Regulation of AMPA receptor surface trafficking and synaptic plasticity by a cognitive enhancer and antidepressant molecule.

The plasticity of excitatory synapses is an essential brain process involved in cognitive functions, and dysfunctions of such adaptations have been linked to psychiatric disorders such as depression. Although the intracellular cascades that are altered in models of depression and stress-related disorders have been under considerable scrutiny, the molecular interplay between antidepressants and glutamatergic signaling remains elusive. Using a combination of electrophysiological and single nanoparticle tracking approaches, we here report that the cognitive enhancer and antidepressant tianeptine (S 1574, [3-chloro-6-methyl-5,5-dioxo-6,11-dihydro-(c,f)-dibenzo-(1,2-thiazepine)-11-yl) amino]-7 heptanoic acid, sodium salt) favors synaptic plasticity in hippocampal neurons both under basal conditions and after acute stress. Strikingly, tianeptine rapidly reduces the surface diffusion of AMPA receptor (AMPAR) through a Ca(2+)/calmodulin-dependent protein kinase II (CaMKII)-dependent mechanism that enhances the binding of AMPAR auxiliary subunit stargazin with PSD-95. This prevents corticosterone-induced AMPAR surface dispersal and restores long-term potentiation of acutely stressed mice. Collectively, these data provide the first evidence that a therapeutically used drug targets the surface diffusion of AMPAR through a CaMKII-stargazin-PSD-95 pathway, to promote long-term synaptic plasticity.

Molecular determinants of kainate receptor trafficking.

Glutamate receptors of the kainate subtype are ionotropic receptors that play a key role in the modulation of neuronal network activity. The role of kainate receptors depends on their precise membrane and subcellular localization in presynaptic, extrasynaptic and postsynaptic domains. These receptors are composed of the combination of five subunits, three of them having several splice variants. The subunits and splice variants show great divergence in their C-terminal cytoplasmic tail domains, which have been implicated in intracellular trafficking of homomeric and heteromeric receptors. Differential trafficking of kainate receptors to specific neuronal compartments likely relies on interactions between the different kainate receptor subunits with distinct subsets of protein partners that interact with C-terminal domains. These C-terminal domains have also been implicated in the degradation of kainate receptors. Finally, the phosphorylation of the C-terminal domain regulates receptor trafficking and function. This review summarizes our knowledge on the regulation of membrane delivery and trafficking of kainate receptors implicating C-terminal domains of the different isoforms and focuses on the identification and characterization of the function of interacting partners.

Kainate receptor-interacting proteins and membrane trafficking.

Kainate receptors are composed of several subunits and splice variants, but the relevance of this diversity is still not well understood. The subunits and splice variants show great divergence in their C-terminal cytoplasmic tail region, which has been identified as a region of interaction with a number of protein partners. Differential trafficking of kainate receptors to neuronal compartments is likely to rely on interactions with distinct subsets of protein partners. This review summarizes our knowledge of the regulation of trafficking of kainate receptors and focuses on the identification and characterization of functions of interacting partners.

Addition of a glycophosphatidylinositol to acetylcholinesterase. Processing, degradation, and secretion.

We introduced various mutations and modifications in the GPI anchoring signal of rat acetylcholinesterase (AChE). 1) The resulting mutants, expressed in transiently transfected COS cells, were initially produced at the same rate, in an active form, but the fraction of GPI-anchored AChE and the steady state level of AChE activity varied over a wide range. 2) Productive interaction with the GPI addition machinery led to GPI anchoring, secretion of uncleaved protein, and secretion of a cleaved protein, in variable proportions. Unproductive interaction led to degradation; poorly processed molecules were degraded rather than retained intracellularly or secreted. 3) An efficient glypiation appeared necessary but not sufficient for a high level of secretion; the cleaved, secreted protein was possibly generated as a by-product of transamidation. 4) Glypiation was influenced by a wider context than the triplet omega/omega + 1/omega + 2, particularly omega - 1. 5) Glypiation was not affected by the closeness of the omega site to the alpha(10) helix of the catalytic domain. 6) A cysteine could simultaneously form a disulfide bond and serve as an omega site; however, there was a mutual interference between glypiation and the formation of an intercatenary disulfide bond, at a short distance upstream of omega. 7) Glypiation was not affected by the presence of an N-glycosylation site at omega or in its vicinity or by the addition of a short hydrophilic, highly charged peptide (FLAG; DYKDDDDK) at the C terminus of the hydrophobic region.

Kainate receptor-mediated synaptic currents in cerebellar Golgi cells are not shaped by diffusion of glutamate.

