Deletion of the glutamate carboxypeptidase II gene in mice reveals a second enzyme activity that hydrolyzes N-acetylaspartylglutamate

Glutamate carboxypeptidase II (GCPII, EC 3.14.17.21) is a membrane-bound enzyme found on the extracellular face ofglia. The gene for this enzyme is designated FOLH1 in humans and Folh1 in mice. This enzyme has been proposed to be responsible for inactivation of the neurotransmitter N-acetylaspartylglutamate (NAAG) following synaptic release. Mice harboring a disruption of the gene for GCPII/Folh1 were generated by inserting into the genome a targeting cassette in which the intron-exon boundary sequences of exons 1 and 2 were removed and stop codons were inserted in exons 1 and 2. Messenger RNA for GCPII was not detected by northern blotting or RT-PCR analysis of RNA from the brains of -/- mutant mice nor was GCPII protein detected on western blots of this tissue. These GCPII null mutant mice developed normally to adulthood and exhibited a normal range of neurologic responses and behaviors including mating, open field activity and retention of position in rotorod tests. No significant differences were observed among responses of wild type, heterozygous mutant and homozygous mutant mice on tail flick and hot plate latency tests. Glutamate, NAAG and mRNA for metabotropic glutamate receptor type 3 levels were not significantly altered in response to the deletion of glutamate carboxypeptidase II. A novel membrane-bound NAAG peptidase activity was discovered in brain, spinal cord and kidney of the GCPII knock out mice. The kinetic values for brain NAAG peptidase activity in the wild type and GCPII nullmutant were Vmax = 45 and 3 pmol/mg/min and Km = 2650 nm and 2494 nm, respectively. With the exception of magnesium and copper, this novel peptidase activity had a similar requirement for metal ions as GCPII. Two potent inhibitors of GCPII, 4,4'-phosphinicobis-(butane-1,3 dicarboxilic acid) (FN6) and 2-(phosphonomethyl)pentanedioic acid (2-PMPA) inhibited the residual activity. The IC50 value for 2-PMPA was about 1 nm for wild-type brain membrane NAAG peptidase activity consistent with its activity against cloned ratand human GCPII, and 88 nm for the activity in brain membranes of the null mutants.     

Biosynthesis of NAAG by an enzyme-mediated process in rat central nervous system neurons and glia

The peptide transmitter N-acetylaspartylglutamate (NAAG) is present in millimolar concentrations in mammalian spinal cord. Data from the rat peripheral nervous system suggest that this peptide is synthesized enzymatically, a process that would be unique for mammalian neuropeptides. To test this hypothesis in the mammalian CNS, rat spinal cords were acutely isolated and used to study the incorporation of radiolabeled amino acids into NAAG. Consistent with the action of a NAAG synthetase, inhibition of protein synthesis did not affect radiolabel incorporation into NAAG. Depolarization of spinal cords stimulated incorporation of radiolabel. Biosynthesis of NAAG by cortical astrocytes in cell culture was demonstrated by tracing incorporation of [3H]-glutamate by astrocytes. In the first test of the hypothesis that NAA is an immediate precursor in NAAG biosynthesis, [3H]-NAA was incorporated into NAAG by isolated spinal cords and by cell cultures of cortical astrocytes. Data from cerebellar neurons and glia in primary culture confirmed the predominance of neuronal synthesis and glial uptake of NAA, leading to the hypothesis that while neurons synthesize NAA for NAAG biosynthesis, glia may take it up from the extracellular space. However, cortical astrocytes in serum-free low-density cell culture incorporated [3H]-aspartate into NAAG, a result indicating that under some conditions these cells may also synthesize NAA. Pre-incubation of isolated spinal cords and cultures of rat cortical astrocytes with unlabeled NAA increased [3H]-glutamate incorporation into NAAG. In contrast, [3H]-glutamine incorporation in spinal cord was not stimulated by unlabeled NAA. These results are consistent with the glutamate-glutamine cycle greatly favoring uptake of glutamine into neurons and glutamate by glia and suggest that NAA availability may be rate-limiting in the synthesis of NAAG by glia under some conditions.     

