Rodríguez-Farré E

The organochlorine pesticides γ-hexachlorocyclohexane (lindane), α-endosulfan and dieldrin differentially interact with GABAA and glycine-gated chloride channels in primary cultures of cerebellar granule cells

The neurotoxic organochlorine pesticides gamma-hexachlorocyclohexane, alpha-endosulfan and dieldrin induce in mammals a hyperexcitability syndrome accompanied by convulsions. They reduce the GABA-induced Cl(-) flux. The strychnine-sensitive glycine receptor also regulates Cl(-)-flux inhibitory responses. We studied the effects of these compounds on Cl(-) channels associated with glycine receptors in cultured cerebellar granule cells in comparison to the GABA(A) receptor. Both GABA (EC(50): 5 microM) and glycine (EC(50): 68 microM) increased (36)Cl(-) influx. This increase was antagonized by bicuculline and strychnine, respectively. Lindane inhibited with similar potency both GABA(A) (IC(50): 6.1 microM) and glycine (5.0 microM) receptors. alpha-Endosulfan and dieldrin inhibited the GABA(A) receptor (IC(50) values: 0.4 microM and 0.2 microM, respectively) more potently than the glycine receptor (IC(50) values: 3.5 microM and 3 microM, respectively). Picrotoxinin also inhibited the glycine receptor, although with low potency (IC(50)>100 microM). A 3D pharmacophore model, consisting of five hydrophobic regions and one hydrogen bond acceptor site in a specific three-dimensional arrangement, was developed for these compounds by computational modelling. We propose that the hydrogen bond acceptor moiety and the hydrophobic region were responsible for the affinity of these compounds at the GABA(A) receptor whereas only the hydrophobic region of the molecules was responsible for their interaction with the glycine receptors. In summary, these compounds could produce neuronal hyperexcitability by blocking glycine receptors besides the GABA(A) receptor. We propose that two zones of the polychlorocycloalkane pesticide molecules (a lipophilic zone and a polar zone) differentially contribute to their binding to GABA(A) and glycine receptors.

Behavioral and monoaminergic changes after lindane exposure in developing rats

The effects of lindane on behavior and central monoaminergic systems were studied in rat pups at 15 days of postnatal age. Pups were previously given nonconvulsant lindane PO doses, either a single 20 mg/kg or 7-day repeated 10 mg/kg doses. Both treatment schedules improved the passive avoidance acquisition but only the acute administration prolonged the step-through latency. Acute lindane decreased the motor activity, whereas the repeated dosing increased it. Increases of the ratio 5-HIAA/serotonin in several brain regions and of the ratio DOPAC/dopamine in the mesencephalon after a single dose of lindane suggest an enhanced monoaminergic turnover. In contrast, repeated lindane doses decreased monoamine/metabolite ratios excluding the striatum, where an increase of DOPAC/dopamine ratio correlates with the higher motor activity of these animals. It is postulated that both the imbalance of the central monoaminergic systems and the lindane-induced GABAergic blockade may be the basis of the behavioral alterations.

Regional concentrations of GABA, serotonin and noradrenaline in brain at onset of seizures induced by lindane (γ-hexachlorocyclohexane)

The main objective of this study was to determine the modifications induced by the pesticide lindane (gamma-hexachlorocyclohexane) in the regional concentration of neurotransmitters in brain, taking the tonic-clonic seizure as the main sign of its neurotoxic action. The animals were given lindane (150 mg/kg p.o. in olive oil) and killed at the onset of the first seizure (mean latency time: 18.3 +/- 5.5 min, n = 16). The inhibitory neurotransmitter gamma-aminobutyric acid (GABA) and the biogenic amines noradrenaline (NA) and serotonin (5-HT), as well as its corresponding acidic metabolite, were determined by high performance liquid chromatography with fluorimetric or electrochemical detection in different areas of the brain: hippocampus, mesencephalon, colliculi, frontal cortex, parietal cortex, striatum and thalamus. The concentration of GABA in whole tissue was only significantly decreased in the colliculi. The concentration of serotonin (5-HT) and its metabolite 5-hydroxyindoleacetic acid (5-HIAA) were decreased in the colliculi, striatum and frontal cortex, suggesting a decreased synthesis and release of 5-HT, but increased in the parietal cortex. The concentration of NA was significantly decreased in the parietal cortex. Thus, the results indicate that lindane induces some modifications of the concentration of neurotransmitters in cortical structures, basal nuclei, midbrain and colliculi. These changes may be consequent upon the proposed interaction of lindane with the GABAergic system, although a direct action of lindane on other neurotransmitter systems cannot be excluded.

