Search

Collybistin (CB) is a guanine nucleotide exchange factor (GEF) selectively localized at GABAergic and glycinergic postsynapses. Analysis of mRNA shows that several isoforms of collybistin are expressed in the brain. Some of the isoforms have a SH3 domain (CBSH3+) and some have no SH3 domain (CBSH3-). The CBSH3+ mRNAs are predominantly expressed over CBSH3-. However, in an immunoblot study of mouse brain homogenates, only CBSH3+ protein isoforms were detected, proposing that CBSH3- protein might not be expressed in the brain. The expression or lack of expression of CBSH3- protein is an important issue because CBSH3- has a strong effect in promoting the postsynaptic clustering of gephyrin and GABA-A receptors (GABAA Rs). Moreover CBSH3- is constitutively active; therefore lower expression of CBSH3- protein might play a relatively stronger functional role than the more abundant but self-inhibited CBSH3+ isoforms, which need to be activated. We are now showing that: (a) CBSH3- protein is expressed in the brain; (b) parvalbumin positive (PV+) interneurons show higher expression of CBSH3- protein than other neurons; (c) CBSH3- is associated with GABAergic synapses in various regions of the brain and (d) knocking down CBSH3- in hippocampal neurons decreases the synaptic clustering of gephyrin and GABAA Rs. The results show that CBSH3- protein is expressed in the brain and that it plays a significant role in the size regulation of the GABAergic postsynapse.

The glutamate-nitric oxide (NO)-cGMP pathway modulates some forms of learning. How glycine modulates this pathway is unclear. Glycine could modulate the pathway biphasically, enhancing its function through NMDA receptor activation or reducing it through glycine receptor activation. Chronic hyperammonemia impairs the glutamate-NO-cGMP pathway in the cerebellum and induces cognitive impairment. The possible alterations in hyperammonemia of glycinergic neurotransmission and of glutamate-NO-cGMP pathway modulation by glycine remain unknown. The aims were to assess, by in vivo microdialysis in cerebellum: (i) the effects of different glycine concentrations, administered through the microdialysis probe, on the glutamate-NO-cGMP pathway function; (ii) the effects of tonic glycine receptors activation on the pathway function, by blocking them with strychnine; (iii) whether hyperammonemia alters the pathway modulation by glycine; (iv) and whether hyperammonemia alters extracellular glycine concentration and/or glycine receptor membrane expression. In control rats, low glycine levels reduce the pathway function, likely by activating glycine receptors, while 20 μM glycine enhances the pathway function, likely by enhancing NMDA receptor activation. In hyperammonemic rats, glycine did not reduce the pathway function, but enhanced it when administered at 1-20 μM. Hyperammonemia reduces extracellular glycine concentration by approximately 50% and glycine receptor membrane expression. However, tonic glycine receptor activation seems to be enhanced in hyperammonemic rats, as indicated by the larger increase in extracellular cGMP induced by strychnine. These data show that glycine modulates the glutamate-NO-cGMP pathway biphasically and that hyperammonemia strongly alters glycinergic neurotransmission and modulation by glycine of the glutamate-NO-cGMP pathway. These alterations may contribute to the cerebellar aspects of cognitive alterations in hyperammonemia. The findings reported in this study show that hyperammonemia alters glycinergic neurotransmission and the glutamate-NO-cGMP pathway modulation by glycine. In control rats, low glycine levels reduced the pathway function, likely by activating glycine receptors, while 20 μM glycine enhanced the pathway, likely by enhancing NMDA receptor activation. In hyperammonemic rats, glycine (administered at 1-20 μM) enhances the pathway, likely by activating NMDA receptors.

The exact effect of glycine pre-treatment on brain ischemic tolerance (IT) remains quite controversial. The objective of this study was to investigate the potential effects of glycine on IT. We used rat models of both in vitro ischemia (oxygen and glucose deprivation) and in vivo ischemia (transient middle cerebral artery occlusion). Low doses of glycine (L-Gly) significantly decreased hippocampal ischemic LTP (i-LTP), infarct volume, and neurological deficit scores which were administered before ischemia was induced in rats, whereas high doses of glycine exerted deteriorative effects under the same condition. These findings suggested that exogenous glycine may induce IT in a dose-dependent manner. Furthermore, L-Gly-dependent neuronal protection was inversed by L689, a selective NMDAR glycine site antagonist both in vitro (abolished i-LTP depression) and in vivo (increased infarct size reduction), but not glycine receptor (GlyR) inhibitor strychnine. Importantly, L-Gly-induced IT was achieved by NR2A-dependent cAMP-response element binding protein phosphorylation. These data imply that glycine pre-treatment may represent a novel strategy for inducing IT based on synaptic NMDAR-dependent neuronal transmission. A model of glycine induced dose-dependent bidirectional regulations in ischemic tolerance. (a) Under low dose of Gly pre-treatment, glycine induces NMDAR potentiation and CREB-dependent neuroprotection through the NMDAR co-agonist binding site. (b) Under high dose of Gly pre-treatment, the excessive glycine in synaptic cleft can activate neighboring extrasynaptic sites and combine to the GlyRs. Then, the deteriorative effects would be triggered after NMDAR endocytosis and synaptic depression. AMPAR, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor; CREB, cAMP response element-binding protein; Gly, glycine; GlyR, glycine receptor; GlyT1, gycine transportor 1; NMDAR, N-methyl-d-aspartate receptor.

Glycine receptor (GlyR) truncations in the intracellular TM3-4 loop, documented in patients suffering from hyperekplexia and in the mouse mutant oscillator, lead to non-functionality of GlyRs. The missing part that contains the TM3-4 loop, TM4 and C-terminal sequences is essential for pentameric receptor arrangements. In vitro co-expressions of GlyRα1-truncated N-domains and C-domains were able to restore ion channel function. An ionic interaction between both domains was hypothesized as the underlying mechanism. Here, we analysed the proposed ionic interaction between GlyR N- and C-domains using C-terminal constructs with either positively or negatively charged N-termini. Charged residues at the N-terminus of the C-domain did interfere with receptor surface expression and ion channel function. In particular, presence of negatively charged residues at the N-terminus led to significantly decreased ion channel function. Presence of positive charges resulted in reduced maximal currents possibly as a result of repulsion of both domains. If the C-domain was tagged by a myc-epitope, low maximal current amplitudes were detected. Intrinsic charges of the myc-epitope and charged N-terminal ends of the C-domain most probably induce intramolecular interactions. These interactions might hinder the close proximity of C-domains and N-domains, which is a prerequisite for functional ion channel configurations. The remaining basic subdomains close to TM3 and 4 were sufficient for domain complementation and functional ion channel formation. Thus, these basic subdomains forming α-helical elements or an intracellular portal represent attractants for incoming negatively charged chloride ions and interact with the phospholipids thereby stabilizing the GlyR in a conformation that allows ion channel opening.