NAAG peptidase inhibitor increases dialysate NAAG and reduces glutamate, aspartate and GABA levels in the dorsal hippocampus following fluid percussion injury in the rat

Traumatic brain injury (TBI) produces a rapid and excessive elevation in extracellular glutamate that induces excitotoxic brain cell death. The peptide neurotransmitter N-acetylaspartylglutamate (NAAG) is reported to suppress neurotransmitter release through selective activation of presynaptic group II metabotropic glutamate receptors. Therefore, strategies to elevate levels of NAAG following brain injury could reduce excessive glutamate release associated with TBI. We hypothesized that the NAAG peptidase inhibitor, ZJ-43 would elevate extracellular NAAG levels and reduce extracellular levels of amino acid neurotransmitters following TBI by a group II metabotropic glutamate receptor (mGluR)-mediated mechanism. Dialysate levels of NAAG, glutamate, aspartate and GABA from the dorsal hippocampus were elevated after TBI as measured by in vivo microdialysis. Dialysate levels of NAAG were higher and remained elevated in the ZJ-43 treated group (50 mg/kg, i.p.) compared with control. ZJ-43 treatment also reduced the rise of dialysate glutamate, aspartate, and GABA levels. Co-administration of the group II mGluR antagonist, LY341495 (1 mg/kg, i.p.) partially blocked the effects of ZJ-43 on dialysate glutamate and GABA, suggesting that NAAG effects are mediated through mGluR activation. The results are consistent with the hypothesis that inhibition of NAAG peptidase may reduce excitotoxic events associated with TBI.     

NAAG peptidase inhibition reduces locomotor activity and some stereotypes in the PCP model of schizophrenia via group II mGluR

Phencyclidine (PCP) administration elicits positive and negative symptoms that resemble those of schizophrenia and is widely accepted as a model for the study of this human disorder. Group II metabotropic glutamate receptor (mGluR) agonists have been reported to reduce the behavioral and neurochemical effects of PCP. The peptide neurotransmitter, N-acetylaspartylglutamate (NAAG), is a selective group II agonist. We synthesized and characterized a urea-based NAAG analogue, ZJ43. This novel compound is a potent inhibitor of enzymes, glutamate carboxypeptidase II (K(i) = 0.8 nM) and III (K(i) = 23 nM) that deactivate NAAG following synaptic release. ZJ43 (100 microM) does not directly interact with NMDA receptors or metabotropic glutamate receptors. Administration of ZJ43 significantly reduced PCP-induced motor activation, falling while walking, stereotypic circling behavior, and head movements. To test the hypothesis that this effect of ZJ43 was mediated by increasing the activation of mGluR3 via increased levels of extracellular NAAG, the group II mGluR selective antagonist LY341495 was co-administered with ZJ43 prior to PCP treatment. This antagonist completely reversed the effects of ZJ43. Additionally, LY341495 alone increased PCP-induced motor activity and head movements suggesting that normal levels of NAAG act to moderate the effect of PCP on motor activation via a group II mGluR. These data support the view that NAAG peptidase inhibitors may represent a new therapeutic approach to some of the components of schizophrenia that are modeled by PCP.     

Quantitative assessment of neurochemical changes in a rat model of long-term alcohol consumption as detected by in vivo and ex vivo proton nuclear magnetic resonance spectroscopy

