.

Male rats were exposed to lead from parturition through weaning. When subjected to electrical hippocampal kindling as adults, although there were no alterations in the kindling rates, lead-treated animals were found to differ from controls in several other respects. In contrast to controls, lead-treated animals did not show an increase in afterdischarge (AD) duration with kindling, and the power spectrum of the AD did not shift to a higher peak frequency. There was a clear dose-related effect of lead on the spectra of the kindled AD, which in lead-treated groups was characterized by greater power in the low-frequency bands. These data indicate that long-lasting dysfunction can occur following a brief neonatal lead exposure. Power spectral techniques may be useful for future studies in neurotoxicology and epilepsy.

Neonatal lead exposure in rats: I. Effects on activity and brain metals.

Upon parturition, Sprague-Dawley dams were administered 1.0 (LL), 2.5 (ML), 5.0 (HL) mg/ml of lead acetate or 1.25 mg/ml sodium acetate (C) in their drinking water. Pups were weaned to tap water at 22 days of age. The developmental pattern of neonatal activity was characterized by monitoring spontaneous activity of single pups for 30 minutes on days 12, 14, 16, 18, 20, and 22, and for one hour in adults. Brains were removed at weaning for subsequent analyses of lead, zinc, and copper. The effects of lead on neonatal activity were considered to be minor because only the ML group differed significantly from C and exhibited hyperactivity on day 16. Adult activity levels, however, were decreased in a dose-dependent fashion. At 22 days of age, whole brain metal analyses revealed a dose-dependent increase in lead levels and a decrease in zinc. These results do not strongly support a hypothesis of lead-induced neonatal hyperactivity.

Neonatal lead exposure in rats: II. Effects on the hippocampal afterdischarge.

Male Sprague-Dawley rats were exposed to lead from parturition to weaning via the dams' milk. Dams were provided with drinking water containing 1.0 (LL), 2.5 (ML), or 5.0 (HL) mg/ml lead acetate or 1.25 (C) mg/ml sodium acetate. Beginning at 15 weeks of age, characteristics of the electrically elicited hippocampal afterdischarge (AD) and its alteration by phenytoin (PHT) were assessed in these rats. A separate group of rats was sacrificed at 20 weeks for hippocampal metal analysis. Increases in primary AD duration were observed in LL and ML, significant in ML. Significantly fewer wet dog shakes occurred in all lead groups. HL animals displayed shorter rebound ADs. All groups responded to PHT with increases in primary AD duration, but the increases in the ML and HL groups were significantly greater than in the C group. Hippocampal lead, zinc, and copper were not different from control. It is concluded that a brief lead exposure can have persistent effects on hippocampal function and that these effects are not due to altered hippocampal metal concentrations. The findings are discussed in relation to the known effects of postnatal lead exposure on hippocampal neuronal development.

Prenatal methyl mercury exposure: I. Alterations in neonatal activity.

Pregnant Long-Evans rats were intubated with 0, 5, or 8 mg/kg of methyl mercury on day 8 or 15 of gestation. Maternal weight gain during gestation was reduced significantly only in those animals that had received 8 mg/kg on day 8 of gestation and had resorbed their litters. In litters that were delivered, the mercury treatment did not affect litter size or weight gain of the pups in the preweaning period. Methyl mercury content of the 1 day old rat brains was directly related to both the dose and time of treatment. Neonatal activity was significantly elevated on postnatal day 4 in rats treated with 5 mg/kg on day 8 of gestation, on postnatal day 8 in the group that received 8 mg/kg on day 8 of gestation, and on postnatal day 8 and 15 in the group that received 5 mg/kg on day 15 of gestation. The results indicate that neonatal activity measures can be used as sensitive indicators of low prenatal neurotoxic exposures.

Prenatal methyl mercury exposure: II. Alterations in learning and psychotropic drug sensitivity in adult offspring.

Male Long-Evans rats that had been exposed in utero to 5 or 8 mg/kg of methyl mercury administered as a single dose on either days 8 or 15 of gestation were tested as adults in two operant tasks. In one task the animals were trained on two-way avoidance to a criterion of 10 consecutive avoidances. Following acquisition the animals were extinguished and 24 hours later re-trained to the previous criterion. Animals treated with 8 mg/kg on day 8 of gestation required significantly more trials to reach criterion during reacquisition than controls. Rats treated on day 15 with either 5 or 8 mg/kg took significantly more trials to reach criterion during acquisition than controls, and of the 8 mg/kg group 55% failed to reach criterion. Rats treated with 8 mg/kg of mercury on day 8 of gestation acquired a DRL-10 sec task at the same rate as controls. When challenged with d-amphetamine the treated animals were less disrupted at the higher dose (1.0 mg/kg) than controls, suggesting a shift in the dose response curve for this psychoactive drug. Activity measures taken simultaneously with the DRL session confirmed this shift in amphetamine effect. Results suggest that a single prenatal exposure to methyl mercury can affect learning and drug sensitivity of the adult animal. Additionally, mercury exposure in late gestation has more deleterious consequences on learning ability than early exposure.

Hippocampal afterdischarges and their post-ictal sequelae in rats: a potential tool for assessment of CNS neurotoxicity.

Two types of AD and post-AD sequences are described. Type 1, which were most common (83%), are followed by a profound post-ictal depression (PID). PIDs can be quantitatively measured by integrating the EEG. The post-ictal EEG power normally returned to 75% of its prestimulation power within 4-5 min. PIDs are usually interrupted by a brief rebound AD, occurring about 1.5 min after the end of the AD. Wet dog shakes (WDSs) often punctuate the end of the AD and the end of the rebound AD. Type II ADs (12% of those recorded) do not have profound PID, but do have irregular post-ictal spikes which may persist for over 30 min. Type IIa ADs do not have the post-ictal spikes or the PID, and represented 5% of the ADs recorded. Females had higher AD thresholds than males, and had more Type IIa ADs than males during threshold testing. AD durations were longest in the subiculum and shortest in the area dentata. Correlations among the different measures of AD activity are discussed. With repeated elicitation, AD and PID durations increased. Stimuli 400% of threshold produced shorter ADs and longer PIDs than stimuli 115% of threshold. High intensity stimuli did not significantly alter frequency of WDSs or rebound AD characteristics. Increasing dosages of sodium pentobarbital decreased AD duration, increased threshold, decreased the probability of a Type I AD and produced depressions which were not accompanied by rebound ADs. At the highest (30 mg/kg) dosage of sodium pentobarbital, the spike frequency within the AD was decreased. When PIDs occurred, they were somewhat longer at higher dosages of sodium pentobarbital. The feasibility of the AD as an index of neurotoxicity is discussed, and it is concluded that it may provide a valuable mirror of dysfunction in the hippocampal formation. Further work must determine its sensitivity to toxicant-induced alterations.

Evoked potential alterations following prenatal methyl mercury exposure.

Pregnant hooded rats were administered either 5 mg/kg CH3, Hg or 0 mg/kg CH3 Hg by gastric intubation on day seven of gestation. Female offspring were implanted with recording electrodes 60 days after birth and had their cortically recorded visual evoked potentials studied at four different flash intensities. Mercury exposed animals had higher P1-N1 and N1-P2 amplitudes and shorter P2 and N2 latencies than controls. The data provides evidence that a single ingestion of CH3 Hg by pregnant rats is sufficient to produce long term alterations in CNS activity.