Role of Ca+2 and other second messengers in excitatory amino acid receptor mediated neurodegeneration: clinical perspectives

Neurodegeneration associated with neurological disorders such as epilepsy, Huntington's Chorea, Alzheimer's disease, and olivoponto cerebellar atrophy or with energy failure such as ischemia, hypoxia, and hypoglycemia proceeds subsequent to overexposure of neurons to excitatory amino acids of which glutamate and aspartate may be quantitatively the most important. The toxic action of glutamate and aspartate is mediated through activation of glutamate receptors of the N-methyl-D-aspartate (NMDA) and non-NMDA subtypes. Antagonists for these receptors can act as neuroprotectants both in in vitro model systems (e.g., cultured neurons) and in vivo. Activation of receptors leads to an increase in the intracellular Ca++ concentration and also to an increase in other second messengers such as cGMP. Thus, Ca++ channel antagonists may have neuroprotective action under certain conditions.

Temporal and spatial differences in intracellular Ca++ changes elicited by K+ and glutamate in single cultured neocortical neurons

Changes as a function of time in the intracellular Ca++ concentration ([Ca++]i) in cultured cerebral cortical neurons were monitored after exposure of the cells to either 55 mM KCl or 100 microM glutamate using the fluorescent Ca++ chelator fura-2. The changes in [Ca++]i were followed in both cell bodies and neurites. Depolarization with K+ led to an immediate increase in [Ca++]i in neurites followed by a slower rise in the cell bodies. In contrast, glutamate elicited a slow increase in [Ca++]i in both neurites and cell bodies, and this increase showed a plateau rather than a peak as that seen after exposure to K+. The Ca++ channel blockers verapamil and nifedipine affecting N- and L-type channels, respectively had differential effects on K+ stimulated increases in [Ca++]i. Nifedipine only affected the increase marginally whereas verapamil inhibited the response by 50-60% both in cell bodies and neurites. The glutamate-induced increase in [Ca++]i was inhibited by nifedipine by 60% in neurites whereas no effect was observed in cell bodies. The results show that depolarization elicited by K+ and glutamate has different effects in different parts of the neurons and that the pharmacological characteristics of voltage gated Ca++ channels are dramatically different in cell bodies and neurites. Moreover, the distribution of L-type channels activated by glutamate differs in cell bodies and neurites. Such differences in the spatial distribution of Ca++ channels are likely to be of major importance for the functional consequences of depolarization coupled increases in [Ca++]i such as transmitter release and neurotoxicity.

Transfected parvalbumin alters calcium homeostasis in teratocarcinoma PCC7 cells

Indirect evidence supports a protective role of some EF-hand calcium-binding proteins against calcium-induced neurotoxicity. Little is known about how these proteins influence cytosolic calcium levels. After cloning the parvalbumin cDNA into an expression vector, teratocarcinoma cells (PCC7) were transfected. Parvalbumin-transfected and mock-transfected cells were loaded with the calcium indicator fura-2 and were exposed, in the same dish, to different concentrations of the calcium ionophore A23187 or to KCI. The results show that parvalbumin-transfected PCC7 cells had much better calcium buffering capacity than control cells.

Differential distribution of GABAA receptor subunits in soma and processes of cerebellar granule cells: effects of maturation and a GABA agonist

Quantitative analysis of the density of alpha 1 and beta 2/3 GABAA receptor subunits was performed at the electron microscope level after indirect pre-embedding immunogold labeling with subunit-specific antibodies of rat cerebellar granule cell cultures grown for 4 or 8 days and in the presence or absence of the GABAA receptor agonist 4,5,6,7-tetrahydroisoxazolo[5,4c]pyridin-3-ol (THIP). THIP (150 microM) induced a 2-fold increase in the number of alpha 1 and beta 2/3 subunits in both cell bodies and processes in 4-day-old cultures. Extending the culture period to 8 days led to a polarization of the receptor expression, since the increase in the number of subunits selectively was observed in the processes. Moreover, a general subcellular differentiation of the receptor population was observed in all culture conditions, since the ratio between the two subunits (beta 2/3; alpha 1) was four times higher in cell bodies compared to processes. A detailed analysis of the less mature (4-day-old) cultures revealed the existence of two populations of neurons exhibiting differences in the average number of receptors. During maturation neurons with few receptors developed into cells with a higher density of receptors resulting in a single population of the latter neurons, a process enhanced by exposure to THIP. This may indicate that receptor development is a discontinuous process with individual neurons following different temporal patterns. In double-labeling experiments, a spatially close association of the alpha 1 and beta 2/3 subunits could be seen, but the subunits were more frequently found separated from each other. In spite of the fact that exposure of the neurons to THIP increased the total number of receptor subunits, its presence apparently prevented formation of receptors with this subunit composition. Interestingly, receptor subunit clusters, consisting of alpha 1 alone, were more frequently observed than composite (alpha 1; beta 2/3) clusters. This substantiates the view that receptors not having alpha 1 and beta 2/3 subunits in the same complex may exist.

