Altered gabaa receptor subunit and splice variant expression in rats treated with chronic intermittent ethanol.

BACKGROUND: Intermittent chronic administration of ethanol to rats has been shown previously to produce a hyperexcitable, kindling-like state, accompanied by reduced inhibitory synaptic transmission in the hippocampus and changes in gamma-aminobutyric acid type A (GABAA) receptors. Further information is needed on the detailed changes in GABAA receptors and their time course and persistence, as is comparison to changes after chronic, continuous ethanol. METHODS: GABAA receptors were analyzed in the rat brain after chronic intermittent ethanol (CIE) by using radioligand binding, photoaffinity labeling of polypeptides, and estimates of messenger RNA (mRNA) levels of receptor subunits by reverse transcriptase-polymerase chain reaction (RT-PCR) and in situ hybridization. RESULTS: CIE rats were confirmed to have increased GABAA receptor binding of the benzodiazepine partial inverse agonist and ethanol antidote ligand Ro15-4513, due to increased expression of the alpha6 subunit polypeptide in the cerebellum, shown by photoaffinity labeling. Estimates of mRNA levels by use of RT-PCR did not reveal any significant increase in alpha6 or in several other receptor subunits in several brain regions, but a decrease in the ratio of the long and short splice variants (L/S) of the gamma2 subunit was detected in the hippocampus, especially the CA1 region. CONCLUSIONS: Changes in GABAA receptors were found in rats given CIE. Increased alpha6 subunit in the cerebellum was demonstrated by using both the binding to diazepam-insensitive sites for [3H]Ro15-4513 and increased levels of the 57-kDa alpha6 polypeptide after photoaffinity labeling with this ligand. This increase appeared after 30 doses of ethanol and decayed to normal 1 week after ethanol was discontinued. The transient change in cerebellar alpha6 subunit-containing receptors, also reportedly seen after chronic continuous ethanol, is thus unlikely to account for the persistently hyperexcitable, kindled, seizure-susceptible state seen in CIE. However, the significant decrease in gamma2 subunit L/S splice variant ratio in the hippocampus implies changes in GABAA receptor function, possibly involving protein phosphorylation by protein kinase C. Altered receptor trafficking and turnover associated with synaptic plasticity may contribute to the observed reduced inhibition in the hippocampus and other signs of alcohol dependence produced by CIE.

Heterogeneity of GABA(A) receptor-mediated responses in the human IMR-32 neuroblastoma cell line.

The gamma-aminobutyric acid (GABA) response profiles of IMR-32 human neuroblastoma cells were examined using whole-cell patch clamp and RT-PCR techniques. GABA activated a concentration-dependent and bicuculline-sensitive current, and RT-PCR revealed the expression of multiple GABA(A) receptor subunit mRNAs (alpha(1), alpha(3), alpha(4), beta(1), beta(3), gamma(2), and delta). A pharmacological profile of the GABA-induced current was derived using several subunit-selective agents. Diazepam, which requires the presence of a gamma subunit in order to modulate GABA(A) receptor-mediated responses, potentiated GABA-induced currents in a subset of IMR-32 cells. Two populations of GABA-activated currents were also evident based on sensitivity to modulation by zinc. Comparison of zinc- and diazepam-induced modulation of GABA-induced current responses in the same cells revealed an inverse correlation between these two modulators. No differences, however, were observed with the GABA(A) receptor modulators loreclezole, allopregnanolone, and pentobarbital. Thus, IMR-32 cells maintained in culture are heterogeneous in terms of expression of GABA(A) receptor isoforms.

Neuregulin induces GABA(A) receptor subunit expression and neurite outgrowth in cerebellar granule cells.

Neuregulin (NRG), a growth and differentiation factor that signals via erbB receptor tyrosine kinases, has been shown to have biological effects in both the CNS and the peripheral nervous system. We report here that erbB4 is expressed in mature cerebellar granule cells, where it appears to be concentrated at the granule cell postsynaptic terminals. We also show that one form of NRG, Ig-NRG, plays a crucial role in aspects of cerebellar granule cell development in vitro. First, Ig-NRG treatment of granule cells in culture selectively induces the expression of the GABA(A) receptor beta2 subunit. This increase in subunit expression is paralleled by an increase in functional GABA(A) receptors. In contrast to its effects on GABA(A) receptor subunit expression, Ig-NRG does not upregulate NMDA receptor N2B and N2C subunit expression. Second, we demonstrate that Ig-NRG also enhances neurite outgrowth from cultured granule cells. Ig-NRG does not, however, act as a survival factor for the granule cells. We have compared the effect of Ig-NRG with the effects of brain-derived neurotrophic factor (BDNF), a neurotrophin that exerts specific effects on granule cells in culture, and found that BDNF does not mimic the effects of Ig-NRG on GABA(A) receptor subunit expression. Our results show that Ig-NRG has specific effects on granule cell development and maturation and may regulate these processes in vivo.