It was previously reported that functional glycine receptors were expressed in neonatal prefrontal cortex; however, the glycine-releasing cells were unknown. We hypothesized that astrocytes might be a major glycine source, and examined the glycine release properties of astrocytes. We also hypothesized that dopamine (DA) might be a trigger for the astrocytic glycine release, as numerous DA terminals localize in the cortex. We combined two different methods to confirm the glycine release from astrocytes. Firstly, we analyzed the supernatant of astrocytes by amino acid analyzer after DA stimulation, and detect significant glycine peak. Furthermore, we utilized a patch-clamp biosensor method to confirm the glycine release from astrocytes by using GlyRα1 and Glyβ-expressing HEK293T cells, and detected significant glycine-evoked current upon DA stimulation. Thus, we clearly demonstrated that DA induces glycine release from astrocytes. Surprisingly, DA caused a functional reversal of astrocytic glycine transporter 1, an astrocytic type of glycine transporter, causing astrocytes to release glycine. Hence, astrocytes transduce pre-synaptic DA signals to glycine signals through a reversal of astrocytic glycine transporter 1 to regulate neuronal excitability. Cover Image for this issue: doi: 10.1111/jnc.13785.

Synaptic glycine receptors (GlyRs) are hetero-pentameric chloride channels composed of α and β subunits, which are activated by agonist binding at subunit interfaces. To examine the pharmacological properties of each potential agonist-binding site, we substituted residues of the GlyR α(1) subunit by the corresponding residues of the β subunit, as deduced from sequence alignment and homology modeling based on the recently published crystal structure of the glutamate-gated chloride channel GluCl. These exchange substitutions allowed us to reproduce the βα, αβ and ββ subunit interfaces present in synaptic heteromeric GlyRs by generating recombinant homomeric receptors. When the engineered α(1) GlyR mutants were expressed in Xenopus oocytes, all subunit interface combinations were found to form functional agonist-binding sites as revealed by voltage clamp recording. The ββ-binding site displayed the most distinct pharmacological profile towards a range of agonists and modulators tested, indicating that it might be selectively targeted to modulate the activity of synaptic GlyRs. The mutational approach described here should be generally applicable to heteromeric ligand-gated ion channels composed of homologous subunits and facilitate screening efforts aimed at targeting inter-subunit specific binding sites.

Glycinergic inhibitory transmission has been described in spinal cord, but rather disregarded in the brain. The spatial-temporal characterization of glycine receptors (GlyR) in the hippocampus over development is herein reported. GlyR expression increases from late embryonic stage (E18) to 7 days postnatal (P7) and decreases from P7 on. Quantitative real-time PCR showed that GlyR subunit expression changes over neuronal maturation with a preponderance of α2 and α3, over α1 and β. In immature stages, GlyR delineate the cell body of neurons at the Dentate Gyrus and Cornus Ammonis 1 and 3 (CA1/CA3) and are composed of α2 and α3 subunits. At P7, synaptic GlyRα2β can already be observed in the dendritic areas of Dentate Gyrus and of CA1/CA3. In the mature hippocampus, synaptic GlyR decrease and, although a few synaptic GlyRα1β can still be detected in the dendritic layers, extrasynaptic α2/α3-containing GlyR and somatic localized GlyRα3 are the most abundant. Our results point towards an important function of a slow tonic activation of extrasynaptic GlyR, over a fast phasic activation of synaptic GlyRα1β. We clearly show that GlyR are widely expressed in hippocampus and that their subcellular localization and subunit composition change over development.

The Cys-loop receptor family consists of nicotinic acetylcholine receptors (nAChR), glycine receptor, GABA-A and some other receptors. They fulfill a plethora of functions, whereas their malfunctioning is associated with many diseases. All three domains - extracellular ligand-binding, membrane and cytoplasmic - of these ligand-gated ion channels play important roles in the receptor assembly, delivery to the membrane surface and functional activity. In this study, we discuss the role of these domains in the assembly of the Cys-loop receptors, most comprehensively for the nAChRs. Heterologous expression and mutations of large N-terminal fragments of various subunits demonstrated their leading role in the assembly, although getting an isolated well-structured pentameric ligand-binding domain is still a problem. The long intracellular loop between transmembrane fragments M3 and M4 participates in modulating the receptor function and in clusterization of the receptor complexes because of interactions with the intracellular proteins. The transmembrane fragments play different functional roles: M2 fragments outline the channel, M4 fragments, the most remote from the channel, modulate the channel function and contact the lipid environment. The interactions of aromatic residues in the M1 and M3 fragments with those of M4 are important for the correct assembly of glycine receptor α1 subunit and for the formation of functional pentaoligomer. The role of the three receptor domains is discussed in the light of electron microscopy structure of the Torpedo nAChR, X-ray structures of agonist and antagonist complexes with the acetylcholine-binding proteins and the X-ray structures of the prokaryotic Cys-loop receptors.

J. Neurochem. (2010) 115, 1245–1255. Abstract Ligand-gated ion channels efficiently couple neurotransmitter binding to the opening of an intrinsic ion channel to generate the post-synaptic potentials that are characteristic of fast synaptic transmission. In the Cys-loop family of ligand-gated ion channels, the ligand-binding site is approximately 60 A above the channel gate. Structural modelling of related proteins and mutagenesis studies led to the hypothesis that loops 2 and 7 of the extracellular domain may couple ligand binding to receptor activation. Mutating loop 2 residues of the glycine receptor to cysteine reveals an alternating pattern of effect upon receptor function. Mutations A52C, T54C and M56C produced a threefold right-shift in EC50. In contrast, a 30-fold right-shift was seen for mutations E53C, T55C and D57C. Loop 2 conformational changes associated with ligand binding were assessed by measuring the rate of covalent modification of substituted cysteines by charged methane thiosulfonate reagents. We show for the first time state-dependent differences in the rate of reaction. A52C and T54C are more accessible in the resting state and M56C is more accessible in the activated state. These results demonstrate that loop 2 does undergo a conformational change as part of the mechanism that couples ligand binding to channel opening.

