Identification of the N-glycosylation sites on glutamate carboxypeptidase II necessary for proteolytic activity

Glutamate carboxypeptidase II (GCPII) is a membrane peptidase expressed in the prostate, central and peripheral nervous system, kidney, small intestine, and tumor-associated neovasculature. The GCPII form expressed in the central nervous system, termed NAALADase, is responsible for the cleavage of N-acetyl-L-aspartyl-L-glutamate (NAAG) yielding free glutamate in the synaptic cleft, and is implicated in various pathologic conditions associated with glutamate excitotoxicity. The prostate form of GCPII, termed prostate-specific membrane antigen (PSMA), is up-regulated in cancer and used as an effective prostate cancer marker. Little is known about the structure of this important pharmaceutical target. As a type II membrane protein, GCPII is heavily glycosylated. In this paper we show that N-glycosylation is vital for proper folding and subsequent secretion of human GCPII. Analysis of the predicted N-glycosylation sites also provides evidence that these sites are critical for GCPII carboxypeptidase activity. We confirm that all predicted N-glycosylation sites are occupied by an oligosaccharide moiety and show that glycosylation at sites distant from the putative catalytic domain is critical for the NAAG-hydrolyzing activity of GCPII calling the validity of previously described structural models of GCPII into question.     

Poster sessions AP11: Neurotransmitters,Transporter and Enzymes. Structure and activity of recombinant human glutamate carboxypeptidase II

Glutamate carboxypeptidase II (GCPII, EC 3.4.17.21) is a membrane peptidase expressed in a number of tissues such as kidney, prostate and brain. The brain form of GCPII (also known as N‐acetylated‐α‐linked‐acidic dipeptidase, NAALADase) cleaves N‐acetyl‐aspartyl glutamate to yield free glutamate. Animal model experiments show that inhibition of GCPII prevents neuronal cell death during experimental ischaemia. GCPII thus represents an important target for the treatment of neuronal damage caused by excess glutamate. We report the mapping of the entire coding region of GCPII and identification of the region sufficient and necessary for the production of active recombinant protein. Extracellular portion of human glutamate carboxypeptidase II (amino acids 44–750) was expressed in Drosophila Schneider's cells and purified to homogeneity. A novel assay for hydrolytic activity of GCPII, based on fluorimetric detection of released alpha‐amino groups was established, and used for enzymological characterization of GCPII. The potential of this assay for high‐throughput inhibitor testing was evaluated and pH dependence for the enzymatic activity have been analysed. Using a complete set of protected dipeptides, substrate specificity of recombinant GCPII was elucidated. Ac‐Glu‐Met, Ac‐Asp‐Met and surprisingly Ac‐Ala‐Met were identified as novel substrates for GCPII. The glycosylation has been found indispensable for the activity of the enzyme. A series of point mutants of the enzyme has been expressed and purified and the glycosylation sites critical for the proteolytic activity have been identified.     

Substrate specificity, inhibition and enzymological analysis of recombinant human glutamate carboxypeptidase II

Glutamate carboxypeptidase II (GCPII, EC 3.4.17.21) is a membrane peptidase expressed in a number of tissues such as kidney, prostate and brain. The brain form of GCPII (also known as NAALADase) cleaves N-acetyl-aspartyl glutamate to yield free glutamate. Animal model experiments show that inhibition of GCPII prevents neuronal cell death during experimental ischaemia. GCPII thus represents an important target for the treatment of neuronal damage caused by excess glutamate. In this paper we report expression of an extracellular portion of human glutamate carboxypeptidase II (amino acids 44-750) in Drosophila Schneider's cells and its purification to homogeneity. A novel assay for hydrolytic activity of recombinant human GCPII (rhGCPII), based on fluorimetric detection of released alpha-amino groups was established, and used for its enzymological characterization. rhGCPII does not show dipeptidylpeptidase IV-like activity assigned to the native form of the enzyme previously. Using a complete set of protected dipeptides, substrate specificity of rhGCPII was elucidated. In addition to the previously described substrates, four novel compounds, Ac-Glu-Met, Ac-Asp-Met and, surprisingly, Ac-Ala-Glu and Ac-Ala-Met were identified as substrates for GCPII, and their respective kinetic constants determined. The glycosylation of rhGCPII was found indispensable for the enzymatic activity.     

