Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Dynamic O-Glycosylation of Nuclear and Cytosolic Proteins

Dynamic O-Glycosylation of Nuclear and Cytosolic Proteins THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 13, Issue of March 30, pp. 9838 –9845, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. CLONING AND CHARACTERIZATION OF A NEUTRAL, CYTOSOLIC b-N-ACETYLGLUCOSAMINIDASE FROM HUMAN BRAIN* Received for publication, November 16, 2000, and in revised form, January 4, 2001 Published, JBC Papers in Press, January 8, 2001, DOI 10.1074/jbc.M010420200 Yuan Gao‡, Lance Wells‡, Frank I. Comer‡, Glendon J. Parker§, and Gerald W. Hart‡¶ From the ‡Department of Biological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205-2185 and the §Department of Internal Medicine, Division of Endocrinology, Metabolism and Diabetes, The University of Utah, Salt Lake City, Utah 84132 Dynamic modification of cytoplasmic and nuclear pro- cover a broad range, including many transcription factors, RNA polymerase II, oncogenes, nuclear pore proteins, viral proteins, teins by O-linked N-acetylglucosamine (O-GlcNAc) on Ser/Thr residues is ubiquitous in higher eukaryotes and and tumor repressors (for details, see Refs. 2 and 3 and cita- is analogous to protein phosphorylation. The enzyme for tions within). Unlike classic O- or N-linked protein glycosyla- the addition of this modification, O-GlcNAc transferase, tions, the O-GlcNAc modification involves only a single GlcNAc has been cloned from several species. Here, we have moiety linked to the hydroxyl group of Ser/Thr residues, gen- cloned a human brain O-GlcNAcase that cleaves O-Glc- erally is not elongated, and is found exclusively in the cyto- NAc off proteins. The cloned cDNA encodes a polypep- plasm and nucleoplasm. tide of 916 amino acids with a predicted molecular mass Protein O-GlcNAcylation is highly dynamic, and the cycle of of 103 kDa and a pI value of 4.63, but the protein mi- addition/removal of the sugar moiety is rapid, analogous to grates as a 130-kDa band on SDS-polyacrylamide gel protein phosphorylation/dephosphorylation catalyzed by ki- electrophoresis. The cloned O-GlcNAcase has a pH opti- nases and phosphatases (2). Indeed, existing evidence suggests mum of 5.5–7.0 and is inhibited by GlcNAc but not by that this modification has a “yin-yang” relationship with pro- b-GlcNAc, but not pNP-b- GalNAc. p-Nitrophenyl (pNP)- tein phosphorylation in some cases (4). Many O-GlcNAcylation a-GlcNAc, is a substrate. The cloned en- GalNAc or pNP- sites have been mapped to phosphorylation sites or adjacent zyme cleaves GlcNAc, but not GalNAc, from glycopep- sites (5–7). Such spatial localization indicates that O-GlcNAc tides. Cell fractionation suggests that the overexpressed may regulate the target protein by competing with protein protein is mostly localized in the cytoplasm. It therefore kinases (4). Recent studies using phosphatase and kinase in- has all the expected characteristics of O-GlcNAcase and hibitors have provided direct evidence for a general reciprocal is distinct from lysosomal hexosaminidases. North- relationship between O-GlcNAcylation and phosphorylation on ern blots show that the transcript is expressed in some proteins (8, 9). every human tissue examined but is the highest in the O-GlcNAcylation appears to be involved in gene transcrip- brain, placenta, and pancreas. An understanding of O- GlcNAc dynamics and O-GlcNAcase may be key to elu- tion. Most transcription factors examined so far, including Sp1, cidating the relationships between O-phosphate and AP1, AP2, AP4 (10), serum response factor (11), the estrogen O-GlcNAc and to the understanding of the molecular receptor (7, 12), the insulin promoter factor-1, and peroxisome mechanisms of diseases such as diabetes, cancer, and proliferator-activated receptor-g, as well as RNA polymerase II neurodegeneration. (13) and chromatin (14) are O-GlcNAcylated. O-GlcNAcylation of Sp1 appears to enhance its activity in transcription, and, conversely, blocking the GlcNAc residues with lectin wheat Since the description of O-linked N-acetylglucosamine (O- germ agglutinin suppresses the transcriptional activity (10). GlcNAc) as an abundant modification in murine lymphocytes O-GlcNAcylation of Sp1 also controls its degradation by the (1), a myriad of cytoplasmic and nuclear proteins in all meta- proteasome (15). Hyperglycemia-induced superoxide produc- zoans have been found to carry this modification. Such proteins tion increases Sp1 glycosylation resulting in the activation of genes that contribute to the pathogenesis of diabetes (16). O-GlcNAc transferase (OGT), which transfers GlcNAc from * Supported by National Institutes of Health Grant HD13563. The the donor substrate UDP-GlcNAc to target proteins, has been costs of publication of this article were defrayed in part by the payment purified and cloned from several species including human, rat, of page charges. This article must therefore be hereby marked “adver- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate and Caenorhabditis elegans (17–19). It does not share any this fact. significant homology with any other known proteins, including To whom correspondence should be addressed: The Johns Hopkins glycosyltransferases, and is highly conserved from C. elegans to University School of Medicine, 725 North Wolfe St., Baltimore, MD human. Disruption of the ogt gene is lethal in mouse embryonic 21205-2185. Tel.: 410-614-5993; Fax: 410-614-8804; E-mail: [email protected]. stem cells, further underscoring the importance of O-GlcNAc The abbreviations used are: O-GlcNAc, O-linked N-acetylglu- modification in cellular functions (20). O-GlcNAcase, the en- cosamine; GalNAc, N-acetylgalactosamine; GlcNAc, N-acetylglu- zyme that removes O-GlcNAc from such proteins, was purified cosamine; OGT, O-GlcNAc transferase; O-GlcNAcase, N-acetyl-b-D-glu- several years ago from rat spleen (21). It is a neutral cytosolic cosaminidase; PAGE, polyacrylamide gel electrophoresis; PCR, b-glucosaminidase or hexosaminidase C (EC 3.2.1.52). To fur- polymerase chain reaction; pNP, p-nitrophenyl; PUGNAc, O-(2-acet- amido-2-deoxy-D-glucopyranosylidene)-amino-N-phenylcarbamate; bp, ther study the function of this modification, we have now ex- base pair(s); CTD, C-terminal domain; kb, kilobase(s); EST, expressed sequence tag; MALDI-TOF, matrix-assisted laser desorption ionization- time of flight; MS, mass spectrometry; ConA, concanavalin A; PMSF, phenylmethylsulfonyl fluroride. Y. Gao and G. W. Hart, manuscript in preparation . 9838 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Cloning of O-GlcNAcase from Human Brain 9839 The tryptic peptides from each protein were analyzed by capillary tensively purified O-GlcNAcase from bovine brain, sequenced reversed phase high pressure liquid chromatography with in-line tan- the protein by mass spectrometry, and cloned the cDNA. The dem MS/MS on a Finnigan LCQ ion-trap mass spectrometer. Proteins O-GlcNAcase is evolutionarily conserved, distinct from lysoso- were identified by the SEQUEST algorithm with sequencing of at least mal acidic hexosaminidases A and B. The recombinant protein seven tryptic peptides for each protein (24). has all the expected characteristics of O-GlcNAcase, including Cloning of O-GlcNAcase—A putative O-GlcNAcase with a theoretical the ability to cleave O-GlcNAc from glycopeptides. length of 916 amino acids in human was identified by the above pro- teomic approach. A cDNA fragment, KIAA0679 (GenBanky accession EXPERIMENTAL PROCEDURES number AB014579), which contains the coding sequence for 767 amino Purification of O-GlcNAcase from Bovine Brain—All chromato- acids of the C terminus of the human O-GlcNAcase and a 2.0-kb 39- graphic materials were purchased from Amersham Pharmacia Biotech untranslated region, was obtained from the Kazusa DNA Research (Piscataway, NJ). Purification procedure is substantially modified from Institute, Japan, in the vector pBluescript. The coding sequence of this that of Dong and Hart (21). All steps were conducted at 4 °C or on ice. fragment was subsequently transferred to pcDNA3.1His A using XhoI Step 1. Tissue Homogenization—Three bovine brains (;1 kg) were and XbaI, which were located within the polycloning cloning site of frozen in liquid N and shipped on dry ice from Pel Freez Biologicals pBluescript and in the 39-untranslated region of the cDNA, respec- (Rogers, AR) and stored at 280 °C until use. The brains were smashed tively. The missing 59-end fragment of the full-length coding cDNA (447 into smaller pieces and homogenized in 5 volumes (v/w) of homogeni- bp) was amplified by PCR from a human brain Marathon cDNA library zation buffer (20 mM sodium phosphate, pH 7.5, 15 mM 2-mecaptoetha- (CLONTECH, Palo Alto, CA) using the forward primer GGATGGTG- nol, 10 mM MgCl ,1mM PMSF, 1 mM EDTA) in a Hamilton Beach 2 CAGAAGGAGAGTCAAGCGAC and the reverse primer TAGAAAC- blender with 5 3 20-s bursts. The homogenate was centrifuged at CTCTTCGATGGACTCTACTGG. The forward primer sequence was 18,000 3 g for 30 min. The pellet was discarded, and the cytosolic based on published data (25), and the reverse primer was located in the supernatant was pooled in a 5-liter beaker. KIAA0679 clone. PCR conditions were 94 °C for 30 s, 63 °C for 30 s, and Step 2. Ammonium Sulfate Precipitation—The cytosolic supernatant 72 °C for 3 min for 30 –35 cycles. A second round of PCR using the first was subjected to 30 –50% ammonium sulfate precipitation. The pellet PCR product as template and a forward primer incorporating a NotI was resuspended in 500 ml of buffer A (20 mM sodium phosphate, pH site (CCGGGCGGCCGCGGATGGTGCAGAAGGAGAG) and the same 7.5, 5 mM 2-mercaptoethanol) and centrifuged to clarify the solution. reverse primer was performed. The product was digested with NotI and The solution was then thoroughly dialyzed against buffer A and centri- HindIII (unique site in the PCR product) and ligated in-frame into the fuged again to eliminate any insoluble materials that had resulted from pcDNA3.1His A-KIAA0679 construct. This gave rise to a full-length dialysis. cDNA in the vector pcDNA3.1His A. The final construct was sequenced. Step 3. DE52 Cellulose Ion Exchange Chromatography—The dia- O-GlcNAcase Assays—Unless stated otherwise, O-GlcNAcase activ- lyzed sample was loaded onto a DE52 column (900-ml bed vol) at a flow ity was assayed as described in 50 mM sodium cacodylate, pH 6.5, 0.3% rate of 2 ml/min using a peristaltic pump. After washing the column bovine serum albumin, 2 mM pNP-b-GlcNAc, 50 mM GalNAc (21). Pu- with 3 liters of buffer A, bound proteins were eluted with a linear rified bovine kidney lysosomal b-hexosaminidase (Roche Molecular Bio- gradient of 0 –1 M NaCl in 4 liters of buffer A at a flow rate of 4 ml/min. chemicals, Indianapolis, IN) activity was assayed in citrate phosphate The protein profile was monitored by absorbance at 280 nm. Fractions buffer, pH 4.5, 0.3% bovine serum albumin, 2 mM pNP-b-GlcNAc. To (16 ml) enriched in O-GlcNAcase activity were