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

J. Biol. Chem., Vol. 277, Issue 48, 46328-46337, November 29, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/48/46328    most recent M208024200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van der Wel, H. Articles by West, C. M. Search for Related Content PubMed PubMed Citation Articles by van der Wel, H. Articles by West, C. M. Social Bookmarking                      What's this?

Molecular Cloning and Expression of a UDP-N-acetylglucosamine (GlcNAc):Hydroxyproline Polypeptide GlcNAc-transferase That Modifies Skp1 in the Cytoplasm of Dictyostelium^*

Hanke van der Wel, Howard R. Morris^§¶, Maria Panico^§, Thanai Paxton^§, Anne Dell^§, Lee Kaplan, and Christopher M. West

From the  Department of Anatomy and Cell Biology, University of Florida College of Medicine, Gainesville, Florida 32610-0235, ^§ Department of Biological Sciences, Imperial College of Science, Technology, and Medicine, London SW7 2AY, United Kingdom, and ^¶ M-SCAN Research and Training Center, Silwood Park, Ascot SL5 7PZ, United Kingdom

Received for publication, August 6, 2002, and in revised form, September 17, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Skp1 is a ubiquitous eukaryotic protein found in several cytoplasmic and nuclear protein complexes, including the SCF-type E3 ubiquitin ligase. In Dictyostelium, Skp1 is hydroxylated at proline 143, which is then modified by a pentasaccharide chain. The enzyme activity that attaches the first sugar, GlcNAc, was previously shown to copurify with the GnT51 polypeptide whose gene has now been cloned using a proteomics approach based on a quadrupole/time-of-flight hybrid mass spectrometer. When expressed in Escherichia coli, recombinant GnT51 exhibits UDP-GlcNAc:hydroxyproline Skp1 GlcNAc-transferase activity. Based on amino acid sequence alignments, GnT51 defines a new family of microbial polypeptide glycosyltransferases that appear to be distantly related to the catalytic domain of mucin-type UDP-GalNAc:Ser/Thr polypeptide -GalNAc-transferases expressed in the Golgi compartment of animal cells. This relationship is supported by the effects of site-directed mutagenesis of GnT51 amino acids associated with its predicted DXD-like motif, DAH. In contrast, GnT51 lacks the N-terminal signal anchor sequence present in the Golgi enzymes, consistent with the cytoplasmic localization of the Skp1 acceptor substrate and the biochemical properties of the enzyme. The first glycosylation step of Dictyostelium Skp1 is concluded to be mechanistically similar to that of animal mucin type O-linked glycosylation, except that it occurs in the cytoplasm rather than the Golgi compartment of the cell.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Skp1 is a small (M_r ~ 21,000), rather remarkable protein found in several multiprotein complexes of the cytoplasm and nucleus of all eukaryotes (1-4). Its best known role is as an adaptor in the SCF-class of E3-ubiquitin ligases that mediate the polyubiquitinylation of numerous regulatory proteins, resulting in their subsequent degradation in the 26 S proteasome (5, 6). The presence of Skp1 in this and other multiprotein complexes suggests a special role for Skp1 at the junction of multiple intracellular regulatory pathways. The number of Skp1 genes ranges from probably one in Saccharomyces cerevisiae and humans to two in Dictyostelium and possibly 20 in Caenorhabditis elegans (7-9). It appears that in some organisms different Skp1 functions are fulfilled by specific Skp1 gene products, but it is unclear how these functions are met in organisms with few Skp1 genes.

In Dictyostelium, Skp1 is modified by a pentasaccharide containing the type 1 blood group H antigen (Fuc1,2Gal1,3GlcNAc-) at its core and a probable peripheral Gal1,6Gal1-disaccharide modification at an unknown position on the Fuc (10). The reducing terminal GlcNAc is linked to a Hyp residue at position 143. Approximately 90% of the Skp1 pool appears to be modified in this manner at steady state (11). Evidence summarized elsewhere indicates that glycosylation is important for Skp1 subcellular localization and function (11, 12), and thus, it would be useful to gain a further understanding about the mechanism of this modification.

We previously characterized a Skp1 GlcNAc-Tase¹ activity, which applies the first sugar to hydroxyproline 143 on the polypeptide chain (13). This activity was found only in the cytosolic fraction and had the biochemical characteristics of a soluble cytoplasmic enzyme, including a requirement for a reducing environment and relatively high affinities for its substrates. After purification, activity was associated with GnT51, a protein with an apparent M_r of 51,000 that could be photoaffinity-labeled with a radiolabeled analog of its donor substrate, UDP-GlcNAc. This activity had properties that were distinct from another UDP-GlcNAc PP O--GlcNAc-Tase that modifies Ser/Thr residues on many proteins in the cytoplasm and nucleus (14, 15). In addition to modifying distinct hydroxyamino acids, the latter enzyme is not divalent cation-dependent, and its subunits have larger apparent M_r values. GnT51 also appeared to be distinct from an activity thought to link GlcNAc to Thr residues of a cell-surface glycosylphosphatidylinositol (GPI)-anchored protein in Dictyostelium (16), because this would be expected to be a membrane-associated Golgi enzyme.

To gain further evidence for the cytoplasmic glycosylation model for Skp1, we have cloned the gene for GnT51 and tested for Skp1 GlcNAc-Tase activity by recombinant expression in Escherichia coli. As recently found for a subsequently acting enzyme in this pathway, the -galactosyltransferase/-fucosyltransferase (17, 18), the sequence of GnT51 predicts it to be a soluble cytoplasmic protein. GnT51 appears to belong to a different GTase superfamily than the Ser/Thr O--GlcNAc-Tase and is distantly related to a family of Golgi type II membrane proteins that initiate mucin-type polypeptide O-glycosylation in the secretory pathway.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

MS Sequencing of GnT51-- Gel-purified GnT51 was purified to chromatographic homogeneity on a Superdex-75 column as described (13) and purified further by SDS-PAGE. The Coomassie Blue-stained gel band was subjected to in-gel digestion with trypsin, and the released peptides were sequenced on a Q-TOF hybrid mass spectrometer using non-automated interpretation of spectra (17).