Mapping of the active site of glutamate carboxypeptidase II by site-directed mutagenesis

Human glutamate carboxypeptidase II [GCPII (EC 3.4.17.21)] is recognized as a promising pharmacological target for the treatment and imaging of various pathologies, including neurological disorders and prostate cancer. Recently reported crystal structures of GCPII provide structural insight into the organization of the substrate binding cavity and highlight residues implicated in substrate/inhibitor binding in the S1' site of the enzyme. To complement and extend the structural studies, we constructed a model of GCPII in complex with its substrate, N-acetyl-l-aspartyl-l-glutamate, which enabled us to predict additional amino acid residues interacting with the bound substrate, and used site-directed mutagenesis to assess the contribution of individual residues for substrate/inhibitor binding and enzymatic activity of GCPII. We prepared and characterized 12 GCPII mutants targeting the amino acids in the vicinity of substrate/inhibitor binding pockets. The experimental results, together with the molecular modeling, suggest that the amino acid residues delineating the S1' pocket of the enzyme (namely Arg210) contribute primarily to the high affinity binding of GCPII substrates/inhibitors, whereas the residues forming the S1 pocket might be more important for the 'fine-tuning' of GCPII substrate specificity.     

Biochemical characterization of human glutamate carboxypeptidase III

Human glutamate carboxypeptidase II (GCPII) is a transmembrane metallopeptidase found mainly in the brain, small intestine, and prostate. In the brain, it cleaves N-acetyl-L-aspartyl-glutamate, liberating free glutamate. Inhibition of GCPII has been shown to be neuroprotective in models of stroke and other neurodegenerations. In prostate, it is known as prostate-specific membrane antigen, a cancer marker. Recently, human glutamate carboxypeptidase III (GCPIII), a GCPII homolog with 67% amino acid identity, was cloned. While GCPII is recognized as an important pharmaceutical target, no biochemical study of human GCPIII is available at present. Here, we report the cloning, expression, and characterization of recombinant human GCPIII. We show that GCPIII lacks dipeptidylpeptidase IV-like activity, its activity is dependent on N-glycosylation, and it is effectively inhibited by several known inhibitors of GCPII. In comparison to GCPII, GCPIII has lower N-acetyl-L-aspartyl-glutamate-hydrolyzing activity, different pH and salt concentration dependence, and distinct substrate specificity, indicating that these homologs might play different biological roles. Based on a molecular model, we provide interpretation of the distinct substrate specificity of both enzymes, and examine the amino acid residues responsible for the differences by site-directed mutagenesis. These results may help to design potent and selective inhibitors of both enzymes.     

The cloning and characterization of a second brain enzyme with NAAG peptidase activity

The peptide neurotransmitter N-acetylaspartylglutamate is inactivated by extracellular peptidase activity following synaptic release. It is speculated that the enzyme, glutamate carboxypeptidase II (GCPII, EC 3.14.17.21), participates in this inactivation. However, CGCPII knockout mice appear normal in standard neurological tests. We report here the cloning and characterization of a mouse enzyme (tentatively identified as glutamate carboxypeptidase III or GCPIII) that is homologous to an enzyme identified in a human lung carcinoma. The mouse peptidase was cloned from two non-overlapping EST clones and mouse brain cDNA using PCR. The sequence (GenBank, AY243507) is 85% identical to the human carcinoma enzyme and 70% homologous to mouse GCPII. GCPIII sequence analysis suggests that it too is a zinc metallopeptidase. Northern blots revealed message in mouse ovary, testes and lung, but not brain. Mouse cortical and cerebellar neurons in culture expressed GCPIII message in contrast to the glial specific expression of GCPII. Message levels of GCPIII were similar in brains obtained from wild-type mice and mice that are null mutants for GCPII. Chinese hamster ovary (CHO) cells transfected with rat GCPII or mouse GCPIII expressed membrane bound peptidase activity with similar V(max) and K(m) values (1.4 micro m and 54 pmol/min/mg; 3.5 micro m and 71 pmol/min/mg, respectively). Both enzymes are activated by a similar profile of metal ions and their activities are blocked by EDTA. GCPIII message was detected in brain and spinal cord by RT-PCR with highest levels in the cerebellum and hippocampus. These data are consistent with the hypothesis that nervous system cells express at least two differentially distributed homologous enzymes with similar pharmacological properties and affinity for NAAG.     