Cytotoxic action of lindane in cerebellar granule neurons is mediated by interaction with inducible GABAB receptors

The cytotoxic action of the γ‐isomer of hexachlorocyclohexane (γ‐HCH, lindane) was studied in cultured mouse cerebellar granule neurons maintained in the presence or absence of the GABAA receptor agonist THIP (4,5,6,7‐tetrahydroisoxazolo[5,4‐c]pyridin‐3‐ol). The cells were exposed for 24 hr to lindane (30‐300 μM) in the culture medium. Changes in mitochondrial function were investigated by using the MTT (3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromide) test. The results showed that lindane‐induced cytotoxicity was concentration‐dependent. In cerebellar granule cells not treated with THIP, lindane‐induced cytotoxicity did not appear to be related to GABAA or GABAB receptors. However, in THIP‐treated cultures, lindane‐induced cytotoxicity was found to be mediated by an action of the insecticide on GABA receptors. In the latter case, GABA reduced the lindane‐induced cytotoxicity, but the protective effect was not potentiated by flunitrazepam. The GABAA receptor agonist muscimol (50 μM) also protected the THIP‐treated cultures against lindane‐induced cytotoxicity. In addition, the GABAB receptor agonist R(+)baclofen protected the cells from lindane‐induced cytotoxicity and the effect of baclofen was blocked by GABAB receptor antagonists. Pertussis toxin was found to reverse the protective effect of baclofen only at the highest lindane concentration (300 μM). The lindane‐induced cytotoxicity could be partly explained as being secondary to excitotoxicity as a mixture of the excitatory amino acid receptor antagonists APV (D‐(‐)‐2‐amino‐5‐phosphonopentanoate) and CNQX (6‐cyano‐7‐nitro‐quinoxaline‐2,3‐dione) shifted the concentration‐response curve for lindane‐induced cytotoxicity to the right. It is suggested that the cytotoxic effects of lindane in THIP‐treated cerebellar granule neurons are primarily related to an action of lindane on GABAB receptors and to a lesser extent on inducible low‐affinity, benzodiazepine insensitive GABAA receptors. J. Neurosci. Res. 52:286–294, 1998.

Cerebrospinal dopamine metabolites in rats after intrastriatal administration of 6-hydroxydopamine or 1-methyl-4-phenylpyridinium ion

Dopamine (DA) and its main cerebral metabolites, dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) were measured in striatum and cerebrospinal fluid (CSF) from cisterna magna in rats bilaterally lesioned by intrastriatal administration of 6-hydroxydopamine (6-OHDA) or 1-methyl-4-phenylpyridinium ion (MPP+). 6-OHDA caused a progressive lesion in striatum that is only moderately reflected in the decrease in dopamine metabolite concentration in CSF. MPP+ caused an acute but less selective lesion in the dopamine striatal system, as indicated by a significant reduction in striatal GABA content, followed by a slow recovery in dopamine striatal metabolism and content. The locomotor activity was dramatically reduced in both groups 48 hours after the treatment but remained significantly decreased after two months only in 6-OHDA lesioned animals. A positive correlation was found between HVA CSF concentration and striatal DA content in MPP+ lesioned rats, but not in 6-OHDA lesioned rats. It is concluded that the concentration of dopamine metabolites in CSF can be altered only after a severe striatal lesion: reduction of striatal dopamine content below 50% of normal values and involvement of neuronal or non-neuronal elements other than the dopaminergic system, similarly to the lesions caused by MPP+. These results may partly explain why CSF dopamine metabolites concentrations were significantly decreased both in advanced stages of parkinsonism and in other neurodegenerative disorders.

Effects of lindane on spontaneous behavior of rats analyzed by multivariate statistics

The effects of subconvulsant doses (10, 15, 30 mg/kg) of the organochlorine insecticide gamma-hexachlorocyclohexane (lindane) on the spontaneous activity of male Wistar rats were assessed in an automated wheel-shaped activity monitor in 4 hr sessions. Most of the activity variables defined on the raw data showed statistically significant differences among treatment groups. These data can be interpreted as a lindane-induced change in the temporal distribution of the activity. The multivariate statistical technique principal component analysis was used to obtain a composite variable that measured this change. The variable obtained was found to be well correlated with dose levels and brain concentration of lindane and was used to estimate a minimal effective dose (MED) of 1.85 mg/kg and a upper limit for a minimal effective concentration (MEC) in brain of 0.84 micrograms/g by a log-linear regression approach. The results obtained with this approach further support the conclusion that the behavioral effects of lindane can be significant at exposure levels far below its LD50.

Does Peer Review Predict the Performance of Research Projects in Health Sciences

Peer review is a basic component of the scientific process, but its performance has seldom been evaluated systematically. To determine whether pre-approval characteristics of research projects predicted the performance of projects, we conducted a retrospective cohort study of all 2744 single-centre research projects financed by the Spanish Health Research Fund since 1988 and completed before 1996. Peer review scores of grant applications were significant predictors of performance of funded projects, and the likelihood of production was also higher for projects with a basic research component, longer duration, higher budget or a financed research fellow. Funding agencies should monitor their selection process and assess the performance of funded projects to design future strategies in supporting health sciences research.