Sequencing the GnT51 Gene-- The gene for GnT51 was sequenced using the approach described previously for the FT85 GTase (17) and explained here under "Results." Methods included querying a database of Dictyostelium DNA sequences (dicty.sdsc.edu) with peptide and nucleotide sequences using tBLASTn and BLASTn programs. This data base contains raw and annotated genomic DNA sequence reads contributed by members of the International Dictyostelium Genome Sequencing Project (19). Peptides and DNA sequences were used to design degenerate and specific oligonucleotides that were employed in pairwise combinations for "touchdown" PCR amplification (20) and linker-mediated PCR (17) of Dictyostelium genomic DNA.

GnT51 Expression Constructs-- Both strands of DNA encoding the first exon of GnT51 (PF9 and PR9) were chemically synthesized (Integrated DNA Technologies, Coralville, IA) with terminal extensions suitable for subsequent ligations: PF9, 5'-AGAAGACATATGAATGAAAATTCTATTTTTGTTTCTATTATAAGTTATAGAGATTCCGCATTCTCGAGATAA. PR9 was the reverse complement of PF9 (see Figs. 3 and 4 for primer locations).

PF9 and PR9 were annealed, incubated for 1 h at 72 °C with 1 mM dATP and 1.25 units of Taq polymerase in 20 µl, cloned into the T/A-cloning site of pCR4-TOPO (Invitrogen) according to the manufacturer's instructions, and amplified in E. coli strain TOP10 One Shot chemically competent cells (Invitrogen). The insert was sequenced in both directions to verify fidelity. Restriction sites present in the insert (underlined) were used to excise and ligate exon 1 between the NdeI and XhoI sites of the expression plasmid pTYB1 (New England Biolabs, MA), resulting in pTYB1-ex1.

The second exon of GnT51 was amplified from Dictyostelium discoideum genomic DNA using the primers PF10, 5'-AGTAGTGAATGCGAATGTCAATGGACTATCAAGAATTTAATTGAA, and PR10, 5'-CTATGGCTCTTCAGCATAAACCGATTTGGGATTTAATTACATACTCCATAA. The amplified DNA was cloned into pCR4-TOPO and sequenced as above. The insert was excised with BsmI and SapI (underlined) and ligated directly into the corresponding sites of pTYB1-ex1 (BsmI site derived from PF9, bold), resulting in an in-frame fusion that conserved the authentic amino acid sequence of GnT51 at the junction between exon-1 and exon-2 and a second in-frame fusion between the C terminus of GnT51 and the IMPACT-CN chitin binding domain (CBD) via an intein linker (Ref. 21; New England Biolabs). This expression plasmid, referred to as pTYB1GnT51, was amplified in TOP10 E. coli cells. pMYB5, which encodes the maltose-binding protein fused to the intein-CBD tag, was used as a control.

pTYB1GnT51 was subjected to site-directed mutagenesis using the QuikChange kit from Stratagene (La Jolla, CA). For the D102A substitution, the sense and antisense primers were 5'-TATCTTCAAATTGCTAGCCATATGAGATTTGTAAAAG and 5'-ATCTCATATGGCTAGCAATTTGAAGATAATATTTTTC, respectively, yielding pTYB1GnT51(D102A). Nucleotide substitutions are in bold, and new diagnostic restrictions sites are underlined. The H104D substitution was created using 5'-TCAAATTGATAGTGACATGAGATTTGTAAAAGATTGG and 5'-ATCTTTTACAAATCTCATGTCACTATCAATTTGAAGA as sense and antisense primers, respectively, yielding pTYB1GnT51(H104D). Note that the sense and antisense primers were staggered slightly because perfectly overlapping primers, created as suggested by the kit, were unsuccessful. pTYB1GnT51(K23R) was created by using a variant of PF10 that contained an A629G substitution (numbered as in Fig. 4).

To create a fusion protein with the CBD-intein at the N terminus of GnT51 (22), the full-length coding region of pTYB1GnT51 was amplified using the following primers: PF11, 5'-ATCCAGCTCTTCCAACATGAATGAAAATTCTATTTTTGTTTCTATTATAAGTTATAG, and PR12, 5'-TTCGCATCGCCTGCAGTTAAATACCGATTTGGGATTTAATTACATACTCCAT. The PCR product was cloned as above, excised with SapI and PstI (underlined), and ligated directly into the corresponding sites of pTYB11 to create pTYB11GnT51(L276S), which encoded a mutant CBD-intein-GnT51 fusion protein. This plasmid contained a PCR-derived t1388c mutation resulting in a L276S amino acid substitution.

Plasmids to encode C-terminal-truncated GnT51 were created using PF11 and either of the antisense primers PR13, 5'-ATTACGCGACCTGCAGTTATTGATTAAATAAAATTAATAACCTTTTTTTTGAA, or PR14, 5'-ATTCGAGGCTCTGCAGTTACAAGGTTATATTTAAGTTCTATTTTAATTCCATCATC. These inserts were cloned into pTYB11 as above to create pTYB11GnT51(1-282) and pTYB11GnT51(1-369+), respectively. pTYB11GnT51(1-369+) contained a nonsense C-terminal extension (VTAGRGSGC) before the stop codon.

Expression and Purification of Recombinant GnT51-- Plasmids were transfected into E. coli strain ER2566 using a CaCl₂ co-precipitate and grown in LB broth containing 100 µg/ml ampicillin at 37 °C on a shaker. When cells achieved an A₆₀₀ of 0.5-0.8, expression was induced by incubation in 0.5 mM isopropyl-1-thio--D-galactopyranoside for 15 h at 10 °C on a gyrator at 200 rpm. Cells were collected by centrifugation, and the pellets were frozen at 80 °C. Cells were lysed by resuspending in 0.5 M NaCl, 20 mM HEPES-NaOH (pH 7.8), 1 mM EDTA, 1 mM tris-(2-carboxyethyl)phosphine, 0.1% Tween 20, and protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, 10 µg/ml aprotinin) and probe-sonicated on ice until the cell suspension became viscous. The lysate was centrifuged at 12,000 × g for 30 min, and the supernatant and pellet were referred to as the S12 and P12 fractions.

To purify expressed protein, each S12 fraction was applied to a 0.6-ml column of chitin equilibrated in wash buffer (0.5 M NaCl, 20 mM HEPES-NaOH (pH 7.8), 0.01% Tween 20, and protease inhibitors (see above)) and washed in the same buffer until the A₂₈₀ fell to below 3% of its initial value. Elution was accomplished by introducing DTT to the column to stimulate autocleavage of the CBD-intein moiety from the GnT51 polypeptide. The column was incubated for 22 h in 50 mM DTT in the wash buffer and eluted with 5 mM DTT in wash buffer. This step was repeated for a second incubation of 36 h, which was then eluted with 50 mM DTT, 0.1% Tween 20 in wash buffer. All manipulations were carried out at 4 °C. Additional elutions were carried out 22 °C or at pH 9.0. Fractions were examined by SDS-PAGE on a 7-20% linear gradient gel followed by staining with Coomassie Blue or Western blotting using anti-CBD from New England Biolabs. M_r standards (SDS-6H) were from Sigma.