The aim of present study was to quantitatively investigate the neurochemical profile of the frontal cortex region in a rat model of long-term alcohol consumption, by using in vivo proton magnetic resonance spectroscopy (¹H-MRS) at 4.7 T and ex vivo¹H high-resolution magic angle spinning (HR-MAS) technique at 11.7 T. Twenty male rats were divided into two groups and fed a liquid diet for 10 weeks. After 10 weeks, in vivo¹H MRS spectra were acquired from the frontal cortex brain region. After in vivo¹H MRS experiments, all animals were sacrificed and 20 frontal cortex tissue samples were harvested. All tissue examinations were performed with the 11.7 T HR-MAS spectrometer and high-resolution spectra were acquired. The in vivo and ex vivo spectra were quantified as absolute metabolite concentrations and normalized ratios of total signal-intensity (i.e., metabolites_Norm), respectively. The absolute quantifications of in vivo spectra showed significantly higher glycerophosphocholine plus phosphocholine (GPC + PCh) and lower myo-inositol (mIns) concentrations in ethanol-treated rats compared to controls. The quantifications of ex vivo spectra showed significantly higher PCh_Norm, Cho_Norm and tCho_Norm, and lower GPC_Norm and mIns_Norm ratio levels in ethanol-treated rats compared to controls. Our findings suggest that reduced mIns concentrations caused by the long-term alcohol consumption may lead to hypo-osmolarity syndrome and astrocyte hyponatremia. In addition, increased choline-containing compound concentrations may reflect an increased cell turnover rate of phosphatidylcholine and other phospholipids, indicating an adaptive mechanism. Therefore, these results might be utilized as key markers in chronic alcohol intoxication metabolism.     

Antipsychotic drugs increase N-acetylaspartate and N-acetylaspartylglutamate in SH-SY5Y human neuroblastoma cells

N -Acetylaspartate (NAA) and N -acetylaspartylglutamate (NAAG) are related neuronal metabolites associated with the diagnosis and treatment of schizophrenia. NAA is a valuable marker of neuronal viability in magnetic resonance spectroscopy, a technique which has consistently shown NAA levels to be modestly decreased in the brains of schizophrenia patients. However, there are conflicting reports on the changes in brain NAA levels after treatment with antipsychotic drugs, which exert their therapeutic effects in part by blocking dopamine D₂ receptors. NAAG is reported to be an agonist of the metabotropic glutamate 2/3 receptor, which is linked to neurotransmitter release modulation, including glutamate release. Alterations in NAAG metabolism have been implicated in the development of schizophrenia possibly via dysregulation of glutamate neurotransmission. In the present study we have used high performance liquid chromatography to determine the effects of the antipsychotic drugs haloperidol and clozapine on NAA and NAAG levels in SH-SY5Y human neuroblastoma cells, a model system used to test the responses of dopaminergic neurons in vitro . The results indicate that the antipsychotic drugs haloperidol and clozapine increase both NAA and NAAG levels in SH-SY5Y cells in a dose and time dependant manner, providing evidence that NAA and NAAG metabolism in neurons is responsive to antipsychotic drug treatment.     

Regulation of N-acetylaspartate and N-acetylaspartylglutamate biosynthesis by protein kinase activators

The neuronal dipeptide N-acetylaspartylglutamate (NAAG) is thought to be synthesized enzymatically from N-acetylaspartate (NAA) and glutamate. We used radiolabeled precursors to examine NAA and NAAG biosynthesis in SH-SY5Y human neuroblastoma cells stimulated with activators of protein kinase A (dbcAMP; N6,2'-O-dibutyryl cAMP) and protein kinase C (PMA; phorbol-12-myristate-13-acetate). Differentiation over the course of several days with dbcAMP resulted in increased endogenous NAA levels and NAAG synthesis from l-[(3)H]glutamine, whereas PMA-induced differentiation reduced both. Exogenously applied NAA caused dose dependent increases in intracellular NAA levels, and NAAG biosynthesis from l-[(3)H]glutamine, suggesting precursor-product and mass-action relationships between NAA and NAAG. Incorporation of l-[(3)H]aspartate into NAA and NAAG occurred sequentially, appearing in NAA by 1 h, but not in NAAG until between 6 and 24 h. Synthesis of NAAG from l-[(3)H]aspartate was increased by dbcAMP and decreased by PMA at 24 h. The effects of PMA on l-[(3)H]aspartate incorporation into NAA were temporally biphasic. Using short incubation times (1 and 6 h), PMA increased l-[(3)H]aspartate incorporation into NAA, but with longer incubation (24 h), incorporation was significantly reduced. These results suggest that, while the neuronal production of NAA and NAAG are biochemically related, significant differences exist in the regulatory mechanisms controlling their biosynthesis.     

