HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 24, 21207-21212, June 14, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/24/21207    most recent M201634200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wilson, I. B. H. Search for Related Content PubMed PubMed Citation Articles by Wilson, I. B. H. Social Bookmarking                      What's this?

Functional Characterization of Drosophila melanogaster Peptide O-Xylosyltransferase, the Key Enzyme for Proteoglycan Chain Initiation and Member of the Core 2/I N-Acetylglucosaminyltransferase Family^*

Iain B. H. Wilson

From the Glycobiology Division, Institut für Chemie, Universität für Bodenkultur, Muthgasse 18, A-1190 Wien, Austria

Received for publication, February 18, 2002, and in revised form, March 27, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Chondroitin and heparan sulfates are essential players in animal development and are synthesized by a series of glycosyltransferases, the first of which is UDP--D-xylose:proteoglycan core protein -D-xylosyltransferase (EC 2.4.2.26). In the present study, a Drosophila melanogaster gene (CG17771), previously designated as a homologue of core 2 and I 1,6-N-acetylglucosaminyltransferases, was shown to encode an active peptide O-xylosyltransferase. A novel coupled assay using matrix-assisted laser desorption ionization time-of-flight mass spectrometry demonstrated transfer of xylose to the peptide DDDSIEGSGGR. Analysis of sequences of various peptide O-xylosyltransferase and 1,6-N-acetylglucosaminyltransferase sequences indicates that they are members of a large multifunctional protein family with a range of roles in -glycosylation of either peptide or glycan substrates. Because in contrast to mammals, there is only one fly peptide O-xylosyltransferase gene, it is anticipated that, given the key roles of proteoglycans, the hereby designated oxt gene is essential for viability.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell-cell and cell-matrix interactions are of profound importance in the development and morphogenesis of multicellular organisms; key molecules in these processes are proteoglycans (1). Animal proteoglycans contain glycosaminoglycan chains that fall into a number of categories as follows: chondroitin sulfate, dermatan sulfate, heparin, heparan sulfate, and keratan sulfate which consist of sulfated disaccharide repeats linked to a core protein via a glycan (2). The first four categories have a glycan core of GlcA1,3Gal1,3Gal1,4Xyl O-linked to the protein, whereas keratan sulfates are attached through typical N- or O-glycans. In plants, arabinogalactans are the functional equivalent (3).

The significance of proteoglycans is shown, for instance, by the correlation of mutations in genes encoding proteoglycan biosynthesizing enzymes with a disease of bone formation (hereditary multiple exostoses) in man (4) or with defects in vulval formation in nematodes (5, 6). As a genetic model animal, Drosophila is an obvious target for exploring the roles of proteoglycans in various aspects of development (7), and from recent studies it appears that, in the fly, proteoglycans are involved in the diffusion and recognition of growth factors such as Wingless (a Wnt homologue), Decapentaplegic (Dpp), Hedgehog (Hh), and fibroblast growth factor. Specifically, mutants of the dally, sugarless, tout-velu, and sulfateless genes are defective in signaling of one or more of these four growth factors. By homology to mammalian proteins, it has been postulated that dally encodes a proteoglycan core protein (8, 9), sugarless (otherwise known as suppenkasker or kiwi) a UDP-Glc dehydrogenase (10, 11), tout-velu an EXT-type heparan sulfate synthetase homologue (12), and sulfateless a homologue of heparan sulfate N-deacetylase/N-sulfotransferase (8, 13). In addition to Sulfateless, a number of fly homologues of mammalian proteoglycan-modifying sulfotransferases have been shown to have roles during development. Whereas heparan sulfate 6-O-sulfotransferase is necessary for primary branching of the tracheal system (14, 15), Pipe, a uronyl 2-sulfotransferase homologue, is required for the definition of embryonic dorsal-ventral polarity (16, 17), and segregation distorter (Sd) protein, a heparan sulfate 2-O-sulfotransferase homologue, is involved in male meiosis (18).