pooled. test the ability of recombinant O-GlcNAcase to cleave O-GlcNAc from Step 4. Concanavalin A-Sepharose 4B Chromatography—MgCl was glycopeptides, two glycopeptides, CTD-GlcNAc (N-YSPTS(Glc- added to the pooled fractions at a final concentration of 1 mM. The NAc)PSK-C) or CTD-GalNAc (N-YSPTS(GalNAc)PSK-C), were synthe- preparation was then applied to a ConA column (60 ml) equilibrated in sized by standard Fmoc (N-(9-fluorenyl)methoxycarbonyl) chemistry. ConA buffer (20 mM sodium phosphate, pH 7.5, 5 mM 2-mercapoethanol, The peptides were purified on a C18 column under reversed phase high 150 mM NaCl, 1 mM MgCl ). The column was washed with 200 ml of pressure liquid chromatography conditions and used as a substrate for ConA buffer. The flow-through and the wash were combined. cloned O-GlcNAcase. The reaction products were analyzed by matrix- Step 5. Affinity Blue A Chromatography—The enzyme solution from assisted laser desorption ionization-time of flight (MALDI-TOF). Step 4 was concentrated by 60% ammonium sulfate precipitation, dia- Overexpression and Purification of O-GlcNAcase from Cos-7 Cells—A lyzed, and applied three times to a Blue A-Sepharose column (25 ml) plasmid of pcDNA3.1His A containing the full-length O-GlcNAcase equilibrated in buffer A. Again the activity was present in the flow- cDNA was prepared using a Qiagen kit (Qiagen, Valencia, CA). Trans- through fraction. The protein was pooled and clarified by fection was mediated by LipofectAMINE Plus (Life Technologies Inc., centrifugation. Gaithersburg, MD) using 50 –90% confluent Cos-7 cells. Cells were Step 6. Re-chromatography on DE52 Column—The sample from Step harvested 2 days post-transfection and sonicated for 2 3 12sin20mM 5 was injected to the DE52 cellulose column (same size as above), and Tris (pH 7.5), 10% glycerol, 150 mM NaCl, 1 mM dithiothreitol, 0.1 mM protein was eluted with a linear gradient of 50 –350 mM NaCl in 4 liters EDTA, 1 mM PMSF and protease inhibitor mixture, and clarified by of buffer A. Activity was recovered as in Step 3 and precipitated with centrifugation. For characterizations, the recombinant protein was pu- 60% ammonium sulfate. The pellet was resuspended in 20 ml of Mono-Q rified over a nickel affinity column. buffer (20 mM Tris, pH 7.5, 5 mM 2-mercaptoethanol, 10% glycerol, 1 mM Cell Fractionation and Western Blot—After transfection with O- EDTA plus protease inhibitor mixture (22) and 1 mM PMSF), dialyzed, GlcNAcase, Cos-7 cells were separated into cytoplasmic and nuclear and clarified by centrifugation. fractions as described (26) with one modification: The 25- to 50-ml Step 7. Native Polyacrylamide Gel Electrophoresis—Native PAGE nuclear pellet was carefully washed in 500 ml of hypotonic buffer A to was performed using a preparative Prepcell apparatus (Bio-Rad, Her- minimize cross-contamination. Immunoblot analysis was performed us- cules, CA). The sample from Step 6 was divided into three equal vol- ing antibodies recognizing the nuclear protein retinoblastoma (Rb) umes (45 mg of protein each) and loaded batch-wise onto a 6% native (Santa Cruz Biotechnology, Santa Cruz, CA), cytoplasmic protein a-tu- polyacrylamide gel (5-cm-long separating gel). The gel was run for 24 h bulin (Sigma Chemical Co., St. Louis, MO), or with anti-Xpress anti- at 12 watts of constant power. Protein was eluted in Mono-Q buffer at body, which is specific for the sequence DLYDDDDK located at the N a flow rate of 0.75 ml/min. 5-min fractions were collected and assayed terminus of the overexpressed O-GlcNAcase fusion protein (Invitrogen, for protein content and enzyme activity. Carlsbad, CA). Step 8. Mono-Q Chromatography—The O-GlcNAcase-containing Northern Blot—Northern blot analyses were performed on a human fractions from each native PAGE run was resolved on a Mono-Q column multiple tissue Northern blot (CLONTECH) using the manufacturer’s (HR10/10) with a linear gradient of 0 –500 mM NaCl in 450 ml of protocol. To prepare an O-GlcNAcase-specific probe, the full-length Mono-Q buffer at a flow rate of 3 ml/min. Fractions (4.5 ml) rich in coding sequence (2.75 kb) was amplified by PCR and labeled by random O-GlcNAcase activity were pooled and then separated for a second time primer using [a- P]dCTP (Stratagene, La Jolla, CA). After stripping in on the Mono-Q column. The final preparation was concentrated using 0.5% SDS at 100 °C for 10 min, the blot was reprobed for b-actin. Millipore concentrators to a final volume of 0.4 ml. Glycerol (40% final) and1mM PMSF and protease inhibitor mixture were added to the preparation. The enzyme was stored at 220 °C. RESULTS Identification of Proteins by Mass Spectrometry—The final prepara- Native Polyacrylamide Gel Electrophoresis Aides in O- tion from Step 8 was separated by 10% SDS-PAGE and stained with GlcNAcase Purification from Bovine Brain—Historically, O- Coomassie Blue G-250 or with silver. The desired protein bands were GlcNAcase, or neutral hexosaminidase C, has been difficult to excised individually, reduced, alkylated, and digested in-gel with mod- ified trypsin (Worthington, Freehold, NJ) as described previously (23). purify. For example, an early report described a purification of 9840 Cloning of O-GlcNAcase from Human Brain FIG.2. Silver staining and protein identification by mass spec- trometry. The final purification was separated on 10% SDS-PAGE and silver stained. Each protein band was individually excised and digested FIG.1. Purification steps of O-GlcNAcase from bovine brain. with modified trypsin, and the resultant peptides were separated and The cytosolic fraction of bovine brain proteins were sequentially puri- sequenced by liquid chromatography-MS/MS. Seven to 22 peptides fied by precipitation with 30 –50% ammonium sulfate, DE52 ion ex- were identified for each protein. change, ConA, and Blue A affinity columns, then re-separated on DE52 cellulose ion exchange chromatography (a), native polyacrylamide gel activity loss (data not shown). electrophoresis (b), and Mono-Q ion exchange chromatography (c). The protein profiles in a and c were continuously monitored by absorbance Mass Spectrometry Identifies O-GlcNAcase on SDS-PAGE at 280 nm with a UV detector during purification on a fast protein Gel—The seven bands on the silver-stained SDS-PAGE gel liquid chromatography system, and are presented by solid lines without were excised individually, digested with trypsin, and se- symbols.In b, the protein contents were manually determined by Bio- quenced by electrospray MS/MS. The fragmentation data were Rad assay (A ) for each fraction after the protein was eluted from the native gel. Therefore, it is presented with symbols. See “Experimental used to search protein and DNA data bases. This approach Procedures” for details. l, protein profile; Œ, O-GlcNAcase activity. identified six proteins with known functions in six of the bands (Fig. 2). Another protein, which runs as a 130-kDa band on only 25- to 40-fold from bovine brain despite the extensive use SDS-PAGE, is a hypothetical protein without any clearly of chromatographic steps (27). However, in rat brain, the en- defined functions (GenBank, KIAA0679). BLAST searches in- zyme has been purified over 2000-fold to a major band (28). dicated that the hypothetical protein shared significant homol- More recently, renewed effort has gone into its purification ogy with a protein from C. elegans called “similar to hyalurono- from rat spleen and bovine brain in the pursuit to cloning the glucosaminidase” (GenBank, AAA68333.1). Because hyaluro- cDNA (21, 29). noglucosaminidase degrades hyaluronic acid, which is a We have taken an approach, partly based on published lit- GlcNAcb1– 4GlcUA polymer, it was possible that a hyalurono- erature (21) but yet incorporating some novel steps in the glucosaminidase may share some homology with O-GlcNAcase. purification of O-GlcNAcase from bovine brain. Notably, we Furthermore, careful comparisons of O-GlcNAcase activity and have discovered that the enzymatic activity survives the harsh protein patterns on the SDS gels of different pools during the conditions of native PAGE (high pH and high ionic strength) purification procedure indicated that this protein was one of and migrates more slowly (R 5 0.28 in 6% native gel) than only two bands that corresponded with activity (the other band most other proteins in the gel (data not shown). This property was Protein 1 in Fig. 2, data not shown). We therefore hypoth- allows the effective separation of O-GlcNAcase from other pro- esized that this may be the O-GlcNAcase, and cloned the cDNA. teins with higher R values on a preparative scale native gel Further characterization of the expression product of the cDNA (Fig. 1b). This step, in conjunction with other chromatographic confirmed that band 2 on Fig. 2 was, indeed, O-GlcNAcase (see steps outlined in the protocol, purified the protein ;1500-fold, below). with a specific activity of 1840 nmol/min/mg of protein. O-GlcNAcase Is Unique and Conserved during Evolution— The final preparation still shows seven well defined bands on BLAST searches of data bases reveal that O-GlcNAcase is SDS-PAGE following silver staining, even after extensive conserved in higher eukaryotic species, and the homologue is purification (Fig. 2). We do not understand the basis for this absent in yeast or prokaryotes. The sequences and alignment of difficulty, but we have observed that the peaks for O-GlcNA- O-GlcNAcase from human, C. elegans, and Drosophila are case activity are very broad throughout the purification proce- shown in Fig. 3. In a pairwise alignment, the human sequence dure (Fig. 1). One example of this is illustrated in the Mono-Q shares 55 and 43% homology with that of Drosophila and C. step, where the general protein peaks are sharp but yet the elegans, respectively, whereas Drosophila and C. elegans are activity peak spreads over 50 ml (Fig. 1c). We have also tried 43% similar. Close inspection of the sequences indicates that ion exchange on Superose Q and hydroxylapatite columns or the N-terminal ;400 and the C-terminal ;350 amino acids in hydrophobic interaction chromatography on a phenyl-Sepha- the human sequence are conserved to a higher degree. These rose column. They, too, give poor separations or the enzyme two domains are separated by a highly variable region of ;150 binds very tightly to phenyl-Sepharose resulting in .50% amino acids. Another feature is that most of the aromatic Cloning of O-GlcNAcase from Human Brain 9841 FIG.3. The predicted amino acid se- quence of O-GlcNAcase and its align- ment with hypothetical proteins in C. elegans (“similar to hyaluronoglu- cosaminidase”; AAA68333.1) and Dro- sophila (AAF55867.1). The alignment was done with ClustalW in MacVector. Identical and conservative residues are indicated by asterisks and dots, respec- tively. The underlined sequences are the typtic peptides that are identified by liquid chromatography-MS/MS. Boldface and underlined letters indicate conserved aromatic residues in the three species. residues are conserved among the