Ultrastructural characterization of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid-induced cell death in embryonic dopaminergic neurons.

Developing neuronal populations undergo significant attrition by natural cell death. Dopaminergic neurons in the substantia nigra pars compacta undergo apoptosis during synaptogenesis. Following this time window, destruction of the anatomic target of dopaminergic neurons results in dopaminergic cell death but the morphology is no longer apoptotic. We describe ultrastructural changes that appear unique to dying embryonic dopaminergic neurons. In primary cultures of mesencephalon, death of dopaminergic neurons is triggered by activation of glutamate receptors sensitive to alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and differs ultrastructurally from both neuronal apoptosis or typical excitotoxicity. AMPA causes morphological changes selectively in dopaminergic neurons, without affecting other neurons in the same culture dishes. Two hours after the onset of treatment swelling of Golgi complexes is apparent. At 3 h, dopaminergic neurons display loss of membrane asymmetry (coinciding with commitment to die), as well as nuclear membrane invagination, irregular aggregation of chromatin, and mitochondrial swelling. Nuclear changes continue to worsen until loss of cytoplasmic structures and cell death begins to occur after 12 h. These changes are different from those described in neurons undergoing either apoptosis or excitotoxic death, but are similar to ultrastructural changes observed in spontaneous death of dopaminergic neurons in the natural mutant weaver mouse.

Covalent and noncovalent interactions mediate metabotropic glutamate receptor mGlu5 dimerization.

Some, perhaps all, G protein-coupled receptors form homo- or heterodimers. We have shown that metabotropic glutamate receptors are covalent dimers, held together by one or more disulfide bonds near the N terminus. Here we report how mutating cysteines in this region affect dimerization and function. Covalent dimerization is preserved when cysteines 57, 93, or 99 are mutated but lost with replacement at 129. Coimmunoprecipitation under nondenaturing conditions indicates that the C[129]S mutant receptor remains a dimer, via noncovalent interactions. Both C[93]S and C[129]S bind [3H]quisqualate, whereas binding to C[57]S or C[99]S mutants is absent or greatly attenuated. The C[93]S and C[129]S receptors have activity similar to wild-type when assayed by fura-2 imaging of intracellular calcium in human embryonic kidney cells or electrophysiologically in Xenopus laevis oocytes. In contrast, C[57]S or C[99]S are less active in both assays but do respond with higher glutamate concentrations in the oocyte assay. These results demonstrate that 1) covalent dimerization is not critical for mGlu5 binding or function; 2) mGlu5 remains a noncovalent dimer even in the absence of covalent dimerization; and 3) high-affinity binding requires Cys-57 and Cys-99.

Ionic selectivity of low-affinity ratiometric calcium indicators: mag-Fura-2, Fura-2FF and BTC.

Accurate measurement of elevated intracellular calcium levels requires indicators with low calcium affinity and high selectivity. We examined fluorescence spectral properties and ionic specificity of three low-affinity, ratiometric indicators structurally related to Fura-2: mag-Fura-2 (furaptra), Fura-2FF, and BTC. The indicators differed in respect to their excitation wavelengths, affinity for Ca2+ (Kd approximately 20 microM, 6 microM and 12 microM respectively) and selectivity over Mg2+ (Kd approximately 2 mM for mag-Fura-2, > 10 mM for Fura-2FF and BTC). Among the tested indicators, BTC was limited by a modest dynamic range upon Ca2+ binding, susceptibility to photodamage, and sensitivity to alterations in pH. All three indicators bound other metal ions including Zn2+, Cd2+ and Gd3+. Interestingly, only in the case of BTC were spectral differences apparent between Ca2+ and other metal ions. For example, the presence of Zn2+ increased BTC fluorescence 6-fold at the Ca2+ isosbestic point, suggesting that this dye may be used as a fluorescent Zn2+ indicator. Fura-2FF has high specificity, wide dynamic range, and low pH sensitivity, and is an optimal low-affinity Ca2+ indicator for most imaging applications. BTC may be useful if experimental conditions require visible wavelength excitation or sensitivity to other metal ions including Zn2+.

Oligodendrocytes from forebrain are highly vulnerable to AMPA/kainate receptor-mediated excitotoxicity.