In contrast to the growing body of genetic information related to proteoglycans in the fly, there remain large gaps in the biochemical knowledge. It is, therefore, appropriate to verify the activity of relevant Drosophila enzymes and not just rely on homologies to determine function. In particular, the enzyme in fly that initiates the formation of those glycosaminoglycan chains with an O-linked xylose core has not yet been the target of any investigation, other than a single report (19) on the effect on fly larvae locomotory behavior of a competitive inhibitor of xylosylation. In vertebrates, this reaction is catalyzed by peptide O-xylosyltransferase, which utilizes UDP-xylose as the donor for a reaction in which the hydroxyl groups of serine residues of proteoglycan core proteins become modified. Despite reports around 30 years ago on the purification of the chicken cartilage enzyme (20), the first peptide sequences derived from a purified form of a xylosyltransferase were only recently published (21). This information allowed the cloning of two human cDNAs, one of which was demonstrated to encode an active xylosyltransferase (22). In the present report, I describe the identification of the apparently single fly peptide O-xylosyltransferase (OXT)¹ and its successful functional expression in Pichia pastoris.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

EST Identification and Cloning of Drosophila OXT cDNA-- The GenBank^TM data base was searched using the tBLASTn program using the sequences of mammalian proteoglycan xylosyltransferases I and II (22). In Drosophila melanogaster, a gene (CG17771) was identified and the predicted sequence used to search the dbEST data base. Portions of the predicted reading frame of 876 residues showed identity with a number of Drosophila ESTs or partial cDNAs from embyro, head, larvae/early pupae, imaginal discs, testes, and Schneider L2 cells. One of these EST sequences (Drosophila gene collection clone LD43716) was identical to the 5'-end of the oxt gene, and because it was considered to be potentially part of a full-length cDNA, this cDNA was obtained from Research Genetics. Sequencing of the clone was completed using the BigDye system (PE Biosystems) using both fly- and vector-specific primers. Comparison of protein sequences was performed using either BLAST (www.ncbi.nlm.nih.gov) or Multalin (prodes.toulouse.inra.fr) software.

A fragment encoding the putative soluble form of the proteoglycan O-xylosyltransferase (i.e. lacking the putative transmembrane domain) was amplified from the clone using Expand DNA polymerase mixture (Roche Molecular Biochemicals), the primers DMXT3/EcoRI 5'-ccggaattccctggacatcgtagg-3' and DMXT2/KpnI/Stop 5'-cggggtacctcatttgagcagggcatc-3', and an Eppendorf MasterCycler PCR machine (3 min at 95 °C, 40 cycles of 1 min at 60 °C, 3 min at 72 °C, and 1 min at 95 °C, followed by a final extension step of 8 min at 72 °C).

PCR products were purified after gel electrophoresis using the Qiagen PCR purification kit and digested with EcoRI and KpnI, gel-purified, and ligated into the pICZC vector (Invitrogen) cut with the same enzymes. Expression vector DNA (~10 µg) from PCR-positive Escherichia coli clones was cut with MssI prior to electroporation into P. pastoris GS115 cells. Positive yeast clones were identified after PCR of genomic DNA using gene-specific primers.

Production of Recombinant Drosophila Proteoglycan O-Xylosyltransferase-- For the screening of oxt clones, expression of recombinant xylosyltransferase was induced by methanol basically as described previously for Drosophila fucosyltransferases (23). Yeast clones were grown in 1 ml of MGYC (medium with glycerol, yeast nitrogen base, and casamino acids) overnight at 30 °C, washed with yeast nitrogen base, and resuspended in 10 ml of MMYC (medium with methanol, yeast nitrogen base, and casamino acids). Methanol (final volume 1%) was added every 24 h thereafter, and the medium was collected after 3 days of induction. The culture medium was concentrated 10-fold using Vivaspin concentrators (M_r 10,000 cut-off).

Assay of Recombinant Drosophila OXT-- UDP-xylose was synthesized enzymatically using the Cryptococcus neoformans UDP-glucuronic acid decarboxylase. In brief, a pET-28a plasmid carrying the decarboxylase (UXS1) cDNA (a kind gift of Dr. Tamara Doering) was used to transform E. coli strain BL21(DE3)pLysS. After induction using isopropyl--D-thiogalactopyranoside, the cells were harvested and lysed as described (24). To generate UDP-xylose, 10 µl of soluble bacterial protein extract was incubated overnight with 250 nmol of UDP-glucuronic acid (UDP-GlcA) and 125 nmol of NAD⁺ in Tris-Cl, pH 7.4, buffer in a total volume of 50 µl at 37 °C. HPLC analyses indicated that UDP-GlcA was quantitatively converted to UDP-Xyl under these conditions.

