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

J. Biol. Chem., Vol. 275, Issue 33, 25315-25321, August 18, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/33/25315    most recent M002630200v1 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 Keusch, J. J. Articles by Baenziger, J. U. Search for Related Content PubMed PubMed Citation Articles by Keusch, J. J. Articles by Baenziger, J. U. Social Bookmarking                      What's this?

Cloning of Gb₃ Synthase, the Key Enzyme in Globo-series Glycosphingolipid Synthesis, Predicts a Family of 1,4-Glycosyltransferases Conserved in Plants, Insects, and Mammals^*

Jeremy J. Keusch^§, Stephen M. Manzella^¶, Kwame A. Nyame, Richard D. Cummings, and Jacques U. Baenziger

From the  Department of Pathology, Washington University School of Medicine, St. Louis, Missouri, 63110 and the  Department of Biochemistry and Molecular Biology, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73190

Received for publication, March 28, 2000, and in revised form, June 13, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have cloned Gb₃ synthase, the key 1,4-galactosyltransferase in globo-series glycosphingolipid (GSL) synthesis, via a phenotypic screen, which previously yielded iGb₃ synthase, the 1,3-galactosyltransferase required in isoglobo-series GSL (Keusch, J. J., Manzella, S. M., Nyame, K. A., Cummings, R. D., and Baenziger, J. U. (2000) J. Biol. Chem. 33). Both transferases act on lactosylceramide, Gal1,4Glc1Cer (LacCer), to produce Gb₃ (Gal1,4LacCer) or iGb₃ (Gal1,3LacCer), respectively. GalNAc can be added sequentially to either Gb₃ or iGb₃ yielding globoside and Forssman from Gb₃, and isogloboside and isoForssman from iGb₃. Gb₃ synthase is not homologous to iGb₃ synthase but shows 43% identity to a human 1,4GlcNAc transferase that transfers a UDP-sugar in an 1,4-linkage to a -linked Gal found in mucin. Extensive homology (35% identity) is also present between Gb₃ synthase and genes in Drosophila melanogaster and Arabidopsis thaliana, supporting conserved expression of an 1,4-glycosyltransferase, possibly Gb₃ synthase, throughout evolution. The isolated Gb₃ synthase cDNA encodes a type II transmembrane glycosyltransferase of 360 amino acids. The highest tissue expression of Gb₃ synthase RNA is found in the kidney, mesenteric lymph node, spleen, and brain. Gb₃ glycolipid, also called P^k antigen or CD77, is a known receptor for verotoxins. CHO cells that do not express Gb₃ and are resistant to verotoxin become susceptible to the toxin following transfection with Gb₃ synthase cDNA.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The glycosphingolipids (GSL)¹ form part of eukaryotic cell membranes. They consist of a hydrophilic carbohydrate moiety linked to a hydrophobic ceramide tail embedded within the lipid bilayer of the membrane. Lactosylceramide, Gal1,4Glc1Cer (LacCer), is the common synthetic precursor to the majority of GSL found in vertebrates. Hundreds of different glycans are synthesized on LacCer with the major structures grouped into six series (1). One such series, globo-GSL, is initiated by the action of Gb₃ synthase on LacCer to produce Gb₃ (Gal1,4Gal1,4Glc1Cer).

Gb₃ is the P^k blood group antigen and has since acquired the designation CD77 (2, 3). Monoclonal antibodies (mAb) against Gb₃ are used as a marker for Burkitt's B-cell lymphoma and are able to initiate apoptosis (4, 5). Gb₃ has also been localized to a subset of tonsillar B-cells in the germinal center, platelets, and uroepithelial cells (2, 6, 7). The B-subunit of Shiga toxins interacts specifically with Gb₃ on the cell surface, thus Gb₃ plays a direct role in toxin entry into the cell (8). The presence of Gb₃ in the endothelial cells of the kidney of pediatric patients accounts for the development of hemolytic uremic syndrome during bacterial infection with Shigella that produce verotoxin (7). Recently, Gb₃ has been implicated in the entry of HIV-1 into cells (9, 10).

In this paper we describe the cloning of Gb₃ synthase, an 1,4-galactosyltransferase, and identify significant homology with an 1,4-N-acetylglucosaminyltransferase that also transfers to a -linked Gal (11). Thus, Gb₃ synthase is a member of an emerging family of 1,4-glycosyltransferases that utilizes UDP donors and transfers to a -linked acceptor. Additional potential members of this family can be identified in plant, insect, and mammalian sequences submitted to GenBank^TM indicate that this is a highly conserved family of 1,4-glycosyltransferases.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression Cloning of Gb₃ Synthase cDNA-- We have described the phenotypic cloning procedure in detail elsewhere (see Keusch et al., companion paper (39)). Briefly, a rat placental cDNA library, RPL18, in pCDM8 was cotransfected with large T-antigen cDNA (plasmid pPSVE1 PyE, a kind gift from Dr. Minoru Fukuda, The Burnham Institute) into CHO host recipient cells and surface expression of downstream products, Gb₄ and Gb₅, were detected using the mAb SMLDN1.1 via flow cytometry (FACS). CHO cells express relatively simple glycosphingolipids, mostly GlcCer, LacCer, and G_M3. Moreover, CHO cells do not express Gb₃ synthase but do express endogenous Gb₄ synthase and Gb₅ synthase. We have previously used this phenotypic screen to clone the iGb₃ synthase (GenBank^TM accession number AF248543, see Keusch et al., companion paper (39)), and further analysis of clones from the iGb₃ FACS sort which were positive with the SMLDN1.1 mAb (12) yet negative for iGb₃ synthase cDNA revealed a clone capable of synthesizing Gb₃. We refer to this latest clone as Gb₃ synthase (accession number AF248544).

cDNA Sequencing-- Gb₃ synthase cDNA insert was fully sequenced in both directions using T7, a pCDM8 reverse primer and gene-specific primers, in a reaction mix containing ABI Big-Dye terminators with AmpliTaq DNA polymerase according to the manufacturer's instructions (PE Applied Biosystems). GenBank^TM data bases were searched for homologous sequences using the BLAST (NCBI) program. Membrane topology of Gb₃ synthase was modeled using the transmembrane prediction program TMHMM (13).