species. For example, out of that the cloned cDNA indeed encoded O-GlcNAcase, we sub- the 13 Trp residues found in the human sequence, 9 are invari- cloned the entire coding region in-frame into the mammalian ant in Drosophila and C. elegans, two are conservative (substi- expression vector pcDNA3.1His and overexpressed for activity tuted by Tyr or Phe), and only two are variable. in Cos-7 cells. Transient transfection resulted in a 6-fold in- The O-GlcNAcase sequence is conserved at a strikingly crease in O-GlcNAcase activity over endogenous activity in the higher level in mammals. Four overlapping EST sequences cells (Fig. 4a). After nickel affinity purification, the activity from cow, which cover 46% of the human protein, show that from the O-GlcNAcase-transfected cells was 230 nmol/min/mg these two species are 100% identical in these regions of protein but was not detectable from control transfected cells (BE481597, BE588694, BF043559, and AW463869). Five EST (Fig. 4b). These data show that the activity is due to overex- entries for mouse, most of which are overlapping, show that pression from the plasmid. Fig. 4c shows that a distinct band of human and mouse are 97.8% identical (AW907793, AW324047, the correct molecular mass (135 kDa) was isolated after nickel AI530529, AW762257, and AA240394). In the case of zebrafish, purification from transfected cells. This band was immunore- two overlapping EST sequences covering 33.8% of the human active with the Xpress antibody, which was specific for a pep- protein indicate that zebrafish and human are 85% identical tide sequence in the overexpressed protein. and 92% similar (AI882982 and AI722710). Recombinant O-GlcNAcase Has Distinct Properties from Apart from the above-described homologues, O-GlcNAcase Lysosomal b-Hexosaminidase—We further characterized the does not show significant homology with any other proteins, properties of the cloned O-GlcNAcase and compared them with including known glycosidases. Short stretches of ;200 amino those of lysosomal b-hexosaminidase purified from bovine kid- acids of the polypeptide do show loose homology to a number of ney. As expected, the lysosomal b-hexosaminidase had an proteins such as hyaluronidase (AAA23259.1), a putative acetyl- acidic pH optimum (pH 3.5–5.5) with little activity at pH 7.0 or transferase (AL158057), eukaryotic translation elongation fac- above (Fig. 5a). On the other hand, the cloned O-GlcNAcase tor-1g (Z11531, S26649), and the 11-1 polypeptide (X07453, had a pH optimum of 5.7–7.0 and retained significant activity S00485). Sequence analyses by a computer program PSORT II (;30%) at pH 7– 8. This pH profile is consistent with the (available on the Web) show that O-GlcNAcase does not possess expected localization of O-GlcNAcase in the cytoplasm and the any known signal peptides, domains, or motifs. The analyses do, nucleus. however, suggest that the endogenous protein is localized in the The two enzymes also responded differently to inhibitors. cytoplasm (p 5 0.522) and the nucleus (p 5 0.391). GalNAc, a widely used inhibitor of acidic b-hexosaminidase, Overexpression of O-GlcNAcase in Cos-7 Cells—To ascertain inhibited the lysosomal enzyme 50% at 5.0 mM and 88% at 50 9842 Cloning of O-GlcNAcase from Human Brain FIG.6. Substrate specificity of cloned O-GlcNAcase. pNP-b- GlcNAc, pNP-b-GalNAc, and pNP-a-GlcNAc (all 2 mM) were tested as substrates for purified O-GlcNAcase from transfected Cos-7 cells. The lysosomal hexosaminidase was used as a control. The assays were done at pH 4.5 for the lysosomal enzyme and at pH 6.5 for O-GlcNAcase. The activity using pNP-b-GlcNAc as substrate is set as 100%. somal b-hexosaminidase in substrate requirements. In the in vitro assays, purified recombinant O-GlcNAcase cleaved only pNP-b-GlcNAc, but not pNP-b-GalNAc or pNP-a-GlcNAc (Fig. FIG.4. Overexpression of O-GlcNAcase in Cos-7 cells. Cos-7 6). The activity using the latter two compounds as substrates cells were transiently transfected with the vector pcDNA3.1His alone or was not detectable. This substrate specificity was in contrast to with pcDNA3.1His containing O-GlcNAcase cDNA using Lipo- the lysosomal b-hexosaminidase, which also cleaved pNP-b- fectAMINE Plus. 48 h post-transfection, soluble proteins were isolated from the cells. a, O-GlcNAcase activity in total cell extract. The activity GalNAc, albeit with slightly lower efficiency compared with in control transfected cells is arbitrarily set as one-fold. b, O-GlcNAcase pNP-b-GlcNAc. activity in control and transfected cells after nickel purification. c, O-GlcNAcase Cleaves O-GlcNAc from Glycopeptides—A true Western blot and Coomassie Blue G-250 staining of nickel affinity- O-GlcNAcase should cleave O-GlcNAc attached to proteins or purified O-GlcNAcase from transfected cells. The primary antibody was peptides. We synthesized two glycopeptides containing one anti-Xpress, which recognized a peptide sequence (DLYDDDDK) pres- ent in the vector pcDNA3.1His. The arrow indicates the band of repeat of the (CTD) C- terminal domain of RNA polymerase II O-GlcNAcase. linked to b-GlcNAc or a-GalNAc through the hydroxyl group of a serine residue (CTD-GlcNAc or CTD-GalNAc). The design of these glycopeptides is based on earlier information that this serine residue is glycosylated in vivo (13). These peptides were tested as substrates for the purified recombinant O-GlcNAcase, and the product peptides were analyzed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF). Cleav- age of GlcNAc or GalNAc from the peptides should result in a downshift of 203 in the molecular weight (the weight of GlcNAc or GalNAc minus 18). O-GlcNAcase did not cleave GalNAc from the peptide (Fig. 7, a and b) but did successfully cleave GlcNAc from the peptide, as judged by the expected shift in molecular weight (1069.8 to 866.6, 1091.8 to 888.6) (Fig. 7, c and d). Overexpressed O-GlcNAcase Is Localized to the Cyto- plasm—As stated earlier, sequence analyses suggest that O- GlcNAcase is localized in the cytoplasm and nucleus. This is consistent with the localization of the O-GlcNAc modification. To obtain direct evidence on its localization, we performed cellular fractionation and assayed O-GlcNAcase activity in the cytoplasm and the nucleus. The data show that, in nontrans- fected cells, O-GlcNAcase activity was distributed in both the FIG.5. Some key properties of the cloned O-GlcNAcase. A lyso- somal b-hexosaminidase purified from bovine kidney was used as a cytoplasm and the nucleus. However, when O-GlcNAcase was control enzyme. a, pH optima of the two enzymes. Inhibition studies by overexpressed, it was predominantly found in the cytoplasm GalNAc (b), GlcNAc (c), and PUGNAc (d). In b, c, and d, the activity was (Fig. 8a). We also probed for its localization by Western blots assayed at pH 4.5 for the lysosomal hexosaminidase and pH 6.5 for the (Fig. 8b). Retinoblastoma protein (Rb) and a-tubulin, which O-GlcNAcase. The highest activity at optimal pH (a) or in the absence were exclusively localized in the nucleus and cytoplasm, re- of any inhibitor (b, c, d) is arbitrarily set as 100%. l, O-GlcNAcase; ƒ, lysosomal b-hexosaminidase. spectively, were used as markers. In agreement with activity assays, overexpressed O-GlcNAcase protein was only detected mM. The cloned O-GlcNAcase was not inhibited at all by Gal- in the cytoplasm of overexpressed cells. NAc up to 50 mM (Fig. 5b). GlcNAc and its synthetic analogue O-GlcNAcase Has One Transcript in Human Tissues—To PUGNAc inhibited both enzymes but were more potent with estimate the number of transcripts of O-GlcNAcase and their the O-GlcNAcase (Fig. 5, c and d). levels of expression, we performed Northern analyses using a Recombinant O-GlcNAcase Shows Strict Substrate Specific- human multiple tissue blot consisting of RNA from eight dif- ity for b-Linked GlcNAc—O-GlcNAcase also differed from lyso- ferent tissues. Labeled PCR product of the entire coding region Cloning of O-GlcNAcase from Human Brain 9843 FIG.7. Recombinant O-GlcNAcase cleaves O-GlcNAc, but not O-GalNAc, from glycopeptides. Sythetic peptides, CTD-GlcNAc (N-YSPTS(GlcNAc)PSK-C) or CTD-GalNAc (N-YSPTS(GalNAc)PSK- C) were tested as substrates for purified recombinant O-GlcNAcase. The reactions containing 0.1 m M peptide were incubated at pH 6.5, 37 °C overnight. The mock re- actions were done using nickel purifica- tion from nontransfected Cos-7 cells. The peptides were then cleaned up by zip tips and analyzed by MALDI-TOF. The values 1091.7 (or 1091.8) and 1069.7 (or 1069.8) represented molecular weights of the Na and H form of the peptides, respectively. After the GlcNAc was cleaved in CTD- GlcNAc peptide by O-GlcNAcase, the mass expectedly shifted down to 888.6 and 866.6, respectively. The numbers on the right-hand side of each spectrum (2744, 2930, 5158, and 2100) are the total ion counts recorded by the detector in MALDI-TOF analysis. FIG.9. O-GlcNAcase transcript is expressed in every human tissue examined but is the highest in the brain, placenta, and pancreas. a, a human multiple tissue Northern blot was probed with [a- P]dCTP-labeled full-length O-GlcNAcase coding sequence and ex- posed to film for 18 h. b, after stripping, the blot was reprobed for b-actin and exposed to film for 7 h. FIG.8. Overexpressed O-GlcNAcase is distributed in the cyto- localized on chromosome 10 (Ref. 25, and available on the Web plasm. Cos-7 cells were transfected with O-GlcNAcase. Two days post- transfection, the cells were fractionated into cytoplasmic or nuclear from Kazusa DNA Research Institute, Japan). fractions. a, O-GlcNAcase activity (n 5 3). Assays were done with 100 DISCUSSION mM GalNAc as inhibitor. b, Western blots with anti-Xpress (for recom- binant O-GlcNAcase detection), anti-a-tubulin (cytoplasmic marker), Two categories of b-hexosaminidases are known to exist in and anti-retinoblastoma (Rb) (nuclear marker). eukaryotic cells. One category, which comprises the A, B, I, S isozymes, is exclusively localized in the lysosomes and are of the O-GlcNAcase cDNA was used as a probe. The Northern responsible for the degradation of complex glycans. This group blot analysis showed only one transcript of ;5.5 kb for O- of hexosaminidases, particularly the A and B isozymes, have GlcNAcase (Fig. 9). Exposing the film for extended time (32 h) been extensively studied because their deficiency leads to Tay- did not reveal any additional bands (data not shown). The gene Sachs and Sandhoff diseases (30, 31). These enzymes are char- was expressed in every tissue on the blot but was the highest in acterized by their acidic pH optima, inhibition by both GlcNAc the brain, followed by placenta, and pancreas. Lung and liver and GalNAc, the ability to use both artificial glucosaminide had the lowest expression (Fig. 9). This pattern of expression and galactosaminide as substrates, and their thermostability largely agrees with that of the ogt gene. ogt is expressed the (32). The second category, however, consists of two neutral highest in the pancreas, followed by heart, brain, and