For the acceptor, a peptide based on part of the sequence of Drosophila syndecan, a proteoglycan core protein homologous to a family of heparan sulfate proteoglycans (25), was designed. The standard assay for xylosyltransferase activities was performed at 37 °C with this synthetic peptide, DDDSIEGSGGR (Thermo Hybaid, Ulm, Germany); 1 µl of enzyme solution was added to 5 µl of 1 mM acceptor, 1 µl of 0.4 M MES, pH 7, 0.5 µl of 0.2 M MnCl₂, and 2.5 µl of the aforementoned decarboxylase reaction. After various time points, 0.5 µl of the reaction mixture was withdrawn and diluted 10-fold with water. 0.8 µl of diluted sample was then mixed with 0.8 µl of 1% (w/v) -cyanohydroxycinnamic acid in 70% (v/v) acetonitrile and allowed to dry on a sample plate prior to MALDI-TOF MS analysis on a ThermoBioanalysis Dynamo spectrometer in delayed extraction mode. Percentage conversion was calculated on the basis of the ratio of relative peak areas. For the determinations of the donor K_m value, UDP-Xyl was purchased from a newly available supplier (CarboSource Services). Kinetic data were analyzed using Hanes plots.

Analysis of Xylosylated Peptides-- Xylosyltransferase incubation mixtures were fractionated by RP-HPLC (Hypersil). The program used is as follows: 0-2 min, 0.05% (v/v) trifluoroacetic acid in water; 2-6 min, linear gradient to 11.2% (v/v) acetonitrile; 6-16 min, linear gradient to 14% (v/v) acetonitrile followed by washing of the column with up to 42% acetonitrile before returning to the original solvent. Peaks were collected which had either the same elution time as the original syndecan peptide or corresponded to a peak of slightly shorter elution time that was found only in incubations to which UDP-GlcA (the UDP-Xyl precursor) had been added. The collected fractions were dried and redissolved in 10 µl of water prior to amino acid analysis (for quantitation) and MALDI-TOF MS. In addition, aliquots of the purified xylosylated peptide were treated for 1 h with methylamine in the vapor phase as described recently (26) for peptides substituted with O-linked GalNAc or digested with endoproteinase Glu-C (Roche Molecular Biochemicals) in 25 mM ammonium bicarbonate, pH 7.8 (1 nmol of peptide with 0.5 µg of protease in a 10-µl final volume with 0.2-µl aliquots removed after 0.5, 2.5, 6, and 24 h at 37 °C).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification and Analysis of the Drosophila oxt Gene and cDNA-- Searching of the Drosophila genome indicated the presence of one gene (CG17771) encoding a protein xylosyltransferase, in contrast to mammals (human, mouse, and rat) which appear to have two such genes (22). The deduced protein sequence (see Fig. 1) of the fly xylosyltransferase displays, if one excludes the first 77 residues (i.e. the probable cytosolic, transmembrane, and stem regions), 36 and 37% identity with the corresponding parts of human xylosyltransferases I and II, respectively. The predicted unprocessed molecular weight (M_r 99,097) of fly OXT is similar to that of the human proteins, whereas of the five possible N-glycosylation sites, only the third and the fourth are conserved as compared with the human, mouse, and rat sequences. By identity of portions of the oxt cDNA with various ESTs, it can be deduced that the gene is widely expressed in the fly with EST clones having been isolated from adult head (4 ESTs), adult testes (1 EST), embryo (15 ESTs), and larvae/early pupae (1 EST).

View larger version (73K): [in this window] [in a new window]   Fig. 1.   Verified sequence of D. melanogaster peptide O-xylosyltransferase (oxt) cDNA. The putative transmembrane domain and N-glycosylation sites are underlined; the DXD motif conserved by comparison to mammalian OXTs and intron/exon boundaries are double-underlined.

The complete cDNA sequence also indicated that the annotation of the Drosophila genome sequence (EBI accession no. AE003474) was correct with three exons following the usual :GU ... AG: rule. In contrast the human xylosyltransferase II gene (AC004707) has 11 exons, whereas the incompletely sequenced human xylosyltransferase I gene (AC021115 and AC106769) has at least 11; furthermore, all identified splice sites are conserved within these two human genes. In addition, as judged from analysis of sequences from the relevant genomes, the same genomic structure appears to be present and also valid for the rat xylosyltransferase I and II genes (AC103474 and AC103440), the murine xylosyltransferase II gene (AL645764), and an estimated two pufferfish (Tetraodon nigroviridis) xylosyltransferase genes. On the other hand, only the second fly oxt exon/intron junction (nucleotides 852-853) corresponds to splice sites within vertebrate xylosyltransferase genes.