Properties of ryanodine receptors in cultured cerebellar granule neurons: Effects of hexachlorocyclohexane isomers and calcium

The binding of [3H]ryanodine was determined in microsomal membrane preparations obtained from cultured cerebellar granule cells. A KD of I nM and a Bmax of 64 fmol/mg protein were calculated from saturation experiments. This binding was calcium dependent and maximum values were obtained at 100–300 μM Ca++. Caffeine increased [3H]ryanodine binding only at Ca++ concentrations lower than optimum. The binding of [3H]ryanodine was inhibited by ruthenium red, procaine and the δ‐isomer of hexachlorocyclohexane (δ‐HCH). Dantrolene, a ryanodine receptor antagonist in skeletal sarcoplasmic reticulum, and the pesticide γ‐HCH (lindane) had no effect on [3H]ryanodine binding. The obtained binding parameters, the Ca++ dependence and the effects of the agents tested agree with previous reports using brain microsomal membranes, further indicating a neuronal localization of [3H]ryanodine binding sites. When the interaction between dantrolene and γ‐ and δ‐HCH was tested, no changes were detected on the effects of HCH isomers on [3H]ryanodine binding. Dantrolene, which inhibits Ca++ release from sarcoplasmic reticulum and from unidentified internal Ca++ stores in neurons, also inhibits the intracellular Ca++ mobilization induced by γ‐HCH but only marginally that induced by δ‐HCH in the same preparation of cerebellar granule cells (Rosa et al.; Toxicol Appl Pharmacol, in press). Thus, the results obtained in this work verify the presence of different intracellular sites of action for the two HCH isomers: the ryanodine Ca++ channel for δ‐HCH and an unidentified dantrolene‐sensitive Ca++ channel for the γ‐HCH isomer. J. Neurosci. Res. 47:27–33, 1997.

Use of Human Central Nervous System Cell Cultures in Neurotoxicity Testing

The nervous system is highly sensitive to toxic damage. Many environmental contaminants can produce acute or chronic neurological effects, and contribute to neural damage and cell death in neurodegenerative diseases. The utilization of primary cultures of neurons and glial cells is an essential step in investigating the specificity of the effects and mechanisms of action of the test chemical. If we take into account interspecies differences, cultures of human central nervous system (CNS) cells would be the best-suited test models for in vitro neurotoxicity testing. For practical and ethical reasons, human neuronal and glial cultures cannot be used for routine neurotoxicity testing, but they may be very useful for validating results from murine cultures and to address specific toxicity questions. For instance, we are investigating the action of agents producing oxygen radical damage in CNS cells. Oxidative stress is known to trigger apoptotic death of neurons and lead to neurodegeneration. A useful model in which to study these processes could be neuronal cultures obtained from CNS tissue with trisomy 21, since these cells suffer oxidative stress and apoptotic cell death in vitro. Besides primary cultures, human-derived clonal cell lines such as neuroblastoma SH-SY5Y can offer a first-step approach in neurotoxicity testing.

Cytotoxic action of lindane in neocortical GABAergic neurons is primarily mediated by interaction with flunitrazepam-sensitive GABAA receptors

The cytotoxic action of the γ‐isomer of hexachlorocyclohexane (γ‐HCH; lindane) was studied in cultured mouse neocortical neurons by measurements of the reduction in mitochondrial function using the MTT (3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromide) test. The cells were exposed to 30‐300 μM lindane in the culture medium for different periods of time and lindane cytotoxicity was found to be time‐ and concentration‐dependent. Lindane cytotoxicity could be ameliorated by addition of gamma aminobutyric acid (GABA) in a concentration‐dependent manner but this effect of GABA was not blocked by bicuculline or picrotoxinin (PTX). Lindane induced cytotoxicity was also reduced by the GABAA receptor agonists muscimol and THIP (4,5,6,7‐tetrahydroisoxazolo[5,4‐c]pyridin‐3‐ol). This effect was enhanced by the simultaneous presence of flunitrazepam but only at the highest lindane concentrations studied (200 and 300 μM). Flunitrazepam by itself had no effect on lindane‐induced cytotoxicity. The protective effect of GABA plus flunitrazepam was blocked by the benzodiazepine receptor antagonist flumazenil and by the GABAA antagonist bicuculline, suggesting the involvement of central benzodiazepine receptors allosterically coupled to the GABA recognition site at the GABAA receptor. When 100 μM PTX was used to suppress the protective effect of GABA and flunitrazepam, a significant effect of PTX was observed only at 300 μM lindane. The GABAB receptor agonist, baclophen, only marginally reduced the cytotoxic effect induced by the highest lindane concentrations. It is concluded that the cytotoxic action of lindane in neocortical neurons in culture is mediated primarily through an interaction with allosterically coupled GABA‐benzodiazepine recognition sites at the GABAA receptor. J. Neurosci. Res. 52:276–285, 1998.