Site-directed Mutagenesis-- The ¹⁹⁹DTD²⁰¹ sequence in the Gb₃ cDNA clone was mutated to ¹⁹⁹ATA²⁰¹ by inverse PCR using high fidelity KlenTaq LA-polymerase mix (Sigma), and completely overlapping primers containing the point mutations (underlined), in the forward direction are: 5'-GGTGGCATCTACTTGGCCACAGCCTTCATCGTCCTCAAG-3' and reverse direction: 5'-CTTGAGGACGATGAAGGCTGTGGCCAAGTAGATGCCACC-3' were used. Following thermocycling and digestion of parental DNA template with DpnI, MC1061/P3 cells were transformed and mini-preps sequenced to verify the desired mutations.

FACS Analysis-- Transfected CHO cells were grown in Ham's F-12 medium supplemented with 10% fetal bovine serum ± 2 µM 1-phenyl-2-hexadecanoylamino-3-pyrrolidino-1-propanol, HCl) (DL-threo-PPPP, Calbiochem), an inhibitor of glucosylceramide synthase (14). Cells were harvested 60 h post-transfection using 0.02% EDTA and stained as described (39). Cell surface GalNAc was detected with the SMLDN1.1 mAb (12) and Gb₃ (Gal1,4Gal1,4Glc1Cer) observed using anti-CD77 mAb clone 38-13 (Biodesign International). Rat mAb M1/22.21 (kindly provided by Dr. Haslam, Washington University, St. Louis, MO) was specific for the Forssman (Gb₅) glycolipid (GalNAc1,3GalNAc1,3Gal1,4Gal1,4Glc1Cer).

Assay of 1,4GalT Activity-- Cell extracts from transfected CHO cells were prepared and assayed as described previously for the iGb₃ synthase (39). Glycolipid acceptor substrates included GlcCer, GalCer, LacCer, and Gb₃ (Sigma). Radiolabeled reaction products were isolated on Sep-Pak C₁₈ columns, then separated via thin layer chromatography (TLC) and exposed to film as described (39).

Metabolic Labeling of CHO Cells-- The day after transfection, CHO cells were labeled with either 35 µCi/ml [³H]Gal or [³H]GlcNAc (NEN Life Science Products), which can also incorporate into GalNAc and sialic acid, for a further 24-48 h. Cells were scraped off the culture plate, and glycolipids were extracted in chloroform/methanol as described (15).

Glycolipid Digestion with Exoglycosidases-- Glycolipid extracts from transfected CHO cells (approximately 1-10 µg or 30,000 cpm) were digested overnight at 37 °C with exoglycosidases in the presence of 0.05% sodium taurodeoxycholate in a final volume of 10 µl as described (39).

Shiga Toxin B-subunit in TLC Blots-- Glycolipids were extracted from CHO cells cotransfected with plasmid pPSVE1 PyE and either Gb₃ or iGb₃ synthase cDNA and separated by TLC as described (39). TLC plates were dried by evaporation and either sprayed with orcinol for total glycolipid detection or processed for immunostaining. TLC plates were blocked overnight in PBS/1% BSA. The B-subunit of Shiga toxin (B-Stx-1) was purified to homogeneity, biotinylated, dialyzed, and generously provided by Dr. Haslam, Washington University. The biotinylated B-Stx-1 was diluted 1:1000 in PBS/1% BSA and added for 1 h at room temperature. After washing with PBS, 3 × 10 min, streptavidin-peroxidase (Zymed Laboratories Inc.) diluted 1:20,000 in PBS/1% BSA was added for 30 min at room temperature. Blots were washed 6 × 10 min in PBS and developed using chemiluminescence (NEN Life Science Products) according to the manufacturer's instructions.

Verocytotoxicity Assay-- Cell sensitivity to verotoxin was measured using an inhibition of protein synthesis assay via Tran³⁵S labeling. CHO cells were transiently cotransfected with plasmid pPSVE1 PyE and either Gb₃ or iGb₃ synthase cDNA or empty pCDM8 vector. Transfectants were harvested 48 h post-transfection using 0.02% EDTA, washed in culture media, and plated out in duplicate at 1 × 10⁶ cells/ml in 24-well plates in the presence of various concentrations of Shiga toxin-1 (Stx-1, kindly provided by Dr. Haslam) for 4 h at 37 °C. The Stx-1 was prepared from a filtered periplasmic extract of cultured bacteria cells containing pNAS13, which carries the Stx-1 operon. Verotoxin and media were washed out and cells incubated in cysteine-, methionine-, SO₄-free minimal essential medium (Earle's) labeling media for 5 min. Fresh labeling media containing 20 µCi/ml Tran³⁵S (ICN) label was added for 30 min. Cells were washed several times in PBS and lysed in PBS/1% SDS/2 mM EDTA and precipitated with ice-cold trichloroacetic acid, final concentration 10%. Following centrifugation, the trichloroacetic acid pellets were solubilized using Protosol (NEN Life Science Products), transferred to scintillation vials containing 3a70B scintillation mixture (Research Products International Corp.), mixed, and counted.