Proteoglycan Xylosyltransferases as Members of a Glycosyltransferase Family-- O-Xylosylation of peptides is apparently animal-specific, and probable proteoglycan xylosyltransferases are present, as judged by analysis of the genomic and EST databases, in fish (T. nigroviridis and Oryzias latipes), amphibians (Xenopus laevis and Silurana tropicalis), birds (Gallus gallus), mammals (Homo sapiens, Mus musculus, Rattus norvegicus, Bos taurus, and Sus scrofa), ascidians (Halocynthia roretzi and Ciona intestinalis), insects (D. melanogaster and Apis mellifera), and Caenorhabditis elegans. Perhaps somewhat surprisingly from the functional viewpoint, however, is the homology of proteoglycan O-xylosyltransferases to mammalian core 2 and I 1,6-N-acetylglucosaminyltransferases, which transfer GlcNAc to either the GalNAc of Gal1,3GalNAc-O-Ser/Thr or internal Gal residues of Gal1,4GlcNAc1,3Gal1,4GlcNAc sequences, respectively.

As shown in Fig. 2, an ~300 amino acid stretch corresponding to residues 250-552 (of 876) of the fly OXT and 123-428 (of 428) of the human core 2 GnT1 sequence shows ~30% identity. It is interesting to note, however, that the cysteine residues conserved in the human core 2/I enzymes (27) are generally absent from the xylosyltransferases suggesting differences in the folding of these domains between the subfamilies. There are also differences in the genomic structure of OXT and core 2/I genes; as mentioned above, the human xylosyltransferase II reading frame is encoded by 11 exons, with the human core 2 genes each contain one coding exon (27). The reading frame of the human I enzyme, however, stretches across three exons (28), with one of these splice sites corresponding to a splice site in the human xylosyltransferase genes.

View larger version (99K): [in this window] [in a new window]   Fig. 2.   Alignment of the deduced Drosophila OXT with homologous human sequences. Included in the alignment are relevant parts of the human xylosyltransferases I and II, core 2 N-acetylglucosaminyltransferases 1-3, and I N-acetylglucosaminyltransferase. Highlighted are the residues that are present in at least one xylosyltransferase and at least one core 2 or I sequence. Double underlined are the amino acids corresponding to mRNA splice sites.

This region of homology is also shared with the 6 C. elegans GLY-1 subfamily, a further 12 Caenorhabditis sequences, many plant sequences, and 1 sequence from Caulobacter crescentus. Apart from GLY-1 itself, which has been shown to be a 1,6-glucosyltransferase capable of modifying O-linked Gal1,3GalNAc (29), no known function is as yet assigned to any of the nematode, plant, or bacterial sequences (30). GLY-1-type sequences are indeed more closely related to the core 2 O-glycan elongating enzymes, as shown by the presence of all but one of the conserved core 2/I cysteine residues, but in contrast to the one or three coding exons of the mammalian core 2/I genes, five of the gly-1-type genes are encoded by eight exons. Reflecting the similarities, the phylogenetic tree presented in Fig. 3 would suggest that the core 2/I/OXT glycosyltransferase family consists of two major halves as follows: one containing the plant, bacterial, and OXT members, and the other with the core 2/I/GLY-1 members. The C-terminal 300 residues of xylosyltransferases, however, are novel and contain the only DXD-type motif that is conserved between the fly and mammalian OXTs. To be exact, the fly sequence has a DFD sequence where the corresponding mammalian sequences are DWD in xylosyltransferases I and (D/E)WD in xylosyltransferases II. The presence of this motif, noted in many glycosyltransferases, in the region absent from other members of the core 2/I/OXT glycosyltransferase family may be compatible with the fact that OXTs are cation-sensitive (see below), whereas Warren et al. (30) noted that core 2/I enzymes are cation-independent and have no DXD motif.

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Phylogenetic tree of the core 2/I/OXT glycosyltransferase family. Included in the Multalin analysis are the following: 11 homologous Arabidopsis thaliana (At) sequences (with the relevant Arabidopsis Information Resource gene number) derived from either full-length cDNAs or predicted proteins; the incomplete homology-predicted C. elegans OXT (CeOXT, based on open reading frame Y50D4C.4); the D. melanogaster OXT (DmOXT); the human (Hs), mouse (Mm), and rat (Rn) xylosyltransferases I and II; the C. crescentus open reading frame CC2860 (CauloGT); the human and mouse core 2 (C2GnT1-3) and I (IGnT); the bovine (Bt) core 2 and bovine herpesvirus (BHV) core 2/4/I 1,6-N-acetylglucosaminyltransferases; and the full-length C. elegans GLY-1, GLY-15, GLY-16, GLY-17, GLY-18, and GLY-19 sequences; in addition to 12 predicted Caenorhabditis family 14 glycosyltransferases (with relevant cosmid reading frame numbers) whose sequences were corrected, where possible, using ESTs. The PAM scale is of percent accepted mutations. Note: murine IGnT and IGnTB mRNAs are alternatively spliced products of the same gene.