Little is known of the molecular mechanisms that trigger oligodendrocyte death and demyelination in many acute central nervous system insults. Since oligodendrocytes express functional alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate-type glutamate receptors, we examined the possibility that oligodendrocyte death can be mediated by glutamate receptor overactivation. Oligodendrocytes in primary cultures from mouse forebrain were selectively killed by low concentrations of AMPA, kainate or glutamate, or by deprivation of oxygen and glucose. This toxicity could be blocked by the AMPA/kainate receptor antagonist 6-nitro-7-sulfamoylbenzo(f)quinoxaline-2,3-dione (NBQX). In vivo, differentiated oligodendrocytes in subcortical white matter expressed AMPA receptors and were selectively injured by microstereotaxic injection of AMPA but not NMDA. These data suggest that oligodendrocytes share with neurons a high vulnerability to AMPA/kainate receptor-mediated death, a mechanism that may contribute to white matter injury in CNS disease.

Distinct roles for sodium, chloride, and calcium in excitotoxic dendritic injury and recovery.

The postsynaptic neuronal dendrite is selectively vulnerable to hypoxic-ischemic brain injury and glutamate receptor overactivation. We explored the glutamate receptor pharmacology and ionic basis of rapid, reversible alterations in dendritic shape which occur in cultured neurons exposed to glutamate. Dendrite morphology was assessed with the fluorescent membrane tracer, DiI, or immunofluorescence labeling of the somatodendritic protein, MAP2. Cortical cultures derived from 15-day-old mouse embryos underwent segmental dendritic beading when exposed to NMDA, AMPA, or kainate, but not to metabotropic glutamate receptor agonists. Varicosity formation in response to NMDA or kainate application was substantially attenuated in reduced sodium buffer (substituted with N-methyl-D-glucamine). Furthermore, veratridine-induced sodium entry mimicked excitotoxic alterations in dendrites and additionally caused varicosity formation in axons. Solutions deficient in chloride (substituted with Na methylsulfate) and antagonists of chloride-permeable GABA/glycine receptors reduced NMDA- or kainate-induced varicosity formation. An increase in dendrite volume was observed as varicosities formed, and varicosity formation was attenuated in sucrose-supplemented hypertonic media. Despite marked structural changes affecting virtually all neurons, dendrite shape returned to normal within 2 h of terminating glutamate receptor agonist application. Neurons exposed to kainate recovered more rapidly than those exposed to NMDA, and neurons exposed to NMDA in calcium-free buffer recovered more rapidly than cells treated with NMDA in normal buffer. While sodium, chloride, and water entry contribute to excitotoxic dendritic injury acutely, calcium entry through NMDA receptors results in lasting structural changes in damaged dendrites.

Neuronal free calcium measurement using BTC/AM, a low affinity calcium indicator.

BTC is a low affinity calcium indicator (Kd approximately 7-26 microM) featuring many desirable properties for cellular calcium imaging, including long excitation wavelengths (400/485 nm), low sensitivity to Mg2+, and accuracy of ratiometric measurement [Iatridou H., Foukaraki E., Kuhn M.A., Marcus E.M., Haugland R.P., Katerinopoulos H.E. The development of a new family of intracellular calcium probes. Cell Calcium 1994; 15: 190-198]. To assess the usefulness of this indicator in cultured neurons, we examined properties of BTC and its acetoxymethyl ester, BTC/AM. BTC/AM had substantial calcium-independent fluorescence at all excitation wavelengths. BTC/AM was readily loaded into neurons and was rapidly hydrolysed. There was little dye compartmentalization, as assessed by digitonin lysis, Co2+ quenching of BTC fluorescence and by confocal microscopy. Despite adequate loading, BTC gradually became unresponsive to [Ca2+]i when cultures were examined under routine imaging conditions. This effect was a function of the cumulative fluorescence illumination and could be minimized by attenuating light intensity or duration. Ratio imaging after exposure of neuronal cultures to 1-50 microM ionomycin revealed distinct sensitivity ranges for BTC and Fura-2. BTC reported graded neuronal [Ca2+]i responses to glutamate receptor stimulation with N-methyl-D-aspartate in the range 10-50 microM, whereas Fura-2 did not distinguish between these stimuli. Under appropriate loading and illumination conditions, bath-loaded BTC/AM may be well suited for measurement of moderate to high calcium concentrations in cultured neurons.

Ionized intracellular calcium concentration predicts excitotoxic neuronal death: observations with low-affinity fluorescent calcium indicators.