Expression of Gb₃ Synthase in Rat Tissues Using RT-PCR-- The levels of Gb₃ synthase were determined as described (39).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of a Rat Gb₃ Synthase cDNA via Expression Cloning-- Using a phenotypic screen based on recognition of terminal GalNAc by the mAb SMLDN1.1, we identified two independent and nonhomologous -galactosyltransferases that act on LacCer. Initially, iGb₃ synthase was cloned, an 1,3-galactosyltransferase that converts LacCer to Gal1,3LacCer (iGb₃) (39). Subsequent analysis of additional positive pools of clones revealed Gb₃ synthase, an 1,4-galactosyltransferase, that adds galactose onto LacCer to produce Gal1,4LacCer. The de novo synthesized products, iGb₃ and Gb₃, serve as acceptor substrates for endogenous Gb₄ and Gb₅ synthases that are present in the CHO recipient host cell line, thus accounting for the reactivity with the mAb SMLDN1.1 following transfection.

cDNA Sequence Analysis-- DNA sequencing of the Gb₃ synthase clone reveals a 1494-base pair insert with a single open reading frame encoding a protein of 360 amino acids (Fig. 1). The TMHMM program (13) predicts an N-terminal transmembrane domain of 23 amino acids with a type II transmembrane topology that is characteristic of Golgi glycosyltransferases (16). The putative catalytic domain of Gb₃ synthase is orientated toward the Golgi, where it is able to interact with its substrate LacCer (17). This domain also contains two potential N-linked glycosylation sites and a ¹⁹⁹DXD²⁰¹ motif common to a number of glycosyltransferases (18).

View larger version (38K): [in this window] [in a new window]   Fig. 1.   cDNA sequence analysis. Complete DNA sequence of the isolated rat Gb3 synthase cDNA clone. The predicted translated protein of 360 amino acids based on the translation of the cDNA sequence is shown in the single-letter code. Underlined is the putative transmembrane signal. Boxed are the two potential N-linked glycosylation sites. The 199DXD201 motif is indicated by three black dots.

Identification of a New 1,4-Glycosyltransferase Gene Family-- A BLAST analysis of the Gb₃ synthase clone with sequences held in the GenBank^TM data base, reveals a number of genes from a diverse number of species with extensive homology. The highest identity (80%) found is to a human expressed sequence tag clone found on chromosome 22 (accession number HSB33B7). We predict that this is the human homologue of the rat Gb₃ synthase. Furthermore, the Gb₃ synthase is 43% identical to a previously cloned human 1,4-N-acetylglucosaminyltransferase (Fig. 2), found on chromosome 3p14.3, that forms GlcNAc1,4Gal1,4-R, a glycan found in class III mucin (11). Long stretches of identity are apparent in the putative catalytic domain, including five conserved cysteines (positions 103, 240, 269, 280, and 349 in rat Gb₃ synthase). Both transferases use a UDP-sugar donor to catalyze the transfer of the sugar in an 1,4-linkage to a -linked Gal. Only limited sequence homology has been found between different families of glycosyltransferases, whereas within families there may be extensive homology. For example, the ABO blood group transferases, which now include both Forssman synthase (15) and the iGb₃ synthase (39), are highly homologous and all produce an 1,3-linkage to a -linked Gal or GalNAc utilizing a UDP-nucleotide sugar donor. The extensive homology between Gb₃ synthase and a number of genes from diverse species, including insects, plants, and mammals is, therefore, notable (Fig. 2). Potential homologues from both Drosophila melanogaster and Arabidopsis thaliana show 35% identity (54% similarity) to Gb₃ synthase, including the location of four cysteine residues in D. melanogaster (equivalent to positions 240, 269, 280, and 349 in rat Gb₃ synthase). Another homologous gene in D. melanogaster (31%, AE003753) is also found that has these four conserved cysteines. Five additional genes with similar degrees of identity to the 1,4-glycosyltransferases are identified in A. thaliana. Two of these genes are clustered on chromosome 2, whereas the others are separated on each of the four remaining chromosomes in the genome. The DXD motif, which may play a role in coordinating the nucleotide sugar donor (38), is maintained in each of the genes homologous to Gb₃ synthase. This suggests that Gb₃ synthase is a member of a family of 1,4-UDP-sugar transferases that is highly conserved during evolution.

View larger version (98K): [in this window] [in a new window]   Fig. 2.   Identification of a new 1,4-glycosyltransferase gene family. ClustalW alignment between known 1,4-glycosyltransferases: rat Gb3 synthase (accession number AF248544); human 1,4GlcNAc transferase (accession number AF141315); and predicted 1,4-glycosyltransferases in D. melanogaster (accession number AC007765) and A. thaliana (accession number AC018908). Boxed regions indicate areas of similarity in light shading and absolute identity shown in dark shading and bold. Dashes represent gaps inserted to optimize the alignment. The 199DXD201 motif is indicated by three black dots.