Purification and primary structure determination of human lysosomal dipeptidase

The lysosomal metallopeptidase is an enzyme that acts preferentially on dipeptides with unsubstituted N- and C-termini. Its activity is highest in slightly acidic pH. Here we describe the isolation and characterization of lysosomal dipeptidase from human kidney. The isolated enzyme has the amino-terminal sequence DVAKAIINLAVY and is a homodimer with a molecular mass of 100 kDa. So far no amino acid sequence has been determined for this metallopeptidase. The complete primary structure as deduced from the nucleotide sequence revealed that the isolated dipeptidase is similar to blood plasma glutamate carboxypeptidase.     

Baculoviral expression and characterization of human recombinant PGCP in the form of an active mature dimer and an inactive precursor protein

The human-blood plasma glutamate carboxypeptidase (PGCP) is a proteinase that acts on the unsubstituted N- and C-termini of dipeptides. It has been suggested that this PGCP is involved in the release of thyroxine. Furthermore, research has suggested that its activity is up-regulated in hepatitis-C-virus-infected patients with hepatocellular carcinoma. In this study expressed human PGCP in the baculovirus expression system was produced by a Sf9 insect cell line with aim to prepare sufficient amounts of active recombinant enzyme for a subsequent biological characterization. Recombinant PGCP was expressed and secreted into the medium in the form of an inactive proenzyme. It was gradually converted into an active form in the medium after three days, with the highest expression of the active form on day six. The protein was sequentially purified by a combination of various liquid chromatographies, such as hydroxyapatite, ion exchange, and gel chromatography, and as final step with affinity chromatography on Phe-Leu-Sepharose. The human PGCP was purified as an active enzyme in the dimer form and as inactive precursor protein. The dipeptidase activity was confirmed by measuring the hydrolysis of the Ser-Met dipeptide at a slightly acidic pH.     

Structure of glutamate carboxypeptidase II, a drug target in neuronal damage and prostate cancer

Membrane-bound glutamate carboxypeptidase II (GCPII) is a zinc metalloenzyme that catalyzes the hydrolysis of the neurotransmitter N-acetyl-L-aspartyl-L-glutamate (NAAG) to N-acetyl-L-aspartate and L-glutamate (which is itself a neurotransmitter). Potent and selective GCPII inhibitors have been shown to decrease brain glutamate and provide neuroprotection in preclinical models of stroke, amyotrophic lateral sclerosis, and neuropathic pain. Here, we report crystal structures of the extracellular part of GCPII in complex with both potent and weak inhibitors and with glutamate, the product of the enzyme's hydrolysis reaction, at 2.0, 2.4, and 2.2 A resolution, respectively. GCPII folds into three domains: protease-like, apical, and C-terminal. All three participate in substrate binding, with two of them directly involved in C-terminal glutamate recognition. One of the carbohydrate moieties of the enzyme is essential for homodimer formation of GCPII. The three-dimensional structures presented here reveal an induced-fit substrate-binding mode of this key enzyme and provide essential information for the design of GCPII inhibitors useful in the treatment of neuronal diseases and prostate cancer.     