Depolarization by K+ and glutamate activates different neurotransmitter release mechanisms in GABAergic neurons: vesicular versus non-vesicular release of GABA

Neurotransmitter release and changes in the concentration of intracellular free calcium ([Ca++]i) were studied in cultured GABAergic cerebral cortical neurons, from mice, upon depolarization with either an unphysiologically high potassium concentration (55 mM) or the physiological excitatory neurotransmitter glutamate (100 microM). Both depolarizing stimuli exerted prompt increases in the release of preloaded [3H]GABA as well as in [Ca++]i. However, the basic properties of transmitter release and the increase in [Ca++]i under a variety of conditions were different during stimulation with K+ or glutamate. Potassium-evoked release of [3H]GABA consisted of two phases, a rapid, large and transient phase followed by a smaller, more persistent second phase. The rapid phase was inhibited (60%) by nocodazole which reduced the number of vesicles in the neurites by 80%. This rapid phase of the GABA release was also reduced by organic (verapamil) and inorganic (Co++) Ca++ channel blockers but was insensitive to the GABA transport inhibitor SKF 89976A. In contrast, the second phase was less sensitive to nocodazole and Ca++ channel antagonists but could be inhibited by SKF 89976A. The glutamate-induced [3H]GABA release, which was mainly mediated by N-methyl-D-aspartate receptors, consisted of a single, sustained phase. This was insensitive to nocodazole, partly inhibited by verapamil and could be blocked by Co++ as well as SKF 89976A. The action of Co++ could be attributed to a block of N-methyl-D-aspartate-associated ion channels. These findings strongly suggest that the majority of the K(+)-stimulated GABA release is dependent upon vesicles whereas the glutamate induced release is non-vesicular and mediated by a depolarization-dependent reversal of the direction of high-affinity GABA transport. The basic differences in the mode of action of the two depolarizing stimuli were reflected in the properties of the increase in [Ca++]i elicited by 55 mM K+ and 100 microM glutamate, respectively. The K(+)-induced increase in [Ca++]i was reduced by both verapamil and Ca(++)-free media whereas the corresponding glutamate response was only sensitive to Ca(++)-free conditions. Exposure of the cells to nocodazole or SKF 89976A had no effect on the ability of K+ or glutamate to increase [Ca++]i. Altogether, the results clearly demonstrate that K(+)-induced transmitter release from these GABAergic neurons is vesicular in nature whereas that induced by the neurotransmitter glutamate is not.

3H-D-aspartate release from cerebellar granule neurons is differentially regulated by glutamate- and K+-stimulation

Neurotransmitter release in response to either 55 mM K+ or 25 microM glutamate as well as its dependency on Ca2+ from different sources was compared in cultured glutamatergic cerebellar granule cells from rat brain. The intracellular Ca2+ concentration was monitored at the single cell level in neurites as well as cell bodies employing the fluorescent Ca2+ indicator fura-2. Transmitter release was assayed using 3H-D-aspartate to label the exogenously accessible glutamate pools, which in these neurons is believed to also include the transmitter pool. In an attempt to distinguish whether transmitter release was dependent on an intact cytoskeleton or not, the colchicine-like drug Nocodazole, which also blocks transport of vesicles, was used. K(+)-stimulated transmitter release consisted for the major part (around 70%) of a Ca(2+)-dependent, Nocodazole sensitive release component and this K(+)-induced release appeared to be almost exclusively dependent on N-type Ca2+ channels. In contrast, 50% of the glutamate-induced Ca(2+)-dependent release was triggered by Ca2+ from a Dantrolene sensitive intracellular Ca2+ pool. Since these neurons undergo a pronounced maturational change in which neurotransmitter vesicles become increasingly prominent, the Ca2+ responses and transmitter release evoked by the two different stimuli were investigated as a function of the culture period. K+ and glutamate were found to increase intracellular [Ca2+] differentially. In 1-day-old cultures K+ elicited a small albeit significant increase in [Ca2+]i while glutamate was completely without effect. In 7-day-old neurons both agents induced a large increase in [Ca2+].(ABSTRACT TRUNCATED AT 250 WORDS)

First direct electron microscopic visualization of a tight spatial coupling between GABAA-receptors and voltage-sensitive calcium channels