placenta hexosaminidases, the GlcNAc-specific glucosaminidase (hex- but the lowest in lung, liver, and kidney (18). Unlike ogt gene, osaminidase C) and the GalNAc-specific galactosaminidase which is on the X chromosome (20), the O-GlcNAcase gene is (hexosaminidase D) (28, 33, 34). In contrast to lysosomal hex- 9844 Cloning of O-GlcNAcase from Human Brain osaminidases, they reside in the cytosol, have neutral pH op- ing an O-GlcNAcase-specific polyclonal antibody based on its tima, and are heat labile. Despite the wide occurrence of these protein sequence. Immunofluorescence microscopic studies will neutral isoforms in tissues, their natural substrates have not determine the relative distribution of the endogenous protein. been previously identified (28). The cloning of O-GlcNAcase provides a valuable tool to fur- ther study the biological function of O-GlcNAc modification and The discovery of O-GlcNAc provides clues to the likely phys- iological function of neutral glucosaminidase (hexosaminidase may help to understand the mechanisms of a number of prev- alent diseases such as the Alzheimer’s and diabetes. Recent C) (2). This modification consists of the addition of a single data suggest that many proteins in the brain, including Syn- GlcNAc moiety to the hydroxyl group of Ser/Thr residues (2). apsin I and neurofilaments (6, 37), are O-glycosylated. Some of The enzyme responsible for GlcNAc addition, OGT, was iden- the glycosylated proteins may be involved in the development tified and cloned several years ago (17–19). On the other hand, of the Alzheimer’s diseases. For example, Tau, whose hyper- the identity of the enzyme for the removal of O-GlcNAc from phosphorylation leads to the formation of neurofibrillary tan- proteins, O-GlcNAcase, has remained obscure. Judged from the gles in the neurons of Alzheimer’s brains, is extensively O- known properties of hexosaminidase C, we speculate that it glycosylated (38). The b-amyloid precursor protein, which gives might in fact be the O-GlcNAcase. We thus have extensively rise to the neurotoxic b-amyloid peptide in Alzheimer’s brains, purified hexosaminidase C from bovine brains. Proteomic analy- is also O-GlcNAcylated (39). Furthermore, Alzheimer’s disease sis allowed us to identify its DNA sequence. The sequence was has been correlated to the glycosylation of at least one protein, originally isolated during screening of a meningioma expres- the chathrin assembly protein-3 (40). Continuing efforts in sion library (25). Based on its initial characterization and se- several laboratories are underway to understand the role of quence homology with an unidentified protein called “similar to O-GlcNAc in the development of such neurodegenerative dis- hyaluronoglucosaminidase” from C. elegans, the authors sug- eases. In addition, strong evidence suggests that O-GlcNAc is gested that it was a new type of hyaluronidase. We have here involved in the development of insulin resistance in diabetes demonstrated that the cloned enzyme has all the expected mellitus (4, 41). Infusion of glucosamine or the overexpression properties of O-GlcNAcase. Most importantly, it specifically of glutamine:fructose-6-phosphate amidotransferase in animal cleaves O-GlcNAc but not O-GalNAc from glycopeptides. We models, both of which increase cellular UDP-GlcNAc levels conclude that we have cloned hexosaminidase C, which we through the hexosamine synthetic pathway (42, 43), leads to have called O-GlcNAcase herein. insulin resistance (44, 45). The cloning of O-GlcNAcase will not The recombinant protein differs from acidic lysosomal hex- only allow the selective disruption of the gene, or tissue-specific osaminidase in a number of ways. It has a neutral, instead of overexpression of the protein in animal models, but will also acidic, pH optimum. This pH optimum is expected due to its make it possible to directly regulate and monitor O-GlcNAc- physiological functions in the cytoplasm and nucleus. The pro- modified proteins and thus facilitate the understanding of the tein displays strict substrate requirement in the linkage of the role of this modification in the development of these diseases. sugar and is also sensitive to the C4 orientation of the sugar. As such, it hydrolyzes only b-linked GlcNAc but is totally ineffec- Acknowledgments—We thank the Kazusa DNA Research Institute, tive toward a-linked GlcNAc or b-linked GalNAc. This sub- Japan for providing the KIAA0679 cDNA and Dr. N. Zachara for read- ing the manuscript. strate specificity is in contrast with that of lysosomal hex- osaminidases, which hydrolyze both GlcNAc and GalNAc REFERENCES substrates. It also determines that GlcNAc, but not GalNAc, is 1. Torres, C. R., and Hart, G. W. (1984) J. Biol. Chem. 259, 3308 –3317 2. Hart, G. W. (1997) Annu. Rev. Biochem. 66, 315–335 a competitive inhibitor of O-GlcNAcase. Unlike lysosomal het- 3. Snow, D. M., and Hart, G. W. (1998) Int. Rev. Cytol. 181, 43–74 ero-oligmeric hexosaminidases A (ab b )orB(2xb b ) (32), the a b a b 4. Comer, F. I., and Hart, G. W. (2000) J. Biol. Chem. 275, 29179 –29182 cloned human O-GlcNAcase has only one polypeptide of 916 5. Chou, T. Y., Hart, G. W., and Dang, C. V. (1995) J. Biol. Chem. 270, 18961–18965 amino acids with a predicted molecular mass of 103 kDa. The 6. Cole, R. N., and Hart, G. W. (1999) J. Neurochem. 73, 418 – 428 apparent size on SDS-PAGE is 130 kDa, which is identical to 7. Cheng, X., Cole, R. N., Zaia, J., and Hart, G. W. (2000) Biochemistry 39, 11609 –11620 the glucosaminidase purified from rat brain (28). The oligo- 8. Griffith, L. S., and Schmitz, B. (1999) Eur. J. Biochem. 262, 824 – 831 meric status of the hexosaminidase C is still not known, be- 9. Lefebvre, T., Alonso, C., Mahboub, S., Dupire, M. J., Zanetta, J. P., Caillet-Boudin, M. L., and Michalski, J. C. (1999) Biochim. Biophys. Acta cause gel filtration, SDS-PAGE, and sucrose density gradient 1472, 71– 81 centrifugation give inconsistent estimations of its native mo- 10. Jackson, S. P., and Tjian, R. (1988) Cell 55, 125–133 lecular size (28). 11. Reason, A. J., Morris, H. R., Panico, M., Marais, R., Treisman, R. H., Haltiwanger, R. S., Hart, G. W., Kelly, W. G., and Dell, A. (1992) J. Biol. O-GlcNAcase is a unique protein with no other obvious fam- Chem. 267, 16911–16921 ily members. It is conserved in evolution down to the nematode 12. Jiang, M. S., and Hart, G. W. (1997) J. Biol. Chem. 272, 2421–2428 13. Kelly, W. G., Dahmus, M. E., and Hart, G. W. (1993) J. Biol. Chem. 268, C. elegans, and shows striking homology in mammals. It does 10416 –10424 not, however, possess any known signal peptides, domains, or 14. Kelly, W. G., and Hart, G. W. (1989) Cell 57, 243–251 15. Han, I., and Kudlow, J. E. (1997) Mol. Cell. Biol. 17, 2550 –2558 motifs. Consistent with the literature that antibodies that rec- 16. Du, W. L., Edelstein, D., Rossetti, L., Fantus, I. G., Goldberg, H., Ziyadeh, F., ognize lysosomal hexosaminidases A or B do not cross-react Wu, J., and Brownlee, M. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, with hexosaminidase C (35), O-GlcNAcase does not share any 12222–12226 17. Haltiwanger, R. S., Blomberg, M. A., and Hart, G. W. (1992) J. Biol. Chem. significant homology with these enzymes, or with any other 267, 9005–9013 known proteins. 18. Lubas, W. A., Frank, D. W., Krause, M., and Hanover, J. A. (1997) J. Biol. Chem. 272, 9316 –9324 The predicted localization of O-GlcNAcase by sequence analy- 19. Kreppel, L. K., Blomberg, M. A., and Hart, G. W. (1997) J. Biol. Chem. 272, sis is consistent with the literature that O-GlcNAc modification 9308 –9315 20. Shafi, R., Iyer, S. P., Ellies, L. G., O’Donnell, N., Marek, K. W., Chui, D., Hart, and OGT are found both in the cytoplasm and nucleus (19, 36). G. W., and Marth, J. D. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 5735–5739 It is therefore somewhat unexpected to find that overexpressed 21. Dong, D. L., and Hart, G. W. (1994) J. Biol. Chem. 269, 19321–19330 O-GlcNAcase in Cos-7 cells is almost exclusively localized to 22. Roquemore, E. P., Chou, T. Y., and Hart, G. W. (1994) Methods Enzymol. 230, 443– 460 the cytoplasm (Fig. 8). Nevertheless, it is not uncommon that 23. Gharahdaghi, F., Weinberg, C. R., Meagher, D. A., Imai, B. S., and Mische, overexpressed proteins often form aggregates in the cytoplasm S. M. (1999) Electrophoresis 20, 601– 605 24. Ducret, A., Van, O.ostveen, I., Eng, J. K., Yates, J. R., and Aebersold, R. (1998) and cannot be correctly translocated. We cannot exclude the Protein Sci. 7, 706 –719 possibility of an unidentified isoform of O-GlcNAcase that is 25. Heckel, D., Comtesse, N., Brass, N., Blin, N., Zang, K. D., and Meese, E. (1998) specifically localized to the nucleus. We are currently generat- Hum. Mol. Genet. 7, 1859 –1872 Cloning of O-GlcNAcase from Human Brain 9845 26. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic 37. Dong, D. L., Xu, Z. S., Chevrier, M. R., Cotter, R. J., Cleveland, D. W., and Acids Res. 17, 6419 Hart, G. W. (1993) J. Biol. Chem. 268, 16679 –16687 27. Overdijk, B., van der Kroef, W. M. J., van Steijn, G. J., and Lisman, J. W. 38. Arnold, C. S., Johnson, G. V., Cole, R. N., Dong, D. L. Y., Lee, M., and Hart, (1981) Biochim. Biophys. Acta 659, 255–266 G. W. (1996) J. Biol. Chem. 271, 28741–28744 28. Izumi, T., and Suzuki, K. (1983) J. Biol. Chem. 258, 6991– 6999 39. Griffith, L. S., Mathes, M., and Schmitz, B. (1995) J. Neurosci. Res. 41, 29. Hanover, J. A., Lai, Z., Lee, G., Lubas, W. A., and Sato, S. M. (1999) Arch. 270 –278 Biochem. Biophys. 362, 38–45 40. Yao, P. J., and Coleman, P. D. (1998) J. Neurosci. 18, 2399 –2411 30. Kustermann-Kuhn, B., and Harzer, K. (1983) Hum. Genet. 65, 172–175 41. McClain, D. A., and Crook, E. D. (1996) Diabetes 45, 1003–1009 31. Neufeld, E. F. (1989) J. Biol. Chem. 264, 10927–10930 42. Yki-Jarvinen, H., Virkamaki, A., Daniels, M. C., McClain, D., and Gottschalk, 32. Emiliani, C., Sciarra, R., Orlacchio, A., and Stirling, J. L. (1990) Biochim. W. K. (1998) Metabolism 47, 449 – 455 Biophys. Acta 1037, 265–273 43. Cooksey, R, C., Hebert, L. F., Jr, Zhu, J. H., Wofford, P., Garvey, W. T., and 33. Overdijk, B., van der Kroef, W. M. J., Veltkamp, W. A., and Hooghwinkel, McClain, D. A. (1999) Endocrinology 140, 1151–1157 G. J. M. (1975) Biochem. J. 151, 257–261 44. Hebert, L. F., Jr, Daniels, M. C., Zhou, J., Crook, E. D., Turner, R. L., 34. Izumi, T., and Suzuki, K. (1980) Biochim. Biophys. Acta 615, 402– 413 Simmons, S. T., Neidigh, J. L., Zhu, J. S., Baron, A. D., and McClain, D. A. 35. Poenaru, L., Weber, A., Vibert, M., and Dreyfus, J. C. (1973) Biomedicine 19, (1996) J. Clin. Invest. 98, 930 –936 538 –540 45. Baron, A. D., Zhu, J. S., Zhu, J. H., Weldon, H., Maianu, L., and Garvey, W. T. 36. Akimoto, Y., Kreppel, L. K., Hirano, H., and Hart, G. W. (1999) Diabetes 48, 2407–2413 (1995) J. Clin. Invest. 96, 2792–2801 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Dynamic O-Glycosylation of Nuclear and Cytosolic Proteins

Gao, Yuan; Wells, Lance; Comer, Frank I.; Parker, Glendon J.; Hart, Gerald W.