Biphasic modulation of GABA(A) receptor binding by steroids suggests functional correlates.

Neuroactive steroids and other positive modulators of GABA(A) receptors showed regional variation in both the efficacy and potency for modulation of [35S]TBPS binding to rat brain membrane homogenates, with biphasic concentration-dependence. GABA present in the binding assays prevented the enhancement phase of the steroid concentration-dependence plot while the antagonists bicuculline and RU5135 prevented the inhibition phase. Using recombinant GABA(A) receptors, expressed in insect cell line Sf9 using baculovirus, enhancement by steroids of [35S]TBPS binding was sensitive to the presence of the gamma2 subunit and the nature of the alpha subunit (alpha1 beta2 gamma2S > alpha1 beta2, alpha6 beta2, alpha6 beta2 gamma2S, and alpha6 beta2 delta). As in cerebellum, addition of RU5135 reduced the inhibitory phase and revealed a small enhancement of TBPS binding by neuroactive steroids. The subunit-dependent interactions of steroid and GABA site ligands are consistent with a three-state model in which the receptor mono-liganded by GABA or steroid has a different affinity for TBPS than the resting state, and the receptor biliganded by GABA, steroid, or both has little affinity for TBPS.

Ethanol-GABAA receptor interactions: a comparison between cell lines and cerebellar Purkinje cells.

This study compared the interaction between ethanol and gamma-aminobutyric acid (GABA)-mediated current responses elicited in several immortalized cell lines and stably transfected cells, as well as in cultured and acutely dissociated rat cerebellar Purkinje cells. Only cell lines that were found previously to possess functional GABAA receptors were examined in this study. Under identical recording conditions, ethanol (10-200 mM) exerted no effect on GABA-induced currents in any of the cell lines or stably transfected cells tested in this study. However, GABA responses monitored in both primary culture and acutely dissociated Purkinje cells were significantly potentiated by ethanol (25 and 50 mM). Mouse pancreatic cells (RINm5F) were insensitive to both diazepam and ethanol suggesting the expression of a GABAA receptor isoform lacking a gamma subunit. Immortalized neuroblastoma IMR-32 cells displayed GABA responses that could be distinguished based on differential sensitivity to diazepam. However, none of the IMR-32 cells displayed GABA responses that were sensitive to modulation by ethanol. GABA responses in the stably transfected cell lines, PA3 (alpha1beta1gamma2L) and WSS-1 (alpha1beta2gamma2), were also unaffected by exposure to ethanol. In Purkinje cells acutely dissociated from the neonatal cerebellum, the ethanol-induced potentiation of GABA-induced current response could be observed before postnatal day 7, when only the gamma2S but not the gamma2L splice variant is expressed. This indicates that the gamma2L subunit is not necessary for an ethanol-induced potentiation of GABAA receptor-mediated response to become manifest. In addition, the results point to inherent differences that should be taken into account in interpreting comparative data between native and recombinant GABAA receptors.

Persistent reduction of GABA(A) receptor-mediated inhibition in rat hippocampus after chronic intermittent ethanol treatment.

GABA(A) receptor-mediated function was studied in rats treated with chronic intermittent ethanol (CIE). Rats were given 60 doses of 6g/kg ethanol every 24 h by gastric intubation, with repeated intoxicating and withdrawal episodes leading to a kindling-like increase in seizure susceptibility (Kokka et al., Alcohol: Clin. Exp. Res., 17 (1993) 525-531). Efflux of 36Cl-, evoked by application of muscimol, a measure of GABA(A) receptor function, was examined in 300 mu m slices obtained from frontal, parietal, and temporal cortex, hippocampus, and inferior colliculus, one day after the last administration of ethanol. Compared to controls, the 36Cl- efflux in hippocampal slices of CIE rats was significantly reduced by 29%, while there were no changes in the other brain regions studied. In hippocampal slices, paired-pulse inhibition in CA1 pyramidal neurons, measured extracellularly using homosynaptic orthodromic stimulation at an interval of 10 ms, was significantly reduced in CIE rats. A significant decrease by 40% both at 2 and 40 days after 60 doses of ethanol was found, implying a persistent decrease in GABA(A) receptor-mediated inhibition in CIE rats. These reductions in paired-pulse inhibition are consistent with the decrease in the pentylenetetrazol (PTZ) seizure threshold which was previously observed in CIE rats. Therefore, we suggest that this reduction of GABA(A) receptor-mediated inhibition contributes to the persistent increase in seizure susceptibility of CIE rats.