The C-terminal region of mitochondrial glycerol-3-phosphate acyltransferase-1 interacts with the active site region and is required for activity

Glycerol phosphate acyltransferase (GPAT) catalyzes the formation of 1-acyl-sn-glycerol-3-phosphate from glycerol-3-phosphate and long chain fatty acyl-CoA substrates. We previously determined the topography of the mitochondrial GPAT1 isoform (mtGPAT1, 828 amino acids). mtGPAT1 has two transmembrane domains (TMDs) (aa 472-493 and aa 576-592) with both the N- and C-termini facing the cytosol and a loop (aa 494-575) facing the intermembrane space. Alignment of amino acid sequences from mtGPAT1 and other acyltransferases and site directed mutagenesis studies have demonstrated that the active site of the enzyme resides in the N-terminal domain of the protein. In this study, we sequentially truncated the C-terminal domain and characterized the properties of the resulting mutants expressed in CHO cells. Although the mutants were overexpressed, none of them conferred GPAT activity. The loss of activity was not due to the miss-targeting of the proteins since immunofluorescence experiments demonstrated their mitochondrial localization. Instead, chemical crosslinking and protein cleavage studies demonstrated that the N- and C-termini of the protein interact. These results suggest that the C-terminal domain is necessary for mtGPAT1 activity, and probably contributes to catalysis or substrate binding.     

Localization of the catalytic activity in restrictocin molecule by deletion mutagenesis.

Restrictocin, produced by the fungus Aspergillus restrictus, is a highly specific ribonucleolytic toxin which cleaves a single phosphodiester bond between G4325 and A4326 in the 28S rRNA. It is a nonglycosylated, single-chain, basic protein of 149 amino acids. The putative catalytic site of restrictocin includes Tyr47, His49, Glu95, Arg120 and His136. To map the catalytic activity in the restrictocin molecule, and to study the role of N- and C-terminus in its activity, we have systematically deleted amino-acid residues from both the termini. Three N-terminal deletions removing 8, 15 and 30 amino acids, and three C-terminal deletions lacking 4, 6, and 11 amino acids were constructed. The deletion mutants were expressed in Escherichia coli, purified to homogeneity and functionally characterized. Removal of eight N-terminal or four C-terminal amino acids rendered restrictocin partially inactive, whereas any further deletions from either end resulted in the complete inactivation of the toxin. The study demonstrates that intact N- and C-termini are required for the optimum functional activity of restrictocin.     

Isolation, purification, and characterization of a new enzyme from Pseudomonas sp. M-27, carboxypeptidase G3.

A new type of carboxypeptidase was found in a strain of Pseudomonas sp. M-27 isolated from soil. The cell-free extract, solubilized by colistin sulfate, was purified to homogeneity. This enzyme had a single peak with a molecular weight of 60,000 on a calibrated Superdex column and consisted of four subunits of identical molecular weights (M(r): 15,000). The enzyme hydrolyzed predominantly acidic peptides and N-acyl amino acids with Glu or Asp in the C-termini. This enzyme was not strongly affected by thiol enzyme inhibitors (PCMB, iodoacetic acid) or serine protease inhibitors (DFP, PMSF), but was inhibited by metal chelators. The enzyme resembles carboxypeptidase G1 or G2 in its glutamate-releasing activity. However, it acts not only on the L-form but also on the D-form of acidic amino acids and shows affinity for the long-chain fatty acyl group but not the benzoyl group. Thus, as this enzyme differs from carboxypeptidase G1 or G2, it was named carboxypeptidase G3.     

Unique gating properties of C. elegans ClC anion channel splice variants are determined by altered CBS domain conformation and the R-helix linker

All eukaryotic and some prokaryotic ClC anion transport proteins have extensive cytoplasmic C-termini containing two cystathionine-β-synthase (CBS) domains. CBS domain secondary structure is highly conserved and consists of two a-helices and three b-strands arranged as b1-a1-b2-b3-a2. ClC CBS domain mutations cause muscle and bone disease and alter ClC gating. However, the precise functional roles of CBS domains and the structural bases by which they regulate ClC function are poorly understood. CLH-3a and CLH-3b are C. elegans ClC anion channel splice variants with strikingly different biophysical properties. Splice variation occurs at cytoplasmic N- and C-termini and includes several amino acids that form a2 of the second CBS domain (CBS2). We demonstrate that interchanging a2 between CLH-3a and CLH-3b interchanges their gating properties. The "R-helix" of ClC proteins forms part of the ion conducting pore and selectivity filter and is connected to the cytoplasmic C-terminus via a short stretch of cytoplasmic amino acids termed the "R-helix linker". C-terminus conformation changes could cause R-helix structural rearrangements via this linker. X-ray structures of three ClC protein cytoplasmic C-termini suggest that a2 of CBS2 and the R-helix linker could be closely apposed and may therefore interact. We found that mutating apposing amino acids in a2 and the R-helix linker of CLH-3b was sufficient to give rise to CLH-3a-like gating. We postulate that the R-helix linker interacts with CBS2 a2, and that this putative interaction provides a pathway by which cytoplasmic C-terminus conformational changes induce conformational changes in membrane domains that in turn modulate ClC function.     