Dopamine (DA)-induced pre-synaptic inhibition of glycinergic transmission was studied from substantia gelatinosa (SG) neurons of the trigeminal nucleus pars caudalis using a conventional whole-cell patch clamp technique. The action potential-dependent glycinergic inhibitory post-synaptic currents (IPSCs) were recorded from SG neurons in the presence of 3 mM kynurenic acid and 10 μM 6-imino-3-(4-methoxyphenyl)-1(6H)-pyridazinebutanoic acid HBr (SR95531). In these conditions, bath applied DA (100 μM) reduced the amplitude of glycinergic IPSCs and increased the paired-pulse ratio, suggesting that DA acts pre-synaptically to reduce the probability of glycine release. However, the inhibitory action of DA on glycinergic IPSCs was not blocked by SCH23390 (10 μM) and spiperone (1 μM), selective D(1) - and D(2) -like receptor antagonists, respectively. In addition, either SKF38393 (100 μM), a selective D(1) -like receptor agonist, or quinpirole (100 μM), a selective D(2) -like receptor agonist, had no pre-synaptic effect on glycinergic IPSCs. The results suggest that both D(1) - and D(2) -like receptors are not involved in the DA-induced decrease in glycinergic IPSCs. On the other hand, tyramine (100 μM), one of representative trace amines, reduced the amplitude of glycinergic IPSCs and increased the paired-pulse ratio, suggesting that tyramine acts pre-synaptically to reduce the probability of glycine release. Considering that DA can activate trace amine (TA) receptors and that TA receptors are exclusively expressed on the trigeminal nucleus pars caudalis, DA might act on putative pre-synaptic TA receptors, rather than classical DA receptors, to inhibit glycinergic transmission onto SG neurons of the trigeminal nucleus pars caudalis.

Glycine transporter 1 (GLYT1) and GLYT2 are the glycine transporters in CNS. While GLYT2 is largely expressed in glycinergic neurons, GLYT1 has long been considered to be exclusively present in glial cells. There is increasing evidence that significant amounts of the 'glial' transporter also exist on neurons, particularly on pre-synaptic nerve endings of glutamatergic neurons. The functions of 'neuronal GLYT1' may be manifold and are discussed in this review. Of major interest are the interactions between neuronal GLYT1 and glutamatergic receptors of the NMDA type the activity of which is modulated not only by astrocytic GLYT1 but also by neuronal GLYT1. Pathophysiological roles and therapeutic implications of neuronal GLYT1 are emerging from recent studies with genetically modified mice, particularly with animals lacking forebrain neuron-specific GLYT1 transporters. These mutant mice exhibit promnesic phenotypes reflecting enhancement of NMDA receptor function, as it occurs following administration of GLYT1 inhibitors. Inactivation of neuronal GLYT1 in the forebrain may represent an effective therapeutic intervention for the treatment of schizophrenia.

The human neurological disorder hyperekplexia is frequently caused by recessive and dominant mutations of the glycine receptor alpha1 subunit gene, GLRA1. Dominant forms are mostly attributed to amino acid substitutions within the ion pore or adjacent loops, resulting in altered channel properties. Here, the biogenesis of glycine receptor alpha1 subunit mutants underlying recessive forms of hyperekplexia was analyzed following recombinant expression in HEK293 cells. The alpha1 mutant S231R resulted in a decrease of surface integrated protein, consistent with reduced maximal current values. Decreased maximal currents shown for the recessive alpha1 mutant I244N were associated with protein instability, rather than decreased surface integration. The recessive mutants R252H and R392H encode exchanges of arginine residues delineating the intracellular faces of transmembrane domains. After expression, the mutant R252H was virtually absent from the cell surface, consistent with non-functionality and the importance of the positive charge for membrane integration. Surface expression of R392H was highly reduced, resulting in residual chloride conductance. Independent of the site of the mutation within the alpha1 polypeptide, metabolic radiolabelling and pulse chase studies revealed a shorter half-life of the full-length alpha1 protein for all recessive mutants as compared to the wild-type. Treatment with the proteasome blocker, lactacystin, significantly increased the accumulation of alpha1 mutants in intracellular membranes. These observations indicated that the recessive alpha1 mutants are recognized by the endoplasmatic reticulum control system, and degraded via the proteasome pathway. Thus, the lack of glycinergic inhibition associated with recessive hyperekplexia may be attributed to sequestration of mutant subunits within the endoplasmatic reticulum quality control system.

cAMP is known to regulate neurotransmitter release via protein kinase A (PKA)-dependent and/or PKA-independent signal transduction pathways at a variety of central synapses. Here we report the cAMP-mediated long-lasting enhancement of glycinergic transmission in developing rat spinal substantia gelatinosa neurons. Forskolin, an adenylyl cyclase activator, elicited a long-lasting increase in the amplitude of nerve-evoked glycinergic inhibitory postsynaptic currents (IPSCs), accompanied by a long-lasting decrease in the paired-pulse ratio in immature substantia gelatinosa neurons, and this forskolin-induced increase in glycinergic IPSCs decreased with postnatal development. Forskolin also decreased the failure rate of glycinergic IPSCs evoked by minimal stimulation, and increased the frequency of glycinergic miniature IPSCs. All of these data suggest that forskolin induces the long-lasting enhancement of glycinergic transmission by increasing in the presynaptic release probability. This pre-synaptic action of forskolin was mediated by hyperpolarization and cyclic nucleotide-activated cation channels and an increase in intraterminal Ca(2+) concentration but independent of PKA. The present results suggest that cAMP-dependent signal transduction pathways represent a dynamic mechanism by which glycinergic IPSCs could potentially be modulated during postnatal development.

Ligand binding to Cys-loop receptors produces either global conformational changes that lead to activation or local conformational changes that do not. We found that the fluorescence of a fluorophore tethered to R271C in the extracellular M2 region of the alpha 1 glycine receptor increases during glycine activation but not during ivermectin activation. This prompted the hypothesis that this signal reports a glycine-mediated conformational change not essential for activation. We tested this by investigating whether the fluorescence signal depended on whether the fluorophore was attached to a glycine-free or a glycine-bound subunit. Agonist-free subunits were created by incorporating T204A and R65K mutations, which disrupted glycine binding to both (+) and (-) subunit interfaces. In heteromeric receptors comprising wild-type and R65K,T204A,R271C triple-mutant subunits, the fluorescence response exhibited a drastically reduced glycine sensitivity relative to the current response. Two conclusions can be drawn from this. First, because the labeled glycine-free subunits were activated by glycine binding to neighboring wild-type subunits, our results provide evidence for a cooperative activation mechanism. However, because the fluorescent label on glycine-free subunits does not reflect movements at the channel gate, we conclude that glycine binding also produces a local non-concerted conformational change that is not essential for receptor activation.