GlcNAc-Tase Activity Assays-- The Skp1 GlcNAc-Tase was assayed using either highly purified Skp1 or peptide acceptor substrates (13): HO-Skp1A1-myc + UDP-[³H]GlcNAc  [³H]GlcNAc-O-Skp1A1-myc + UDP and HO-peptide133-155 + UDP-[³H]GlcNAc  [³H]GlcNAc-O-peptide + UDP. The reaction mixture contained 50 mM HEPES-NaOH (pH 7.8), 0.01% (v/v) Tween 80, 10 mM MgCl₂, 2 mM MnCl₂, 5 mM DTT, 0.5 mg/ml bovine serum albumin, and donor and acceptor substrates. UDP-[³H]GlcNAc (PerkinElmer Life Sciences) was present at 1 µM, about 4-fold above its K_m value for GnT51. Skp1A1-myc is a modified form of Skp1 expressed recombinantly with a C-terminal Myc epitope tag. Skp1A1-myc has 2 amino acid substitutions (I34T,D71G) that render it inefficiently modified by GlcNAc in vivo. Purified Skp1A1-myc has only about 15% of the normal level of sugar based on composition analyses, and mass spectrometric analysis suggests approximately equal parts of peptides containing Pro-143 that are hydroxylated or not hydroxylated (10). Skp1A1-myc was present at ~2.5 µM, above its apparent K_m of 0.6 µM (13). Peptide 133-155 (containing Hyp-143) was present at 0.67-1 mM, which is below its apparent K_m of 1.6 mM (13). Reactions containing S12 extracts were supplemented with 1 mM NaF and 1 mM ATP, and those containing purified recombinant GnT51 were supplemented with 2 mM ATP. Reactions were incubated at 22-30 °C for 15-240 min and verified to be linear with respect to time (data not shown). Results presented in the same panel were obtained under the same conditions and are representative of at least two independent trials on different enzyme preparations. Incorporation was assayed by either trichloroacetic acid precipitation or SDS-PAGE followed by scintillation counting of gel slices (13) as indicated.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Sequencing the GnT51 Polypeptide-- GnT51 was purified to near homogeneity and applied to an SDS-PAGE gel that was stained with Coomassie Blue (13). The GnT51 band was cleaved with porcine trypsin in an in-gel digest (17). The released peptides were extracted, cleaned by microcartridge C₁₈ purification, and introduced into a Q-TOF tandem mass spectrometer (23). A doubly charged ion detected in the mass spectrum at m/z 510 was passed for collisional-activated decomposition (CAD) MS/MS. The resulting product ion spectrum (Fig. 1A) could be interpreted via a series of b and y'' ions (Fig. 1B) to give a high confidence sequence for peptide 3 (Table I). A further doubly charged ion at m/z 976 was passed for CAD MS/MS, yielding the product ion mass spectrum shown in Fig. 2A, which was interpreted as deriving from the sequence given in Fig. 2B, designated peptide 1 in Table I. Altogether, this approach yielded five peptide sequences (see Table I) that were candidates for the design of degenerate primers to be used in a PCR approach to amplify GnT51-coding DNA.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Sequencing of peptide 3. A, Q-TOF CAD MS/MS spectrum of the doubly charged 510 ion found in the tryptic digest of gel-purified GnT51 showing the location of fragment ions used in the de novo sequence interpretation. B, The b and y" product ions identified in panel A were assigned as shown to give the sequence FGGYYEER.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Sequencing of peptide 1. A, Q-TOF CAD MS/MS spectrum of the doubly charged 976 ion found in the tryptic digest of gel-purified GnT51 showing the location of fragment ions used in the de novo sequence interpretation. B, The b and y" product ions identified in panel A were assigned as shown to give the sequence (L/I)(L/I)(L/I)VASGFGENDGF(L/I)R (leucine, L, and isoleucine, I, have the same mass).

Sequencing the GnT51 Gene-- The peptide sequences were initially used to screen a conceptually translated (tBLASTn) data base containing DNA sequences obtained from randomly fragmented Dictyostelium DNA. The sequence of a 137-nt fragment was found to encode peptide 3 and also the partially characterized neighboring peptide 6 in the same frame (labeled in Fig. 4). The nucleotide sequence of this fragment was used to design PCR primers in both forward and reverse orientations to be used in association with degenerate primers from peptides 1 and 2, also designed in both orientations, to amplify GnT51 DNA from genomic DNA. Degenerate primers PF1 (5'-MAACCAAAYGATGATAAYAAYATGG) and PF2 (5'-TCWGGTTTYGGTGAAAAYGATGGTTT), when combined with reverse-oriented primers from the 137-nt fragment (PR1, 5'-ATAATCATCTAAACTTTTAATTTTACC, and PR2, 5'-CTCCATAATGGGATTCATAAAAAATGT), yielded discrete products with lengths of ~770 and 750 nt, respectively, as depicted in Fig. 3A. Interestingly, these PCR products were amplified despite significant sequence differences between primer and genomic DNA template sequences. These differences arose in one case from an inability of the MS analysis to resolve between an Asn residue and a Gly-Gly pair in peptide 2 (which have the same mass), albeit this residue was assigned as beyond the high confidence region of the interpretation in Table I, and in the other case from a probable error in the one-pass sequencing of the original 137-bp fragment from the Dictyostelium sequencing project. The success of the approach despite these problems, which are inherent in the methodologies involved here, attests to the robustness of the touchdown PCR strategy employed (20).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   GnT51 gene cloning strategy, domain model, and expression constructs. A, the sequence of peptide 3 (from Fig. 1) matched an ORF found in a 137-nt sequence read (fragment 1) in a database maintained by the Genome Sequencing Database at Jena, Germany. A nucleotide sequence corresponding to peptide 3 and degenerate oligonucleotides designed from peptides 1 (from Fig. 2) and 2 (Table I) were used to amplify additional genomic DNA using touchdown-PCR. Adjacent DNA sequences were obtained using linker-mediated PCR and were confirmed by sequence data from the International Dictyostelium Genome Project (fragments 1, 2, and 3). B, the two-exon model predicted to encode GnT51, showing the locations of the sequenced GnT51 peptides (from Table I). Predicted flanking intergenic DNA and adjacent ORFs, with the directions of transcription indicated with arrows, are indicated. Positions of primers used to construct a full-length cDNA are shown. C, predicted domains of GnT51 based on sequence alignments and motifs observed. Primers used to construct expression constructs in panel D are shown. D, DNA constructs for expressing normal and mutant forms of GnT51 in E. coli strain ER2566. Positions of site-directed mutations are indicated. All proteins were initially expressed as either C- or N-terminal fusions with intein-CBD (as shown). The size of the self-cleaving intein-CBD domain (Mr 56,000) is not drawn to scale.