Modulation of GABAA receptor binding in human brain by neuroactive steroids: species and brain regional differences.

Allosteric modulation by neuroactive steroids of radioligand binding sites on the GABAA receptor complex was demonstrated by autoradiography in vitro in several regions of human brain and the effects compared to those in rat brain. Comparing human and rat, two steroids known to be active in enhancing GABA-mediated postsynaptic inhibition, 5 alpha-pregnane-3 alpha,21-diol-20-one (tetrahydro-deoxycorticosterone, THDOC) and alphaxalone (5 alpha-pregnane-3 alpha-hydroxy-11,20-dione), allosterically inhibited [35S]T-BPS binding to the picrotoxin/convulsant site in both species in several regions including the hippocampus. Unlike rat, human brain binding of [3H]flunitrazepam to the benzodiazepine site was not enhanced by alphaxalone (at any concentration), but was unaffected in many regions and inhibited in others. Binding of [3H]muscimol to high and low affinity GABA sites were enhanced by both steroids in all tested regions of rat brain, although to varying degrees. However, several lobes of human cortex showed no modulation of muscimol binding by either steroid, and THDOC, but not alphaxalone, inhibited in some areas. Comparing regions, THDOC at high concentrations (10 microM) enhanced in human frontal lobe and primary sensory and motor cortex, with greater effect in deep layers than superficial. This steroid had no effect in other parts of parietal lobe and inhibited muscimol binding in temporal lobe, primary visual cortex, and other parts of occipital lobe. Concentration-dependence curves for THDOC showed regional variation, e.g., in the hippocampal formation and surrounding neocortex. These regional and species differences are consistent with the existence of multiple GABAA receptor subtypes that differ in pharmacology. This heterogeneity provides both the opportunity and the difficulty of targeting clinically useful medications such as antiepileptic drugs to the appropriate human brain regions, and the species differences in regional subtype expression suggest caution in use of animal models.

Neuroactive steroid modulation of GABAA receptors.

In summary, the discovery of neuroactive steroids has led to a new appreciation of endocrine-nervous system interactions. The modulation of GABAA receptors by steroid hormone metabolites has far-reaching physiological and pharmacological implications, including relevance to human psychiatric and neurological problems (Fig. 7): I. MECHANISM OF GENERAL ANESTHESIA (Alphaxalone) II. GABA TONE IN SOME REGIONS (e.g., caudate) III. ANTI-ANXIETY ACTIVITY: (Post-partum depression, premenstrual syndrome?) (New drugs?) IV. CHILDBIRTH V. ANTIEPILEPTIC ACTIVITY (New drugs?) VI. REGIONAL HETEROGENEITY (Receptor subtypes) VII. PARTIAL AGONISTS (Better clinical profile?) FIG. 7. Implications of GABAA receptor as a nongenomic CNS target of steroids.

GABA alters GABAA receptor mRNAs and increases ligand binding.

In addition to its role as an inhibitory neurotransmitter, gamma-aminobutyric acid (GABA) influences the cytodifferentiation of developing neurons both in culture and in vivo. Here, we report some of the targets of GABA action and the mechanism through which GABA acts. In primary cultures of cerebellar granule cells, GABA specifically stimulates an increase in the levels of mRNAs for alpha 1 and beta 2 GABAA receptor subunits. The GABAA agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP) mimics this effect, and the GABAA antagonist bicuculline prevents it. In addition, GABA and THIP trigger an increase in the number of GABA binding sites. This increase parallels that seen in vivo, where the total number of GABAA receptor sites increases during postnatal cerebellar development. It is interesting that the period of the greatest increase in the number of receptor sites coincides with the development of the granule cells. Taken together, our data suggest that GABA may play an important role during maturation of cerebellar granule cells by influencing the number and composition of its own receptors.

The kindling model of alcohol dependence: similar persistent reduction in seizure threshold to pentylenetetrazol in animals receiving chronic ethanol or chronic pentylenetetrazol.

