Transcriptional regulation of alpha1,3-galactosyltransferase in embryonal carcinoma cells by retinoic acid. Masking of Lewis X antigens by alpha-galactosylation

Application of topologically constrained mini-proteins as ligands, substrates, and inhibitors

Protein-protein interactions are governed by a variety of structural features. The sequence specificities of such interactions are usually easier to establish than the "topological specificities," whereby interactions may be classified based on recognition of distinct three-dimensional structural motifs. Approaches to explore topological specificities have been based primarily on assembly of mini-proteins with well defined secondary, tertiary, and/or quarternary structures. The present chapter focuses on three approaches for constructing topologically well defined mini-proteins: template-assembled synthetic proteins (TASPs), disulfide-stabilized structures, and peptide-amphiphiles (PAs). Specific examples are given for applying each approach to explore topologically-dependent protein-protein interactions. TASPs are utilized to identify a metastatic melanoma receptor that binds to the alpha1(IV)1263-1277 region of basement membrane (type IV) collagen. A disulfide-stabilized structure incorporating a sarafotoxin (SRT) 6b model was examined as a matrix metalloproteinase (MMP)-3 inhibitor. PAs were developed as (a) fluorogenic triple-helical or polyPro II substrates for MMPs and aggrecanase members of the a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) family and (b) glycosylated and nonglycosylated ligands for metastatic melanoma cells. Topologically constrained mini-proteins have proved to be quite versatile, helping to define critical primary, secondary, and tertiary structural elements that modulate enzyme and receptor functions.

Differentiating N-linked glycan structural isomers in metastatic and nonmetastatic tumor cells using sequential mass spectrometry

In an effort to understand the role of molecular glycosylation in cancer a murine model has been used to characterize and fingerprint malignancies in established cell lines that manifest all the hallmarks of metastatic disease: spontaneous development, local invasion, intravasation, immune system survival, extravasation, and secondary tumor formation involving liver, kidney, spleen, lung, and brain. Using astrocyte cell controls, we compared N-linked glycosylation from a nonmetastatic brain tumor cell line and two different metastatic brain tumor cells. Selected ions in each profile were disassembled by ion trap mass spectrometry (MS(n)) which exhibited multiple structural differences between each tissue. These unique structures were identified within isomeric compositions as pendant nonreducing termini of di- and trisaccharide fragments, probably transparent to a tandem MS approach but distinctively not to sequential ion trap MS(n) detection.

Cancer Stem Cells and Differentiation Therapy

Cancers arise from stem cells in adult tissues and the cells that make up a cancer reflect the same stem cell --> progeny --> differentiation progression observed in normal tissues. All adult tissues are made up of lineages of cells consisting of tissue stem cells and their progeny (transit-amplifying cells and terminally differentiated cells); the number of new cells produced in normal tissue lineages roughly equals the number of old cells that die. Cancers result from maturation arrest of this process, resulting in continued proliferation of cells and a failure to differentiate and die. The biological behavior, morphological appearance, and clinical course of a cancer depend on the stage of maturation at which the genetic lesion is activated. This review makes a comparison of cancer cells to embryonic stem cells and to adult tis sue stem cells while addressing two basic questions: (1) Where do cancers come from?, and (2) How do cancers grow? The answers to these questions are critical to the development of approaches to the detection, prevention, and treatment of cancer.