Adjacent sequences were obtained by linker-mediated PCR using primers shown in Figs. 3A and 4, with partial confirmation coming from additional sequences deposited later in the International Dictyostelium DNA Sequencing Consortium database. PCR-4 spanned upstream from GnT51 coding DNA across 443 nt of A/T-rich (84%) putative intergenic DNA into a new ORF oriented in the opposite direction. PCR-5 spanned downstream of GnT51 coding DNA across 567 nt of A/T-rich (88%) intergenic DNA into a new ORF. A large, central ORF consisting of 1220 nt encoded all five peptides from the MS sequencing (Table I) but lacked a suitable 5'-start codon. Its A/T content of 78% was within the normal range of Dictyostelium coding DNA (19). In addition, a second, short 49-nt ORF (78% A/T content) was identified upstream of the main ORF, separated by 118 nt of DNA with an A/T content of 93%. This intervening DNA was terminated by a consensus intron motif (GT ... AG) that fell between the first and second positions of the codon predicted to join the two candidate exons. These characteristics are typical of Dictyostelium introns (19). The initial Met residue of the candidate exon-1 lies in a context typical of Dictyostelium start codons (24) and is preceded by multiple in-frame stop codons. The 2-exon model yielded a predicted coding region of 423 amino acids with an M_r of 49,114, similar to the apparent M_r of GnT51 of 51,000, based on SDS-PAGE (13). 28% of the amino acid sequence could be verified by comparison with the original MS/MS analysis, which had been optimized to identify and sequence peptides for PCR primers (Fig. 4).

View larger version (44K): [in this window] [in a new window]   Fig. 4.   GnT51 gene sequence. The DNA sequence of the region between the start codon of the reverse-oriented, predicted upstream ORF, and the start codon of the downstream ORF, as illustrated in Fig. 3B, is shown. The predicted GnT51 ORF consists of two exons. Tryptic peptide sequences obtained from the initial MS sequencing (Table I) are bold underlined, and peptides subsequently confirmed from the MS data are in bold. PCR primer positions shown in Fig. 3 are underlined (note: the degeneracy of PF1 and PF2, the 5'-extensions of PF5 and PR5, and mismatches in PR1 and PR2 (see "Results") are not shown). Predicted polyadenylation signals in bold are downstream of the stop codon. The position of the intron was concluded as described under "Results." The sequence shown corresponds to GenBankTM accession AF375997.

Properties of the Predicted GnT51 Sequence-- The first 280 amino acids of the predicted GnT51 sequence show distant similarity to the catalytic domain of family GT27 animal mucin-type PP -GalNAc-Tases (25). Two regions of greatest similarity are shown in Fig. 5A. The first is similar to the previously identified NRD-2 (nucleotide recognition domain-2) seen in family GT2 and GT27 GTases (26) and includes a highly conserved DXH tripeptide motif (denoted with #) corresponding to the metal binding DXD motif seen in many GTase families. The second region corresponds to but extends in both directions beyond the so-called Gal/GalNAc motif shared between 4-GalTases, PP -GalNAc-Tases, and other GTase families (25). Together these regions are predicted to comprise the catalytic domain of GnT51 (Fig. 3C). This similarity is consistent with their related functions, the addition of a HexNAc residue to a protein hydroxyamino acid. An even greater similarity is seen to sequences from four predicted genes and one known gene in prokaryotic and eukaryotic microbe genomes, as discussed in West et al. (12 and 27). A region of high similarity with these microbial sequences lies in the predicted exon-1 and extends through the predicted exon-1/exon-2 boundary (Fig. 5B), providing additional support for the 2-exon model.

View larger version (63K): [in this window] [in a new window]   Fig. 5.   Alignment of GnT51-like sequences. A, sequences from selected regions of GnT51 were aligned with the nucleotide recognition domain-2 (NRD-2)-like and Gal/GalNAc-like regions of the catalytic domains of seven animal family GT27 PP -GalNAc-Tases. The amino acids are color-coded based on their hydrophobic (green), basic (dark red), acidic (blue), structure-breaking (red), or polar (black) characters. Residues that are identical in the majority of -GalNAc-Tase sequences are bold and that have chemical qualities that are related in the majority of -GalNAc-Tase sequences are highlighted as yellow (L, I, V, M, F, Y, or sometimes A), dark gray (R, K, or H), light gray (D, E, N, or Q), or turquoise (S, T, C, A, G, or P). The perfectly conserved DXD-like DXH-motif is denoted with number (#) symbols, and other perfectly conserved residues are denoted with asterisks. B, the sequence surrounding the predicted exon 1/exon 2 splice junction from GnT51 is aligned with the corresponding region of predicted gene sequences from bacteria and eukaryotic microbes. An overall alignment of these sequences is given in West (27). The arrow denotes the exon 1/exon 2 boundary of GnT51 as deduced in Fig. 4. Numbers in parentheses represent the length of the upstream coding region in amino acids (where known). Origin of sequences: C. elegans (C. e.) Gly3, gi3047186; C. elegans Gly5a, gi3047190; C. elegans Gly7, gbAF031841; C. elegans Gly8, gbAF031842; C. elegans Gly9, gbAF031843; Mus musculus (M. m.) GalNac-Tase-T1, gi2149049; Homo sapiens (H. s.) GalNAc-Tase-T3, gi4758413; Synechecoccus sp. (S. sp.) WH8102 putative GT-A, DOE 84588, Contig48.gene455; Yersinia pestis (Y. p.) putative GT-A, gbAJ414154.1, gi15980810; D. discoideum (D. d.) cis4c, gbAF509501; Leishmania major (L. m.) putative GT-A, gbAC084317.4, gi18640648; Trypanosoma brucei (T. b.) putative GT-A (assembled from gi11836577, gbAQ951144, gbAQ942881, gbAZ219465, gbAQ639524).