A novel isoform of a kallikrein-like protease, TLSP/hippostasin, (PRSS20), is expressed in the human brain and prostate

cDNAs encoding two splicing variants of a serine protease, termed hippostasin, were isolated by a PCR-based cloning strategy. The difference of 5' nucleotide sequence resulted in the variation in the amino terminal ends of the two, brain and prostate, types of human hippostasin. The longest ORF of the brain-type was 250 amino acids with a putative signal peptide, while that of the prostate-type was 282 amino acids. Homology search using the amino acid sequence revealed that prostate-type hippostasin was identical to TLSP (PRSS20), which is expressed in human primary keratinocytes (1). Transient expression analysis showed that both brain- and prostate-type TLSP/hippostasin were secreted into the conditioned medium as about 40 kDa proteins. Human TLSP/hippostasin showed 47% and 45% identity to trypsinogen II and kallikrein, respectively. In fact, the recombinant human TLSP/hippostasin efficiently cleaved Bz-Phe-Arg-4-methylcoumaryl-7-amide, a kallikrein substrate, and weakly cleaved other substrates for kallikrein and trypsin. Northern blot analysis detected a 1.3 kb band in the whole brain and a 1.4 kb band in the prostate and the lung. In situ hybridization revealed that it was expressed preferentially by the pyramidal neurons in the human hippocampus and secretory epithelial cells in the prostate. These results indicated that TLSP/hippostasin is involved in the functions of the human central nervous system and prostate and that it is a multifunctional protease present in various organs.     

Analysis of membrane topology of the human reduced folate carrier protein by hemagglutinin epitope insertion and scanning glycosylation insertion mutagenesis

The human reduced folate carrier (RFC) is the major membrane transport system for both reduced folates and chemotherapeutic antifolate drugs, such as methotrexate (MTX). Although the RFC protein has been subjected to intensive study in order to identify critical structural and functional determinants of transport, it is impossible to assess the significance of these studies without characterizing the essential domain structure and membrane topology. The primary amino acid sequence from the cloned cDNAs predicts that the human RFC protein has 12 transmembrane domains (TMDs) with a large cytosolic loop between TMDs 6 and 7, and cytosolic-facing N- and C-termini. To establish the RFC membrane topology, a hemagglutinin (HA) epitope was inserted into the individual predicted intracellular and extracellular loops. HA insertions into putative TMD interconnecting loops 3/4, 6/7, 7/8, and 8/9, and the N- and C-termini all preserved MTX transport activity upon expression in transport-impaired K562 cells. Immunofluorescence detection with HA-specific antibody under both permeabilized and non-permeabilized conditions confirmed extracellular orientations for loops 3/4 and 7/8, and cytosolic orientations for loops 6/7 and 8/9, and the N- and C-termini. Insertion of a consensus N-glycosylation site [NX(S/T)] into putative loops 5/6, 8/9, and 9/10 of deglycosylated RFC-Gln⁵⁸ had minimal effects on MTX transport. Analysis of glycosylation status on Western blots suggested an extracellular orientation for loop 5/6, and intracellular orientations for loops 8/9 and 9/10. Our findings strongly support the predicted topology model for TMDs 1-8 and the C-terminus of human RFC. However, our results raise the possibility of an alternative membrane topology for TMDs 9-12.     

