Synaptic plasticity of kainate receptors.

Synaptic plasticity of ionotropic glutamate receptors has been extensively studied with a particular focus on the role played by NMDA (N-methyl-D-aspartate) receptors in the induction of synaptic plasticity and the subsequent movement of AMPA (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid) receptors. The third subtype of ionotropic glutamate receptor, kainate receptors, has not been studied to the same extent, but recent evidence shows that these receptors also exhibit synaptic plasticity in response to activity. There is also a growing body of data on the mechanisms underlying kainate receptor trafficking and the proteins they interact with. This review summarizes the current state of knowledge on this topic, focusing on the evidence for the removal or insertion of functional kainate receptors in response to synaptic activity and the cellular mechanisms that underlie this regulation of neuronal kainate receptor function.

Synaptically released neurotransmitter fails to desensitize postsynaptic GABA(A) receptors in cerebellar cultures.

GABA concentration jump experiments performed on membrane patches predict that postsynaptic GABA(A) receptors will become desensitized following the release of the contents of a single GABA-containing synaptic vesicle. To examine this we used a single synaptic bouton stimulation technique to directly examine whether postsynaptic GABA(A) receptors in cultured cerebellar granule cells exhibit transmitter-induced desensitization. In a large number of recordings, no evidence was found for desensitization of postsynaptic GABA(A) receptors by vesicularly released transmitter. This was the case even when as many as 40 vesicles were released from a single bouton within 1.5 s. In addition, postsynaptic depolarization and application of the benzodiazepine flunitrazepam, manipulations previously shown to enhance desensitization of GABA(A) receptors, failed to unmask transmitter-induced desensitization. In contrast, a single 2- to 3-s application of a high concentration of exogenous GABA was able to depress synaptic responsiveness for up to 70 s. Furthermore, pharmacological depletion of GABA eliminated inhibitory synaptic communication, suggesting that GABA is the transmitter and the desensitization-resistant inhibitory postsynaptic currents are not mediated by a "nondesensitizing" ligand such as beta-alanine. Overall our data indicate that a specific desensitization-resistant population of GABA(A) receptors are present at postsynaptic sites on cultured cerebellar granule cells.

Somato-synaptic variation of GABA(A) receptors in cultured murine cerebellar granule cells: investigation of the role of the alpha6 subunit.

Electrophysiological investigation of cultured cerebellar murine granule cells revealed differences between the GABA(A) receptors at inhibitory synapses and those on the cell body. Specifically, mIPSCs decayed more rapidly than cell body receptors deactivated, the mean single channel conductance at the synapse (32 pS) was greater than that at cell body (21 pS) and only cell body receptors were sensitive to Zn(2+) (150 microM), which depressed response amplitude by 82+/-5% and almost doubled the rate of channel deactivation. The GABA(A) receptor alpha6 subunit is selectively expressed in cerebellar granule cells. Although concentrated at synapses, it is also found on extrasynaptic membranes. Using a mouse line (Deltaalpha6lacZ) lacking this subunit, we investigated its role in the somato-synaptic differences in GABA(A) receptor function. All differences between cell body and synaptic GABA(A) receptors observed in wild-type (WT) granule cells persisted in Deltaalpha6lacZ cells, thus demonstrating that they are not specifically due to the cellular distribution of the alpha6 subunit. However, mIPSCs from WT and Deltaalpha6lacZ cells differed in both their kinetics (faster decay in WT cells) and underlying single channel conductance (32 pS WT, 25 pS Deltaalpha6lacZ). This provides good evidence for a functional contribution of the alpha6 subunit to postsynaptic GABA(A) receptors in these cells. Despite this, deactivation kinetics of mIPSCs in WT and Deltaalpha6lacZ granule cells exhibited similar benzodiazepene (BDZ) sensitivity. This suggests that the enhanced BDZ-induced ataxia seen in Deltaalpha6lacZ mice may reflect physiological activity at extrasynaptic receptors which, unlike those at synapses, display differential BDZ-sensitivity in WT and Deltaalpha6lacZ granule cells (Jones, A.M., Korpi, E.R., McKernan, R.M., Nusser, Z., Pelz, R., Makela, R., Mellor, J.R., Pollard, S., Bahn, S., Stephenson, F.A., Randall, A.D., Sieghart, W., Somogyi, P., Smith, A.J.H., Wisden, W., 1997. Ligand-gated ion channel partnerships: GABA(A) receptor alpha(6) subunit inactivation inhibits delta subunit expression. Journal of Neuroscience 17, 1350-1362).

The taurine uptake inhibitor guanidinoethyl sulphonate is an agonist at gamma-aminobutyric acid(A) receptors in cultured murine cerebellar granule cells.