The region of GnT51 aligned in Fig. 5B beginning with amino acid 5 appears to lie at the N terminus of the catalytic domain based on similarity to family GT2 and GT27 catalytic domains (12). Thus, there is no signal anchor motif at the N terminus of GnT51 that could direct the protein into the secretory pathway, consistent with the cytoplasmic localization of GnT51 predicted by its biochemical properties and cell fractionation studies (13).

At the C-terminal end of the putative catalytic domain (after Phe-280, Fig. 4) lies a stretch of 67 amino acids that consists of several homopolymeric tracts of Asn interrupted by Ser, Asp, Ile, and Thr. Poly-Asn tracts have been seen in other Dictyostelium proteins (e.g. Ref. 17), and the coding DNA for this sequence exhibits an unusually low threshold for unwinding in model studies (28). Beyond this, at the C-terminal end of the protein lies a 66-residue segment with a novel sequence and unknown function. Family GT27 GTases also have a C-terminal domain of about 150 residues (25), which has homology to lectin domains and might be involved in acceptor substrate recognition (29).

Recombinant GnT51 Exhibits Skp1 GlcNAc-Tase Activity-- To verify that the predicted protein product of this gene exhibits Skp1 GlcNAc-Tase activity, an expression plasmid (pTYB1GnT51) was designed that fused the two exons with a C-terminal, autocleavable, intein-bridged chitin binding domain to facilitate purification (Fig. 3D). This construct was expected to yield, after cleavage, the authentic amino acid sequence of GnT51. pTYB1GnT51 was transfected into E. coli strain ER2566, and expression was induced with isopropyl-1-thio--D-galactopyranoside (see "Experimental Procedures"). A maltose-binding protein fusion fused to intein-CBD was expressed as a negative control. Soluble (S12) and insoluble (P12) extracts of these cells, which contained about 75 and 25% of total cell protein, respectively, were analyzed by SDS-PAGE and staining for total protein with Coomassie Blue and by Western blot analysis with an anti-CBD antibody (Fig. 6, A, lanes a and k, and B, lanes a and k). All strains exhibited novel bands in both the soluble and insoluble fractions corresponding to the predicted M_r values of the proteins expressed (Table II). Pilot studies showed that substantially higher levels of GnT51-related proteins were present in the soluble fraction when E. coli were grown at 10 °C compared with 22 and 37 °C (data not shown), although the level of recombinant protein still appeared lower than in the insoluble fractions.

View larger version (66K): [in this window] [in a new window]   Fig. 6.   Expression of GnT51 constructs in E. coli and their purification. A and B, soluble S12 and insoluble P12 extracts from E. coli strains expressing normal and mutant GnT51 constructs were prepared, and 75 µg of protein from each was applied to an SDS-PAGE gel and analyzed by either Coomassie Blue (panel A) or Western blotting using anti-CBD antibody (panel B). Positions of GnT51s and other proteins are indicated. F.L., full-length GnT51. The light intensities of the full-length bands in lanes a and b in panel B result from a blotting artifact in this experiment. Asterisks (*) mark the positions of truncated GnT51s and are labeled in panel B. Mr standards in lanes i and k correspond to the top four markers in panel C. C, the S12 fractions were adsorbed to chitin columns. GnT51s were eluted after self-cleavage of the intein-CBD domain in the presence of 50 mM DTT and analyzed by SDS-PAGE followed by staining for protein with Coomassie Blue. At the bottom, relative volumes of the preparation assayed in Fig. 8, adjusted to correct for different GnT51 levels, are indicated.

The S12 extract from E. coli cells expressing the GnT51-intein-CBD fusion exhibited a level of Skp1 GlcNAc-Tase activity that was increased about 6-fold over a control that lacked added Skp1A1-myc and a control extract from cells expressing maltose-binding protein-intein-CBD (Fig. 7A). GnT51 was subjected to purification by application of the extract onto a chitin column and elution with 50 mM DTT. As shown in the next section, this treatment induced self-cleavage of the intein bridge from the C terminus of the GnT51 sequence and release of GnT51 together with other proteins thought to be the GroEL and DnaK chaperone proteins. As expected, the specific activity of this fraction using Skp1A1-myc as the acceptor was increased (Fig. 7B). The specific activity in the presence of saturating levels of UDP-GlcNAc (1 µM) and Skp1A1-myc (2.5 µM) was estimated as 8.4 nmol/h/mg of GnT51 protein, similar to the value of 12.6 nmol/h/mg reported for GnT51 purified from Dictyostelium (13). These numbers may not, however, represent true V_max values for GnT51, as the Skp1A1-myc utilized for these assays is a mutant form that is inefficiently modified by GlcNAc in vivo (13).

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Skp1 GlcNAc-Tase activity of recombinant GnT51 preparations. A, crude soluble E. coli lysates (S12 fraction) from cells expressing GnT51 (pTYB1GnT51) or maltose-binding protein (MBP) (pMYB5) were assayed in the presence or absence of 2 µM Skp1A1-myc, as indicated, using the trichloroacetic acid precipitation assay. B, the S12 fraction was applied to a chitin column, and cleaved and eluted using 50 mM DTT. The GlcNAc-Tase activity of the S12 and eluted fractions was determined using the acceptors Skp1A1-myc (2.5 µM) or peptide133-155 (Hyp143) (1 mM) by the SDS-PAGE assay. Values from reaction blanks containing no added substrate were subtracted. Protein content of the S12 fraction was determined using a modification of the Bradford reaction, and the protein content of the eluted material was estimated from the Coomassie blue staining intensity of the Mr 51,000 band after SDS-PAGE, relative to Mr standards. C, comparison of ratios of activity of GnT51 preparations with respect to Skp1A1-myc and peptide133-155 (Hyp143). All reactions contained 1 µM UDP-[3H]GlcNAc, and were carried out for 15 min (S12 fractions) or 2 h (chitin-affinity fractions).

The chitin-purified fraction was stable for at least 24 h at 5 °C in the presence of 1 mM DTT, 10 mM MgCl₂, MnCl₂ but was unstable in the absence of the divalent cations. Divalent cations were also required for activity (see below). Dilution of DTT to 0.02 mM resulted in an 86% loss of activity compared with dilution in 1 mM DTT. These characteristics are similar to the native enzyme purified from Dictyostelium (13).