Journal of Biological Chemistry – Mar 1, 2001
Download PDF
8 pages

Loading next page...
 
/lp/unpaywall/dynamic-o-glycosylation-of-nuclear-and-cytosolic-proteins-BfQDrbbUXY

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.m010420200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 13, Issue of March 30, pp. 9838 –9845, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. CLONING AND CHARACTERIZATION OF A NEUTRAL, CYTOSOLIC b-N-ACETYLGLUCOSAMINIDASE FROM HUMAN BRAIN* Received for publication, November 16, 2000, and in revised form, January 4, 2001 Published, JBC Papers in Press, January 8, 2001, DOI 10.1074/jbc.M010420200 Yuan Gao‡, Lance Wells‡, Frank I. Comer‡, Glendon J. Parker§, and Gerald W. Hart‡¶ From the ‡Department of Biological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205-2185 and the §Department of Internal Medicine, Division of Endocrinology, Metabolism and Diabetes, The University of Utah, Salt Lake City, Utah 84132 Dynamic modification of cytoplasmic and nuclear pro- cover a broad range, including many transcription factors, RNA polymerase II, oncogenes, nuclear pore proteins, viral proteins, teins by O-linked N-acetylglucosamine (O-GlcNAc) on Ser/Thr residues is ubiquitous in higher eukaryotes and and tumor repressors (for details, see Refs. 2 and 3 and cita- is analogous to protein phosphorylation. The enzyme for tions within). Unlike classic O- or N-linked protein glycosyla- the addition of this modification, O-GlcNAc transferase, tions, the O-GlcNAc modification involves only a single GlcNAc has been cloned from several species. Here, we have moiety linked to the hydroxyl group of Ser/Thr residues, gen- cloned a human brain O-GlcNAcase that cleaves O-Glc- erally is not elongated, and is found exclusively in the cyto- NAc off proteins. The cloned cDNA encodes a polypep- plasm and nucleoplasm. tide of 916 amino acids with a predicted molecular mass Protein O-GlcNAcylation is highly dynamic, and the cycle of of 103 kDa and a pI value of 4.63, but the protein mi- addition/removal of the sugar moiety is rapid, analogous to grates as a 130-kDa band on SDS-polyacrylamide gel protein phosphorylation/dephosphorylation catalyzed by ki- electrophoresis. The cloned O-GlcNAcase has a pH opti- nases and phosphatases (2). Indeed, existing evidence suggests mum of 5.5–7.0 and is inhibited by GlcNAc but not by that this modification has a “yin-yang” relationship with pro- b-GlcNAc, but not pNP-b- GalNAc. p-Nitrophenyl (pNP)- tein phosphorylation in some cases (4). Many O-GlcNAcylation a-GlcNAc, is a substrate. The cloned en- GalNAc or pNP- sites have been mapped to phosphorylation sites or adjacent zyme cleaves GlcNAc, but not GalNAc, from glycopep- sites (5–7). Such spatial localization indicates that O-GlcNAc tides. Cell fractionation suggests that the overexpressed may regulate the target protein by competing with protein protein is mostly localized in the cytoplasm. It therefore kinases (4). Recent studies using phosphatase and kinase in- has all the expected characteristics of O-GlcNAcase and hibitors have provided direct evidence for a general reciprocal is distinct from lysosomal hexosaminidases. North- relationship between O-GlcNAcylation and phosphorylation on ern blots show that the transcript is expressed in some proteins (8, 9). every human tissue examined but is the highest in the O-GlcNAcylation appears to be involved in gene transcrip- brain, placenta, and pancreas. An understanding of O- GlcNAc dynamics and O-GlcNAcase may be key to elu- tion. Most transcription factors examined so far, including Sp1, cidating the relationships between O-phosphate and AP1, AP2, AP4 (10), serum response factor (11), the estrogen O-GlcNAc and to the understanding of the molecular receptor (7, 12), the insulin promoter factor-1, and peroxisome mechanisms of diseases such as diabetes, cancer, and proliferator-activated receptor-g, as well as RNA polymerase II neurodegeneration. (13) and chromatin (14) are O-GlcNAcylated. O-GlcNAcylation of Sp1 appears to enhance its activity in transcription, and, conversely, blocking the GlcNAc residues with lectin wheat Since the description of O-linked N-acetylglucosamine (O- germ agglutinin suppresses the transcriptional activity (10). GlcNAc) as an abundant modification in murine lymphocytes O-GlcNAcylation of Sp1 also controls its degradation by the (1), a myriad of cytoplasmic and nuclear proteins in all meta- proteasome (15). Hyperglycemia-induced superoxide produc- zoans have been found to carry this modification. Such proteins tion increases Sp1 glycosylation resulting in the activation of genes that contribute to the pathogenesis of diabetes (16). O-GlcNAc transferase (OGT), which transfers GlcNAc from * Supported by National Institutes of Health Grant HD13563. The the donor substrate UDP-GlcNAc to target proteins, has been costs of publication of this article were defrayed in part by the payment purified and cloned from several species including human, rat, of page charges. This article must therefore be hereby marked “adver- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate and Caenorhabditis elegans (17–19). It does not share any this fact. significant homology with any other known proteins, including To whom correspondence should be addressed: The Johns Hopkins glycosyltransferases, and is highly conserved from C. elegans to University School of Medicine, 725 North Wolfe St., Baltimore, MD human. Disruption of the ogt gene is lethal in mouse embryonic 21205-2185. Tel.: 410-614-5993; Fax: 410-614-8804; E-mail: [email protected]. stem cells, further underscoring the importance of O-GlcNAc The abbreviations used are: O-GlcNAc, O-linked N-acetylglu- modification in cellular functions (20). O-GlcNAcase, the en- cosamine; GalNAc, N-acetylgalactosamine; GlcNAc, N-acetylglu- zyme that removes O-GlcNAc from such proteins, was purified cosamine; OGT, O-GlcNAc transferase; O-GlcNAcase, N-acetyl-b-D-glu- several years ago from rat spleen (21). It is a neutral cytosolic cosaminidase; PAGE, polyacrylamide gel electrophoresis; PCR, b-glucosaminidase or hexosaminidase C (EC 3.2.1.52). To fur- polymerase chain reaction; pNP, p-nitrophenyl; PUGNAc, O-(2-acet- amido-2-deoxy-D-glucopyranosylidene)-amino-N-phenylcarbamate; bp, ther study the function of this modification, we have now ex- base pair(s); CTD, C-terminal domain; kb, kilobase(s); EST, expressed sequence tag; MALDI-TOF, matrix-assisted laser desorption ionization- time of flight; MS, mass spectrometry; ConA, concanavalin A; PMSF, phenylmethylsulfonyl fluroride. Y. Gao and G. W. Hart, manuscript in preparation . 9838 This paper is available on line at http://www.jbc.org This is an Open Access article under the CC BY license. Cloning of O-GlcNAcase from Human Brain 9839 The tryptic peptides from each protein were analyzed by capillary tensively purified O-GlcNAcase from bovine brain, sequenced reversed phase high pressure liquid chromatography with in-line tan- the protein by mass spectrometry, and cloned the cDNA. The dem MS/MS on a Finnigan LCQ ion-trap mass spectrometer. Proteins O-GlcNAcase is evolutionarily conserved, distinct from lysoso- were identified by the SEQUEST algorithm with sequencing of at least mal acidic hexosaminidases A and B. The recombinant protein seven tryptic peptides for each protein (24). has all the expected characteristics of O-GlcNAcase, including Cloning of O-GlcNAcase—A putative O-GlcNAcase with a theoretical the ability to cleave O-GlcNAc from glycopeptides. length of 916 amino acids in human was identified by the above pro- teomic approach. A cDNA fragment, KIAA0679 (GenBanky accession EXPERIMENTAL PROCEDURES number AB014579), which contains the coding sequence for 767 amino Purification of O-GlcNAcase from Bovine Brain—All chromato- acids of the C terminus of the human O-GlcNAcase and a 2.0-kb 39- graphic materials were purchased from Amersham Pharmacia Biotech untranslated region, was obtained from the Kazusa DNA Research (Piscataway, NJ). Purification procedure is substantially modified from Institute, Japan, in the vector pBluescript. The coding sequence of this that of Dong and Hart (21). All steps were conducted at 4 °C or on ice. fragment was subsequently transferred to pcDNA3.1His A using XhoI Step 1. Tissue Homogenization—Three bovine brains (;1 kg) were and XbaI, which were located within the polycloning cloning site of frozen in liquid N and shipped on dry ice from Pel Freez Biologicals pBluescript and in the 39-untranslated region of the cDNA, respec- (Rogers, AR) and stored at 280 °C until use. The brains were smashed tively. The missing 59-end fragment of the full-length coding cDNA (447 into smaller pieces and homogenized in 5 volumes (v/w) of homogeni- bp) was amplified by PCR from a human brain Marathon cDNA library zation buffer (20 mM sodium phosphate, pH 7.5, 15 mM 2-mecaptoetha- (CLONTECH, Palo Alto, CA) using the forward primer GGATGGTG- nol, 10 mM MgCl ,1mM PMSF, 1 mM EDTA) in a Hamilton Beach 2 CAGAAGGAGAGTCAAGCGAC and the reverse primer TAGAAAC- blender with 5 3 20-s bursts. The homogenate was centrifuged at CTCTTCGATGGACTCTACTGG. The forward primer sequence was 18,000 3 g for 30 min. The pellet was discarded, and the cytosolic based on published data (25), and the reverse primer was located in the supernatant was pooled in a 5-liter beaker. KIAA0679 clone. PCR conditions were 94 °C for 30 s, 63 °C for 30 s, and Step 2. Ammonium Sulfate Precipitation—The cytosolic supernatant 72 °C for 3 min for 30 –35 cycles. A second round of PCR using the first was subjected to 30 –50% ammonium sulfate precipitation. The pellet PCR product as template and a forward primer incorporating a NotI was resuspended in 500 ml of buffer A (20 mM sodium phosphate, pH site (CCGGGCGGCCGCGGATGGTGCAGAAGGAGAG) and the same 7.5, 5 mM 2-mercaptoethanol) and centrifuged to clarify the solution. reverse primer was performed. The product was digested with NotI and The solution was then thoroughly dialyzed against buffer A and centri- HindIII (unique site in the PCR product) and ligated in-frame into the fuged again to eliminate any insoluble materials that had resulted from pcDNA3.1His A-KIAA0679 construct. This gave rise to a full-length dialysis. cDNA in the vector pcDNA3.1His A. The final construct was sequenced. Step 3. DE52 Cellulose Ion Exchange Chromatography—The dia- O-GlcNAcase Assays—Unless stated otherwise, O-GlcNAcase activ- lyzed sample was loaded onto a DE52 column (900-ml bed vol) at a flow ity was assayed as described in 50 mM sodium cacodylate, pH 6.5, 0.3% rate of 2 ml/min using a peristaltic pump. After washing the column bovine serum albumin, 2 mM pNP-b-GlcNAc, 50 mM GalNAc (21). Pu- with 3 liters of buffer A, bound proteins were eluted with a linear rified bovine kidney lysosomal b-hexosaminidase (Roche Molecular Bio- gradient of 0 –1 M NaCl in 4 liters of buffer A at a flow rate of 4 ml/min. chemicals, Indianapolis, IN) activity was assayed in citrate phosphate The protein profile was monitored by absorbance at 280 nm. Fractions buffer, pH 