Verification of Drosophila OXT as a Proteoglycan Xylosyltransferase-- A soluble form of fly OXT was expressed under control of the AOX1 alcohol oxidase promoter in P. pastoris as a fusion protein with the -mating factor secretion signal. By using a novel mass spectrometric coupled assay method, a number of yeast clones were screened for secreted xylosyltransferase activity. As acceptor, a peptide (DDDSIEGSGGR) corresponding to residues 55-65 of Drosophila syndecan (25) was used, which contains the features of an acidic region and a Ser-Gly sequence found around chondroitin and heparan sulfate attachment sites (31, 32), as well as an arginine residue which apparently aids detection by MALDI-TOF mass spectrometry (33). For most experiments, the UDP-Xyl donor was synthesized, without subsequent purification, using recombinant Cryptococcus UDP-GlcA decarboxylase (24).

A number of clones of Pichia transformed with oxt-containing constructs secreted active xylosyltransferase as judged by conversion of the syndecan peptide substrate to a form with an m/z ratio of 1239.9, the difference of 132.8 compared with the original peptide (m/z 1107.1) approximating to the mass of a xylose residue (see Fig. 4B). The percentage incorporation as determined by MALDI-TOF MS was time-dependent and comparable with, although consistently slightly lower than, that calculated from chromatogram peak areas of the same samples analyzed by RP-HPLC (Fig 4C).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Assay of recombinant D. melanogaster OXT by MALDI-TOF MS. The syndecan peptide substrate (m/z 1108) was incubated with an extract of E. coli expressing Cryptococcus UDP-GlcA decarboxylase in the presence of UDP-glucuronic acid and supernatant of Pichia expressing either tomato Lewis-type 1,4-fucosyltransferase (A) or Drosophila OXT (B). Conversion to a product containing transferred xylose is indicated by the presence of a species of m/z ~1240. The time-dependent conversion (C) of substrate to product was determined using peak height ratios from MALDI-TOF MS (solid line, solid circles) and RP-HPLC (dashed line, open circles) analysis of the same samples. Kinetic data were analyzed by Hanes plots for both UDP-Xyl (D) and the syndecan peptide (E).

Despite the presence of two serine residues in the peptide, only a maximum of one xylose residue was transferred. No xylosylated peak was seen if UDP-GlcA (as UDP-xylose precursor; data not shown) was absent from the added UDP-GlcA decarboxylase incubation mixture nor when supernatants of Pichia expressing other glycosyltransferases were tested (e.g. tomato 1,4-fucosyltransferase; Fig. 4A) in the presence of the E. coli extract. These data indicate a lack of endogenous OXT activity in both Pichia and E. coli. The activity of the enzyme was also verified using UDP-Xyl from a commercial source (data not shown), whereas no activity was detectable with other UDP-sugars tested. Respective K_m values of ~100 and 500 µM were determined for the donor and acceptor (Fig. 4, D and E); these values can be compared with those found (180 for UDP-Xyl and 60-790 µM for various peptide substrates) for partially purified rat chondrosarcoma xylosyltransferase (34).

As judged by MALDI-TOF assays using unpurified enzyme, fly OXT is more active in the presence of 10 or 20 mM Ca(II), Mg(II), and Mn(II) ions than in the presence of 10 or 20 mM EDTA or of no added cations, whereas it is completely inhibited by 10 mM Ni(II) and Zn(II) (for an HPLC comparison of incubations with Mn(II) and Ni(II), see Fig. 5). These results can be compared with those for the rat ear cartilage xylosyltransferase, which indicated activation in the presence of Mn(II), Mg(II), and Ca(II), but inhibition by Zn(II) (35). Similar to the rat enzyme was also the pH optimum of the fly OXT (in the range pH 7-8). In addition, whereas fly OXT was initially most active (i.e. over a few hours) at 37 °C, the highest activities overnight were determined after incubation at 30 °C or at room temperature (possibly due to stability of the enzyme in the crude supernatant). In contrast, virtually no activity was detectable at 16 °C.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Fractionation of the OXT enzymatic product by RP-HPLC. Overnight incubations of supernatant of Pichia expressing Drosophila OXT and an extract of E. coli expressing Cryptococcus UDP-GlcA decarboxylase in the presence of UDP-glucuronic acid and MnCl2 (A) or NiCl2 (B) were fractionated by RP-HPLC as described. The elution of the unxylosylated syndecan peptide is shown in trace C.