Rats on a chronic intermittent ethanol (CIE) regimen showed a persistent reduction in seizure threshold to the convulsant drug pentylenetetrazol (PTZ). CIE rats were given ethanol by intubation on an alternate day schedule and tested at selected intervals for seizure threshold with PTZ. A significant reduction in seizure threshold, a sign of withdrawal, was observed 20 hr after the first dose. The severity of withdrawal intensified on repetition of the ethanol administration and depression-hyperexcitability cycle, with the seizure threshold reaching a maximum decrease after 12 doses and remaining reduced up to 60 doses. The reduction in seizure threshold persisted for at least 40 days of no alcohol following the 60th dose. The long-lasting decrease in seizure threshold following CIE treatment resembled the "kindling" phenomenon produced by chronic administration of PTZ (25 mg/kg, 3 times/week). The CIE rats developed, in addition, a tolerance to the anticonvulsant action of ethanol, which occurred well after the decrease in PTZ seizure threshold, and a tolerance to the hypothermic effect of ethanol, which developed rapidly. PTZ kindled rats that had never been exposed to ethanol also exhibited tolerance to the hypothermic effect of ethanol. We propose that kindling contributes to the mechanism of the development of dependence on central nervous system depressants like benzodiazepines, barbiturates, and alcohol, drugs that act on the gamma-aminobutyric acid-A receptor chloride ion channel complex. Repeated episodes of depression and withdrawal hyperexcitability are postulated to produce kindling during the repeated withdrawal episodes.(ABSTRACT TRUNCATED AT 250 WORDS)

Sex differences in sensitivity to pentylenetetrazol but not in GABAA receptor binding.

Female rats have a higher threshold than males for seizures induced by the convulsant pentylenetetrazol, a GABAA receptor-chloride channel complex blocker. No sex difference was observed for the anticonvulsant activities of ethanol or diazepam to protect against pentylenetetrazol seizures. Ovariectomy reduces the pentylenetetrazol seizure threshold of females to that of males. In contrast, females have a lower threshold than males to electroshock seizures. Pentylenetetrazol receptors were compared in males and females and gonadectomized animals by binding of several radioligands to the GABAA receptor complex. No differences were found for these four groups of animals in the binding of [3H]flunitrazepam to the benzodiazepine sites and [35S]t-butyl bicyclophosphorothionate ([35S]TBPS) to the chloride channel/convulsant sites in membrane homogenates, nor for allosteric modulation of binding by GABA, the steroid anesthetic alphaxalone, or the benzodiazepine Ro 5-4864. In tissue section autoradiography, no difference was observed for these same assays nor for the binding of [3H]muscimol in the presence and absence of alphaxalone in several major regions. We conclude that circulating female sex hormones, possibly neurosteroid metabolites of progesterone, known to interact directly with the GABAA receptor complex, are involved in the sex differences in pentylenetetrazol seizure susceptibility.

Regional variation in steroid anesthetic modulation of [35S]TBPS binding to gamma-aminobutyric acidA receptors in rat brain.

Steroids that enhance gamma-aminobutyric acid (GABA)A receptor function in the central nervous system allosterically modulate the binding of the convulsant chloride channel ligand [35S]-t-butyl bicyclophosphorothionate. When assayed in membrane homogenates and in tissue sections by autoradiography, concentration-dependence curves vary with respect to both brain region and the nature of the steroid. Alphaxalone and endogenous steroid hormone metabolites inhibit the binding of [35S]-t-butyl bicyclophosphorothionate in some regions, enhance it in others and give biphasic concentration-dependence in others, apparently the result of algebraic summation of two effects involving regional-dependent enhancement or inhibition. The alphaxalone effect is additive with that produced by adding GABA to the binding assays in some regions, but synergistic in other areas. Likewise, the effect of GABA is inhibited completely by saturating concentrations of the antagonist bicuculline methochloride in some areas but only partially in others, and completely or partially reversed by the convulsant benzodiazepine Ro5-4864, depending on region. The granule cell and molecular layers of cerebellum are particularly different in these allosteric interactions. The heterogeneity of binding behavior is consistent with the presence of multiple GABAA receptor subtypes in the brain. Regional variation in subunit gene expression apparently produces a family of hetero-oligomeric GABAA receptors with different biological and pharmacological properties, including qualitative and quantitative differences in modulation by neuroactive steroids.

The benzodiazepine/alcohol antagonist Ro 15-4513: binding to a GABAA receptor subtype that is insensitive to diazepam.