The present studies used increased atmospheric pressure in place of a traditional pharmacological antagonist to probe the molecular sites and mechanisms of ethanol action in glycine receptors (GlyRs). Based on previous studies, we tested the hypothesis that physical-chemical properties at position 52 in extracellular domain Loop 2 of alpha1GlyRs, or the homologous alpha2GlyR position 59, determine sensitivity to ethanol and pressure antagonism of ethanol. Pressure antagonized ethanol in alpha1GlyRs that contain a non-polar residue at position 52, but did not antagonize ethanol in receptors with a polar residue at this position. Ethanol sensitivity in receptors with polar substitutions at position 52 was significantly lower than GlyRs with non-polar residues at this position. The alpha2T59A mutation switched sensitivity to ethanol and pressure antagonism in the WTalpha2GlyR, thereby making it alpha1-like. Collectively, these findings indicate that (i) polarity at position 52 plays a key role in determining sensitivity to ethanol and pressure antagonism of ethanol; (ii) the extracellular domain in alpha1- and alpha2GlyRs is a target for ethanol action and antagonism and (iii) there is structural-functional homology across subunits in Loop 2 of GlyRs with respect to their roles in determining sensitivity to ethanol and pressure antagonism of ethanol. These findings should help in the development of pharmacological agents that antagonize ethanol.

The beta-amino acid, taurine, is a full agonist of the human glycine receptor alpha1 subunit when recombinantly expressed in a mammalian (HEK293) cell line, but a partial agonist of the same receptor when expressed in Xenopus oocytes. Several residues in the Ala101-Thr112 domain have previously been identified as determinants of beta-amino acid binding and gating mechanisms in Xenopus oocyte-expressed receptors. The present study used the substituted cysteine accessibility method to investigate the role of this domain in controlling taurine-specific binding and gating mechanisms of glycine receptors recombinantly expressed in mammalian cells. Asn102 and Glu103 are identified as taurine and glycine binding sites, whereas Ala101 is eliminated as a possible binding site. The N102C mutation also abolished the antagonistic actions of taurine, indicating that this site does not discriminate between the putative agonist- and antagonist-bound conformations of beta-amino acids. The effects of mutations from Lys104-Thr112 indicate that the mechanism by which this domain controls beta-amino acid-specific binding and gating processes differs substantially depending on whether the receptor is expressed in mammalian cells or Xenopus oocytes. Thr112 is the only domain element in mammalian cell-expressed GlyRs which was demonstrated to discriminate between glycine and taurine.

Poly(A)+ mRNA was isolated from cultured mouse cerebellar granule cells and injected into Xenopus oocytes. This led to the expression of receptors that evoked large membrane currents in response to glycine. Current-responses were also obtained after application of beta-alanine and taurine, but these were very low relative to that of glycine (maximal beta-alanine and taurine responses were 8 and 3% of that of glycine, respectively). The role of glycine receptors on K(+)-evoked transmitter release in cultured cerebellar granule cells was also assayed. Release of preloaded D-[3H]aspartate evoked by 40 mM K+ was dose dependently inhibited by glycine, and the concentration producing half-maximal inhibition was 50 microM. Taurine, beta-alanine, and the specific GABAA receptor agonist isoguvacine also inhibited K(+)-evoked release, and the maximal inhibition was similar for all agonists (approximately 40%). The EC50 value was 200 microM for taurine, 70 microM for beta-alanine, and 4 microM for isoguvacine. Bicuculline (150 microM) antagonized the inhibitory effect of isoguvacine (150 microM) but not that of glycine (1 mM). In contrast, strychnine (20 microM) antagonized the inhibitory effect of glycine (1 mM) but not that of isoguvacine (150 microM). The pharmacology of the responses to beta-alanine and taurine showed that these agonists activate both glycine and GABAA receptors. The results indicate that cultured cerebellar granule cells translate the gene for the glycine receptor and that activation of glycine receptors produces neuronal inhibition.

We report the isolation of a full-length clone from a Drosophila melanogaster head cDNA library that encodes a 614-residue polypeptide that exhibits all of the features of a ligand-gated chloride-channel/receptor subunit. This polypeptide, which has been named GRD (denoting that the polypeptide is a GABAA and glycine receptor-like subunit of Drosophila), displays between 33 and 44% identity to vertebrate GABAA and glycine receptor subunits and 32–37% identity to the GABAA receptor-like polypeptides from Drosophila and Lymnaea. It is interesting that the large amino-terminal, presumed extracellular domain of the GRD protein contains an insertion, between the dicysteine loop and the first putative membrane-spanning domain, of 75 amino acids that is not found in any other ligand-gated chloride-channel subunit. Analysis of cDNA and genomic DMA reveals that these residues are encoded by an extension of an exon that is equivalent to exon 6 of vertebrate GABAA and glycine receptor genes. The gene (named Grd) that encodes the Drosophila polypeptide has been mapped, by in situ hybridization, to position 75A on the left arm of chromosome 3.

Ginkgolides, active constituents of Ginkgo biloba extracts, potently block the glycine receptor chloride channel (GlyR). Ginkgolides A, B, C and J are structurally similar, varying only by the presence or absence of oxygens at their R1 and R2 positions. The aim of this study was to understand how variable ginkgolide groups bind to pore-lining 2' and 6' residues in the alpha1 GlyR. Ginkgolide potency was not affected by G2'A or G2'S mutations, suggesting 2' residues are not important for ginkgolide coordination. Analysis of the alpha1(T6'S) GlyR suggests that ginkgolides bind to this receptor via hydrogen bonds between T6'S and ginkgolide R1 hydroxyls. The abolition of block by the T6'A and T6'V mutations but not by the T6'S mutation implies the existence a second transmembrane domain alpha-helical kink formed by hydrogen bonding between 6' threonine and serine sidechains and backbone carbonyl oxygens. We also found that ginkgolide A binds in different orientations in the closed and open states of a mutant GlyR, possibly reflecting its enhanced flexibility relative to other ginkgolides. Together these results indicate that small variations in ginkgolide structure or pore structure can lead to drastic potency variations. This property may be exploited to create improved pharmacological probes for discriminating among anionic Cys-loop receptor isoforms with 6' structural variations.