In patch clamp experiments the beta-amino acid uptake inhibitor guanidinoethyl sulphonate (GES) activated currents in intact cultured murine cerebellar granule neurones. These responses could be attenuated by the gamma-aminobutyric acid(A) (GABA(A)) receptor antagonists bicuculline and picrotoxin. With intracellular chloride concentrations of either 20 or 130 mM, GES-induced current responses reversed polarity near the chloride equilibrium potential. When fast applications of agonist were made to excised granule cell macropatches GES responses were dose-dependent and exhibited significant outward rectification. Like taurine (but unlike GABA and beta-alanine) responses, macroscopic desensitisation of GES-induced currents was slow. Our data indicate that care should be exercised when using GES as a taurine uptake inhibitor in systems that also contain GABA(A) receptors.

Tonic benzodiazepine-sensitive GABAergic inhibition in cultured rodent cerebellar granule cells.

Recent studies have demonstrated that granule cells in rat cerebellar slices exhibit a tonic form of GABAergic inhibition. The presence of a similar constitutive GABAergic conductance was investigated in synaptically coupled cultures of neonatal rat cerebellum. In cells exhibiting spontaneous inhibitory postsynaptic currents (IPSCs), application of the GABA(A) receptor antagonist bicuculline (10 microM) eliminated the IPSCs and also produced a significant decrease in holding current. This latter effect was lacking in cells that did not exhibit IPSCs. Application of TTX (1 microM) and Cd(2+) (100 microM) decreased the IPSC frequency and also produced a change in holding current; these effects were eliminated by the prior application of bicuculline. In the presence of TTX, application of the benzodiazepine (BDZ) Flunitrazepam (1 microM) caused a 85+/-15% increase in the component of holding current that arose from GABA(A) receptor activity. Noise analysis indicated that the GABA(A) receptors underlying this tonic form of GABAergic inhibition exhibited a mean single channel conductance close to 14 pS, a value similar to that seen for somatic GABA(A) receptors in these cells. Thus, like their counterparts in cerebellar slices, cerebellar granule cells in culture exhibit a background GABAergic conductance. The most likely source of this tonic current is GABA spilt over from active inhibitory synapses. As this conductance was sensitive to benzodiazepine receptor agonists it is unlikely to arise entirely from GABA(A) receptors containing the alpha6 subunit.

Properties of GABA(A) receptors in cultured rat oligodendrocyte progenitor cells.

We have studied the properties of GABA responses in oligodendrocyte-type 2 astrocyte (O-2A) progenitor cells derived from primary cultures of the neonatal rat brain. In whole cell voltage clamp recordings, rapid application of 1-10 mM GABA elicited current responses in > 85% of the cells examined. The dose-response relationship pooled from nine progenitor cells was best fit by a logistic function of EC50=113 microM and Hill coefficient=0.9. In contrast to the rate of current deactivation, the rate of current activation exhibited marked concentration-dependence. Pharmacologically, GABA, muscimol and ZAPA ((Z)-3[(aminiiminomethyl)thio]prop-2-enoic acid sulphate) produced responses with ligand-specific kinetics, whereas glycine and the GABA(C) receptor agonist CACA were without effect; bicuculline methochloride acted as a competitive antagonist. Neither the amplitude nor the kinetics of currents produced by 100 microM GABA were affected by the benzodiazepine flunitrazepam (1 microM). Similarly the benzodiazepine receptor inverse agonist DMCM (1 microM) was also without effect. GABA-activated currents reversed polarity within 2 mV of the calculated Cl- equilibrium potential. With brief agonist pulses deactivation was monoexponential, however, unlike neurones the rate of deactivation was voltage-independent. Desensitisation of responses to 10 mM GABA was bi-exponential and accelerated at depolarised membrane potentials. Increasing the amount of GABA(A) receptor desensitisation (by increasing the duration of the agonist exposure) consistently produced a slowing of deactivation.

Mouse cerebellar granule cell differentiation: electrical activity regulates the GABAA receptor alpha 6 subunit gene.

GABAA receptor alpha6 subunit gene expression marks cerebellar granule cell maturation. To study this process, we used the Deltaalpha6lacZ mouse line, which has a lacZ reporter inserted into the alpha6 gene. At early stages of postnatal cerebellar development, alpha6-lacZ expression is mosaic; expression starts at postnatal day 5 in lobules 9 and 10, and alpha6-lacZ is switched on inside-out, appearing first in the deepest postmigratory granule cells. We looked for factors regulating this expression in cell culture. Membrane depolarization correlates inversely with alpha6-lacZ expression: granule cells grown in 25 mM [K+]o for 11-15 d do not express the alpha6 gene, whereas cultures grown for the same period in 5 mM [K+]o do. This is influenced by a critical early period: culturing for >/=3 d in 25 mM [K+]o curtails the ability to induce the alpha6 gene on transfer to 5 mM [K+]o. If the cells start in 5 mM [K+]o, however, they still express the alpha6-lacZ gene in 25 mM [K+]o. In contrast to granule cells grown in 5 mM [K+]o, cells cultured in 25 mM [K+]o exhibit no action potentials, mEPSCs, or mIPSCs. In chronic 5 mM [K+]o, factors may therefore be released that induce alpha6. Blockade of ionotropic and metabotropic GABA and glutamate receptors or L-, N-, and P/Q-type Ca2+ channels did not prevent alpha6-lacZ expression, but inhibition of action potentials with tetrodotoxin blocked expression in a subpopulation of cells.