We report the presence of kainate receptors (KARs) in cerebellar Golgi cells of wild-type but not GluR6-deficient mice. Parallel fiber stimulation activates KAR-mediated synaptic currents [KAR-excitatory postsynaptic currents (EPSCs)] of small amplitude. KAR-EPSCs greatly differ from synaptic currents mediated by alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors (AMPAR-EPSCs) at the same synapse. KAR-EPSCs display slow rise and decay time and summate in response to a train of stimulations. By using PDA, a low-affinity competitive antagonist and agents that modify the clearance of glutamate, we show that these properties cannot be explained by diffusion of glutamate outside of the synaptic cleft and activation of extrasynaptic KARs. These data suggest that the slow kinetic of KAR-EPSCs is due to intrinsic properties of KARs being localized at postsynaptic sites. The contrasting properties of KAR- and AMPAR-EPSCs in terms of kinetics and summation offer the possibility for a glutamatergic synapse to integrate excitatory inputs over two different time scales.

Quaternary associations of acetylcholinesterase. II. The polyproline attachment domain of the collagen tail.

In transfected COS cells, we analyzed the formation of heteromeric associations between rat acetylcholinesterase of type T (AChET) and various constructions derived from the NH2-terminal region of the collagen tail of asymmetric forms, QN. Using a series of deletions and point mutations in QN, we showed that the binding of AChET to QN does not require the cysteines that normally establish intersubunit disulfide bonds with catalytic subunits and that it essentially relies on the presence of stretches of successive prolines, although adjacent residues also contribute to the interaction. We thus defined a proline-rich attachment domain or PRAD, which recruits AChET subunits to form heteromeric associations. Such molecules, consisting of one PRAD associated with a tetramer of AChET, are exported efficiently by the cells. Using the proportion of AChET subunits engaged in heteromeric tetramers, we ranked the interaction efficiency of various constructions. From these experiments we evaluated the contribution of different elements of the PRAD to the quaternary assembly of AChET subunits in the secretory pathway. The PRAD remained functional when reduced to six residues followed by a string of 10 prolines (Glu-Ser-Thr-Gly3-Pro10). We then showed that synthetic polyproline itself can associate with AChET subunits, producing well defined tetramers, when added to live transfected cells or even to cell extracts. This is the first example of an in vitro assembly of AChE tetramers from monomers and dimers. These results open the way to a chemical-physical exploration of the formation of these quaternary associations, both in the secretory pathway and in vitro.

Stable expression of acetylcholinesterase and associated collagenic subunits in transfected RBL cell lines: production of GPI-anchored dimers and collagen-tailed forms.

We obtained a stable expression of acetylcholinesterase (AChE, E.C. 3.1.1.7) in the rat basoleukemia cell line, RBL-2H3, which possesses a well developed secretory pathway, but expresses only very little endogenous AChE. Metabolic labeling showed that AChEH and AChET, differing by C-terminal peptides encoded by alternatively spliced exons, were synthesized at a similar rate. When transfected with AChEH, RBL cells efficiently produced GPI-anchored dimers, which were mostly exposed at the cell surface, as shown both by activity and immunofluorescence labeling. In contrast, when transfected with AChET, RBL cells produced about tenfold less activity, which was essentially retained in the cell, and the enzyme could not be detected at the cell surface by immunolabeling. The fate of the enzyme is therefore determined by its C-terminal alternative peptides. We were also able to coexpress the AChET subunit with the collagenic Q subunit. The cells produced small but significant amounts of collagen-tailed forms, essentially A4. The expression of these different catalytic and structural subunits in stably transfected RBL cells will be useful to explore the regulated posttranslational processes involved in protein maturation and transport.

Cloning and expression of a rat acetylcholinesterase subunit: generation of multiple molecular forms and complementarity with a Torpedo collagenic subunit.