Using cerebellar granule neurons in culture it was demonstrated that exposure of the cells to the GABAA receptor agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP) leads to an increase in the number of voltage-gated calcium channels as revealed by quantitative preembedding indirect immunogold labelling using a monoclonal antibody specific for phenylalkylamine and dihydropyridine sensitive Ca2+ channels. Using the same technique and a monoclonal antibody (bd-17) to the beta 2/beta 3-subunit of the GABAA-receptor, double labelling of Ca2+ channels and GABAA-receptors with gold particles of different and well defined sizes were performed. This showed that in THIP-treated cultures 20% of GABAA-receptors in cell processes were located in close proximity (i.e. within 40 nm) of Ca2+ channels in the plasma membrane. This was not observed in non-treated cultures nor was it observed in cell bodies of THIP-treated cultures. This suggests that primarily low affinity GABAA-receptors are closely associated with Ca2+ channels and this may be important for the ability of these receptors to mediate an inhibitory action on transmitter release even under extreme depolarizing conditions.

Effect of a GABA agonist on the expression and distribution of GABAA receptors in the plasma membrane of cultured cerebellar granule cells: an immunocytochemical study

The effect of the gamma-aminobutyric acid (GABA) agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP, 150 microM) on the localization and density of GABAA receptors in the plasma membrane of rat cerebellar granule cells in primary cultures was studied at the electron microscope (EM) level by preembedding immunogold staining using the monoclonal antibody bd-17 directed against the beta-subunit of the GABAA receptor complex. In THIP-treated as well as untreated control cultures, GABAA receptors were found to be evenly distributed in the plasma membrane of cell bodies as well as processes. However, the density of the GABAA receptors was significantly increased in the THIP-treated cultures as compared to the control cultures and this effect of THIP was particularly pronounced in the processes. GABAA receptors were occasionally observed to form 'hot spots' in process-like structures and again the frequency of these areas with an extremely high density of GABAA receptors was greatly increased in the THIP-treated cultures compared to the controls. It has thus been demonstrated that the ability of GABA agonists to induce formation of low-affinity GABA receptors can be directly visualized and quantified at the EM level using the preembedding immunogold technique. It is likely that low-affinity GABAA receptors are preferentially located in the cell processes and to a considerable extent in the form of 'hot spots'. However, these 'hot spots' also contain high-affinity receptors.

Light and electron microscopic localization of GABAA-receptors on cultured cerebellar granule cells and astrocytes using immunohistochemical techniques

GABAA-receptors were localized in explant cultures of rat cerebellum and in dissociated primary cultures of rat cerebellar granule cells and rat cerebellar astrocytes using the monoclonal antibody bd-17 directed against the beta-subunit of the GABAA/benzodiazepine/chloride channel complex. At the light microscope level specific staining of GABAA-receptors was localized in various types of neurones in explant cultures of rat cerebellum using the indirect peroxidase-antiperoxidase (PAP) technique, whereas no specific staining was found in astrocytes. At the electron microscope level labeling of GABAA-receptors was observed in the plasma membrane of both the cell bodies and processes in dissociated primary cultures of cerebellar granule cells using an indirect preembedding immunogold staining technique which in contrast to the classical PAP technique allows quantitative estimations to be performed. Quantification of the labeling intensity revealed a higher concentration of GABAA-receptors per microns plasma membrane in the cell bodies than in the processes. In discrete areas an extremely high density of the GABAA-receptors was observed. No specific labeling of GABAA-receptors was observed in dissociated primary cultures of cerebellar astrocytes.

Effects of inhibitors of protein synthesis and intracellular transport on the gamma-aminobutyric acid agonist-induced functional differentiation of cultured cerebellar granule cells

The effect of inhibitors of protein synthesis (actinomycin D, cycloheximide), proteases (leupeptin), and intracellular transport (colchicine, monensin) on the gamma-aminobutyric acid (GABA) agonist [4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP)]-induced changes in morphological differentiation and GABA receptor expression was investigated in cultured cerebellar granule cells. After 4 days in culture the neurons were exposed to the inhibitors for 6 h in the simultaneous presence of THIP. Subsequently, cultures were either fixed for electron microscopic examination or used for preparation of membranes for [3H]GABA binding assays. In some experiments the functional activity of the newly induced low-affinity GABA receptors was assessed by investigation of the ability of GABA to inhibit neurotransmitter release from the neurons. These experiments were performed to differentiate between an intracellular and a plasma membrane localization of the receptors. In all experiments cultures treated with THIP alone served as controls. The inhibitors of protein synthesis totally abolished the ability of THIP to induce low-affinity GABA receptors. In contrast, the inhibitors of intracellular transport as well as the protease inhibitor did not affect this parameter. However, studies of effects of GABA on transmitter release from monensin-treated cultures showed that transmitter release could not be inhibited by GABA in these cells in spite of the presence of low-affinity GABA receptors in the membrane preparations. This indicates that the low-affinity receptors were not located in the plasma membrane. This is in good agreement with the corresponding morphological findings, that monensin treatment led to an intense vacuolization of the Golgi apparatus, thereby preventing intracellular transport of the newly synthesized GABA receptors.(ABSTRACT TRUNCATED AT 250 WORDS)