Cell Surface Expression of Globo-series GSL in CHO Cells Transfected with Gb₃ Synthase cDNA-- The expression of GSL bearing terminal GalNAc at the surface of CHO cells was not expected as these cells have not been reported to synthesize Gb₃, Gb₄, or Gb₅. FACS analysis shows strong GalNAc staining on the surface of CHO cells transfected with the Gb₃ synthase cDNA compared with mock-transfectants using SMLDN1.1 and anti-Forssman mAb (Fig. 3). Culture of transfected cells in the presence of PPPP, an inhibitor of glucosylceramide synthase (14), prevented the appearance of surface material reactive with any of these reagents indicating the products were confined to GSL. A major proportion of the surface GalNAc is due to Forssman glycolipid (Gb₅) expression, suggesting that conversion of Gb₃ to Gb₅ is an efficient process (Fig. 3). Note that CHO cells transfected with Forssman (Gb₅) synthase cDNA are unable to synthesize Gb₅ and do not express any terminal GalNAc (Fig. 3). Thus, the conversion of LacCer to Gb₃ is essential for the subsequent production of Gb₄ and Gb₅ by Gb₄ and Gb₅ synthases that are endogenous to the CHO cells. The conversion of Gb₃ to Gb₄ and Gb₅ is not complete, because surface expression of Gb₃ is detectable using anti-Gb₃ (CD77) mAb (Fig. 3).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Cell surface staining of globo-series GSL in CHO cells transfected with Gb3 synthase cDNA is abolished by treatment with PPPP. CHO cells were transfected with Gb3 synthase cDNA (top three plots), with empty pCDM8 vector (middle three plots), or with Gb5 (Forssman) synthase cDNA (bottom two plots). Cells were cultured in the absence (thick line) or presence of PPPP (thin line) and probed for cell surface expression of: GalNAc using SMLDN1.1 mAb (left panels); Gb3 (Gal1,4Gal1,4Glc1Cer) using CD77 clone 38-13 mAb (middle panels); Gb5 (GalNAc1,3GalNAc1,3Gal1,4Gal1,4Glc1Cer) using anti-Gb5 (Forssman) antibody, M1/22.21 (right panels). Anti-IgM-FITC was used as the secondary reagent. Isotype control primary antibodies are shown as dotted lines.

Identification of GSL Synthesized in CHO Cells Transfected with Gb₃ Synthase cDNA-- Parent CHO cells synthesize GlcCer, LacCer, and G_M3 but no GSL containing terminal GalNAc (37). The GSL expressed by CHO cells transfected with Gb₃ synthase cDNA were characterized by metabolically labeling with [³H]Gal or [³H]GlcNAc. The glycolipid extracts were treated with exoglycosidases, isolated via Sep-Pak C₁₈, and analyzed by autoradiography after TLC. [³H]GalNAc-labeled doublets that migrate at the positions of Gb₄ and Gb₅ are seen in the CHO transfectants (Fig. 4, lane 1). In addition to the doublets migrating at the position of Gb₄ and Gb₅, doublets migrating at the positions of Gb₃, LacCer, and GalCer are seen when the GSL are metabolically labeled with [³H]Gal (Fig. 4, lane 3). Based on the relative intensities of the GSL labeled with [³H]Gal, the majority of the Gb₃ is converted to Gb₄ and Gb₅, consistent with the pattern of staining seen during FACS analysis (Fig. 3). Digestion of labeled GSL with -N-acetylgalactosaminidase increased the amount of [³H]Gal label migrating at the position of Gb₄ at the expense of the material migrating as Gb₅ (Fig. 4, lane 4). In contrast, digestion with -N-acetylhexosaminidase increases the material migrating as Gb₃ at the expense of the material migrating as Gb₄ (Fig. 4, lane 5). Digestion with a combination of -N-acetylgalactosaminidase, -N-acetylhexosaminidase and -galactosidase converts all the labeled material to LacCer (Fig. 4, lane 6). The pattern of digestion is typical of that obtained with authentic Gb₅, GalNAc1,3GalNAc1,3Gal1,4Gal1,4Glc1Cer (19). The material remaining near the position of Gb₅ (indicated with an asterisk in Fig. 4) is attributable to G_M3, which is seen when CHO cells are transfected with the ¹⁹⁹ATA²⁰¹ mutant of Gb₃ synthase (Fig. 4, lane 2), and in nontransfected parent CHO cells (not shown). An interesting feature seen in the Gb₃ synthase sequence is the presence of a DXD motif, previously identified as catalytically important in a number of different glycosyltransferases (18). Conversion of the ¹⁹⁹DTD²⁰¹ of Gb₃ synthase to ¹⁹⁹ATA²⁰¹ by site-directed mutagenesis results in a loss of Gb₃ activity manifested by a lack of globo-series glycolipid expression following transfection into CHO cells (Fig. 4, lane 2). Assuming that both the wild-type and ¹⁹⁹ATA²⁰¹ mutant Gb₃ synthase are expressed at similar levels, the mutant expresses a similar set of GSL as found in parent CHO cells. The results confirm that functional Gb₃ synthase is required in CHO cells for the complete production of globo-series GSL.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Identification of globo-series GSL in CHO cells transfected with Gb3 synthase cDNA. CHO cells transfected with Gb3 synthase cDNA (lanes 1, 3-6) or mutant 199ATA201 Gb3 synthase cDNA (lane 2) were metabolically labeled with either [3H]GlcNAc (lane 1) or [3H]Gal (lanes 2-6). Glycolipids were extracted and digested with exoglycosidases prior to separation by TLC and autoradiography. Lane 1, [3H]GlcNAc, Gb3 synthase; lane 2, [3H]Gal, 199ATA201 mutant; lane 3, [3H]Gal, Gb3 synthase; lane 4, [3H]Gal Gb3 synthase + -N-acetylgalactosaminidase; lane 5, [3H]Gal Gb3 synthase + -N-acetylhexosaminidase; lane 6, [3H]Gal Gb3 synthase + -N-acetylgalactosaminidase + -N-acetylhexosaminidase + -galactosidase. The square brackets ([, ])indicate putative Gb5, Gb4, Gb3, and LacCer products. The asterisks indicate the location of the GM3 doublet. The angled brackets (<, >)show, in descending order, the location of authentic GalCer, LacCer, Gb3, Gb4, and Gb5 standards.