4.5, 0.3% bovine serum albumin, 2 mM pNP-b-GlcNAc. To (16 ml) enriched in O-GlcNAcase activity were pooled. test the ability of recombinant O-GlcNAcase to cleave O-GlcNAc from Step 4. Concanavalin A-Sepharose 4B Chromatography—MgCl was glycopeptides, two glycopeptides, CTD-GlcNAc (N-YSPTS(Glc- added to the pooled fractions at a final concentration of 1 mM. The NAc)PSK-C) or CTD-GalNAc (N-YSPTS(GalNAc)PSK-C), were synthe- preparation was then applied to a ConA column (60 ml) equilibrated in sized by standard Fmoc (N-(9-fluorenyl)methoxycarbonyl) chemistry. ConA buffer (20 mM sodium phosphate, pH 7.5, 5 mM 2-mercapoethanol, The peptides were purified on a C18 column under reversed phase high 150 mM NaCl, 1 mM MgCl ). The column was washed with 200 ml of pressure liquid chromatography conditions and used as a substrate for ConA buffer. The flow-through and the wash were combined. cloned O-GlcNAcase. The reaction products were analyzed by matrix- Step 5. Affinity Blue A Chromatography—The enzyme solution from assisted laser desorption ionization-time of flight (MALDI-TOF). Step 4 was concentrated by 60% ammonium sulfate precipitation, dia- Overexpression and Purification of O-GlcNAcase from Cos-7 Cells—A lyzed, and applied three times to a Blue A-Sepharose column (25 ml) plasmid of pcDNA3.1His A containing the full-length O-GlcNAcase equilibrated in buffer A. Again the activity was present in the flow- cDNA was prepared using a Qiagen kit (Qiagen, Valencia, CA). Trans- through fraction. The protein was pooled and clarified by fection was mediated by LipofectAMINE Plus (Life Technologies Inc., centrifugation. Gaithersburg, MD) using 50 –90% confluent Cos-7 cells. Cells were Step 6. Re-chromatography on DE52 Column—The sample from Step harvested 2 days post-transfection and sonicated for 2 3 12sin20mM 5 was injected to the DE52 cellulose column (same size as above), and Tris (pH 7.5), 10% glycerol, 150 mM NaCl, 1 mM dithiothreitol, 0.1 mM protein was eluted with a linear gradient of 50 –350 mM NaCl in 4 liters EDTA, 1 mM PMSF and protease inhibitor mixture, and clarified by of buffer A. Activity was recovered as in Step 3 and precipitated with centrifugation. For characterizations, the recombinant protein was pu- 60% ammonium sulfate. The pellet was resuspended in 20 ml of Mono-Q rified over a nickel affinity column. buffer (20 mM Tris, pH 7.5, 5 mM 2-mercaptoethanol, 10% glycerol, 1 mM Cell Fractionation and Western Blot—After transfection with O- EDTA plus protease inhibitor mixture (22) and 1 mM PMSF), dialyzed, GlcNAcase, Cos-7 cells were separated into cytoplasmic and nuclear and clarified by centrifugation. fractions as described (26) with one modification: The 25- to 50-ml Step 7. Native Polyacrylamide Gel Electrophoresis—Native PAGE nuclear pellet was carefully washed in 500 ml of hypotonic buffer A to was performed using a preparative Prepcell apparatus (Bio-Rad, Her- minimize cross-contamination. Immunoblot analysis was performed us- cules, CA). The sample from Step 6 was divided into three equal vol- ing antibodies recognizing the nuclear protein retinoblastoma (Rb) umes (45 mg of protein each) and loaded batch-wise onto a 6% native (Santa Cruz Biotechnology, Santa Cruz, CA), cytoplasmic protein a-tu- polyacrylamide gel (5-cm-long separating gel). The gel was run for 24 h bulin (Sigma Chemical Co., St. Louis, MO), or with anti-Xpress anti- at 12 watts of constant power. Protein was eluted in Mono-Q buffer at body, which is specific for the sequence DLYDDDDK located at the N a flow rate of 0.75 ml/min. 5-min fractions were collected and assayed terminus of the overexpressed O-GlcNAcase fusion protein (Invitrogen, for protein content and enzyme activity. Carlsbad, CA). Step 8. Mono-Q Chromatography—The O-GlcNAcase-containing Northern Blot—Northern blot analyses were performed on a human fractions from each native PAGE run was resolved on a Mono-Q column multiple tissue Northern blot (CLONTECH) using the manufacturer’s (HR10/10) with a linear gradient of 0 –500 mM NaCl in 450 ml of protocol. To prepare an O-GlcNAcase-specific probe, the full-length Mono-Q buffer at a flow rate of 3 ml/min. Fractions (4.5 ml) rich in coding sequence (2.75 kb) was amplified by PCR and labeled by random O-GlcNAcase activity were pooled and then separated for a second time primer using [a- P]dCTP (Stratagene, La Jolla, CA). After stripping in on the Mono-Q column. The final preparation was concentrated using 0.5% SDS at 100 °C for 10 min, the blot was reprobed for b-actin. Millipore concentrators to a final volume of 0.4 ml. Glycerol (40% final) and1mM PMSF and protease inhibitor mixture were added to the preparation. The enzyme was stored at 220 °C. RESULTS Identification of Proteins by Mass Spectrometry—The final prepara- Native Polyacrylamide Gel Electrophoresis Aides in O- tion from Step 8 was separated by 10% SDS-PAGE and stained with GlcNAcase Purification from Bovine Brain—Historically, O- Coomassie Blue G-250 or with silver. The desired protein bands were GlcNAcase, or neutral hexosaminidase C, has been difficult to excised individually, reduced, alkylated, and digested in-gel with mod- ified trypsin (Worthington, Freehold, NJ) as described previously (23). purify. For example, an early report described a purification of 9840 Cloning of O-GlcNAcase from Human Brain FIG.2. Silver staining and protein identification by mass spec- trometry. The final purification was separated on 10% SDS-PAGE and silver stained. Each protein band was individually excised and digested FIG.1. Purification steps of O-GlcNAcase from bovine brain. with modified trypsin, and the resultant peptides were separated and The cytosolic fraction of bovine brain proteins were sequentially puri- sequenced by liquid chromatography-MS/MS. Seven to 22 peptides fied by precipitation with 30 –50% ammonium sulfate, DE52 ion ex- were identified for each protein. change, ConA, and Blue A affinity columns, then re-separated on DE52 cellulose ion exchange chromatography (a), native polyacrylamide gel activity loss (data not shown). electrophoresis (b), and Mono-Q ion exchange chromatography (c). The protein profiles in a and c were continuously monitored by absorbance Mass Spectrometry Identifies O-GlcNAcase on SDS-PAGE at 280 nm with a UV detector during purification on a fast protein Gel—The seven bands on the silver-stained SDS-PAGE gel liquid chromatography system, and are presented by solid lines without were excised individually, digested with trypsin, and se- symbols.In b, the protein contents were manually determined by Bio- quenced by electrospray MS/MS. The fragmentation data were Rad assay (A ) for each fraction after the protein was eluted from the native gel. Therefore, it is presented with symbols. See “Experimental used to search protein and DNA data bases. This approach Procedures” for details. l, protein profile; Œ, O-GlcNAcase activity. identified six proteins with known functions in six of the bands (Fig. 2). Another protein, which runs as a 130-kDa band on only 25- to 40-fold from bovine brain despite the extensive use SDS-PAGE, is a hypothetical protein without any clearly of chromatographic steps (27). However, in rat brain, the en- defined functions (GenBank, KIAA0679). BLAST searches in- zyme has been purified over 2000-fold to a major band (28). dicated that the hypothetical protein shared significant homol- More recently, renewed effort has gone into its purification ogy with a protein from C. elegans called “similar to hyalurono- from rat spleen and bovine brain in the pursuit to cloning the glucosaminidase” (GenBank, AAA68333.1). Because hyaluro- cDNA (21, 29). noglucosaminidase degrades hyaluronic acid, which is a We have taken an approach, partly based on published lit- GlcNAcb1– 4GlcUA polymer, it was possible that a hyalurono- erature (21) but yet incorporating some novel steps in the glucosaminidase may share some homology with O-GlcNAcase. purification of O-GlcNAcase from bovine brain. Notably, we Furthermore, careful comparisons of O-GlcNAcase activity and have discovered that the enzymatic activity survives the harsh protein patterns on the SDS gels of different pools during the conditions of native PAGE (high pH and high ionic strength) purification procedure indicated that this protein was one of and migrates more slowly (R 5 0.28 in 6% native gel) than only two bands that corresponded with activity (the other band most other proteins in the gel (data not shown). This property was Protein 1 in Fig. 2, data not shown). We therefore hypoth- allows the effective separation of O-GlcNAcase from other pro- esized that this may be the O-GlcNAcase, and cloned the cDNA. teins with higher R values on a preparative scale native gel Further characterization of the expression product of the cDNA (Fig. 1b). This step, in conjunction with other chromatographic confirmed that band 2 on Fig. 2 was, indeed, O-GlcNAcase (see steps outlined in the protocol, purified the protein ;1500-fold, below). with a specific activity of 1840 nmol/min/mg of protein. O-GlcNAcase Is Unique and Conserved during Evolution— The final preparation still shows seven well defined bands on BLAST searches of data bases reveal that O-GlcNAcase is SDS-PAGE following silver staining, even after extensive conserved in higher eukaryotic species, and the homologue is purification (Fig. 2). We do not understand the basis for this absent in yeast or prokaryotes. The sequences and alignment of difficulty, but we have observed that the peaks for O-GlcNA- O-GlcNAcase from human, C. elegans, and Drosophila are case activity are very broad throughout the purification proce- shown in Fig. 3. In a pairwise alignment, the human sequence dure (Fig. 1). One example of this is illustrated in the Mono-Q shares 55 and 43% homology with that of Drosophila and C. step, where the general protein peaks are sharp but yet the elegans, respectively, whereas Drosophila and C. elegans are activity peak spreads over 50 ml (Fig. 1c). We have also tried 43% similar. Close inspection of the sequences indicates that ion exchange on Superose Q and hydroxylapatite columns or the N-terminal ;400 and the C-terminal ;350 amino acids in hydrophobic interaction chromatography on a phenyl-Sepha- the human sequence are conserved to a higher degree. These rose column. They, too, give poor separations or the enzyme two domains are separated by a highly variable region of ;150 binds very tightly to phenyl-Sepharose resulting in .50% amino acids. Another feature is that most of the aromatic Cloning of O-GlcNAcase from Human Brain 9841 FIG.3. The predicted amino acid se- quence of O-GlcNAcase and its align- ment with hypothetical proteins in C. elegans (“similar to hyaluronoglu- cosaminidase”; AAA68333.1) and Dro- sophila (AAF55867.1). The alignment was done with ClustalW in MacVector. Identical and conservative residues are indicated by asterisks and dots, respec- tively. The underlined sequences are the typtic peptides that are identified by liquid chromatography-MS/MS. Boldface and underlined