Implications of galactocerebrosidase and galactosylcerebroside metabolism in cancer cells

Galactosylcerebroside is known to be overexpressed upon the cellular surface of a variety of cancers. In squamous cell carcinomas of the head and neck, one explanation for galactosylcerebroside accumulation has been identified as a transcriptional repression of the galactocerebrosidase gene. Galactocerebrosidase is the enzyme responsible for degrading galactosylcerebroside to ceramide. Ceramide is an important apoptosis activator, whereas galactosylcerebroside functions as an inhibitor. A shift of the ceramide metabolism balance in favor of glycosylated forms has been identified as a mechanism of drug resistance for several antineoplastic agents. Our review elaborates on possible explanations for galactocerebrosidase suppression and on other explanations for increased glycosphingolipid concentration within cancer cell membranes. Furthermore, conjecturable influences of a repressed galactocerebrosidase expression on tumor biology are to be explained. The inhibiting transcription factors YY1 and AP2 have been identified as potential galactocerebrosidase gene suppressors. The resulting accumulation of galactosylcerebroside promotes a reduction of cellular adhesion and inhibits apoptosis, leading to increased cellular growth, migration and prolonged cell survival contributing to carcinogenesis.

Stem cell origin of cancer and differentiation therapy

Our forefathers in pathology, on observing cancer tissue under the microscope in the mid-19th century, noticed the similarity between embryonic tissue and cancer, and suggested that tumors arise from embryo-like cells [Recherches dur le Traitement du Cancer, etc. Paris. (1829); Editoral Archiv fuer pathologische Anatomie und Physiologie und fuer klinische Medizin 8 (1855) 23]. The concept that adult tissues contain embryonic remnants that generally lie dormant, but that could be activated to become cancer was later formalized by Cohnheim [Path. Anat. Physiol. Klin. Med. 40 (1867) 1-79; Virchows Arch. 65 (1875) 64] and Durante [Arch. Memori ed Osservazioni di Chirugia Practica 11 (1874) 217-226], as the "embryonal rest" theory of cancer. An updated version of the embryonal rest theory of cancer is that cancers arise from tissue stem cells in adults. Analysis of the cellular origin of carcinomas of different organs indicates that there is, in each instance, a determined stem cell required for normal tissue renewal that is the most likely cell of origin of carcinomas [Lab. Investig. 70 (1994) 6-22]. In the present review, the nature of normal stem cells (embryonal, germinal and somatic) is presented and their relationships to cancer are further expanded. Cell signaling pathways shared by embryonic cells and cancer cells suggest a possible link between embryonic cells and cancer cells. Wilm's tumors (nephroblastomas) and neuroblastomas are presented as possible tumors of embryonic rests in children. Teratocarcinoma is used as the classic example of the totipotent cancer stem cell which can be influenced by its environment to differentiate into a mature adult cell. The observation that "promotion" of an epidermal cancer may be accomplished months or even years after the initial exposure to carcinogen ("initiation"), implies that the original carcinogenic event occurs in a long-lived epithelial stem cell population. The cellular events during hepatocarcinogenesis illustrate that cancers may arise from cells at various stages of differentiation in the hepatocyte lineage. Examples of genetic mutations in epithelial and hematopoietic cancers show how specific alterations in gene expression may be manifested as maturation arrest of a cell lineage at a specific stage of differentiation. Understanding the signals that control normal development may eventually lead us to insights in treating cancer by inducing its differentiation (differentiation therapy). Retinoid acid (RA) induced differentiation therapy has acquired a therapeutic niche in treatment of acute promyelocytic leukemia and the ability of RA to prevent cancer is currently under examination.

Microarray analysis of retinoid-dependent gene activity during rat embryogenesis: increased collagen fibril production in a model of retinoid insufficiency