Reaction mixtures were subject to RP-HPLC to purify the putatively xylosylated peptide. Comparison of the chromatograms showed that only in samples to which both UDP-GlcA and "activating" cations (e.g. Mn(II)) had been added was there a new peak with a retention time 1 min earlier than the peak with the same retention time as the original syndecan peptide (Fig. 5). The relevant fractions were collected, and this new peak was verified to also contain a species of m/z 1239.9 (Fig. 6A).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Mass spectrometric analyses of the xylosylated peptide. Positive mode MALDI-TOF spectra of the HPLC-purified xylosylated peptide untreated (m/z ~1240) (A) or treated for 1 h with methylamine (m/z ~1120) (B) and of endoproteinase Glu-C treated (24 h) unxylosylated peptide (C) or xylosylated peptide (D). C and D, the asterisks indicate peaks also present in a protease-only blank. Arrows indicate positions of ions corresponding to the unxylosylated and xylosylated forms of the peptide.

Mass Spectrometric Analysis of the Purified Xylosylated Peptide-- To analyze the HPLC-purified product further, vapor-phase -elimination using methylamine, a method previously only used for the removal of O-linked GalNAc from peptides and involving replacement of the sugar on the serine by a methylamine group (26), was employed. As expected, treatment of the xylosylated peptide resulted in the expected mass increment of +13 as compared with the original non-xylosylated peptide and of 119 as compared with the xylosylated form (Fig. 6B). This is indicative of a single glycosylation event and suggests that this method should be applicable to the analysis of glycosaminoglycan attachment sites in general.

Subsequent analysis of the peptide was directed at determining which of the two serine residues is xylosylated. Initially, MALDI-TOF analysis after mild acid hydrolysis of the methylamine-treated peptide was performed, but this method did not yield unambiguous data. It was then decided to exploit the presence of a single glutamate residue in the DDDSIEGSGGR substrate peptide by use of endoproteinase Glu-C to digest the xylosylated and unxylosylated forms. The samples were thereafter analyzed by MALDI-TOF in both positive and negative modes for the time-dependent appearance of species absent from incubations containing only the protease. A species of m/z 565, compatible with being a xylosylated form of GSGGR and whose intensity increased with time as the amount of the original m/z 1240 species decreased, was found only with the xylosylated sample in positive mode. In addition, a species of m/z 433, possibly corresponding to the unmodified form of GSGGR, was consistently found to a more significant extent in the digest of the unxylosylated peptide than in that of the xylosylated peptide. The other theoretical product of the proteolysis incubations, DDDSIE, was not detected in positive or negative modes in either a xylosylated or an unxylosylated form. Although not absolutely ruling out some degree of xylosylation of the first serine in the substrate peptide in a manner mutually exclusive of xylosylation of the second serine, the presence of the m/z 565 ion suggests that the enzyme xylosylates the second serine; such a result was expected due to the data cited above showing that xylosylation tends to occur at Ser-Gly sequences (31, 32, 34). This conclusion is also supported by MS/MS experiments (performed on a ThermoFinnigan LCQ Deca XP Quadrupole Ion Trap Mass Spectrometer; data not shown) in which a y fragment ion (m/z 694.3) compatible with xylosylation of the second serine was detected; no ion corresponding to a theoretical fragment of a peptide xylosylated on the first serine was found.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The proteoglycan xylosyltransferase from D. melanogaster is the first such enzyme from invertebrates to be functionally characterized in either recombinant or native form. It is also the first enzyme relevant to the biosynthesis of the GlcA1,3Gal1,3Gal1,4Xyl--O-Ser core common to both chondroitin and heparan sulfates to be characterized from the fly. Although examples of the other core-synthesizing enzymes, i.e. galactosyltransferase I (4GalTVII) (5, 36, 37), galactosyltransferase II (3GalTVI), (38) and glucuronyltransferase I (5, 39), have been characterized from mammalian and nematode sources, the relevant fly genes have not yet been described. For example, it is not known which of the three 1,4-galactosyltransferase homologues from fly acts as galactosyltransferase I and which may have roles in elongation of Fringe-modified O-fucosyl glycans, for example (40).