We have previously shown that the glutamate receptor interacting protein 1 (GRIP1) splice forms GRIP1a/b and GRIP1c4–7 are present at the GABAergic post-synaptic complex. Nevertheless, the role that these GRIP1 protein isoforms play at the GABAergic post-synaptic complex is not known. We are now showing that GRIP1c4–7 and GRIP1a/b interact with gephyrin, the main post-synaptic scaffold protein of GABAergic and glycinergic synapses. Gephyrin coprecipitates with GRIP1c4–7 or GRIP1a/b from rat brain extracts and from extracts of human embryonic kidney 293 cells that have been cotransfected with gephyrin and one of the GRIP1 protein isoforms. Moreover, purified gephyrin binds to purified GRIP1c4–7 or GRIP1a/b, indicating that gephyrin directly interacts with the common region of these GRIP1 proteins, which includes PDZ domains 4–7. An engineered deletion construct of GRIP1a/b (GRIP1a4–7), which both contains the aforementioned common region and binds to gephyrin, targets to the post-synaptic GABAergic complex of transfected cultured hippocampal neurons. In these hippocampal cultures, endogenous gephyrin colocalizes with endogenous GRIP1c4–7 and GRIP1a/b in over 90% of the GABAergic synapses. Double-labeling electron microscopy immunogold reveals that in the rat brain GRIP1c4–7 and GRIP1a/b colocalize with gephyrin at the post-synaptic complex of individual synapses. These results indicate that GRIP1c4–7 and GRIP1a/b colocalize and interact with gephyrin at the GABAergic post-synaptic complex and suggest that this interaction plays a role in GABAergic synaptic function.

Studies on hippocampal glycine release are extremely rare. We here investigated release from mouse hippocampus glycinergic terminals selectively pre-labelled with [3H]glycine through transporters of the GLYT2 type. Purified synaptosomes were incubated with [3H]glycine in the presence of the GLYT1 blocker NFPS to abolish uptake (∼ 30%) through GLYT1. The non-GLYT1-mediated uptake was entirely sensitive to the GLYT2 blocker Org25543. Depolarization during superfusion with high-K+ (15–50 mmol/L) provoked overflows totally dependent on external Ca2+, whereas in the spinal cord the 35 or 50 mmol/L KCl-evoked overflow (higher than that in hippocampus) was only partly dependent on extraterminal Ca2+. In the hippocampus, the Ca2+-dependent 4-aminopyridine (1 mmol/L)-evoked overflow was five-fold lower than that in spinal cord. The component of the 10 μmol/L veratridine-induced overflow dependent on external Ca2+ was higher in the hippocampus than that in spinal cord, although the total overflow in the hippocampus was only half of that in the spinal cord. Part of the veratridine-evoked hippocampal overflow occurred by GLYT2 reversal and part by bafilomycin A1-sensitive exocytosis dependent on cytosolic Ca2+ generated through the mitochondrial Na+/Ca2+ exchanger. As glycine sites on NMDA receptors are normally not saturated, understanding mechanisms of glycine release should facilitate pharmacological modulation of NMDA receptor function.

The neuronal glycine transporter GLYT2 is a plasma membrane protein that removes the neurotransmitter glycine from the synaptic cleft, thereby aiding the pre-synaptic terminal reloading and the termination of the glycinergic signal. Missense mutations in the gene encoding GLYT2 (SLC6A5) cause hyperekplexia in humans. The activity of GLYT2 seems to be highly regulated. In this report, we demonstrate that GLYT2 is associated with membrane rafts in the plasma membrane of brainstem terminals and neurons. The transporter is localized to Triton X-100-insoluble light synaptosomal membranes together with flotillin-1, a marker protein for membrane rafts, in a methyl-β-cyclodextrin (MβCD)-sensitive manner. In brainstem primary neurons, the GLYT2 punctuate pattern visualized by confocal microscopy was modified by cholesterol depletion with MβCD, unlike other non-raft neuronal markers. GLYT2-associated gold particles were observed by electron microscopy on purified rafts from brainstem synaptosomes. Furthermore, either in brainstem terminals and cultured neurons, the pharmacological reduction of the levels of raft components, cholesterol and sphingomyelin, impairs both the association of GLYT2 with membrane rafts and its transport activity. Thus, GLYT2 may require membrane raft location for optimal function, and therefore the lipid environment may constitute a new mechanism to modulate GLYT2.

The glycine receptor is a member of the Cys-loop, ligand-gated ion channel family and is responsible for inhibition in the CNS. We examined the orientation of amino acids I229 in transmembrane 1 (TM1) and A288 in TM3, which are both critical for alcohol and volatile anesthetic action. We mutated these two amino acids to cysteines either singly or in double mutants and expressed the receptors in Xenopus laevis oocytes. We tested whether disulfide bonds could form between A288C in TM3 paired with M227C, Y228C, I229C, or S231C in TM1. Application of cross-linking (mercuric chloride) or oxidizing (iodine) agents had no significant effect on the glycine response of wild-type receptors or the single mutants. In contrast, the glycine response of the I229C/A288C double mutant was diminished after application of either mercuric chloride or iodine only in the presence of glycine, indicating that channel gating causes I229C and A288C to fluctuate to be within 6 A apart and form a disulfide bond. Molecular modeling was used to thread the glycine receptor sequence onto a nicotinic acetylcholine receptor template, further demonstrating that I229 and A288 are near-neighbors that can cross-link and providing evidence that these residues contribute to a single binding cavity.

Picrotoxin, an antagonist of structurally-rated GABA(A) receptors (GABA(A)Rs) and glycine receptors (GlyRs), is an equimolar mixture of picrotoxinin (PTXININ) and picrotin (PTN). These compounds share a common structure except that PTN contains a slightly larger dimethylmethanol in place of the PTXININ isopropenyl group. Although the homomeric alpha1 GlyR is equally sensitive to both compounds, we show here that homomeric alpha2 and alpha3 GlyRs, like most GABA(A)Rs, are selectively inhibited by PTXININ. As conservative mutations to pore-lining 6' threonines equally affect the sensitivity of the alpha1 GlyR to both compounds, we conclude that PTXININ and PTN bind to 6' threonines by hydrogen bonding with exocyclic oxygens common to both molecules. In contrast, substitution of the 2' pore-lining glycine by serine selectively reduces PTN sensitivity, whereas the introduction of 2' alanines selectively increases PTXININ sensitivity. These results define the orientation of PTXININ and PTN binding in the alpha1 GlyR pore and allow us to conclude that the relatively reduced sensitivity of PTN at GABA(A)Rs and alpha2 and alpha3 GlyRs is due predominantly to its larger size and reduced ability to form hydrophobic interactions with 2' alanines.