In Vitro Enzyme Activity of Gb₃ Synthase-- Triton X-100 extracts of CHO cells transfected with Gb₃ synthase cDNA transfer galactose from UDP-[³H]galactose to exogenously added LacCer, whereas parent CHO cells do not (Fig. 5, lane 3, lower arrow). Following the enzyme reaction, radiolabeled glycolipid products were isolated via Sep-Pak C₁₈ and separated by TLC. Low, but detectable, levels of galactose are transferred to GalCer (Fig. 5, lane 2, upper arrow). No transfer to other GSL tested was detected (Fig. 5, lanes 1 and 4). Bands present near the origin of the TLC plate are found in assays using extracts from both the parent and Gb₃ synthase transfected CHO cells and are therefore nonspecific.

View larger version (60K): [in this window] [in a new window]   Fig. 5.   Gb3 synthase activity from CHO cell extracts transfected with Gb3 synthase cDNA. Transfer of [3H]Gal from UDP-[3H]Gal to different GSL acceptors was detected by autoradiography following separation by TLC. The enzyme source is from extracts of CHO cells transfected with Gb3 synthase cDNA. GSL acceptor substrates: GlcCer, lane 1; GalCer, lane 2; LacCer, lane 3; Gb3, lane 4. Arrows indicate the position of GSL standards, LacCer (upper arrow) and Gb3 (lower arrow).

Functional Gb₃ Is Expressed by CHO Cells Transfected with Gb₃ Synthase cDNA-- The B-subunit of the verotoxin, Shiga toxin (B-Stx-1), which specifically recognizes Gb₃, reacts with material migrating at the position of Gb₃ when crude extracts from CHO cells transfected with Gb₃ synthase but not iGb₃ synthase were examined (Fig. 6). Gb₃ has been identified as the receptor for Shiga toxin, and is required for the transport of the toxin into the cell leading to cell death (8). Following transfection with Gb₃ synthase cDNA, CHO cells become sensitive to killing by verotoxin. Approximately 65% of Gb₃ synthase cDNA transfectants are killed compared with <15% of control transfectants (Fig. 7). Considering that a transfection efficiency of 65-70% is routinely achieved with CHO cells, the verotoxin is highly toxic to the transfected cells.

View larger version (32K): [in this window] [in a new window]   Fig. 6.   B-subunit of Shiga toxin (B-Stx-1) specifically detects Gb3 in GSL extracts from CHO cells transfected with Gb3 synthase cDNA. Crude GSL extracts from CHO cells transfected with either Gb3 synthase (lane 3) or iGb3 synthase cDNA (lane 2) were separated by TLC. A biotinylated form of the purified B-Stx-1 was overlaid on the TLC followed by streptavidin-peroxidase. Blots were developed using chemiluminescence. Neutral GSL standards, which includes Gb3, are in lane 1.

View larger version (11K): [in this window] [in a new window]   Fig. 7.   CHO cells transfected with Gb3 synthase cDNA are susceptible to killing by verotoxin. CHO cells were transfected with empty vector, pCDM8 (open boxes), iGb3 synthase cDNA (open circles), or Gb3 synthase cDNA (filled circles). 48 h post-transfection, cells were plated out in duplicate and incubated with several dilutions of verotoxin for 4 h, washed, and labeled with Tran35S. A decrease in the percentage of trichloroacetic acid-precipitable counts indicates inhibition of protein synthesis and is shown as percentage cell viability.

Gb₃ Synthase RNA Expression in Rat Tissues-- The tissue distribution of Gb₃ synthase RNA was examined by RT-PCR using gene-specific primers. The expected product size of 500 base pairs is found in most tissues, however, the apparent levels of expression varies significantly (Fig. 8). Tissues with the highest expression include the brain, kidney, mesenteric lymph node, and spleen. Intermediate levels of Gb₃ synthase RNA expression are seen in the adrenal gland, uterus, pituitary, and thymus. All other tissues tested show low/undetectable expression. The expression of glyceraldehyde-3-phosphate dehydrogenase RNA is uniform in these samples.

View larger version (60K): [in this window] [in a new window]   Fig. 8.   Gb3 synthase RNA expression in rat tissues. Total RNA was isolated from rat tissues and assayed for the presence of Gb3 synthase expression (upper panel) using RT-PCR and gene-specific primers. The Gb3 synthase PCR product (arrow) has the predicted 500-base pair size. Gene-specific primers for glyceraldehyde-3-phosphate dehydrogenase, a housekeeping gene, yielded similar amounts of product from each tissue (lower panel). *, repeat analysis of mesenteric lymph node (mes. LN) and kidney showed the presence of glyceraldehyde-3-phosphate dehydrogenase product. pop. LN, popliteal lymph node; and skel. muscle, skeletal muscle.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have cloned Gb₃ synthase, an 1,4-galactosyltransferase that acts on LacCer to form Gb₃ GSL, also known as CD77 or blood group P^k antigen (2, 4). Gb₃ synthase initiates the synthesis of the globo-series GSL that includes Gb₄ and Gb₅, the two products that are detected in the expression cloning procedure. Thus, Gb₃ (P^k antigen) is the precursor to Gb₄, the P antigen, which can then be converted to Gb₅ (Forssman, GalNAc1,3Gb₄). Gb₃ synthase activity toward LacCer is substantially higher than to another -linked Gal substrate, GalCer, indicating that the correct underlying carbohydrate structure is necessary for efficient enzymatic activity. The Gb₃ synthase sequence should prove useful in the understanding of the genetic defects in the very rare blood group p phenotype, where there is a lack of Gb₃/P^k galactosyltransferase activity and consequently an absence of P^k, P, and P¹ antigens (20, 21).