Of enzymes required later in the biosynthesis of glycosaminoglycan chains in the fly, the activities of only the heparan sulfate 6-O-sulfotransferase (14), the Sugarless UDP-Glc dehydrogenase (required for the biosynthesis of UDP-GlcA, the donor for glucuronyltransferases) (41), and DEXT3, a heparan sulfate-initiating N-acetylglucosaminyltransferase encoded by the botv gene (42), have been determined in vitro. In two other cases, structural analysis of the glycosaminoglycans of relevant fly mutants (specifically tout-velu and sulfateless) allows the activity in vivo of homologues of mammalian enzymes to be inferred (43, 44).

The key to the verification of the fly OXT as an active enzyme was the development of a coupled enzymatic assay. Due to a temporary lack of commercially available non-radioactive UDP-xylose, it was decided to exploit a fungal enzyme (UDP-GlcA decarboxylase) to generate the donor substrate. Also the use of MALDI-TOF mass spectrometry allowed simultaneous product quantification and identification. By using this method it was possible to estimate a yield of secreted Pichia-expressed fly xylosyltransferase of ~1 unit/liter (assuming that 1 µl of 10-fold concentrated supernatant xylosylates around 1 nmol of peptide in 2 h). This is around 2000 times that determined, using a peptide substrate and only 1 µM UDP-[¹⁴C]Xyl, for the activity of the human xylosyltransferase I expressed in CHO-K1 cells (22). Considering also that the expression in Pichia of fly OXT was not even optimized, it is to be expected that improvements on the relatively good yield can be achieved. The fly xylosyltransferase is therefore a realistic tool for generating sufficient xylosylated peptide for studying the effect of the growing glycosaminoglycan chain on the conformation of the core peptide and for the preparation of biological substrates suitable in assays of other enzymes involved in proteoglycan biosynthesis.

Analysis of the sequences of peptide O-xylosyltransferases indicates that they belong to the same group of enzymes as the core 2 and I-synthesizing 1,6-N-acetylglucosaminyltransferases. Although this homology was apparently not noted previously, an examination of the CAZy data base does show that the proteoglycan xylosyltransferases and core 2/I enzymes share a common membership of glycosyltransferase family 14.² The homology of these enzymes of rather contrasting specificities (both for acceptor and donor) indicates a rather flexible process of evolution within this protein family. Interestingly, OXT (whose gene, CG17771, is presently listed in Flybase as encoding a core 2/I-branching enzyme) is the only member of glycosyltransferase family 14 to be identifiable from searching of the Drosophila genome. This is in distinct contrast to mammals (at least two xylosyltransferases and at least 4 functional core 2/I genes) and Caenorhabditis (1 xylosyltransferase, 6 GLY-1 homologues, and 12 other core 2/I-like sequences). It obviously remains to be seen whether other insects and other non-nematode invertebrates also have only one xylosyltransferase and no core 2/I-like sequences. One would also assume from the apparent importance of proteoglycans that oxt is an essential gene and that it would be of interest to perform experiments on this locus at a genetic level to explore the biological repercussions of the abolition of both heparan and chondroitin sulfate biosynthesis.

	ACKNOWLEDGEMENTS

I am most grateful to Dr. Tamara Doering for the UDP-GlcA decarboxylase plasmid, to Katharina Paschinger for critically reading the manuscript, and to Andreas Roitinger of Spectronex GmbH for ion-trap MS/MS analysis. Carbosource is supported in part by National Science Foundation Grant RCN 0090281.

	FOOTNOTES

^* This work was supported by Fonds zur Förderung der Wissenschaftlichen Forschung Grant P13810-GEN and by a Neose Technologies Glycoscience Research award.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EBI Data Bank with accession number(s) AJ430595.

To whom correspondence should be addressed. Tel.: 43-1-36006-6065; Fax: 43-1-36006-6059; E-mail: iwilson@edv2.boku.ac.at.

Published, JBC Papers in Press, April 2, 2002, DOI 10.1074/jbc.M201634200

² P. M. Coutinho and B. Henrissat, Carbohydrate-active enzymes server, afmb.cnrs-mrs.fr/~cazy/CAZY/index.html.