GABA agonist induced changes in ultrastructure and GABA receptor expression in cerebellar granule cells is linked to hyperpolarization of the neurons

GABA has been shown to exert a neurotrophic like activity by enhancing the morphological and functional maturation of neurons. Mechanisms involved in this effect of GABA are largely unknown but since GABA has been shown to mediate a hyperpolarizing action on neurons it can be assumed that this action might be important. In order to investigate this possibility, the ability to mimic the trophic actions of GABA of different agents known to influence the membrane potential or the GABA gated chloride channels was studied. Hence, GABA receptor expression as well as the ultrastructure of cerebellar granule cells were monitored after exposure of the cells in culture to either bromide, valinomycin or picrotoxin. It was found that cells which at early developmental stages (4 days in culture) were exposed to bromide or valinomycin expressed low affinity GABA receptors similar to cells treated with the GABA agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol(THIP). This was in contrast to untreated cultures or cultures treated with both THIP and picrotoxin which expressed only high affinity GABA receptors and no low affinity receptors. In cultures at more mature stages (14 days in culture) bromide treatment did not lead to formation of low affinity GABA receptors. Studies of the ultrastructure of the cells (4-day-old cultures) showed that exposure to bromide or valinomycin mimicked the ability of THIP to enhance the cytoplasmic density of rough endoplasmic reticulum, Golgi apparatus, vesicles and coated vesicles. Again, in 14-day-old cultures treatment with bromide had no effect on the ultrastructure.(ABSTRACT TRUNCATED AT 250 WORDS)

Glutamate receptor activation in cultured cerebellar granule cells increases cytosolic free Ca2+ by mobilization of cellular Ca2+ and activation of Ca2+ influx

The Ca2+ sensitive fluorescent probe, fura-2 has been used to monitor cytosolic free calcium levels in mature primary cultures of cerebellar granule cells during exposure to L-glutamate and other excitatory amino acids: quisqualate (QA) kainate (KA) and N-methyl-D-aspartate (NMDA). Glutamate at micromolar concentrations produced a prompt and dose-related increase in the intracellular concentration of free Ca2+, ([Ca2+]i), whereas QA, KA and NMDA had no effect. This increase was also seen in the absence of extracellular Ca2+, suggesting that L-glutamate promotes mobilization of Ca2+ from intracellular stores. In the presence of extracellular calcium, the elevation of [Ca2+]i was, in part, mediated by an increase in the plasma membrane permeability to Ca2+. This Ca2+ influx was not affected by the Ca2+-channel antagonist l-Verapamil. However, L-Verapamil did block the increase in [Ca2+]i seen after depolarization of the cells with potassium. The Ca2+ response elicited by glutamate was partially blocked by the excitatory amino acid antagonist glutamate diethyl ester (GDEE). Furthermore, glutamate stimulated the formation of inositol mono-, bis-, tris- and tetrakisphosphates (IP1, IP2, IP3, and IP4) suggesting a role for these compounds for the increase in [Ca2+]i.

Gamma-aminobutyric acid agonist-induced alterations in the ultrastructure of cultured cerebellar granule cells is restricted to early development

The effect of 4,5,6,7-tetrahydroisoxazolo[5,4-c]-pyridin-3-ol (THIP) on the ultrastructural composition of cultured cerebellar granule cells was investigated during development by quantitative electron microscopy (morphometric analysis). Granule cells were exposed to THIP (150 microM) for 6 h after 7 and 14 days, respectively, in culture. THIP treatment of 7-day-old cultures led to a statistically significant increase in the cytoplasmic density of rough endoplasmic reticulum, Golgi apparatus, vesicles, and coated vesicles, whereas no significant increase in the cytoplasmic density of these organelles was observed in 14-day-old cultures exposed to THIP for 6 h. These findings show that the effect of THIP on the ultrastructural composition of cultured cerebellar granule cells is restricted to early development.