These fractions also transferred ³H to the Skp1-derived peptide 133-155 (Hyp-143) (Fig. 7B). Because peptide 133-155 (Pro-143) was inactive (data not shown; see below), Hyp appears to be the site of attachment as previously described (13). The chitin column purification resulted in a smaller enrichment in peptide compared with Skp1 GlcNAc-Tase activity. Possibly the C-terminal intein CBD had selectively interfered with recognition of the Skp1 acceptor before cleavage because of its bulky size. The Skp1A1-myc:peptide 133-155 activity ratio of the purified recombinant GnT51 was very similar to that of authentic GnT51 purified from Dictyostelium (Fig. 7C). Taking this result together with their similar specific activities, the recombinant enzyme, which is expected to have the identical sequence to native GnT51, appears to exhibit normal GlcNAc-Tase function.

Site-directed Mutagenesis of GnT51-- To further examine the apparent similarity between GnT51 and the family GT27 mucin-PP -GalNAc-Tase enzymes, the dependence of GnT51 GlcNAc-Tase activity on the highly conserved DXH motif was determined. A previous study showed that the activity of murine PP -GalNAc-Tase T1 is abolished when the Asp and His residues are separately changed to Ala and Asp, respectively (25). The same Asp  Ala and His  Asp mutations were introduced into the Gnt51-intein-CBD construct described above, as described under "Experimental Procedures" and depicted in Fig. 3D. In addition, the activities of two constructs deleted for C-terminal regions that extend beyond the predicted catalytic domain and two serendipitous point mutants were tested.

The mutant plasmid constructs were expressed in E. coli, and soluble S12 and insoluble P12 extracts were examined by SDS-PAGE followed by staining with Coomassie Blue or by Western blotting and probing with an anti-CBD antibody, as described above. Each soluble extract contained comparable levels of a novel protein that could be identified in the Coomassie-stained gel and detected with the anti-CBD antibody in the Western blot (Fig. 6, A and B; listed in Table II). These extracts were applied to chitin columns and the eluted fractions containing cleaved GnT51 were examined by SDS-PAGE and staining with Coomassie Blue (Fig. 6C). Each sample contained a new band corresponding to the expected position of the GnT51 isoform intended to be produced in that strain (Table II). Higher levels were seen for the constructs that originally had intein-CBD appended to the C terminus (lanes a-d) than to the N terminus (lanes e-g), but the similar levels of protein in each group suggested that the mutations did not alter the solubility or stability of the GnT51 polypeptide. The difference in level between groups may be due to differences in cleavage efficiency of the peptide bond linking intein to GnT51. In addition, each sample contained a prominent band at the M_r 60,000 position and a minor band at M_r 70,000. These are likely to be the chaperone proteins GroEL and DnaK, which have been found to copurify with other intein-CBD fusion proteins (30). The high level of GroEL may be due to expression of GnT51 in E. coli cells at 10 °C or cell lysis in the presence of EDTA, which might deplete GnT51 of an essential divalent cation. The minor band at M_r 54,000 is likely to be cleaved intein-CBD that leached from the chitin column during elution, based on Western blot analysis with anti-CBD antibody (data not shown), as previously reported to occur for some fusion proteins (21). Because the purified preparation of recombinant GnT51 exhibited the expected specific activity and ratio of activities with respect to its Skp1 and peptide substrates (see above), it appears that the presence of GroEL, DnaK, and intein-CBD did not interfere with GnT51 GlcNAc-Tase activity, possibly because of the inclusion of 2 mM ATP in the reaction to support chaperone folding functions. This is consistent with reports that the kinase activity of C-terminal Src kinase (30) and the DNA binding activity of Sinorhizobium NodD transcriptional factor (31) were not affected by high levels of co-purifying GroEL.

The level of Skp1A1-myc GlcNAc-Tase activity with respect to Skp1A1-myc was not affected by the conservative K23R mutation (Fig. 8A). In contrast, the D102A and H104D substitutions each reduced activity to undetectable levels (<0.2% of wild type), equivalent to the effect of these mutations on murine PP GalNAc-Tase T1 (25). Each of the two C-terminal truncations was equally deleterious to GnT51 activity. Finally, the L276S mutation reduced activity to about 7% or 30-fold above the detection threshold. A hydrophobic residue is conserved at this position in other GnT51-related sequences, and thus, this residue may be important for GnT51 folding.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Skp1 GlcNAc-Tase activity of mutant GnT51s. A, aliquots of the GnT51 preparations shown in Fig. 6C, normalized with respect to GnT51 protein level (as listed in Fig. 6C), were assayed for GlcNAc-Tase activity with respect to Skp1A1-myc (2.5 µM) using the SDS-PAGE assay. B, after dialysis of the GnT51 preparations shown in Fig. 6C, 3-fold larger volumes were used to assay activity with respect to peptide 133-155 (Hyp-143) and peptide 133-155 (Pro-43), present at 0.67 µM, using the SDS-PAGE assay. Note that a logarithmic scale is used in panel B to facilitate display of the full range of activities. UDP[3H]GlcNAc was present at 1 µM, and reactions were incubated for 4 h. MBP, maltose-binding protein. w/t, wild type.