letters indicate conserved aromatic residues in the three species. residues are conserved among the species. For example, out of that the cloned cDNA indeed encoded O-GlcNAcase, we sub- the 13 Trp residues found in the human sequence, 9 are invari- cloned the entire coding region in-frame into the mammalian ant in Drosophila and C. elegans, two are conservative (substi- expression vector pcDNA3.1His and overexpressed for activity tuted by Tyr or Phe), and only two are variable. in Cos-7 cells. Transient transfection resulted in a 6-fold in- The O-GlcNAcase sequence is conserved at a strikingly crease in O-GlcNAcase activity over endogenous activity in the higher level in mammals. Four overlapping EST sequences cells (Fig. 4a). After nickel affinity purification, the activity from cow, which cover 46% of the human protein, show that from the O-GlcNAcase-transfected cells was 230 nmol/min/mg these two species are 100% identical in these regions of protein but was not detectable from control transfected cells (BE481597, BE588694, BF043559, and AW463869). Five EST (Fig. 4b). These data show that the activity is due to overex- entries for mouse, most of which are overlapping, show that pression from the plasmid. Fig. 4c shows that a distinct band of human and mouse are 97.8% identical (AW907793, AW324047, the correct molecular mass (135 kDa) was isolated after nickel AI530529, AW762257, and AA240394). In the case of zebrafish, purification from transfected cells. This band was immunore- two overlapping EST sequences covering 33.8% of the human active with the Xpress antibody, which was specific for a pep- protein indicate that zebrafish and human are 85% identical tide sequence in the overexpressed protein. and 92% similar (AI882982 and AI722710). Recombinant O-GlcNAcase Has Distinct Properties from Apart from the above-described homologues, O-GlcNAcase Lysosomal b-Hexosaminidase—We further characterized the does not show significant homology with any other proteins, properties of the cloned O-GlcNAcase and compared them with including known glycosidases. Short stretches of ;200 amino those of lysosomal b-hexosaminidase purified from bovine kid- acids of the polypeptide do show loose homology to a number of ney. As expected, the lysosomal b-hexosaminidase had an proteins such as hyaluronidase (AAA23259.1), a putative acetyl- acidic pH optimum (pH 3.5–5.5) with little activity at pH 7.0 or transferase (AL158057), eukaryotic translation elongation fac- above (Fig. 5a). On the other hand, the cloned O-GlcNAcase tor-1g (Z11531, S26649), and the 11-1 polypeptide (X07453, had a pH optimum of 5.7–7.0 and retained significant activity S00485). Sequence analyses by a computer program PSORT II (;30%) at pH 7– 8. This pH profile is consistent with the (available on the Web) show that O-GlcNAcase does not possess expected localization of O-GlcNAcase in the cytoplasm and the any known signal peptides, domains, or motifs. The analyses do, nucleus. however, suggest that the endogenous protein is localized in the The two enzymes also responded differently to inhibitors. cytoplasm (p 5 0.522) and the nucleus (p 5 0.391). GalNAc, a widely used inhibitor of acidic b-hexosaminidase, Overexpression of O-GlcNAcase in Cos-7 Cells—To ascertain inhibited the lysosomal enzyme 50% at 5.0 mM and 88% at 50 9842 Cloning of O-GlcNAcase from Human Brain FIG.6. Substrate specificity of cloned O-GlcNAcase. pNP-b- GlcNAc, pNP-b-GalNAc, and pNP-a-GlcNAc (all 2 mM) were tested as substrates for purified O-GlcNAcase from transfected Cos-7 cells. The lysosomal hexosaminidase was used as a control. The assays were done at pH 4.5 for the lysosomal enzyme and at pH 6.5 for O-GlcNAcase. The activity using pNP-b-GlcNAc as substrate is set as 100%. somal b-hexosaminidase in substrate requirements. In the in vitro assays, purified recombinant O-GlcNAcase cleaved only pNP-b-GlcNAc, but not pNP-b-GalNAc or pNP-a-GlcNAc (Fig. FIG.4. Overexpression of O-GlcNAcase in Cos-7 cells. Cos-7 6). The activity using the latter two compounds as substrates cells were transiently transfected with the vector pcDNA3.1His alone or was not detectable. This substrate specificity was in contrast to with pcDNA3.1His containing O-GlcNAcase cDNA using Lipo- the lysosomal b-hexosaminidase, which also cleaved pNP-b- fectAMINE Plus. 48 h post-transfection, soluble proteins were isolated from the cells. a, O-GlcNAcase activity in total cell extract. The activity GalNAc, albeit with slightly lower efficiency compared with in control transfected cells is arbitrarily set as one-fold. b, O-GlcNAcase pNP-b-GlcNAc. activity in control and transfected cells after nickel purification. c, O-GlcNAcase Cleaves O-GlcNAc from Glycopeptides—A true Western blot and Coomassie Blue G-250 staining of nickel affinity- O-GlcNAcase should cleave O-GlcNAc attached to proteins or purified O-GlcNAcase from transfected cells. The primary antibody was peptides. We synthesized two glycopeptides containing one anti-Xpress, which recognized a peptide sequence (DLYDDDDK) pres- ent in the vector pcDNA3.1His. The arrow indicates the band of repeat of the (CTD) C- terminal domain of RNA polymerase II O-GlcNAcase. linked to b-GlcNAc or a-GalNAc through the hydroxyl group of a serine residue (CTD-GlcNAc or CTD-GalNAc). The design of these glycopeptides is based on earlier information that this serine residue is glycosylated in vivo (13). These peptides were tested as substrates for the purified recombinant O-GlcNAcase, and the product peptides were analyzed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF). Cleav- age of GlcNAc or GalNAc from the peptides should result in a downshift of 203 in the molecular weight (the weight of GlcNAc or GalNAc minus 18). O-GlcNAcase did not cleave GalNAc from the peptide (Fig. 7, a and b) but did successfully cleave GlcNAc from the peptide, as judged by the expected shift in molecular weight (1069.8 to 866.6, 1091.8 to 888.6) (Fig. 7, c and d). Overexpressed O-GlcNAcase Is Localized to the Cyto- plasm—As stated earlier, sequence analyses suggest that O- GlcNAcase is localized in the cytoplasm and nucleus. This is consistent with the localization of the O-GlcNAc modification. To obtain direct evidence on its localization, we performed cellular fractionation and assayed O-GlcNAcase activity in the cytoplasm and the nucleus. The data show that, in nontrans- fected cells, O-GlcNAcase activity was distributed in both the FIG.5. Some key properties of the cloned O-GlcNAcase. A lyso- somal b-hexosaminidase purified from bovine kidney was used as a cytoplasm and the nucleus. However, when O-GlcNAcase was control enzyme. a, pH optima of the two enzymes. Inhibition studies by overexpressed, it was predominantly found in the cytoplasm GalNAc (b), GlcNAc (c), and PUGNAc (d). In b, c, and d, the activity was (Fig. 8a). We also probed for its localization by Western blots assayed at pH 4.5 for the lysosomal hexosaminidase and pH 6.5 for the (Fig. 8b). Retinoblastoma protein (Rb) and a-tubulin, which O-GlcNAcase. The highest activity at optimal pH (a) or in the absence were exclusively localized in the nucleus and cytoplasm, re- of any inhibitor (b, c, d) is arbitrarily set as 100%. l, O-GlcNAcase; ƒ, lysosomal b-hexosaminidase. spectively, were used as markers. In agreement with activity assays, overexpressed O-GlcNAcase protein was only detected mM. The cloned O-GlcNAcase was not inhibited at all by Gal- in the cytoplasm of overexpressed cells. NAc up to 50 mM (Fig. 5b). GlcNAc and its synthetic analogue O-GlcNAcase Has One Transcript in Human Tissues—To PUGNAc inhibited both enzymes but were more potent with estimate the number of transcripts of O-GlcNAcase and their the O-GlcNAcase (Fig. 5, c and d). levels of expression, we performed Northern analyses using a Recombinant O-GlcNAcase Shows Strict Substrate Specific- human multiple tissue blot consisting of RNA from eight dif- ity for b-Linked GlcNAc—O-GlcNAcase also differed from lyso- ferent tissues. Labeled PCR product of the entire coding region Cloning of O-GlcNAcase from Human Brain 9843 FIG.7. Recombinant O-GlcNAcase cleaves O-GlcNAc, but not O-GalNAc, from glycopeptides. Sythetic peptides, CTD-GlcNAc (N-YSPTS(GlcNAc)PSK-C) or CTD-GalNAc (N-YSPTS(GalNAc)PSK- C) were tested as substrates for purified recombinant O-GlcNAcase. The reactions containing 0.1 m M peptide were incubated at pH 6.5, 37 °C overnight. The mock re- actions were done using nickel purifica- tion from nontransfected Cos-7 cells. The peptides were then cleaned up by zip tips and analyzed by MALDI-TOF. The values 1091.7 (or 1091.8) and 1069.7 (or 1069.8) represented molecular weights of the Na and H form of the peptides, respectively. After the GlcNAc was cleaved in CTD- GlcNAc peptide by O-GlcNAcase, the mass expectedly shifted down to 888.6 and 866.6, respectively. The numbers on the right-hand side of each spectrum (2744, 2930, 5158, and 2100) are the total ion counts recorded by the detector in MALDI-TOF analysis. FIG.9. O-GlcNAcase transcript is expressed in every human tissue examined but is the highest in the brain, placenta, and pancreas. a, a human multiple tissue Northern blot was probed with [a- P]dCTP-labeled full-length O-GlcNAcase coding sequence and ex- posed to film for 18 h. b, after stripping, the blot was reprobed for b-actin and exposed to film for 7 h. FIG.8. Overexpressed O-GlcNAcase is distributed in the cyto- localized on chromosome 10 (Ref. 25, and available on the Web plasm. Cos-7 cells were transfected with O-GlcNAcase. Two days post- transfection, the cells were fractionated into cytoplasmic or nuclear from Kazusa DNA Research Institute, Japan). fractions. a, O-GlcNAcase activity (n 5 3). Assays were done with 100 DISCUSSION mM GalNAc as inhibitor. b, Western blots with anti-Xpress (for recom- binant O-GlcNAcase detection), anti-a-tubulin (cytoplasmic marker), Two categories of b-hexosaminidases are known to exist in and anti-retinoblastoma (Rb) (nuclear marker). eukaryotic cells. One category, which comprises the A, B, I, S isozymes, is exclusively localized in the lysosomes and are of the O-GlcNAcase cDNA was used as a probe. The Northern responsible for the degradation of complex glycans. This group blot analysis showed only one transcript of ;5.5 kb for O- of hexosaminidases, particularly the A and B isozymes, have GlcNAcase (Fig. 9). Exposing the film for extended time (32 h) been extensively studied because their deficiency leads to Tay- did not reveal any additional bands (data not shown). The gene Sachs and Sandhoff diseases (30, 31). These enzymes are char- was expressed in every tissue on the blot but was the highest in acterized by their acidic pH optima, inhibition by both GlcNAc the brain, followed by placenta, and pancreas. Lung and liver and GalNAc, the ability to use both artificial glucosaminide had the lowest expression (Fig. 9). This pattern of expression and galactosaminide as substrates, and their thermostability largely agrees with that of the ogt gene. ogt is expressed the (32). The second category, however, consists