We obtained a cDNA clone encoding one type of catalytic subunit of acetylcholinesterase (AChE) from rat brain (T subunit). The coding sequence shows a high frequency of (G+C) at the third position of the codons (66%), as already noted for several AChEs, in contrast with mammalian butyrylcholinesterase. The predicted primary sequence of rat AChE presents only 11 amino acid differences, including one in the signal peptide, from that of the mouse T subunit. In particular, four alanines in the mouse sequence are replaced by serine or threonine. In northern blots, a rat AChE probe indicates the presence of major 3.2- and 2.4-kb mRNAs, expressed in the CNS as well as in some peripheral tissues, including muscle and spleen. In vivo, we found that the proportions of G1, G2, and G4 forms are highly variable in different brain areas. We did not observe any glycolipid-anchored G2 form, which would be derived from an H subunit. We expressed the cloned rat AChE in COS cells: The transfected cells produce principally an amphiphilic G1a form, together with amphiphilic G2a and G4a forms, and a nonamphiphilic G4na form. The amphiphilic G1a and G2a forms correspond to type II forms, which are predominant in muscle and brain of higher vertebrates. The cells also release G4na, G2a, and G1a in the culture medium. These experiments show that all the forms observed in the CNS in vivo may be obtained from the T subunit. By co-transfecting COS cells with the rat T subunit and the Torpedo collagenic subunit, we obtained chimeric collagen-tailed forms. This cross-species complementarity demonstrates that the interaction domains of the catalytic and structural subunits are highly conserved during evolution.

Molecular architecture of acetylcholinesterase collagen-tailed forms; construction of a glycolipid-tailed tetramer.

Asymmetric forms of Torpedo acetylcholinesterase (AChE) are produced in COS cells by the simultaneous expression of collagenic subunits (Q) and catalytic T subunits (AChET). Truncated AChET delta subunits, from which most of the C-terminal peptide (TC) had been deleted by mutagenesis, did not associate with Q subunits. The TC peptide is therefore necessary for the association of the AChET and Q subunits. In order to determine the orientation of the Q subunit in the collagen-tailed forms, we have developed an antiserum against its non-collagenic C-terminal domain, expressed as a fusion protein in Escherichia coli. This antiserum, which recognized the Q subunit in Western blots, was found to react with intact asymmetric forms, but not with collagenase-treated forms, from which the distal part of the tail had been cleaved, suggesting that the N-terminal non-collogenic domain (QN) is responsible for the interaction with the AChET subunits. This was confirmed by creating a chimeric subunit (QN/HC), in which QN was linked to the C-terminal peptide of the H subunit of Torpedo AChE, which contains the glycophosphatidylinositol (GPI) cleavage/attachment signal: co-expression of AChET and QN/NC produced GPI-anchored tetramers, which were sensitive to PI-PLC and largely exposed to the external surface of the cells. We thus demonstrate that: (i) the HC peptide is sufficient to determine the addition of a glycolipid anchor and (ii) the QN domain is sufficient to bind a catalytic AChET tetramer by interacting with the TC peptide.

Primary structure of a collagenic tail peptide of Torpedo acetylcholinesterase: co-expression with catalytic subunit induces the production of collagen-tailed forms in transfected cells.

The asymmetric forms of cholinesterases are synthesized only in differentiated muscular and neural cells of vertebrates. These complex oligomers are characterized by the presence of a collagen-like tail, associated with one, two or three tetramers of catalytic subunits. The collagenic tail is responsible for ionic interactions, explaining the insertion of these molecules in extracellular basal lamina, e.g. at neuromuscular endplates. We report the cloning of a collagenic subunit from Torpedo marmorata acetylcholinesterase (AChE). The predicted primary structure contains a putative signal peptide, a proline-rich domain, a collagenic domain, and a C-terminal domain composed of proline-rich and cysteine-rich regions. Several variants are generated by alternative splicing. Apart from the collagenic domain, the AChE tail subunit does not present any homology with previously known proteins. We show that co-expression of catalytic AChE subunits and collagenic subunits results in the production of asymmetric, collagen-tailed AChE forms in transfected COS cells. Thus, the assembly of these complex forms does not depend on a specific cellular processing, but rather on the expression of the collagenic subunits.

Adenylyl cyclase amino acid sequence: possible channel- or transporter-like structure.

Complementary DNA's that encode an adenylyl cyclase were isolated from a bovine brain library. Most of the deduced amino acid sequence of 1134 residues is divisible into two alternating sets of hydrophobic and hydrophilic domains. Each of the two large hydrophobic domains appears to contain six transmembrane spans. Each of the two large hydrophilic domains contains a sequence that is homologous to a single cytoplasmic domain of several guanylyl cyclases; these sequences may represent nucleotide binding sites. An unexpected topographical resemblance between adenylyl cyclase and various plasma membrane channels and transporters was observed. This structural complexity suggests possible, unappreciated functions for this important enzyme.

Antibodies directed against transducin beta subunits interfere with the regulation of adenylate cyclase activity in brain membranes.