The mutant GnT51s were also assayed using peptide-133-155 (Hyp-143) acceptor because enzyme recognition of this small substrate might be distinct from that of the full-length Skp1 protein. As seen using the Skp1 substrate, high activity levels were observed for the normal and K23R mutant (Fig. 8B). Incorporation was specific for Hyp-143, because peptide 133-155 (Pro-143) was inactive. No activity could be detected for the D102A, H104D, GnT51(1-282), and the GnT51(1-369+) mutants, whose incorporation values were indistinguishable from controls that contained either a chitin eluate from the maltose-binding protein strain or EDTA to inactivate the enzyme. The L276S mutant exhibited about 3% of normal activity. Thus, the activities of the mutant GnT51s were qualitatively similar with respect to the two acceptor substrates.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The Skp1-Hyp GlcNAc-Tase GnT51 was previously proposed to be encoded by GnT51 based on its co-purification to apparent homogeneity coordinate with 130,000-fold purification of the enzyme activity (13). This was tested here by cloning and analyzing the gene that encodes GnT51. The gene was identified beginning with a sensitive MS-based approach to obtain GnT51 tryptic peptide sequences from the original purified material (Figs. 1 and 2; Table I). One of the tryptic sequences was used to identify a candidate partial coding DNA fragment previously sequenced in a random genomic DNA sequencing study (see Ref. 32). The other sequences were used to design degenerate oligonucleotide primers to extend this sequence using PCR methods (Fig. 3). This approach predicted the existence of a 2-exon open reading frame that encodes a putative M_r 49,114 protein (Figs. 3 and 4), similar to predictions for GnT51 of 51,000 and 45,000 from SDS-PAGE and size exclusion chromatography (13). 28% of the predicted amino acid sequence was accounted for in the original MS study, optimized to obtain sequence information to design the PCR primers (Fig. 4). Its role as the Skp1 GlcNAc-Tase is supported by the finding that after recombinant expression in and purification from E. coli, GnT51 was similar in certain catalytic properties to GnT51 isolated from Dictyostelium (Fig. 8). Other approaches will be required to verify that GnT51 is required for the GlcNAc modification on Skp1 in vivo, although no closely related genes have been detected in Dictyostelium DNA sequence databases, which have fairly complete coverage of genomic DNA (32). Up to now, the Gnt51 locus has resisted gene replacement² using an approach that was successful in inactivating the FT85 locus (17), so it might be an essential gene.

GnT51 was previously predicted to be a conventional GTase based on its use of the standard sugar nucleotide donor UDP-GlcNAc in vitro (13), but its soluble nature was unusual. Like the soluble FT85 GTase that acts subsequently to GnT51 to add -galactose and -fucose (17, 18), both native and recombinant GnT51 requires a divalent cation, reducing conditions, and near-neutral pH for optimal activity. Although sequences significantly related to GnT51 could not be found in GenBank^TM using standard BLAST analysis, an initial alignment with certain family GT2 GTases produced by Bernard Henrissat (AFMB-CNRS, Marseilles, France) led subsequently to the recognition of closer similarities to the related GT27 family of mucin-type PP -GalNAc-Tases (33)³ found as type II membrane proteins in the Golgi apparatus of animal cells (25). This relationship can also be detected using the PHI-BLAST algorithm (35) if GnT51-like sequences from prokaryotes and lower eukaryotes (12) are included to develop the hit pattern. Similarity extends throughout the length of the previously described 300-amino acid catalytic domain of the family GT27 GTases. Although the amino acid identity is quite low (14% compared with murine -GalNAc-Tase T1), the similarity is greater (45%) and is high at several highly conserved regions (Fig. 5A) and is supported by a phylogenetic analysis of the sequences (12, 27). An association with family GT27 is supported by the finding that mutagenesis of either of two residues, Asp and His of the highly conserved DXD-like DXH motif (denoted by # in Fig. 5A), had similar incapacitating effects on activity without affecting the levels of the recombinant protein expressed in E. coli (Figs. 6 and 8). The related sequence motifs between GnT51 and the family GT27 enzymes are consistent with their similar functions in transferring a HexNAc residue from a UDP-HexNAc donor to a hydroxyamino acid acceptor site in the recipient protein. The marked sequence divergence between GnT51 and family GT27 sequences is likely to be due in part to the long evolutionary time separating Dictyostelium and animals.

The sequence of the predicted catalytic domain of GnT51 starts very near to the N terminus of the protein (residue 5) unlike conventional eukaryotic Golgi GTases, which have an N-terminal signal anchor motif and a spacer domain preceding the catalytic domain (36). There are no sequences for targeting the protein to the rough endoplasmic reticulum and Golgi, which is congruent with the soluble nature of GnT51, its presence in the cytosolic fraction of cell extracts, and its catalytic dependence on reducing conditions as found in the cytoplasm in vivo. Consistent with this model, Cys residues thought to form disulfide bonds in the family GT27 proteins are not conserved in GnT51.

GnT51 is most closely related to sequences that potentially encode cytoplasmic GTases in a proteobacterium and a cyanobacterium (12, 27), and together these sequences define family GT60 (33).³ A phylogenetic analysis of the related sequences from prokaryotes and lower and higher eukaryotes suggests that GnT51 radiated evolutionarily from an ancient lineage of GTases that later gave rise to two classes of Golgi mucin-type PP HexNAc-Tases, -GlcNAc-Tases in eukaryotic microbes and family GT27 -GalNAc-Tases in animals. Thus, the GnT51 lineage may have originated and remained as a cytoplasmic enzyme, whereas other branches of the lineage acquired an N-terminal signal anchor motif and became compartmentalized into the Golgi.

The C-terminal end of the family GT27 PP GalNAc-Tases contains a lectin-like sequence of about 150 residues, which in the case of human GalNAc-T4 has been suggested to direct specificity toward partially glycosylated polypeptide acceptors based on point mutational analysis (29). Thus the C-terminal domain may be involved in acceptor recognition. GnT51 also contains a C-terminal region of about 150 amino acids consisting of 80 residues of mostly Asn and 65 residues of novel sequence at the very C terminus (Fig. 3C). Poly-Asn stretches are present in other Dictyostelium proteins including the di-GTase FT85, where it separates the two catalytic domains (17, 18), and it may play a spacer role in GnT51. No catalytic activity was detected in the constructs lacking the C-terminal 65 or 145 amino acids using either the peptide or Skp1 substrates (Fig. 8), suggesting that in GnT51 this region of the protein is essential for protein structure or catalysis. The C-terminal domain may be oriented toward the acceptor substrate because the form containing the C-terminal intein-CBD domain prefers the peptide over the full-length Skp1 substrate (Fig. 7C), suggesting that the larger Skp1 is sterically blocked. Further study will be required to determine whether the C-terminal domains of GnT51 and family GT27 enzymes have homologous functions.

The family GT27 PP -GalNAc-Tases retains the configuration of the GalNAc linkage in the sugar nucleotide donor, suggesting that this may also occur for the linkage formed by GnT51. This would differentiate the GlcNAc-Hyp linkage from that formed by the ubiquitous O--GlcNAc-Tase of the cytoplasm and nucleus (14, 15) and may help explain why the GlcNAc on Skp1 is further modified, whereas -linked GlcNAc residues are not. However, this is not the only explanation because the GlcNAc1-Ser/Thr structure formed by the -toxin of Clostridium novyi on Rho-related proteins in animal cells (37) is also not extended. There is also very little sequence similarity between the -toxin and GnT51 except for their DXD-like motifs. However, sequences related to GnT51 are found in the genomes of Dictyostelium and Trypanosoma cruzi (12, 27), and these are predicted to be type II membrane proteins that are therefore candidates for forming the GlcNAc1-Thr/Ser linkages formed on glycosylphosphatidylinositol (GPI)-anchored cell surface proteins produced in the Golgi of these organisms (16, 34, 38). Recent studies indicate the Dictyostelium gene is necessary and sufficient for glycosylating cell surface proteins that bear this structure in this organism,⁴ supporting the interpretation that GnT51 attaches GlcNAc in -linkage onto Hyp143 in Skp1.