iGb₃ synthase, an 1,3-galactosyltransferase (39), and Gb₃ synthase were cloned using the same phenotypic screen, which shows that globoside and Forssman products may be synthesized on either iGb₃ or Gb₃ core structures, to yield iGb₄ and iGb₅ or Gb₄ and Gb₅, respectively. The presence of isoglobo- and globo-GSL, which differ only in a positional linkage (1,3 or 1,4) of the internal -linked Gal, raises questions as to their possible significance. Are globoside and Forssman found in the isoglobo series functionally different from their globo series counterparts? Because products based on globoside, including Forssman, have been identified as stage-specific embryonic antigens (22), it raises important questions about the role of the underlying core structure (iGb₃/Gb₃). To address these questions we need to examine, in greater detail, the expression of each GSL pathway in different tissues and ultimately different cell types. A comparison of the Gb₃ synthase and iGb₃ synthase expression in rat tissues revealed a broad distribution with varying levels. Six out of the 13 tissues examined show a preferential expression of one of the enzymes, almost to exclusivity, perhaps arguing for distinct functional roles, not only for Gb₃ and iGb₃ but also for their downstream products. In contrast, the spleen and, to a lesser extent, the thymus show high expression of both -galactosyltransferases. GSL structures containing iGb₃ and Gb₃ have been isolated from the spleen (23, 24). It is notable that the CHO cells transfected with iGb₃ synthase cDNA produced a greater amount of product compared with CHO cells transfected with Gb₃ synthase cDNA in in vitro assays and by FACS analysis. At present, it is unclear what the molecular basis is for this difference, although it raises the possibility that these two enzymes differ in their catalytic efficiency.

The expression of Gb₃ varies with Gb₃ synthase activity in development and cellular differentiation (25, 26). Although it is unclear what signals these changes, Gb₃ expression has been linked to various pathological events, including hemolytic uremic syndrome, HIV-1 cell fusion, lymphomas, and Fabry's disease (4, 9, 10, 27-29). The high expression of Gb₃ synthase activity in the kidney correlates well with Gb₃ expression (30). Children, in contrast to adults, express high levels of Gb₃ in their kidney and as such are at risk to infection by the Shiga toxins (7, 27). We have shown that CHO cells become susceptible to verotoxin following expression of Gb₃ on their cell surface. Hence, de novo synthesized Gb₃, in contrast to exogenously added Gb₃ (31), is functional in CHO cells. Shiga toxin binds specifically to Gb₃ and not to the related iGb₃ GSL. The ability of iGb₃ synthase to out-compete any Gb₃ synthase activity in the same cell may afford the animal protection against infection by bacterial toxins, adhesins, and viruses that interact with Gb₃. Some bacterial pathogens recognize internal Gal1,4Gal structures (iso-receptors) via their P-fimbriae as well as terminal epitopes (32, 33), so an alternative core (iGb₃ over Gb₃) rather than simply capping terminal Gal1,4Gal by Gb₄ and/or Gb₅ synthases would be required for resistance.

It is interesting to note that, although Gb₃ synthase and iGb₃ synthase catalyze very similar reactions using the same acceptor substrate (LacCer), they are not homologous. iGb₃ synthase is found to have very high sequence identity to the existing ABO blood group glycosyltransferases, which all catalyze the transfer of a UDP-sugar in an 1,3-linkage to either a -linked Gal or GalNAc (39). In contrast, Gb₃ appears to be a member of a distinct family of glycosyltransferases. These transferases utilize UDP-sugars as the donor to add the sugar in 1,4-linkage to a -linked sugar acceptor, most commonly a -linked Gal. Other structures with this linkage have been described, such as the GSL P¹ blood group antigen (Gal1,4Gal1,3GlcNAc1-R) (34), GlcNAc1,4Gal1-R found in type III mucin in humans (11), and Gal1,4Gal1,3GalNAc (35) and GalNAc1,4GalNAc1,4GlcNAc1-R found in insects (36). The homologues in mammals and insects may thus represent the transferases responsible for the synthesis of these structures. All the homologues to Gb₃ synthase contained the catalytically important DXD motif. Genes encoding sequences with 30-35% identity with Gb₃ synthase can also be identified on the five chromosomes of the plant Arabidopsis, supporting the possibility that the transferases producing structures with  1,4-linked sugars have been highly conserved during evolution and are therefore of biological importance.

	ACKNOWLEDGEMENTS

We thank Dr. D. Haslam for his helpful suggestions and generous gifts of the anti-Forssman (Gb₅) mAb, B-Stx-1, and verotoxin. We are also indebted to Dr. P. Smith for providing the RPL18 cDNA library.

	Addendum

While this manuscript was being reviewed, the following two independent groups reported the cloning of the human Gb₃ synthase (HSB33B7): Kojima, et al. (40) and Steffensen, et al. (41).

	FOOTNOTES

^* This work was supported by National Institutes of Health Grant R01-DK 41738 (to J. U. B.).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/EMBL Data Bank with accession number(s) AF248544.

^§ To whom correspondence should be addressed. 19714-6101: Tel.: 314-362-8733; Fax: 314-362-8888; E-mail: jkeusch@pathbox.wustl.edu.

^¶ Current address: Dade-Behring Inc., Newark, DE 19714-6101.

