Green fluorescent protein as a quantitative tool.

Manipulating the expression of a protein can provide a powerful tool for understanding its function, provided that the protein is expressed at physiologically-significant concentrations. We have developed a simple method to measure (1) the concentration of an overexpressed protein in single cells and (2) the covariation of particular physiological properties with a protein's expression. As an example of how this method can be used, teratocarcinoma cells were transfected with the neuron-specific calcium binding protein calretinin (CR) tagged with green fluorescent protein (GFP). By measuring GFP fluorescence in microcapillaries, we created a standard curve for GFP fluorescence that permitted quantification of CR concentrations in individual cells. Fura-2 measurements in the same cells showed a strong positive correlation between CR-GFP fusion protein expression levels and calcium clearance capacity. This method should allow reliable quantitative analysis of GFP fusion protein expression.

Developmental changes in the subcellular localization of calretinin.

Brainstem auditory neurons in the chick nucleus magnocellularis (NM) express high levels of the neuron-specific calcium-binding protein calretinin (CR). CR has heretofore been considered a diffusible calcium buffer that is dispersed uniformly throughout the cytosol. Using high-resolution confocal microscopy and complementary biochemical analyses, we have found that during the development of NM neurons, CR changes from being expressed diffusely at low concentrations to being highly concentrated beneath the plasma membrane. This shift in CR localization occurs at the same time as the onset of spontaneous activity, synaptic transmission, and synapse refinement in NM. In the chick brainstem auditory pathway, this subcellular localization appears to occur only in NM neurons and only with respect to CR, because calmodulin remains diffusely expressed in NM. Biochemical analyses show the association of calretinin with the membrane is detergent-soluble and calcium-independent. Because these are highly active neurons with a large number of Ca2+-permeable synaptic AMPA receptors, we hypothesize that localization of CR beneath the plasma membrane is an adaptation to spatially restrict the calcium influxes.

Calcium influx via ionotropic glutamate receptors causes long lasting inhibition of metabotropic glutamate receptor-coupled phosphoinositide hydrolysis.

Functional interaction between ionotropic and metabotropic glutamate receptors (iGluR and mGluR respectively) was studied in cerebellar granule cell cultures using quisqualate (QA), the most potent agonist of phosphoinositide hydrolysis coupled mGluR, and N-methyl-D-aspartate (NMDA) or kainate (KA) that activate different classes of iGluR. Two h exposure to NMDA or KA resulted in a marked reduction (about 75%) of QA-evoked PI hydrolysis. The efficacy of the two agonists was about the same, but the potencies were different (IC50 for NMDA about 35 microM and for KA about 70 microM). NMDA-induced depression of QA-stimulated PI hydrolysis was relatively long lasting but reversible. Recovery required protein synthesis. In nominally Ca2+-free medium both NMDA and KA failed to attenuate QA-stimulated PI hydrolysis. The effect of NMDA was prevented by the NMDA receptor antagonist MK801, but not by the wide spectrum protein kinase inhibitor staurosporin nor by the nitric oxide synthase inhibitor N omega-nitro-L-arginine. Cycloheximide and concanavalin A were also ineffective. The effect of KA was prevented by the selective non-NMDA receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX). Voltage sensitive Ca2+ channel antagonists together with MK801 did not counteract the inhibition by KA of the QA response. Both NMDA and KA attenuated PI hydrolysis evoked by the muscarinic receptor agonist carbachol (about 30%), indicating that the activation of iGluRs exerts a relatively general inhibitory effect on the function of different PLC-coupled metabotropic receptors. Consistent with this observation is that treatments either with KA and NMDA induced an inhibition (about 30%) of NaF-stimulated PI hydrolysis which occurs through the direct activation of G proteins. Our observations show that ionotropic glutamate receptor stimulation induces a long lasting suppression of QA-evoked PI breakdown through a Ca2+ dependent mechanism which seems to involve receptor coupled transduction systems downstream from mGluR. Such a Ca2+-dependent cross-talk involving ionotropic and metabotropic receptors may play a role in certain events of synaptic plasticity.

Development of voltage-activated potassium currents in cultured cerebellar granule neurons under different growth conditions.