of two neutral highest in the pancreas, followed by heart, brain, and placenta hexosaminidases, the GlcNAc-specific glucosaminidase (hex- but the lowest in lung, liver, and kidney (18). Unlike ogt gene, osaminidase C) and the GalNAc-specific galactosaminidase which is on the X chromosome (20), the O-GlcNAcase gene is (hexosaminidase D) (28, 33, 34). In contrast to lysosomal hex- 9844 Cloning of O-GlcNAcase from Human Brain osaminidases, they reside in the cytosol, have neutral pH op- ing an O-GlcNAcase-specific polyclonal antibody based on its tima, and are heat labile. Despite the wide occurrence of these protein sequence. Immunofluorescence microscopic studies will neutral isoforms in tissues, their natural substrates have not determine the relative distribution of the endogenous protein. been previously identified (28). The cloning of O-GlcNAcase provides a valuable tool to fur- ther study the biological function of O-GlcNAc modification and The discovery of O-GlcNAc provides clues to the likely phys- iological function of neutral glucosaminidase (hexosaminidase may help to understand the mechanisms of a number of prev- alent diseases such as the Alzheimer’s and diabetes. Recent C) (2). This modification consists of the addition of a single data suggest that many proteins in the brain, including Syn- GlcNAc moiety to the hydroxyl group of Ser/Thr residues (2). apsin I and neurofilaments (6, 37), are O-glycosylated. Some of The enzyme responsible for GlcNAc addition, OGT, was iden- the glycosylated proteins may be involved in the development tified and cloned several years ago (17–19). On the other hand, of the Alzheimer’s diseases. For example, Tau, whose hyper- the identity of the enzyme for the removal of O-GlcNAc from phosphorylation leads to the formation of neurofibrillary tan- proteins, O-GlcNAcase, has remained obscure. Judged from the gles in the neurons of Alzheimer’s brains, is extensively O- known properties of hexosaminidase C, we speculate that it glycosylated (38). The b-amyloid precursor protein, which gives might in fact be the O-GlcNAcase. We thus have extensively rise to the neurotoxic b-amyloid peptide in Alzheimer’s brains, purified hexosaminidase C from bovine brains. Proteomic analy- is also O-GlcNAcylated (39). Furthermore, Alzheimer’s disease sis allowed us to identify its DNA sequence. The sequence was has been correlated to the glycosylation of at least one protein, originally isolated during screening of a meningioma expres- the chathrin assembly protein-3 (40). Continuing efforts in sion library (25). Based on its initial characterization and se- several laboratories are underway to understand the role of quence homology with an unidentified protein called “similar to O-GlcNAc in the development of such neurodegenerative dis- hyaluronoglucosaminidase” from C. elegans, the authors sug- eases. In addition, strong evidence suggests that O-GlcNAc is gested that it was a new type of hyaluronidase. We have here involved in the development of insulin resistance in diabetes demonstrated that the cloned enzyme has all the expected mellitus (4, 41). Infusion of glucosamine or the overexpression properties of O-GlcNAcase. Most importantly, it specifically of glutamine:fructose-6-phosphate amidotransferase in animal cleaves O-GlcNAc but not O-GalNAc from glycopeptides. We models, both of which increase cellular UDP-GlcNAc levels conclude that we have cloned hexosaminidase C, which we through the hexosamine synthetic pathway (42, 43), leads to have called O-GlcNAcase herein. insulin resistance (44, 45). The cloning of O-GlcNAcase will not The recombinant protein differs from acidic lysosomal hex- only allow the selective disruption of the gene, or tissue-specific osaminidase in a number of ways. It has a neutral, instead of overexpression of the protein in animal models, but will also acidic, pH optimum. This pH optimum is expected due to its make it possible to directly regulate and monitor O-GlcNAc- physiological functions in the cytoplasm and nucleus. The pro- modified proteins and thus facilitate the understanding of the tein displays strict substrate requirement in the linkage of the role of this modification in the development of these diseases. sugar and is also sensitive to the C4 orientation of the sugar. As such, it hydrolyzes only b-linked GlcNAc but is totally ineffec- Acknowledgments—We thank the Kazusa DNA Research Institute, tive toward a-linked GlcNAc or b-linked GalNAc. This sub- Japan for providing the KIAA0679 cDNA and Dr. N. Zachara for read- ing the manuscript. strate specificity is in contrast with that of lysosomal hex- osaminidases, which hydrolyze both GlcNAc and GalNAc REFERENCES substrates. It also determines that GlcNAc, but not GalNAc, is 1. Torres, C. R., and Hart, G. W. (1984) J. Biol. Chem. 259, 3308 –3317 2. Hart, G. W. (1997) Annu. Rev. Biochem. 66, 315–335 a competitive inhibitor of O-GlcNAcase. Unlike lysosomal het- 3. Snow, D. M., and Hart, G. W. (1998) Int. Rev. Cytol. 181, 43–74 ero-oligmeric hexosaminidases A (ab b )orB(2xb b ) (32), the a b a b 4. Comer, F. I., and Hart, G. W. (2000) J. Biol. Chem. 275, 29179 –29182 cloned human O-GlcNAcase has only one polypeptide of 916 5. Chou, T. Y., Hart, G. W., and Dang, C. V. (1995) J. Biol. Chem. 270, 18961–18965 amino acids with a predicted molecular mass of 103 kDa. The 6. Cole, R. N., and Hart, G. W. (1999) J. Neurochem. 73, 418 – 428 apparent size on SDS-PAGE is 130 kDa, which is identical to 7. Cheng, X., Cole, R. N., Zaia, J., and Hart, G. W. (2000) Biochemistry 39, 11609 –11620 the glucosaminidase purified from rat brain (28). The oligo- 8. Griffith, L. S., and Schmitz, B. (1999) Eur. J. Biochem. 262, 824 – 831 meric status of the hexosaminidase C is still not known, be- 9. Lefebvre, T., Alonso, C., Mahboub, S., Dupire, M. J., Zanetta, J. P., Caillet-Boudin, M. L., and Michalski, J. C. (1999) Biochim. Biophys. Acta cause gel filtration, SDS-PAGE, and sucrose density gradient 1472, 71– 81 centrifugation give inconsistent estimations of its native mo- 10. Jackson, S. P., and Tjian, R. (1988) Cell 55, 125–133 lecular size (28). 11. Reason, A. J., Morris, H. R., Panico, M., Marais, R., Treisman, R. H., Haltiwanger, R. S., Hart, G. W., Kelly, W. G., and Dell, A. (1992) J. Biol. O-GlcNAcase is a unique protein with no other obvious fam- Chem. 267, 16911–16921 ily members. It is conserved in evolution down to the nematode 12. Jiang, M. S., and Hart, G. W. (1997) J. Biol. Chem. 272, 2421–2428 13. Kelly, W. G., Dahmus, M. E., and Hart, G. W. (1993) J. Biol. Chem. 268, C. elegans, and shows striking homology in mammals. It does 10416 –10424 not, however, possess any known signal peptides, domains, or 14. Kelly, W. G., and Hart, G. W. (1989) Cell 57, 243–251 15. Han, I., and Kudlow, J. E. (1997) Mol. Cell. Biol. 17, 2550 –2558 motifs. Consistent with the literature that antibodies that rec- 16. Du, W. L., Edelstein, D., Rossetti, L., Fantus, I. G., Goldberg, H., Ziyadeh, F., ognize lysosomal hexosaminidases A or B do not cross-react Wu, J., and Brownlee, M. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, with hexosaminidase C (35), O-GlcNAcase does not share any 12222–12226 17. Haltiwanger, R. S., Blomberg, M. A., and Hart, G. W. (1992) J. Biol. Chem. significant homology with these enzymes, or with any other 267, 9005–9013 known proteins. 18. Lubas, W. A., Frank, D. W., Krause, M., and Hanover, J. A. (1997) J. Biol. Chem. 272, 9316 –9324 The predicted localization of O-GlcNAcase by sequence analy- 19. Kreppel, L. K., Blomberg, M. A., and Hart, G. W. (1997) J. Biol. Chem. 272, sis is consistent with the literature that O-GlcNAc modification 9308 –9315 20. Shafi, R., Iyer, S. P., Ellies, L. G., O’Donnell, N., Marek, K. W., Chui, D., Hart, and OGT are found both in the cytoplasm and nucleus (19, 36). G. W., and Marth, J. D. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 5735–5739 It is therefore somewhat unexpected to find that overexpressed 21. Dong, D. L., and Hart, G. W. (1994) J. Biol. Chem. 269, 19321–19330 O-GlcNAcase in Cos-7 cells is almost exclusively localized to 22. Roquemore, E. P., Chou, T. Y., and Hart, G. W. (1994) Methods Enzymol. 230, 443– 460 the cytoplasm (Fig. 8). Nevertheless, it is not uncommon that 23. Gharahdaghi, F., Weinberg, C. R., Meagher, D. A., Imai, B. S., and Mische, overexpressed proteins often form aggregates in the cytoplasm S. M. (1999) Electrophoresis 20, 601– 605 24. Ducret, A., Van, O.ostveen, I., Eng, J. K., Yates, J. R., and Aebersold, R. (1998) and cannot be correctly translocated. We cannot exclude the Protein Sci. 7, 706 –719 possibility of an unidentified isoform of O-GlcNAcase that is 25. Heckel, D., Comtesse, N., Brass, N., Blin, N., Zang, K. D., and Meese, E. (1998) specifically localized to the nucleus. We are currently generat- Hum. Mol. Genet. 7, 1859 –1872 Cloning of O-GlcNAcase from Human Brain 9845 26. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic 37. Dong, D. L., Xu, Z. S., Chevrier, M. R., Cotter, R. J., Cleveland, D. W., and Acids Res. 17, 6419 Hart, G. W. (1993) J. Biol. Chem. 268, 16679 –16687 27. Overdijk, B., van der Kroef, W. M. J., van Steijn, G. J., and Lisman, J. W. 38. Arnold, C. S., Johnson, G. V., Cole, R. N., Dong, D. L. Y., Lee, M., and Hart, (1981) Biochim. Biophys. Acta 659, 255–266 G. W. (1996) J. Biol. Chem. 271, 28741–28744 28. Izumi, T., and Suzuki, K. (1983) J. Biol. Chem. 258, 6991– 6999 39. Griffith, L. S., Mathes, M., and Schmitz, B. (1995) J. Neurosci. Res. 41, 29. Hanover, J. A., Lai, Z., Lee, G., Lubas, W. A., and Sato, S. M. (1999) Arch. 270 –278 Biochem. Biophys. 362, 38–45 40. Yao, P. J., and Coleman, P. D. (1998) J. Neurosci. 18, 2399 –2411 30. Kustermann-Kuhn, B., and Harzer, K. (1983) Hum. Genet. 65, 172–175 41. McClain, D. A., and Crook, E. D. (1996) Diabetes 45, 1003–1009 31. Neufeld, E. F. (1989) J. Biol. Chem. 264, 10927–10930 42. Yki-Jarvinen, H., Virkamaki, A., Daniels, M. C., McClain, D., and Gottschalk, 32. Emiliani, C., Sciarra, R., Orlacchio, A., and Stirling, J. L. (1990) Biochim. W. K. (1998) Metabolism 47, 449 – 455 Biophys. Acta 1037, 265–273 43. Cooksey, R, C., Hebert, L. F., Jr, Zhu, J. H., Wofford, P., Garvey, W. T., and 33. Overdijk, B., van der Kroef, W. M. J., Veltkamp, W. A., and Hooghwinkel, McClain, D. A. (1999) Endocrinology 140, 1151–1157 G. J. M. (1975) Biochem. J. 151, 257–261 44. Hebert, L. F., Jr, Daniels, M. C., Zhou, J., Crook, E. D., Turner, R. L., 34. Izumi, T., and Suzuki, K. (1980) Biochim. Biophys. Acta 615, 402– 413 Simmons, S. T., Neidigh, J. L., Zhu, J. S., Baron, A. D., and McClain, D. A. 35. Poenaru, L., Weber, A., Vibert, M., and Dreyfus, J. C. (1973) Biomedicine 19, (1996) J. Clin. Invest. 98, 930 –936 538 –540 45. Baron, A. D., Zhu, J. S., Zhu, J. H., Weldon, H., Maianu, L., and Garvey, W. T. 36. Akimoto, Y., Kreppel, L. K., Hirano, H., and Hart, G. W. (1999) Diabetes 48, 2407–2413 (1995) J. Clin. Invest. 96, 2792–2801

Journal

Journal of Biological Chemistry – Unpaywall

Published: Mar 1, 2001

There are no references for this article.