Cultured hippocampal slices retain many in vivo features with regard to circuitry, synaptic plasticity, and pathological responsiveness, while remaining accessible to a variety of experimental manipulations. The present study used ligand binding, immunostaining, and in situ hybridization assays to determine the stability of AMPA- and NMDA-type glutamate receptors and other synaptic proteins in slice cultures obtained from 11 day postnatal rats and maintained in culture for at least 4 weeks. Binding of the glutamate receptor ligands [3H]AMPA and [3H]MK-801 exhibited a small and transient decrease immediately after slice preparation, but the binding levels recovered by culture day (CD) 5-10 and remained stable for at least 30 days in culture. Autoradiographic analyses with both ligands revealed labeling of dendritic fields similar to adult tissue. In addition, slices at CD 10-20 expressed a low to high affinity [3H]AMPA binding ratio that was comparable with that in the adult hippocampus (10:1). AMPA receptor subunits GluR1 and GluR2/3 and an NMDA receptor subunit (NMDAR1) exhibited similar postcutting decreases as that exhibited by the ligand binding levels, followed by stable recovery. The GluR4 AMPA receptor subunit was not evident during the first 10 CDs but slowly reached detectable levels thereafter in some slices. Immunocytochemistry and in situ hybridization techniques revealed adult-like labeling of subunit proteins in dendritic processes and their mRNAs in neuronal cell body layers. Long-term maintenance was evident for other synapse-related proteins, including synaptophysin, neural cell adhesion molecule isoforms (NCAMs), and an AMPA receptor related antigen (GR53), as well as for certain structural and cytoskeletal components (e.g., myelin basic protein, spectrin, microtubule-associated proteins). In summary, following an initial and brief depression, many synaptic components were expressed at steady-state levels in long-term hippocampal slices, thus allowing the use of such a culture system for investigations into mechanisms of brain synapses.

Protein oligomerization and aggregation are key events in age-related neurodegenerative disorders, causing neuronal disturbances including microtubule destabilization, transport failure and loss of synaptic integrity that precede cell death. The abnormal buildup of proteins can overload digestive systems and this, in turn, activates lysosomes in different disease states and stimulates the inducible class of lysosomal protein degradation, macroautophagy. These responses were studied in a hippocampal slice model well known for amyloidogenic species, tau aggregates, and ubiquitinated proteins in response to chloroquine-mediated disruption of degradative processes. Chloroquine was found to cause a pronounced appearance of prelysosomal autophagic vacuoles in pyramidal neurons. The vacuoles and dense bodies were concentrated in the basal pole of neurons and in dystrophic neurites. In hippocampal slice cultures treated with Abeta(142), ultrastructural changes were also induced. Autophagic responses may be an attempt to compensate for protein accumulation, however, they were not sufficient to prevent axonopathy indicated by swellings, transport deficits, and reduced expression of synaptic components. Additional chloroquine effects included activation of cathepsin D and other lysosomal hydrolases. Abeta(142) produced similar lysosomal activation, and the effects of Abeta(142) and chloroquine were not additive, suggesting a common mechanism. Activated levels of cathepsin D were enhanced with the lysosomal modulator Z-Phe-Ala-diazomethylketone (PADK). PADK-mediated lysosomal enhancement corresponded with the restoration of synaptic markers, in association with stabilization of microtubules and transport capability. To show that PADK can modulate the lysosomal system in vivo, IP injections were administered over a 5-day period, resulting in a dose-dependent increase in lysosomal hydrolases. The findings indicate that degradative responses can be modulated to promote synaptic maintenance.

The coupling of GABA B receptors to G-protein-gated inwardly rectifying potassium (GIRK) channels constitutes an important inhibitory pathway in the brain. Here, we examined the mechanism underlying desensitization of agonist-evoked currents carried by homomeric GIRK2 channels expressed in HEK-293T cells. The canonical GABA B receptor agonist baclofen produced GIRK2 currents that decayed by 57.3±1.4% after 60 s of stimulation, and then deactivated rapidly (time constant of 3.90±0.21 s) upon removal of agonist. Surface labeling studies revealed that GABA B receptors, in contrast to l opioid receptors (MOR), did not internalize with a sustained stimulation for 10 min, excluding receptor redistribution as the primary mechanism for desensitization. Furthermore, heterologous desensitization was observed between GABA B receptors and MOR, implicating downstream proteins, such G-proteins or the GIRK channel. To investigate the G-protein turnover cycle, the non-hydrolyzable GTP analogue (GTPcS) was included in the intracellular solution and found to attenuate desensitization to 38.3±2.0%. The extent of desensitization was also reduced (45.3±1.3%) by coexpressing a mutant form of the Gaq G-protein subunit that has been designed to sequester endogenous RGS proteins. Finally, reconstitution of GABA B receptors with Gao G-proteins rendered insensitive to RGS resulted in significantly less desensitization (28.5±3.2%). Taken together, our results demonstrate that endogenous levels of RGS proteins effectively enhance GABA B receptor-dependent desensitization of GIRK currents.

Homomeric AMPA (a-amino-3-hydroxy-5methyl-4-isoxazolepropionic acid) -type glutamate receptors (GluRs) were stably expressed in kidney cells from cDNA5 encoding Glu Ri flop, GluR2 flip, GIuR2 flop, and GIuR3 flop subunits. The recombinant receptors were of the expected size and showed functional properties in whole-cell recording as previously reported.[ 3H]AMPA binding to all subunits was increased to a similar extent by the chaotropic ion thiocyanate (SCN ). Significant differences were found in the Scatchard plots, however, which were linear and of high affinity for GIuR1 and -3 receptors (K 0 values of 33 and 52 nM, respectively) but showed curvature for GIuR2 receptors, indicating the presence of two components with distinct affinities. As with brain AMPA receptors, solubilization of GIuR2 receptors reduced the number of lower-affinity sites and correspondingly increased the number of higher-affinity sites. The suif hydryl reagent p-chloromercuriphenylsulfonic acid, which increases binding to brain receptors, produced only minor changes except in the case of GIuR2 flip. These results indicate that GIuR2, among the subunits examined here, most closely resembles the native AMPA receptors in brain membranes. [ 3H]AMPAbinding was inhibited n a noncompetitive manner by two drugs that change the desensitization kinetics of the AMPA receptor. In agreement with physiological observations, the apparent affinity of cyclothiazide for GIuR2 flip (EC 50 = 7 tiM) was higher than that for receptors made of flop subunits (49-130 pM). In contrast, BDP-37, a member of the benzamide family of drugs, exhibited a lower potency for GIuR2 flip (58 [LM) than for any of the flop isoforms (18-40 pM). These results predict that the action of centrally active AMPA-receptor modulators varies across brain regions depending on their flip/flop composition.