	ABBREVIATIONS

The abbreviations used are: OXT, proteoglycan O-xylosyltransferase; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight spectrometry; HPLC, high pressure liquid chromatography; RP-HPLC, reverse phase-HPLC; MES, 4-morpholineethanesulfonic acid; EST, expressed sequence tag.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				M. Ueyama, H. Takemae, Y. Ohmae, H. Yoshida, H. Toyoda, R. Ueda, and S. Nishihara Functional Analysis of Proteoglycan Galactosyltransferase II RNA Interference Mutant Flies J. Biol. Chem., March 7, 2008; 283(10): 6076 - 6084. [Abstract] [Full Text] [PDF]

				R. R. Dinglasan, A. Alaganan, A. K. Ghosh, A. Saito, T. H. van Kuppevelt, and M. Jacobs-Lorena Plasmodium falciparum ookinetes require mosquito midgut chondroitin sulfate proteoglycans for cell invasion PNAS, October 2, 2007; 104(40): 15882 - 15887. [Abstract] [Full Text] [PDF]

				H. Wang, K. Julenius, J. Hryhorenko, and F. K. Hagen Systematic Analysis of Proteoglycan Modification Sites in Caenorhabditis elegans by Scanning Mutagenesis J. Biol. Chem., May 11, 2007; 282(19): 14586 - 14597. [Abstract] [Full Text] [PDF]

				J. Voglmeir, R. Voglauer, and I. B. H. Wilson XT-II, the Second Isoform of Human Peptide-O-xylosyltransferase, Displays Enzymatic Activity J. Biol. Chem., March 2, 2007; 282(9): 5984 - 5990. [Abstract] [Full Text] [PDF]

				C. Ponighaus, M. Ambrosius, J. C. Casanova, C. Prante, J. Kuhn, J. D. Esko, K. Kleesiek, and C. Gotting Human Xylosyltransferase II Is Involved in the Biosynthesis of the Uniform Tetrasaccharide Linkage Region in Chondroitin Sulfate and Heparan Sulfate Proteoglycans J. Biol. Chem., February 23, 2007; 282(8): 5201 - 5206. [Abstract] [Full Text] [PDF]

				J. Kuhn, C. Prante, S. Schon, C. Gotting, and K. Kleesiek Measurement of Fibrosis Marker Xylosyltransferase I Activity by HPLC Electrospray Ionization Tandem Mass Spectrometry Clin. Chem., December 1, 2006; 52(12): 2243 - 2249. [Abstract] [Full Text] [PDF]

				E. Goda, S. Kamiyama, T. Uno, H. Yoshida, M. Ueyama, A. Kinoshita-Toyoda, H. Toyoda, R. Ueda, and S. Nishihara Identification and Characterization of a Novel Drosophila 3'-Phosphoadenosine 5'-Phosphosulfate Transporter J. Biol. Chem., September 29, 2006; 281(39): 28508 - 28517. [Abstract] [Full Text] [PDF]

				J. E. Pak, P. Arnoux, S. Zhou, P. Sivarajah, M. Satkunarajah, X. Xing, and J. M. Rini X-ray Crystal Structure of Leukocyte Type Core 2 beta1,6-N-Acetylglucosaminyltransferase: EVIDENCE FOR A CONVERGENCE OF METAL ION-INDEPENDENT GLYCOSYLTRANSFERASE MECHANISM J. Biol. Chem., September 8, 2006; 281(36): 26693 - 26701. [Abstract] [Full Text] [PDF]

				C. Gotting, S. Muller, M. Schottler, S. Schon, C. Prante, T. Brinkmann, J. Kuhn, and K. Kleesiek Analysis of the DXD Motifs in Human Xylosyltransferase I Required for Enzyme Activity J. Biol. Chem., October 8, 2004; 279(41): 42566 - 42573. [Abstract] [Full Text] [PDF]

				H. Takemae, R. Ueda, R. Okubo, H. Nakato, S. Izumi, K. Saigo, and S. Nishihara Proteoglycan UDP-Galactose:beta -Xylose beta 1,4-Galactosyltransferase I Is Essential for Viability in Drosophila melanogaster J. Biol. Chem., April 25, 2003; 278(18): 15571 - 15578. [Abstract] [Full Text] [PDF]

				S. Yamada, Y. Okada, M. Ueno, S. Iwata, S. S. Deepa, S. Nishimura, M. Fujita, I. Van Die, Y. Hirabayashi, and K. Sugahara Determination of the Glycosaminoglycan-Protein Linkage Region Oligosaccharide Structures of Proteoglycans from Drosophila melanogaster and Caenorhabditis elegans J. Biol. Chem., August 23, 2002; 277(35): 31877 - 31886. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/24/21207    most recent M201634200v1