Cytosolic calcium ([Ca2+]i) is an important mediator of neuronal signal transduction, participating in diverse biochemical reactions that elicit changes in synaptic efficacy, metabolic rate, and gene transcription. Excessive [Ca2+]i also has been implicated as a cause of acute neuronal injury, although measurement of [Ca2+]i in living neurons by fluorescent calcium indicators has not consistently demonstrated a correlation between [Ca2+]i and the likelihood of neuronal death after a variety of potentially lethal insults. Using fluorescence videomicroscopy and microinjected calcium indicators, we measured [Ca2+]i in cultured cortical neurons during intense activation with either NMDA (300 microM) or AMPA (450 microM). At these concentrations NMDA killed >80% of the cultured neurons by the next day, whereas neuronal death from AMPA was <20%. Using the conventional calcium indicator, fura-2/AM, we estimated [Ca2+]i elevations to be approximately 300-400 nM during exposure to either glutamate agonist. In contrast, indicators with lower affinity for calcium, benzothiazole coumarin (BTC), and fura-2/dextran reported [Ca2+]i levels >5 microM during lethal NMDA exposure, but [Ca2+]i levels were <1.5 microM during nonlethal activation of AMPA receptors or voltage-gated calcium channels. Fura-2 reported [Ca2+]i responses during brief exposure to glutamate, NMDA, AMPA, kainate, and elevated extracellular K+ between 0.5 and 1 microM. With the use of BTC, only NMDA and glutamate exposures resulted in micromolar [Ca2+]i levels. Neurotoxic glutamate receptor activation is associated with sustained, micromolar [Ca2+]i elevation. The widely used calcium indicator fura-2 selectively underestimates [Ca2+]i, depending on the route of entry, even at levels that appear to be within its range of detection.

Electrophoretic heterogeneity of pigmented hamster melanoma cells.

Pigmented hamster melanoma tumors growing in situ contain two subpopulations of melanoma cells that have different electrophoretic mobilities (EPM). A mild neuraminidase treatment, which removes sialic acid residues from the cell surface glycoproteins, reduces the EPM of both groups of melanoma cells yielding an electrophoretically uniform population. This shows that the differences in the EPM between the subpopulations of pigmented melanoma cells stem from the different content of sialic acid residues on the cell surface. The relationship between the different EPM melanoma cell subpopulations was, therefore, examined during tumor growth, development, and formation of metastases. The relative content of cells having high electrophoretic mobility, the "fast moving" cells, increases as the tumors grow larger. However, tumors of the same diameter contain nearly the same fraction of "fast moving" cells despite their age. The proportion of the "fast moving" cells is significantly higher in the central part than in the outermost layer of pigmented melanoma tumors. These data suggest that the development of "fast moving" cells is promoted by some size-dependent changes in the intratumor environment. In vivo selection of melanoma cells for their ability to colonize lungs renders tumors that reveal elevated metastatic potential and contain a significantly higher fraction of cells possessing high electrophoretic mobility than the parent tumor. Moreover, the metastatic nodules contain a remarkably elevated fraction of the "fast moving" cells. The reported correlation between the "fast moving" cell fraction and the metastatic potential suggests that the relative content of cells having high electrophoretic mobility may determine the metastaticity of pigmented hamster melanoma.

The action of v-src on gap junctional permeability is modulated by pH.

The product of the viral src gene (v-src) is the protein tyrosine kinase pp60v-src. Among the known consequences of pp60v-src activity is the reduction in permeability of gap junctions, an effect that is counteracted by the calcium antagonist TMB-8 (8-N,N-[diethylamino]octyl-3,4,5-trimethoxybenzoate). We show here that a decrease in intracellular pH (pHi) also counteracts the v-src effect: junctional permeability of cells containing active v-src kinase rose with decreasing pHi in the range 7.15 to 6.75, whereas junctional permeability of cells containing inactive v-src kinase or no v-src at all was insensitive to pH in that range. Low pH also counteracted the known action of diacylglycerol on junction, but only when pp60v-src kinase was inactive. Immunoblots of whole-cell lysates using an antibody against phosphotyrosine show that phosphorylation on tyrosine of at least one cellular protein, specific for pp60v-src kinase activity, was reduced by low pH but not by TMB-8. These results suggest that TMB-8 does not inhibit v-src action on junctional permeability by interfering with tyrosine phosphorylation of a protein crucial for closure of gap junction channels, but that the inhibition by low pH may be via this mechanism.