Considerable evidence indicates that ethanol acts on specific residues in the transmembrane domains of glycine receptors (GlyRs). In this study, we tested the hypothesis that the extracellular domain is also a target for ethanol action by investigating the effect of cysteine substitutions at positions 52 (extracellular domain) and 267 (transmembrane domain) on responses to n-alcohols and propyl methanethiosulfonate (PMTS) in α1GlyRs expressed in Xenopus oocytes. In support of the hypothesis: (i) The A52C mutation changed ethanol sensitivity compared to WT GlyRs; (ii) PMTS produced irreversible alcohol-like potentiation in A52C GlyRs; and (iii) PMTS binding reduced the n-chain alcohol cutoff in A52C GlyRs. Further studies used PMTS binding to cysteines at positions 52 or 267 to block ethanol action at one site in order to determine its effect at other site(s). In these situations, ethanol caused negative modulation when acting at position 52 and positive modulation when acting at position 267. Collectively, these findings parallel the evidence that established the TM domain as a target for ethanol, suggest that positions 52 and 267 are part of the same alcohol pocket and indicate that the net effect of ethanol on GlyR function reflects the summation of its positive and negative modulatory effects on different targets.

The 5-hydroxytryptamine type-3 receptor antagonist tropisetron is in clinical use as an anti-emetic drug. This compound also exerts both potentiating and inhibitory effects on the glycine receptor chloride channel. The inhibitory effects occur at micromolar concentrations, whereas the potentiating effects are shown here to occur at femtomolar concentrations at the homomeric alpha1 receptor. Potentiation occurred only when tropisetron was applied in the presence of glycine. We also sought to identify molecular determinants of tropisetron inhibition at the alpha1 glycine receptor by serially mutating residues located in or near known ligand-binding sites. We discovered that conservative mutations to N102 ablated tropisetron inhibition without affecting the magnitude or sensitivity of tropisetron potentiation. Several lines of evidence, including a structure-activity analysis of tropisetron, atropine and SB203186, suggest that N102 may bind to the tropisetron tropane nitrogen via H-bonding. Mutation of the N125 residue in the beta subunit, which corresponds to N102 in the alpha1 subunit, had little effect on tropisetron inhibitory potency. These results show that N102 is required for tropisetron inhibition but not potentiation and that inhibitory tropisetron binds in different orientations at different subunit interfaces. To our knowledge, tropisetron is the most exquisitely sensitive modulator yet identified for a cys-loop receptor.

The inhibitory glycine receptor (GlyR), a cys-loop ion channel receptor, mediates rapid synaptic inhibition in spinal cord, brainstem and higher centres of the mammalian central nervous system. Here, modulation of GlyR function by glucose and fructose was examined in recombinant alpha1 and alpha1/beta GlyRs using patch-clamp methods. Glucose was a positive modulator of the receptor, reducing the average EC50 for glycine up to 4.5-fold. Glucose reduced cell-to-cell variability of glycine-mediated currents by stabilizing receptors with low EC50. Pre-incubation with sugars for several hours also produced augmentation of current responses that persisted after sugar removal. Potentiation by sugars was most significant in the range between 5 and 20 mM, with EC50 values ~ 10 mM, i.e. at physiological levels. Addition of glucose had no significant influence on responses mediated by the other GlyR agonists like taurine, β-alanine or ivermectin, indicating that glucose specifically augmented glycine receptor-mediated responses, and did not act through indirect metabolic effects. Receptor modulation by glucose may account for differences in constants reported in the literature and may be clinically relevant for disorders with elevated blood glucose levels. Glucose and related sugars are essential metabolites. We identified glucose and fructose as positive modulators of the human inhibitory glycine receptor, a neuronal ligand-gated ion channel. Receptor-mediated currents were enhanced at physiological concentrations (~ 10 mM of sugar). Direct modulation of a synaptic receptor by glucose is relevant in clinical cases of elevated blood glucose, and may be considered in experimental protocols.

In this study, we describe a novel form of anti-homeostatic plasticity produced after culturing spinal neurons with strychnine, but not bicuculline or 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). Strychnine caused a large increase in network excitability, detected as spontaneous synaptic currents and calcium transients. The calcium transients were associated with action potential firing and activation of gamma-aminobutyric acid (GABA(A)) and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors as they were blocked by tetrodotoxin (TTX), bicuculline, and CNQX. After chronic blockade of glycine receptors (GlyRs), the frequency of synaptic transmission showed a significant enhancement demonstrating the phenomenon of anti-homeostatic plasticity. Spontaneous inhibitory glycinergic currents in treated cells showed a fourfold increase in frequency (from 0.55 to 2.4 Hz) and a 184% increase in average peak amplitude compared with control. Furthermore, the augmentation in excitability accelerated the decay time constant of miniature inhibitory post-synaptic currents. Strychnine caused an increase in GlyR current density, without changes in the apparent affinity. These findings support the idea of a post-synaptic action that partly explains the increase in synaptic transmission. This phenomenon of synaptic plasticity was blocked by TTX, an antibody against brain-derived neurotrophic factor (BDNF) and K252a suggesting the involvement of the neuronal activity-dependent BDNF-TrkB signaling pathway. These results show that the properties of GlyRs are regulated by the degree of neuronal activity in the developing network.

[3H]Strychnine binding to rat pons + medulla membranes was used as a measure of glycine receptors or glycine receptor-coupled chloride channels in vitro. A series of compounds structurally related to 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP), which previously were shown to antagonize glycine responses in cat spinal cord, inhibited [3H]strychnine binding in micromolar concentrations. The most potent of these glycine antagonists, 5,6,7,8-tetrahydro-4H-isoxazolo[3,4-d]azepin-3-ol (iso-THAZ), was also the most potent inhibitor of [3H]strychnine binding, with a Ki of 1,400 nM. The Ki value for strychnine was 7.0 nM, whereas the Ki value for the mixed gamma-aminobutyric acid (GABA)/glycine antagonist 3 alpha-hydroxy-16-imino-5 beta-17-aza-androstan-11-one (RU 5135) was only 4.6 nM. Sodium chloride (1,000 mM) enhanced the affinity of strychnine, brucine, isostrychnine, and the nonselective GABA antagonist pitrazepin for [3H]strychnine binding sites, whereas the affinities of glycine, beta-alanine, and taurine were reduced. These sodium chloride shifts, however, were not predictive of antagonist or agonist properties, since the sodium chloride shift for the glycine antagonist iso-THAZ and of the other THIP-related antagonists were similar to those of the glycine-like agonists. The various sodium chloride shifts show that different groups of ligands bind to glycine receptor sites in different ways.