GABA agonist promoted formation of low affinity GABA receptors on cerebellar granule cells is restricted to early development

The ability of the GABA receptor agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP) to promote formation of low affinity GABA receptors on cerebellar granule cells was tested using primary cultures of these neurons. Granule cells were exposed to THIP (150 microM) for 6 hr after, respectively, 4, 7, 10 and 14 days in culture. It was found that THIP treatment of 4- and 7-day-old cultures led to formation of low affinity GABA receptors, whereas such receptors could not be detected after THIP treatment in the older cultures (10 and 14 days) in spite of the fact that these cultured granule cells expressed a high density of high affinity GABA receptors. It is concluded that the ability of THIP to promote formation of low affinity GABA receptors on cerebellar granule cells is restricted to an early developmental period.

Differences between seizure-prone and non-seizure-prone mice with regard to glutamate and GABA receptor binding in the hippocampus and other regions of the brain

Quisqualate-preferring glutamate receptors were determined in membranes from frontal cortex, occipital cortex, hippocampus and cerebellum, from seizure-prone DBA/2J BOM and seizure-resistant C57/BL mice. The animals were studied 21, 27 and 40 days postnatally, i.e., before, during and after the age at which DBA mice are most susceptible to seizures. Radio-binding assays were performed using [3H]AMPA in the presence of 100 nM glutamate. Except for the occipital cortex, where no significant differences between the two strains were observed, all areas of the brain of DBA mice exhibited significantly (P less than 0.001, t test) higher AMPA binding than the corresponding areas of C57/BL mice at 27 days of age. At pre- and post-susceptible ages, the two strains showed no significant differences in the hippocampus and occipital cortex. A significant difference was observed, however, in the frontal cortex and cerebellum at the ages of 21 and 40 days, respectively, although this difference was considerably less than at 27 days. In addition to determination of glutamate receptors, GABA-receptor binding was also studied in membranes from the same cerebral areas and at the above-mentioned ages. Binding characteristics, using [3H]GABA as the ligand, were essentially identical in the two strains at all ages investigated, i.e., both low and high affinity GABA receptors could be identified with KD values of 6-16 nM and 100-800 nM, respectively.

Temporal development of GABA agonist induced alterations in ultrastructure and GABA receptor expression in cultured cerebellar granule cells

The temporal development of the effect of THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) on the ultrastructure composition and GABA receptor expression in cerebellar granule cells was investigated by quantitative electron microscopy (morphometric analysis) and GABA binding assays. It was found that the cytoplasmic density of smooth endoplasmic reticulum was decreased, while the cytoplasmic density of rough endoplasmic reticulum, Golgi apparatus, vesicles and coated vesicles was greatly enhanced after exposure of the cells to THIP (150 microM) for only 1 hr. In cerebellar granule cells exposed to THIP (150 microM) for 3 hr low affinity GABA receptors were induced. These findings show that the effect of THIP on the ultrastructure composition and GABA receptor expression in cultured cerebellar granule cells may be interrelated and moreover it is likely that the turn-over of GABA receptors is extremely fast.

GABA-agonists induce the formation of low-affinity GABA-receptors on cultured cerebellar granule cells via preexisting high affinity GABA receptors

The kinetics of specific GABA-binding to membranes isolated from cerebellar granule cells, cultured for 12 days from dissociated cerebella of 7-day-old rats was studied using [3H]GABA as the ligand. The granule cells were cultured in the presence of the specific GABA receptor agonist 4, 5, 6, 7-tetrahydroisoxazolo [5,4-c]pyridin-3-ol (THIP, 150 microM) or THIP plus the antagonist bicuculline methobromide (150 microM of each) or in the absence of the agonist or antagonist. Membranes isolated from granule cells cultured in a medium without the GABA agonist revealed a single binding site for GABA with a binding constant (KD) of 7.9 +/- 0.4 nM and a Bmax of 3.42 +/- 0.08 pmol X mg-1 protein. Membranes from cells cultured in the presence of THIP had two binding sites for GABA with KD-values of 6.8 +/- 0.9 nM and 476 +/- 311 nM, respectively. The corresponding Bmax values were 4.41 +/- 0.42 pmol X mg-1 and 5.81 +/- 1.20 pmol X mg-1. The effect of culturing the cells in THIP was antagonized by the simultaneous presence of bicuculline in the culture media, i.e. no significant low-affinity binding for GABA was found on the membranes from granule cells cultured in both THIP and bicuculline. The KD value (14.3 +/- 1.4 nM) for the high affinity binding site was, however, slightly increased compared to the non-treated cells. These findings suggest that the ability of THIP to induce formation of low-affinity GABA receptors is mediated by preexisting high-affinity GABA-receptors on the granule cells.