Differential negative coupling of type 3 metabotropic glutamate receptor to cyclic GMP levels in neurons and astrocytes

Metabotropic receptors may couple to different G proteins in different cells or perhaps even in different regions of the same cell. To date, direct studies of group II and group III metabotropic glutamate receptors' (mGluRs) relationships to second messenger cascades have reported negative coupling of these receptors to cyclic AMP (cAMP) levels in neurons, astrocytes and transfected cells. In the present study, we found that the peptide neurotransmitter N-acetylaspartylglutamate (NAAG), an mGluR3-selective agonist, decreased sodium nitroprusside (SNP)-stimulated cyclic GMP (cGMP) levels in cerebellar granule cells and cerebellar astrocytes. The mGluR3 and group II agonists FN6 and LY354740 had similar effects on cGMP levels. The mGluR3 and group II antagonists beta-NAAG and LY341495 blocked these actions. Treatment with pertussis toxin inhibited the effects of NAAG on SNP-stimulated cGMP levels in rat cerebellar astrocytes but not in cerebellar neurons. These data support the conclusion that mGluR3 is also coupled to cGMP levels and that this mGluR3-induced reduction of cGMP levels is mediated by different G proteins in cerebellar astrocytes and neurons. We previously reported that this receptor is coupled to a cAMP cascade via a pertussis toxin-sensitive G protein in cerebellar neurons, astrocytes and transfected cells. Taken together with the present data, we propose that mGluR3 is coupled to two different G proteins in granule cell neurons. These data greatly expand knowledge of the range of second messenger cascades induced by mGluR3, and have implications for clinical conditions affected by NAAG and other group II mGluR agonists.     

Regulation of glutamate carboxypeptidase II hydrolysis of N-acetylaspartylglutamate (NAAG) in crayfish nervous tissue is mediated by glial glutamate and acetylcholine receptors

Glutamate carboxypeptidase II (GCPII), a glial ectoenzyme, is responsible for N-acetylaspartylglutamate (NAAG) hydrolysis. Its regulation in crayfish nervous tissue was investigated by examining uptake of [3H]glutamate derived from N-acetylaspartyl-[3H]glutamate ([3H]NAAG) to measure GCPII activity. Electrical stimulation (100 Hz, 10 min) during 30 min incubation with [3H]NAAG increased tissue [3H]glutamate tenfold. This was prevented by 2-(phosphonomethyl)-pentanedioic acid (2-PMPA), a GCPII inhibitor, suggesting that stimulation increased the hydrolysis of [3H]NAAG and metabolic recycling of [3H]glutamate. Antagonists of glial group II metabotropic glutamate receptors (mGLURII), NMDA receptors and acetylcholine (ACh) receptors that mediate axon-glia signaling in crayfish nerve fibers decreased the effect of stimulation by 58-83%, suggesting that glial receptor activation leads to stimulation of GCPII activity. In combination, they reduced [3H]NAAG hydrolysis during stimulation to unstimulated control levels. Agonist stimulation of mGLURII mimicked the effect of electrical stimulation, and was prevented by antagonists of GCPII or mGLURII. Raising extracellular K+ to three times the normal level stimulated [3H]NAAG release and GCPII activity. These effects were also blocked by antagonists of GCPII and mGLUR(II). No receptor antagonist or agonist tested or 2-PMPA affected uptake of [3H]glutamate. We conclude that NAAG released from stimulated nerve fibers activates its own hydrolysis via stimulation of GCPII activity mediated through glial mGLURII, NMDA and ACh receptors.