Voltage-dependent deactivation and desensitization of GABA responses in cultured murine cerebellar granule cells.

1. Electrophysiological recordings of GABAergic IPSCs and responses to applications of exogenous GABA were made from cultured murine cerebellar granule cells. In both the presence and absence of tetrodotoxin, depolarization of the postsynaptic cell consistently produced a broadening of the IPSC. This voltage-dependent change in kinetics arose entirely from a slowing of the rate of current decay. The duration of miniature IPSCs was increased by a significant but lesser amount by the GABA uptake inhibitor nipecotic acid (300 microM). 2. Five millisecond applications of 1 mM GABA elicited rapidly activating, biexponentially deactivating currents in patches derived from granule cell bodies. Deactivation of these responses was slowed by membrane depolarization. This effect arose from an increased fractional participation of the slow component of deactivation. The benzodiazepine flunitrazepam (1 microM) slowed deactivation at a holding potential of -70 mV but not at +50 mV. 3. Longer-lasting applications of GABA produced substantial biexponential macroscopic desensitization. The rate of desensitization was faster at a holding potential of +50 mV than at -70 mV. The speeding of desensitization at depolarized membrane potentials arose from an increase in the fractional contribution of the fast component of desensitization. 4. When two 5 ms, 1 mM GABA applications were made at an interstimulus latency of 150 ms, the second response was consistently smaller than the first. The depression of the second response was significantly heightened when the membrane potential was depolarized from -70 to +50 mV. 5. The degree of desensitization produced was closely linked to receptor occupancy. The rate of current deactivation was also voltage dependent when non-saturating, and therefore less desensitizing, applications of GABA were analysed. In contrast, both the GABA EC50 (approximately 30 microM) and the current activation kinetics at near EC50 agonist concentrations appeared to be voltage independent.

Frequency-dependent actions of benzodiazepines on GABAA receptors in cultured murine cerebellar granule cells.

1. Miniature IPSCs recorded from cultured murine cerebellar granule cells increased in half-width and amplitude following application of the benzodiazepine (BDZ) Flunitrazepam (Flu, 1 microM). The increase in the half-width was much greater than that in the amplitude. 2. Five-millisecond applications of 1 mM GABA to nucleated outside-out patches elicited rapidly rising biexponentially decaying responses that resembled IPSCs. Flu had no effect on the amplitude of such responses, but consistently slowed their deactivation by approximately 50%. This effect was reversed by Flu washout or application of the BDZ antagonist Ro15-1788. The partial inverse agonist. Ro15-4513 speeded deactivation and depressed peak current amplitude by 23 +/- 12%. 3. The EC50 for GABA was between 45 and 50 microM. At submaximally effective agonist concentrations, Flu increased response amplitude and slowed response deactivation. Both effects were present in all cells taken from young cultures (4-7 days in vitro) but the latter was absent in 55% of the neurones obtained from older cultures (14-27 days in vitro). 4. With 120 ms applications of 20 microM GABA, responses activated monoexponentially (time constant, 39.8 +/- 2.8 ms) and deactivated biexponentially (time constants, 40.4 +/- 2.1 and 251 +/- 15 ms). Application of Flu slowed both activation and deactivation. The latter effect arose from an increased contribution of the slower component of decay. 5. Desensitization of responses to 1 mM GABA was biexponential, with time constants of 47 +/- 11 and 479 +/- 49 ms. Flu speeded desensitization by decreasing both fast and slow time constants. GABAA receptor desensitization consistently slowed subsequent deactivation. No significant relationship between the level of desensitization and the amount of slowing of deactivation produced by Flu was found. 6. Responses to paired 5 ms applications of 1 mM GABA indicated that the slowing of deactivation and the speeding of desensitization produced by Flu combine to generate a marked frequency dependence in the actions of this BDZ. Thus when compared with control responses, GABA-induced charge transfer was only enhanced by Flu during the first of two successive agonist applications.

Ligand-gated ion channel subunit partnerships: GABAA receptor alpha6 subunit gene inactivation inhibits delta subunit expression.

Cerebellar granule cells express six GABAA receptor subunits abundantly (alpha1, alpha6, beta2, beta3, gamma2, and delta) and assemble various pentameric receptor subtypes with unknown subunit compositions; however, the rules guiding receptor subunit assembly are unclear. Here, removal of intact alpha6 protein from cerebellar granule cells allowed perturbations in other subunit levels to be studied. Exon 8 of the mouse alpha6 subunit gene was disrupted by homologous recombination. In alpha6 -/- granule cells, the delta subunit was selectively degraded as seen by immunoprecipitation, immunocytochemistry, and immunoblot analysis with delta subunit-specific antibodies. The delta subunit mRNA was present at wild-type levels in the mutant granule cells, indicating a post-translational loss of the delta subunit. These results provide genetic evidence for a specific association between the alpha6 and delta subunits. Because in alpha6 -/- neurons the remaining alpha1, beta2/3, and gamma2 subunits cannot rescue the delta subunit, certain potential subunit combinations may not be found in wild-type cells.