Published, JBC Papers in Press, June 14, 2000, DOI 10.1074/jbc.M002630200

¹ The abbreviations for the glycosphingolipids are in accordance with the 1997 recommendations of the IUPAC-IUB Joint Commission of Biochemical Nomenclature. The abbreviations used are: GSL, glycosphingolipid; LacCer, lactosylceramide (Gal1,4Glc1Cer); Cer, ceramide; Gal, galactose; GalNAc, N-acetylgalactosamine; Glc, glucose; Gb₃, P^k, CD77, or globotriaosylceramide (Gal1,4Gal1,4Glc1 Cer); iGb₃, isoglobotriaosylceramide (Gal1,3Gal1,4Glc1Cer); Gb₄, globoside (GalNAc1,3Gal1,4Gal1,4Glc1Cer); iGb₄, isogloboside (GalNAc1,3Gal1,3Gal1,4Glc1Cer); Gb₅, Forssman (GalNAc1,3GalNAc1,3Gal1,4Gal1,4Glc1Cer); iGb₅, isoForssman (GalNAc1,3GalNAc1,3Gal1,3Gal1,4Glc1Cer); mAb, monoclonal antibody; PPPP, 1-phenyl-2-hexadecanoylamino-3-pyrrolidino-1-propanol, hydrochloride; TLC, thin layer chromatography; CHO, Chinese hamster ovary; FACS, fluorescence-activated cell sorting; RT-PCR, reverse transcriptase-polymerase chain reaction; PBS, phosphate-buffered saline.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

1.	Stults, C. L., Sweeley, C. C., and Macher, B. A. (1989) Methods Enzymol. 179, 167-214
2.	Marcus, D. M., Kundu, S. K., and Suzuki, A. (1981) Semin. Hematol. 18, 63-71
3.	Kishimoto, T. (ed) (1997) Leukocyte Typing VI , pp. 175-177, Garland Publishing Inc., London
4.	Nudelman, E., Kannagi, R., Hakomori, S., Parsons, M., Lipinski, M., Wiels, J., Fellous, M., and Tursz, T. (1983) Science 220, 509-511
5.	Mangeney, M., Lingwood, C. A., Taga, S., Caillou, B., Tursz, T., and Wiels, J. (1993) Cancer Res. 53, 5314-5319
6.	Cooling, L. L., Walker, K. E., Gille, T., and Koerner, T. A. (1998) Infect. Immun. 66, 4355-4366
7.	Lingwood, C. A. (1994) Nephron. 66, 21-28
8.	Lindberg, A. A., Brown, J. E., Stromberg, N., Westling-Ryd, M., Schultz, J. E., and Karlsson, K. A. (1987) J. Biol. Chem. 262, 1779-1785
9.	Puri, A., Hug, P., Jernigan, K., Barchi, J., Kim, H. Y., Hamilton, J., Wiels, J., Murray, G. J., Brady, R. O., and Blumenthal, R. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 14435-14440
10.	Hammache, D., Yahi, N., Maresca, M., Pieroni, G., and Fantini, J. (1999) J. Virol. 73, 5244-5248
11.	Nakayama, J., Yeh, J. C., Misra, A. K., Ito, S., Katsuyama, T., and Fukuda, M. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 8991-8996
12.	Nyame, A. K., Leppanen, A. M., DeBose-Boyd, R., and Cummings, R. D. (1999) Glycobiology 9, 1029-1035
13.	Glasgow, J., Littlejohn, T., Major, F., Lathrop, R., Sankoff, D., and Sensen, C. (eds) (1998) Proceedings of the Sixth International Conference on Intelligent Systems for Molecular Biology , pp. 175-182, AAAI Press, Menlo Park, CA
14.	Abe, A., Radin, N. S., Shayman, J. A., Wotring, L. L., Zipkin, R. E., Sivakumar, R., Ruggieri, J. M., Carson, K. G., and Ganem, B. (1995) J. Lipid Res. 36, 611-621
15.	Haslam, D. B., and Baenziger, J. U. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 10697-10702
16.	Colley, K. J. (1997) Glycobiology 7, 1-13
17.	Lannert, H., Gorgas, K., Meissner, I., Wieland, F. T., and Jeckel, D. (1998) J. Biol. Chem. 273, 2939-2946
18.	Wiggins, C. A., and Munro, S. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 7945-7950
19.	Li, Y.-T., and Li, S.-C. (1999) Anal. Biochem. 273, 1-11
20.	Wiels, J., Taga, S., Tetaud, C., Cedergren, B., Nilsson, B., and Clausen, H. (1996) Glycoconj. J. 13, 529-535
21.	Bailly, P., Piller, F., Gillard, B., Veyrieres, A., Marcus, D., and Cartron, J. P. (1992) Carbohydr. Res. 228, 277-287
22.	Kannagi, R., Cochran, N. A., Ishigami, F., Hakomori, S., Andrews, P. W., Knowles, B. B., and Solter, D. (1983) EMBO J. 2, 2355-2361
23.	Stoffyn, P., Stoffyn, A., and Hauser, G. (1973) Biochim. Biophys. Acta 306, 283-286
24.	Makita, A., and Yamakawa, T. (1962) J. Biochem. (Tokyo) 51, 124-133
25.	Mobassaleh, M., Koul, O., Mishra, K., McCluer, R. H., and Keusch, G. T. (1994) Am. J. Physiol. 267, G618-G624
26.	Taga, S., Tetaud, C., Mangeney, M., Tursz, T., and Wiels, J. (1995) Biochim. Biophys. Acta 1254, 56-65
27.	Obrig, T. G., Louise, C. B., Lingwood, C. A., Boyd, B., Barley-Maloney, L., and Daniel, T. O. (1993) J. Biol. Chem. 268, 15484-15488
28.	Lingwood, C. A. (1999) Biochim. Biophys. Acta 1455, 375-386
29.	Kolter, T., and Sandhoff, K. (1998) Brain Pathol. 8, 79-100
30.	Makita, A. (1964) J. Biochem. (Tokyo) 55, 269-276
31.	Jacewicz, M. S., Mobassaleh, M., Gross, S. K., Balasubramanian, K. A., Daniel, P. F., Raghavan, S., McCluer, R. H., and Keusch, G. T. (1994) J. Infect. Dis. 169, 538-546
32.	Karlsson, K. A. (1995) Curr. Opin. Struct. Biol. 5, 622-635
33.	Roberts, J. A., Marklund, B. I., Ilver, D., Haslam, D., Kaack, M. B., Baskin, G., Louis, M., Mollby, R., Winberg, J., and Normark, S. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 11889-11893
34.	Yang, Z., Bergstrom, J., and Karlsson, K. A. (1994) J. Biol. Chem. 269, 14620-14624
35.	Lopez, M., Gazon, M., Juliant, S., Plancke, Y., Leroy, Y., Strecker, G., Cartron, J. P., Bailly, P., Cerutti, M., Verbert, A., and Delannoy, P. (1998) J. Biol. Chem. 273, 33644-33651
36.	Dennis, R. D., Geyer, R., Egge, H., Peter-Katalinic, J., Li, S.-C., Stirm, S., and Wiegandt, H. (1985) J. Biol. Chem. 260, 5370-5375
37.	Rosales Fritz, V. M., Daniotti, J. L., and Maccioni, H. J. (1997) Biochim. Biophys. Acta 1354, 153-158
38.	Gastinel, L. N., Cambillau, C., and Bourne, Y. (1999) EMBO J. 18, 3546-3557
39.	Keusch, J. J., Manzella, S. M., Nyame, K. A., Cummings, R. D., and Baenziger, J. U. (2000) J. Biol. Chem. 33, 25308-25314
40.	Kojima, Y., Fukumoto, S., Furukawa, K., Okajima, T., Wiels, J., Yokoyama, K., Suzuki, Y., Urano, T., Ohta, M., and Furukawa, K. (2000) J. Biol. Chem. 275, 15152-15156.
41.	Steffensen, R., Carier, K., Wiels, J., Levery, S. B., Stroud, M., Cedergren, B., Sojka, B. N., Bennett, E. P., Jersild, C., and Clausen, H. (2000) J. Biol. Chem. 275, 16723-16729.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				T. Okuda and K.-i. Nakayama Identification and characterization of the human Gb3/CD77 synthase gene promoter Glycobiology, December 1, 2008; 18(12): 1028 - 1035. [Abstract] [Full Text] [PDF]