The imidazobenzodiazepinone Ro 15-4513 has been shown previously to bind to central benzodiazepine receptors as well as to a second, uncharacterized class of sites that do not bind diazepam, differentiating them from the normal benzodiazepine-binding site on the gamma-aminobutyric acid (GABA)-A (GABAA) receptor. This study describes the characterization of these unique diazepam-insensitive (DZ-IS) sites. Ro 15-4513 binding to DZ-IS sites was abundant in cerebellum from cow, rat and human and detectable in cortex, hippocampus and striatum by autoradiography on rat brain sections. These sites represented approximately 20% of the total binding in bovine cerebellar membranes, but only 2 to 3% of the total in cortex. Ro 15-4513 binds with the same affinity (Kd approximately 4.5 nM) to both diazepam-sensitive and DZ-IS sites in the cerebellum. A number of compounds which bind to the classical benzodiazepine receptors also bind to the DZ-IS sites. These compounds include: the pyrazoloquinoline CGS 8216, the beta-carbolines methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate, ZK 95962, ZK 94326 and ZK 93126, as well as the classical benzodiazepine receptor antagonist, Ro 15-1788. Besides binding diazepam poorly, the DZ-IS sites demonstrate a very low affinity for other benzodiazepines. Ligands which bind to the various drug receptor sites on the GABA receptor complex do not directly modulate the binding of Ro 15-4513 to DZ-IS sites nor does ethanol. However, antagonism of Ro 15-4513 binding to the DZ-IS sites by methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate and by CGS 8216 is modulated by the presence of GABA.(ABSTRACT TRUNCATED AT 250 WORDS)

[3H]hemicholinium-3 binding in rabbit striatum: correspondence with patchy acetylcholinesterase staining and a method for quantifying striatal compartments.

Comparisons were made of the distribution of acetylcholinesterase (AChE) staining and [3H]hemicholinium-3 binding in adjacent tissue sections through the rabbit striatum. The distribution of these two markers matched closely in all regions examined. Circumscribed AChE-poor zones (striosomes) were visible throughout the rostral one-third of the caudate nucleus and were aligned with areas of low [3H]hemicholinium-3 binding in adjacent sections. Quantitative analysis of the [3H]hemicholinium-3 film autoradiographs revealed a 2- to 3-fold lower binding of [3H]hemicholinium-3 to sites within the striosome compartment compared with the darker background (matrix). Further analysis revealed at least 3 distinct levels of binding throughout the striatum. The results suggest a novel autoradiographic approach for identifying and quantifying striatal heterogeneities. Because [3H]hemicholinium-3 apparently labels the high-affinity choline transport complex associated with cholinergic neurons, the patchy striatal distribution of AChE/[3H]hemicholinium-3 apparently marks zones with differing densities of cholinergic neuron processes.

Human striatal dopamine receptors are organized in compartments.

Dopamine (D2) receptors visualized in postmortem human striatum by quantitative autoradiography of [3H]spiroperidol binding are organized into circumscribed zones of low receptor density separated from other such zones by regions of higher D2 density. The D2-rich zones of the caudate nucleus and putamen contain twice the binding of D2-poor zones. The Hill coefficient, obtained from saturation analysis of [3H]spiroperidol binding to thin sections of human striatum, gave a value near unity, indicating the binding was occurring to a single type of site. The patchiness of [3H]spiroperidol binding was unaltered by postincubation removal of lipid from the tissue sections, indicating that a differential absorption of tritium in white and grey matter does not account for the heterogeneous distribution. The D2-rich and D2-poor regions appear to form labyrinths oriented in the anterior-posterior axis and are typically aligned with, respectively, acetylcholinesterase-rich and -poor compartments as visualized on stained adjacent sections. Thus, the distribution of dopamine D2 receptors conforms to the "striosomal" organization of the human caudate-putamen, a finding that suggests that this receptor subtype may mediate the influence of dopamine on distinct neurochemical compartments within the structure.

Age-related regional loss of caudate-putamen dopamine receptors revealed by quantitative autoradiography.

Senescent (26-28 months) Fischer 344 rats were shown to have a lower density of D2 sites (-36%) without any change in affinity in membranes prepared from homogenized caudate-putamen (CPU), as compared to young adult (5-6 months) rats. In additional rats prepared for quantitative autoradiography of [3H]spiroperidol binding, the loss was shown to be specific for and quite extensive (50-60%) in the lateral to ventrolateral region of the CPU. The region of the CPU that shows the major age-related loss of D2 sites is critically involved in sensorimotor functions; therefore this regional loss of D2 sites may contribute importantly to the sensorimotor deficits in aged rats.