Analysis of membrane topology of the human reduced folate carrier protein by hemagglutinin epitope insertion and scanning glycosylation insertion mutagenesis

The human reduced folate carrier (RFC) is the major membrane transport system for both reduced folates and chemotherapeutic antifolate drugs, such as methotrexate (MTX). Although the RFC protein has been subjected to intensive study in order to identify critical structural and functional determinants of transport, it is impossible to assess the significance of these studies without characterizing the essential domain structure and membrane topology. The primary amino acid sequence from the cloned cDNAs predicts that the human RFC protein has 12 transmembrane domains (TMDs) with a large cytosolic loop between TMDs 6 and 7, and cytosolic-facing N- and C-termini. To establish the RFC membrane topology, a hemagglutinin (HA) epitope was inserted into the individual predicted intracellular and extracellular loops. HA insertions into putative TMD interconnecting loops 3/4, 6/7, 7/8, and 8/9, and the N- and C-termini all preserved MTX transport activity upon expression in transport-impaired K562 cells. Immunofluorescence detection with HA-specific antibody under both permeabilized and non-permeabilized conditions confirmed extracellular orientations for loops 3/4 and 7/8, and cytosolic orientations for loops 6/7 and 8/9, and the N- and C-termini. Insertion of a consensus N-glycosylation site [NX(S/T)] into putative loops 5/6, 8/9, and 9/10 of deglycosylated RFC-Gln(58) had minimal effects on MTX transport. Analysis of glycosylation status on Western blots suggested an extracellular orientation for loop 5/6, and intracellular orientations for loops 8/9 and 9/10. Our findings strongly support the predicted topology model for TMDs 1-8 and the C-terminus of human RFC. However, our results raise the possibility of an alternative membrane topology for TMDs 9-12.     

Isolation, tissue distribution, and chromosomal localization of a novel testis-specific human four-transmembrane gene related to CD20 and FcepsilonRI-beta

CD20 and the beta subunit of the high affinity receptor for IgE (FcepsilonRIbeta) are related four-transmembrane molecules that are expressed on the surface of hematopoietic cells and play crucial roles in signal transduction. Herein, we report the identification and characterization of a human gene, TETM4, that encodes a novel four-transmembrane protein related to CD20 and FcepsilonRIbeta. The predicted TETM4 protein is 200 amino acids and contains four putative transmembrane regions, N- and C-terminal cytoplasmic domains, and three inter-transmembrane loop regions. TETM4 shows 31.0 and 23.2% overall identity with CD20 and FcepsilonRIbeta respectively, with the highest identity in the transmembrane regions, whereas the N- and C-termini and inter-transmembrane loops are more divergent. Northern blot and RT-PCR analysis suggest that TETM4 mRNA has a highly restricted tissue distribution, being expressed selectively in the testis. Using fluorescence in situ hybridization and radiation hybrid analysis, the TETM4 gene has been localized to chromosome 11q12. The genes for CD20 and FcepsilonRIbeta have also been mapped to the same region of chromosome 11 (11q12-13.1), suggesting that these genes have evolved by duplication to form a family of four-transmembrane genes. TETM4 is the first nonhematopoietic member of the CD20/FcepsilonRIbeta family, and like its hematopoietic-specific relatives, it may be involved in signal transduction as a component of a multimeric receptor complex.     

Catalytic and regulatory sites of yeast plasma membrane H(+)-ATPase studied by directed mutagenesis

More than 35 site-directed mutants of the plasma membrane H(+)-ATPase of the yeast Saccharomyces cerevisiae have been constructed and expressed to investigate the function of N- and C-termini and of conserved amino acids. Conserved motif TGES seems to form part of both the catalytic machinery for the hydrolysis of the phosphorylated intermediate and the vanadate binding site. In addition, it is involved in the coupling of ATP hydrolysis to H+ transport. The phosphorylated intermediate is also essential for this coupling, but not for ATP hydrolysis. The aspartate residues of conserved motifs DPPR, TGD and TGDGVND (the last one) seem to form part of the ATP binding site. The positive charge of the conserved motif KGAP is important for the kinase or phosphorylating activity. A conserved proline and a conserved aspartate predicted to have a transmembrane location are essential for activity. The N-terminus contains a conserved acidic region which may be involved in assembly into the plasma membrane. All the hydrophobic stretches at the C-terminus are also required for assembly. The last 11 amino acids constitute a non-essential inhibitory domain involved in regulation of the enzyme by glucose metabolism.