				Y. Fujii, S.-i. Numata, Y. Nakamura, T. Honda, K. Furukawa, T. Urano, J. Wiels, M. Uchikawa, N. Ozaki, S.-i. Matsuo, et al. Murine glycosyltransferases responsible for the expression of globo-series glycolipids: cDNA structures, mRNA expression, and distribution of their products Glycobiology, December 1, 2005; 15(12): 1257 - 1267. [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]

				K. Iwamura, K. Furukawa, M. Uchikawa, B. N. Sojka, Y. Kojima, J. Wiels, H. Shiku, T. Urano, and K. Furukawa The Blood Group P1 Synthase Gene Is Identical to the Gb3/CD77 Synthase Gene: A CLUE TO THE SOLUTION OF THE P1/P2/p PUZZLE J. Biol. Chem., November 7, 2003; 278(45): 44429 - 44438. [Abstract] [Full Text] [PDF]

				S. G. Taylor, I. F.C. Mckenzie, and M. S. Sandrin Characterization of the rat {alpha}(1,3)galactosyltransferase: evidence for two independent genes encoding glycosyltransferases that synthesize Gal{alpha}(1,3)Gal by two separate glycosylation pathways Glycobiology, May 1, 2003; 13(5): 327 - 337. [Abstract] [Full Text] [PDF]

				T. G. Obrig, R. M. Seaner, M. Bentz, C. A. Lingwood, B. Boyd, A. Smith, and W. Narrow Induction by Sphingomyelinase of Shiga Toxin Receptor and Shiga Toxin 2 Sensitivity in Human Microvascular Endothelial Cells Infect. Immun., February 1, 2003; 71(2): 845 - 849. [Abstract] [Full Text] [PDF]

				A. Hellberg, J. Poole, and M. L. Olsson Molecular Basis of the Globoside-deficient Pk Blood Group Phenotype. IDENTIFICATION OF FOUR INACTIVATING MUTATIONS IN THE UDP-N-ACETYLGALACTOSAMINE: GLOBOTRIAOSYLCERAMIDE 3-beta -N-ACETYLGALACTOSAMINYLTRANSFERASE GENE J. Biol. Chem., August 9, 2002; 277(33): 29455 - 29459. [Abstract] [Full Text] [PDF]

				I. Majoul, T. Schmidt, M. Pomasanova, E. Boutkevich, Y. Kozlov, and H.-D. Soling Differential expression of receptors for Shiga and Cholera toxin is regulated by the cell cycle J. Cell Sci., February 15, 2002; 115(4): 817 - 826. [Abstract] [Full Text] [PDF]

				M. X. Zhang, J. Nakayama, E. Hidaka, S. Kubota, J. Yan, H. Ota, and M. Fukuda Immunohistochemical Demonstration of {{alpha}}1,4-N-acetylglucosaminyltransferase that Forms GlcNAc{{alpha}}1,4Gal{beta} Residues in Human Gastrointestinal Mucosa J. Histochem. Cytochem., May 1, 2001; 49(5): 587 - 596. [Abstract] [Full Text]

				J. Li, D. M. Rancour, M. L. Allende, C. A. Worth, D. S. Darling, J. B. Gilbert, A. K. Menon, and W. W. Young Jr. The DXD motif is required for GM2 synthase activity but is not critical for nucleotide binding Glycobiology, March 1, 2001; 11(3): 217 - 229. [Abstract] [Full Text] [PDF]