In an attempt to study the mechanisms of action of membrane-bound adenylate cyclase, we have applied to rat brain synaptosomal membranes antibodies raised against purified bovine transducin (T) beta gamma subunits. The antibodies recognized one 36-kDa protein in Western blots of the membranes. Adenylate cyclase activation by GTP non-hydrolyzable analogues was greatly decreased in immune, as compared to preimmune, antibody-treated membranes, whereas the enzyme basal activity was unaffected by both types of antibodies. The inhibition of forskolin-stimulated adenylate cyclase by guanine 5'-(beta, gamma-imino)triphosphate (Gpp-(NH)p) was decreased in membranes preincubated with immune, but not preimmune, antibodies. Anti-T beta antibodies moderately decreased the extent of subsequent adenylate cyclase activation by forskolin, while not affecting activation by Al3+/F-. The enzyme activation by Gpp(NH)p in untreated membranes remained the same upon further incubation in the presence of either type of antibodies. Such results were consistent with the decreased exchange of guanine nucleotides which occurred in membrane treated with immune, but not preimmune antibodies, upon addition of GTP. The blockade of the regulation of adenylate cyclase by Gpp(NH)p observed in membranes pretreated by anti-T beta antibodies thus appears to be caused by the impairment of the guanine nucleotide exchange occurring on Gs alpha subunits. The G beta subunits in the adenylate cyclase complex seem to be instrumental in the guanine nucleotide exchange on G alpha subunits, just as T beta subunits are in the transducin complex.

Purification of the catalytic subunit of adenylate cyclase in vertebrates: state of the art in 1987.

After 30 years of effort, the mammalian adenylate cyclase catalytic subunit has now been purified. It is a glycoprotein of 155 kDa, representing less than 0.01% of synaptosomal membrane protein. As measured in the presence of forskolin, its specific activity is 10-20 mumol of cAMP X mg-1 X min-1. The enzyme obtained is completely devoid of Gs alpha subunits, and is calmodulin-dependent. The purification procedures involve an affinity chromatography step, either with calmodulin, or with forskolin, or both. If gel filtration precedes the affinity chromatography, two different fractions with high specific enzyme activity are obtained. One contains the 155 kDa protein as the sole component. The other contains, as its major component, a 105 kDa protein. The relationship between the 2 proteins remains to be defined.

Evidence for two distinct adenylate cyclase catalysts in rat brain.

The Lubrol-soluble adenylate cyclase activity of brain synaptosomal membranes appeared, upon gel filtration or sucrose gradient centrifugation, as two overlapping peaks. Fractions corresponding to the peak of the largest Stokes radius (Biogel pool 1) or highest s value (gradient pool 1) contained an adenylate cyclase activity which could be detected whatever the enzyme assay conditions. In contrast, in fractions from the second peak (Biogel pool 2 or gradient pool 2), forskolin was needed to reveal adenylate cyclase activity. The enzyme activity of each Biogel pool was retained by forskolin-agarose and eluted by forskolin with a 34-83% yield. A polypeptide of 155 kDa made up 80% of the forskolin-agarose eluate 1, whereas it was almost absent from eluate 2. Since data from various groups point to the 155 kDa polypeptide as a brain adenylate cyclase catalyst, still another distinct catalyst of lower molecular mass is likely to be present in brain.

Identification of the catalytic subunit of brain adenylate cyclase: a calmodulin binding protein of 135 kDa.

The partial purification of the eukaryote adenylate cyclase [ATP pyrophosphate-lyase (cyclizing), EC 4.6.1.1] catalytic subunit has been achieved by a procedure based on the calmodulin (CaM) sensitivity of the enzyme. Small amounts of rat brain synaptosomal membranes depleted of CaM were solubilized with Lubrol and subjected to a three-step chromatographic procedure involving gel filtration, a CaM-Sepharose affinity step, and fast protein liquid chromatography. About 20% of the adenylate cyclase activity contained in the membranes was recovered in the final enriched fraction with a specific activity of 200 nmol X mg-1 X min-1. The alpha subunits of the adenylate cyclase stimulatory proteins NS were absent from this final fraction. The addition of CaM, of forskolin, or of preactivated NS-containing fractions to this preparation greatly increased the enzyme activity. A CaM-binding polypeptide of 135,000 Da copurified with the adenylate cyclase activity in each of the three steps. Polyacrylamide gel electrophoresis of the final fraction showed that this polypeptide represented 35% of the total protein. We propose that this polypeptide is likely to be the adenylate cyclase catalytic subunit. This enzyme would represent close to 0.5% of the synaptosomal membrane proteins. Its low turnover number would be due to the absence of the alpha subunits of the NS regulatory proteins and would correspond to the enzymic basal level.