Ginkgolides are potent blockers of the glycine receptor Cl- channel (GlyR) pore. We sought to identify their binding sites by comparing the effects of ginkgolides A, B and C and bilobalide on alpha1, alpha2, alpha1beta and alpha2beta GlyRs. Bilobalide sensitivity was drastically reduced by incorporation of the beta subunit. In contrast, the sensitivities to ginkgolides B and C were enhanced by beta subunit expression. However, ginkgolide A sensitivity was increased in the alpha2beta GlyR relative to the alpha2 GlyR but not in the alpha1beta GlyR relative to the alpha1 GlyR. We hypothesised that the subunit-specific differences were mediated by residue differences at the second transmembrane domain 2' and 6' pore-lining positions. The increased ginkgolide A sensitivity of the alpha2beta GlyR was transferred to the alpha1beta GlyR by the G2'A (alpha1 to alpha2 subunit) substitution. In addition, the alpha1 subunit T6'F mutation abolished inhibition by all ginkgolides. As the ginkgolides share closely related structures, their molecular interactions with pore-lining residues were amenable to mutant cycle analysis. This identified an interaction between the variable R2 position of the ginkgolides and the 2' residues of both alpha1 and beta subunits. These findings provide strong evidence for ginkgolides binding at the 2' pore-lining position.

Functional activation of NMDA receptors requires co-activation of glutamate- and glycine-binding sites. D-serine is considered to be an endogenous ligand for the glycine site of NMDA receptors. Using a combination of a rat formalin-induced conditioned place avoidance (F-CPA) behavioral model and whole-cell patch-clamp recording in rostral anterior cingulate cortex (rACC) slices, we examined the effects of d-amino acid oxidase (DAAO), an endogenous D-serine-degrading enzyme, and 7-chlorokynurenate (7Cl-KYNA), an antagonist of the glycine site of NMDA receptors, on pain-related aversion. Degradation of endogenous D-serine with DAAO, or selective blockade of the glycine site of NMDA receptors by 7Cl-KYNA, effectively inhibited NMDA-evoked currents in rACC slices. Intra-rACC injection of DAAO (0.1 U) and 7Cl-KYNA (2 and 0.2 mM, 0.6 microL per side) 20 min before F-CPA conditioning greatly attenuated F-CPA scores, but did not affect formalin-induced acute nociceptive behaviors and electric foot shock-induced conditioned place avoidance. This study reveals for the first time that endogenous D-serine plays a critical role in pain-related aversion by activating the glycine site of NMDA receptors in the rACC. Furthermore, these results extend our hypothesis that activation of NMDA receptors in the rACC is necessary for the acquisition of specific pain-related negative emotion. Thus a new and promising strategy for the prevention of chronic pain-induced emotional disturbance might be raised.

Protein kinase C (PKC) phosphorylates the NR1 and NR2A subunits of NMDARs at consensus sites located within their intracellular C-terminal tails. However, the functional consequences of these biochemical events are not well understood. In HEK293 cells expressing NR1/NR2A, activation of endogenous PKC by 4β-phorbol 12-myristate 13-acetate (PMA) increased NMDAR desensitization as evidenced by a reduced steady-state current without any change in peak. The effects of PMA on NMDAR-mediated responses were prevented by specific PKC inhibitors and were not mimicked by an inactive enantiomer of PMA. The effects of PMA were preserved despite mutagenesis of the major PKC sites on the NR1 subunit (S889A, S890A, S896A and S897A) or removal of the entire NR1 C-terminal tail (NR1stop838). When co-expressing NR1stop838/NR2A the effects of PMA could only be observed with agonist concentrations sufficient to induce glycine-insensitive desensitization. Moreover, the effects of PMA were observed in receptors composed of NR1/NR2A and NR1/NR2B, but not NR1/NR2C, a subunit combination in which desensitization is absent. The NR2 subunit dependence suggested that the actions of PMA might require specific PKC sites previously identified within NR2A. However, a C-terminal truncated form of NR2A (NR2Astop905) remained responsive to PMA. We conclude that activation of PKC increases NMDAR glycine-insensitive desensitization independently of previously identified sites located within the NR1 C-terminus and distal segment of the NR2A C-terminus.

Glycine residues can introduce flexibility in proteins, give rise to turns and breaks in secondary structure and are key components of some nucleotide binding motifs. In the P2X receptor extracellular ATP binding domain, 11 glycine residues are completely conserved and an additional five are conserved in at least five of the seven family members. We have mutated individual conserved glycine residues and determined their effect on the ATP sensitivity and time-course of P2X1 receptors expressed in Xenopus oocytes. In the majority of cases, replacement by alanine had no or a less than 3-fold effect on ATP sensitivity and time-course of responses. G71A resulted in a 6-fold decrease in ATP potency and ATP (10 mM) failed to evoke functional responses from G96A, G250A and G301A mutant receptors. However, proline or cysteine could substitute for glycine at positions 96 and 301, giving receptors that were essentially normal. At glycine 250 substitution by serine gave functional responses to ATP with no effect on ATP sensitivity but a reduction in peak amplitude; in contrast, functional responses were not recorded when glycine 250 was replaced by the amino acids alanine, cysteine, aspartate, phenylalanine, isoleucine, lysine, proline or asparagine. These results suggest that glycine 250 plays an important role in determining the function of P2X receptors.

The inferior colliculus (IC) is a major center of integration in the ascending as well as descending auditory pathways, where both excitatory and inhibitory amino acid neurotransmitters play a key role. When normal input to the auditory system is decreased, the balance between excitation and inhibition in the IC is disturbed. We examined global changes in gene expression in the rat IC 3 and 21 days following bilateral deafening, using Affymetrix GeneChip arrays and focused our analysis on changes in expression of neurotransmission-related genes. Over 1400 probe sets in the Affymetrix Rat Genome U34A Array were identified as genes that were differentially expressed. These genes encoded proteins previously reported to change as a consequence of deafness, such as calbindin, as well as proteins not previously reported to be modulated by deafness, such as clathrin. A subset of 19 differentially expressed genes was further examined using quantitative RT–PCR at 3, 21 and 90 days following deafness. These included several GABA, glycine, glutamate receptor and neuropeptide-related genes. Expression of genes for GABA-A receptor subunits β2, β3, and γ2, plus ionotropic glutamate receptor subunits AMPA 2, AMPA 3, and kainate 2, increased at all three times. Expression of glycine receptor α1 initially declined and then later increased, while α2 increased sharply at 21 days. Glycine receptor α3 increased between 3 and 21 days, but decreased at 90 days. Of the neuropeptide-related genes tested with qRT–PCR, tyrosine hydroxylase decreased approximately 50% at all times tested. Serotonin receptor 2C increased at 3, 21, and 90 days. The 5B serotonin receptor decreased at 3 and 21 days and returned to normal by 90 days. Of the genes tested with qRT–PCR, only glycine receptor α2 and serotonin receptor 5B returned to normal levels of expression at 90 days. Changes in GABA receptor β3, GABA receptor γ2, glutamate receptor 2/3, enkephalin, and tyrosine hydroxylase were further confirmed using immunocytochemistry.