1. The functional expression of two potassium currents in cultured cerebellar granule cells was investigated with the whole cell patch-clamp technique in relation to development and growth condition. Cells were grown in medium containing different concentrations of potassium: 25 mM (K25) and 40 mM (K40), together referred to as "high K+"; 10 mM (K10) or "low K+"; and K10 with 100 microM N-methyl-D-aspartate (KNMDA). All conditions are known to influence maturation and survival of granule cells in culture. 2. At 2 days in vitro (DIV) the membrane capacitance, taken as index of membrane surface area, was the same for cells grown in each growth condition. At 7-9 DIV it had increased in each condition, but to a substantially larger extent in cells grown in KNMDA, K25, and K40 than in cells grown in K10. During development the input resistance only decreased in cells grown in KNMDA and high K+. 3. A delayed potassium current (IK) and a fast transient potassium current (IA) could both be recorded at 2 DIV in each growth condition, although a few neurons only expressed the IK. The IK was partially suppressed by tetraethylammonium (5 mM), whereas IA was predominantly sensitive to 4-aminopyridine (5 mM). 4. Normalized for cell capacitance, the specific IA conductance hardly changed during development in cells grown in high K+ and KNMDA. Cells in K10, however, displayed an IA with totally different properties in 23 of 24 cells; the specific IA conductance in these cells was considerably smaller at 7-9 DIV, suggesting a deletion of these channels during development.(ABSTRACT TRUNCATED AT 250 WORDS)

AMPA receptors in cerebellar granule cells during development in culture.

The survival and maturation of differentiating cerebellar granule cells in culture are known to be promoted by excitatory amino acids (EAAs) which, however, compromise the survival of mature cells. In contrast to the trophic effect, the toxic effect of alpha-amino-3-hydroxy-5-methyl-4-isoxasolepropiate (AMPA) could only be elicited when the desensitisation of AMPA receptors was blocked, cyclothiazide being used in this study. Nevertheless, even under these conditions, toxicity induced by AMPA in contrast to kainate was, at 9 DIV, only half of the maximal toxicity attained by 13-16 DIV. Since cellular responses to AMPA depend so dramatically on the maturational stage of granule cells, we examined here whether this characteristic is related to developmental changes in AMPA receptor properties, which may result from changes in the subunit composition of the receptor. In contrast to toxicity, AMPA-induced 45Ca2+ influx (determined in the presence of cyclothiazide and the NMDA receptor blocker MK-801) reached a maximum already at 9 DIV. This also applied to a fraction of the 45Ca2+ uptake which persisted either after Cd2+ application or under Na(+)-free conditions and therefore presumably was mediated directly through AMPA receptor channels. Quantitative analysis of Western blots showed that the amounts of GluR4 and to a lesser extent GluR2/3/4c are substantial already at 2 DIV, remaining fairly constant until 9 DIV, followed by an increase by 16 DIV. However GluR1, which is hardly detectable in granule cells in vivo and is also low early in vitro, increased almost linearly with cultivation time.(ABSTRACT TRUNCATED AT 250 WORDS)

Subcellular localization and properties of lipase activities in human polymorphonuclear leukocytes.

A fluorimetric assay for lipase activity has been optimized for measurement of the enzyme in human neutrophils. Activity was maximal at acid (4.5) and alkaline (9.5) pH, although there was also a neutral peak of activity at pH 6.5. Neutrophils were homogenised in isotonic sucrose and subjected to analytical subcellular fractionation by sucrose density gradient centrifugation. The gradient fractions were assayed for acid, neutral and alkaline lipase activity and for the principal organelle marker enzymes. Neutral lipase showed a unimodal distribution with an equilibrium density of 1.19 g . cm-3, corresponding to the distribution of particulate leucine aminopeptidase. Acid and alkaline lipase activities showed very similar distribution profiles to each other with both soluble components and a broad peak of particulate activity. The broad modal density of 1.19-1.22 g . cm-3 suggests that acid and alkaline lipase activities could be localised to more than one population of cytoplasmic granule. Fractionation experiments with neutrophils homogenised in sucrose medium containing digitonin confirmed the localisation of neutral lipase and leucine aminopeptidase to the same cytoplasmic granule, and suggested that at least part of the acid lipase activity was localised to the specific granule. No lipase activity could be attributed to the alkaline phosphatase-containing granule. Neutrophils were isolated from control subjects, patients with chronic granulocytic leukaemia and women in the third trimester of pregnancy. The specific activity of acid, neutral and alkaline lipase, and leucine aminopeptidase, in contrast to that of alkaline phosphatase, were similar in the three patient groups.