	ACKNOWLEDGEMENTS

We are grateful to Sharon Compton for helping to clone the GnT51 gene, Wang Fei for assistance in the plasmid constructions, and Bernard Henrissat (AFMB-CNRS, Marseilles, France) for the initial analysis of the GnT51 sequence. Data accessed at genome.imb-jena.de/dictyostelium were obtained at the Institute of Biochemistry I, Cologne and the Genome Sequencing Centre Jena (G. Glöckner, A. Rosenthal, L. Eichinger, and A. Noegel, Jena, Germany) with support by Deutsche Forschungsgemeinschaft Grant 113/10-1 and 10-2. Sequences from the Baylor Sequencing Center (A. Kuspa and R. Gibbs, Houston, TX) were obtained through the support of the NICHD, National Institutes of Health. Some searches were performed courtesy of National Biomedical Computation Resource (National Institutes of Health Grant P41-RR80605) at the San Diego Supercomputer Center.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grant R01-GM37539 (to C. M. W.) and the Wellcome Trust and Biotechnology and Biological Sciences Research Council, Swindon, United Kingdom (to H. R. M. and A. D.).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.

To whom correspondence should be addressed: Dept. of Anatomy and Cell Biology, University of Florida College of Medicine, 1600 S. W. Archer Rd., Gainesville, FL 32610-0235. Tel.: 352-392-3329; Fax: 352-392-3305; E-mail: westcm@ufl.edu.

Published, JBC Papers in Press, September 19, 2002, DOI 10.1074/jbc.M208024200

² Kaplan, L., Compton, S., and West, C. M., unpublished data.

³ P. M. Coutinho and B. Henrissat (1999) Carbohydrate-active Enzymes server at URL: afmb. cnrs-mrs. fr/~cazy/CAZY/index. html.

⁴ F. Wang, H. van der Wel, T. Metcalf, L. Kaplan, and C. M. West, unpublished data.

	ABBREVIATIONS

The abbreviations used are: GlcNAc-Tase, N-acetylglucosaminyltransferase; GalNAc-Tase, N-acetylgalactosaminyltransferase; PP, polypeptide; CBD, chitin binding domain; DTT, dithiothreitol; GTase, glycosyltransferase; Hyp, hydroxyproline; MS, mass spectrometry; CAD, collisional-activated decomposition; contig, group of overlapping clones; ORF, open reading frame; nt, nucleotide(s).

	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:

				N. Heise, D. Singh, H. van der Wel, S. O Sassi, J. M Johnson, C. L Feasley, C. M Koeller, J. O Previato, L. Mendonca-Previato, and C. M West Molecular analysis of a UDP-GlcNAc:polypeptide {alpha}-N-acetylglucosaminyltransferase implicated in the initiation of mucin-type O-glycosylation in Trypanosoma cruzi Glycobiology, August 1, 2009; 19(8): 918 - 933. [Abstract] [Full Text] [PDF]

				C. M. West, H. van der Wel, and Z. A. Wang Prolyl 4-hydroxylase-1 mediates O2 signaling during development of Dictyostelium Development, September 15, 2007; 134(18): 3349 - 3358. [Abstract] [Full Text] [PDF]

				T. Metcalf, H. van der Wel, R. Escalante, L. Sastre, and C. M. West Role of SP65 in Assembly of the Dictyostelium discoideum Spore Coat Eukaryot. Cell, July 1, 2007; 6(7): 1137 - 1149. [Abstract] [Full Text] [PDF]

				H. H. Ho, C. Y. He, C. L. de Graffenried, L. J. Murrells, and G. Warren Ordered assembly of the duplicating Golgi in Trypanosoma brucei PNAS, May 16, 2006; 103(20): 7676 - 7681. [Abstract] [Full Text] [PDF]

				A. Ercan and C. M. West Kinetic analysis of a Golgi UDP-GlcNAc:polypeptide-Thr/Ser N-acetyl-{alpha}-glucosaminyltransferase from Dictyostelium Glycobiology, May 1, 2005; 15(5): 489 - 500. [Abstract] [Full Text] [PDF]

				H. van der Wel, A. Ercan, and C. M. West The Skp1 Prolyl Hydroxylase from Dictyostelium Is Related to the Hypoxia-inducible Factor-{alpha} Class of Animal Prolyl 4-Hydroxylases J. Biol. Chem., April 15, 2005; 280(15): 14645 - 14655. [Abstract] [Full Text] [PDF]

				C. Ketcham, F. Wang, S. Z. Fisher, A. Ercan, H. van der Wel, R. D. Locke, k. Sirajud-Doulah, K. L. Matta, and C. M. West Specificity of a Soluble UDP-Galactose:Fucoside {alpha}1,3-Galactosyltransferase That Modifies the Cytoplasmic Glycoprotein Skp1 in Dictyostelium J. Biol. Chem., July 9, 2004; 279(28): 29050 - 29059. [Abstract] [Full Text] [PDF]

				F. Wang, T. Metcalf, H. van der Wel, and C. M. West Initiation of Mucin-type O-Glycosylation in Dictyostelium Is Homologous to the Corresponding Step in Animals and Is Important for Spore Coat Function J. Biol. Chem., December 19, 2003; 278(51): 51395 - 51407. [Abstract] [Full Text] [PDF]

				H. van der Wel, S. Z. Fisher, and C. M. West A Bifunctional Diglycosyltransferase Forms the Fucalpha 1,2Galbeta 1,3-Disaccharide on Skp1 in the Cytoplasm of Dictyostelium J. Biol. Chem., November 22, 2002; 277(48): 46527 - 46534. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/48/46328    most recent M208024200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van der Wel, H. Articles by West, C. M. Search for Related Content PubMed PubMed Citation Articles by van der Wel, H. Articles by West, C. M. Social Bookmarking                      What's this?