Contact points between transmembrane segments (TMs) two and three of the glycine receptor are undefined and may play an important role in channel gating. We tested whether two amino acids in TM2 (S267) and TM3 (A288), known to be critical for alcohol and volatile anesthetic action, could cross-link by mutating both to cysteines and expressing the receptors in Xenopus laevis oocytes. In contrast with the wild-type receptor and single cysteine mutants, the S267C/A288C double mutant displayed unusual responses, including a tonic leak activity that was closed by strychnine and a run-down of the response upon repeated applications of glycine. We hypothesized that these characteristics were due to cross-linking of the two cysteines on opposing faces of these adjacent, alpha helical TMs. This would alter the movement of these two regions required for normal gating. To test this hypothesis, we used dithiothreitol to reduce the putative S267C-A288C disulfide bond. Reduction abolished the leak current and provided normal responses to glycine. Subsequent application of the cross-linking agent mercuric chloride caused the initial characteristics to return. These data demonstrate that S267 and A288 are near-neighbors and provide insight towards the location and role of the TM2-TM3 interface in ligand-gated ion channels.

The current study used an ethanol antagonist, increased atmospheric pressure, to test the hypothesis that ethanol acts on multiple sites in glycine receptors (GlyRs). The effects of 12 times normal atmospheric pressure of helium-oxygen gas (pressure) on ethanol-induced potentiation of GlyR function in Xenopus oocytes expressing human alpha1, alpha2 or the mutant alpha1(A52S) GlyRs were measured using two-electrode voltage clamp. Pressure reversibly antagonized potentiation of glycine in alpha1 GlyR by 40-200 mm ethanol, but did not antagonize 10 and 25 mm ethanol in the same oocytes. In contrast, pressure did not significantly affect potentiation of glycine by 25-100 mm ethanol in alpha2 GlyRs, nor did pressure alter ethanol response in the A52S mutant. Pressure did not affect baseline receptor function or response to glycine in the absence of ethanol. These findings provide the first direct evidence for multiple sites of ethanol action in GlyRs. The sites can be differentiated on the basis of ethanol concentration, subunit and structural composition and sensitivities to pressure antagonism of ethanol. Parallel studies with butanol support this conclusion. The mutant alpha1(A52S) GlyR findings suggest that increased attention should be focused on the amino terminus as a potential target for ethanol action.

This study investigated the role of beta subunits in the activation of alphabeta heteromeric glycine receptor (GlyR) chloride channels recombinantly expressed in HEK293 cells. The approach involved incorporating mutations into corresponding positions in alpha and beta subunits and comparing their effects on receptor function. Although cysteine-substitution mutations to residues in the N-terminal half of the alpha subunit M2-M3 loop dramatically impaired the gating efficacy, the same mutations exerted little effect when incorporated into corresponding positions of the beta subunit. Furthermore, although the alpha subunit M2-M3 loop cysteines were modified by a cysteine-specific reagent, the corresponding beta subunit cysteines showed no evidence of reactivity. These observations suggest structural or functional differences between alpha and beta subunit M2-M3 loops. In addition, a threonine-->leucine mutation at the 9' position in the beta subunit M2 pore-lining domain dramatically increased the glycine sensitivity. By analogy with the effects of the same mutation in other ligand-gated ion channels, it was concluded that the mutation affected the GlyR activation mechanism. This supports the idea that the GlyR beta subunit is involved in receptor gating. In conclusion, this study demonstrates that beta subunits contribute to the activation of the GlyR, but that their involvement in this process is significantly different to that of the alpha subunit.

The glycine transporter 2 (GlyT2) belongs to the family of Na+/CL–-dependent plasma membrane transporters and is localized on the presynaptic terminals of glycinergic neurons. GlyT2 differs from other family members by its extended N-terminal cytoplasmic region. We report that activation of a Ca2+-dependent protease, most likely calpain, in spinal cord synaptosomes or cultured spinal cord neurons, results in partial proteolysis of GlyT2. Regions sensitive to calpain cleavage in vivo are located in the N-terminal and, to a lesser extent, C-terminal regions of the transporter protein. Incubation of a GlyT2 N-terminal fusion protein with spinal cord extract in the presence of calcium followed by protein sequence analysis localized the major N-terminal cleavage site after methionine 156, with a second cleavage site being situated after glycine 164. Interestingly, the size of the N-terminally truncated GlyT2 protein (70 kDa) is similar to that of most other transporter family members, and truncated GlyT2 displayed full transport activity upon expression in HEK293 cells. Our data suggest that Ca2+-triggered proteolysis may contribute to the regulation of GlyT2 trafficking and/or function in the neuronal plasma membrane.

Although glycine has been assumed to be the sole endogenous coagonist at NMDA-associated glycine receptors, recent descriptions of endogenous D-serine in the brain indicate that this assumption is probably not valid. D-Serine is a stereospecific agonist of the NMDA-associated glycine receptor, with an affinity equal to or greater than that of glycine but with no affinity for the strychnine-sensitive glycine receptor. In the current studies, we assessed the levels and metabolic sources of D-serine in rat neocortical synaptosomal preparations. Previous studies have demonstrated that CNS serine and glycine are synthesized de novo primarily via a phosphorylated pathway, originating with the glycolytic intermediate phosphoglycerate. The rate-limiting step in the synthesis of serine is the hydrolysis of phosphoserine by phosphoserine phosphatase (EC 3.1.3.3). In synaptosomal preparations we have demonstrated high endogenous levels of D-serine and the uptake of L-phosphoserine along with its hydrolysis to both L-serine and D-serine, which are preferentially released into the medium. Experiments with both intact and lysed synaptosomal preparations demonstrated hydrolysis of D-phosphoserine to only D-serine and inhibition of hydrolysis by the phosphoserine phosphatase inhibitor 2-amino-3-phosphonopropionic acid (AP3). The lack of stereospecificity for synaptosomal hydrolysis of phosphoserine and the inhibitory actions of AP3 are consistent with the presence of phosphoserine phosphatase in synaptosomes and further indicate that epimerization of serine can occur during or subsequent to the hydrolysis of L-phosphoserine but not D-phosphoserine. In conclusion, these studies demonstrate that phosphoserine phosphatase may be an important enzyme in regulating the steady-state levels of D-serine in neocortical synaptosomes.