Retinoic acid (RA) is an essential mediator of embryogenesis. Some, but not all, of its targets have been identified. We previously developed a rat model of gestational retinoid deficiency (RAD; Power et al. [1999] Dev. Dyn. 216:469-480) and generated embryos with developmental impairments that closely resemble genetic and dietary models of retinoid insufficiency. Here, we used microarray analysis and expression profiling to identify 88 transcripts whose abundance was altered under conditions of retinoid insufficiency, as compared with normal embryos. Among these, the induction by RAD of genes involved in collagen I synthesis (COL1A1, IA2 and VA2, prolyl-4-hydroxylase-alpha1) and protein galactosylation (galactokinase, ABO galactosyltransferase, UDP-galactose transporter-related protein) was especially noteworthy because extracellular matrix regulates many developmental events. We also identified several genes involved with stress responses (cathepsin H, UBC2E, IGFBP3, smoothelin). Real-time polymerase chain reaction analysis of selected candidates revealed excellent agreement with the array findings. Further validation came from the demonstration that these genes were similarly dysregulated in two genetic models of retinoid insufficiency, the retinol binding protein null-mutant embryo and the Raldh2 null-mutant embryo. In situ hybridization of RAD embryos found increased collagen IA1 and IGFBP3 mRNA within the connective mesenchyme and vasculature, respectively, and a failure to repress the growth factor midkine within the RAD neural tube. Many of the identified genes were not known previously to respond to retinoid status and will provide new insights to retinoid roles and to the consequences of retinoid insufficiency.

Genomic structure and alternative splicing of the insulin receptor tyrosine kinase substrate of 53-kDa protein

Insulin receptor tyrosine kinase substrate of 53-kDa protein (IRSp53) is now known to be a key factor in cytoskeleton reorganization. The human IRSp53 was identified as a binding partner with DRPLA protein, a product of the gene responsible for a neurodegenerative disorder, dentatorubral pallidoluysian atrophy, as well as a binding partner with brain-specific angiogenesis inhibitor 1. Previous studies identified at least four isoforms (L-, M-, S- and T-forms) in human, where 511 amino acid residues from the N-terminus were identical, followed by unique sequences of 9-41 amino acid residues. As each isoform had a distinct function, the unique sequences at the C-terminus had a vital role in its function. Here we report that these isoforms were indeed generated by alternative splicing, which was established by experimental and computational studies on human and rodent genomes. Previous biochemical reports suggested that rodents may lack one of the isoforms (L-form). This study solved this issue, as a nucleotide substitution occurred at a splice donor site followed by a large deletion in the rodent genome compared with human, which made the generation of the L-form impossible. This study also revealed overlapping of the IRSp53 and AATK genes coded for by complementary strands.

Neurogenin3 triggers beta-cell differentiation of retinoic acid-derived endoderm cells

Neurogenin3 is a member of the basic helix-loop-helix ('bHLH') family of transcription factors. It plays a crucial role in the commitment of embryonic endoderm into the pancreatic differentiation programme. This factor is considered to act upstream of a cascade of other transcription factors, leading to the fully differentiated endocrine phenotype. Direct observation of the sequential activation of these factors starting from Neurogenin3 had never been demonstrated. By using retinoic acid-derived-endoderm F9 cells as a model, the present study indicates that the ectopic expression of Neurogenin3 is able to start the differentiation pathway of endocrine pancreas. Neurogenin3 triggers the expression of several pancreatic transcription factors following a well defined temporal activation sequence. By reverse transcriptase PCR, immunohistochemistry and RIA, it is shown that stable transfected cells are able to form embryod bodies that produce insulin in response to glucose stimulation. This is the first report of a differentiation event induced by the ectopic expression of a transcription factor in embryonic pluripotent stem cells.

Melanoma cell CD44 interaction with the alpha 1(IV)1263-1277 region from basement membrane collagen is modulated by ligand glycosylation

Invasion of the basement membrane is believed to be a critical step in the metastatic process. Melanoma cells have been shown previously to bind distinct triple-helical regions within basement membrane (type IV) collagen. Additionally, tumor cell binding sites within type IV collagen contain glycosylated hydroxylysine residues. In the present study, we have utilized triple-helical models of the type IV collagen alpha1(IV)1263-1277 sequence to (a) determine the melanoma cell receptor for this ligand and (b) analyze the results of single-site glycosylation on melanoma cell recognition. Receptor identification was achieved by a combination of methods, including (a) cell adhesion and spreading assays using triple-helical alpha1(IV)1263-1277 and an Asp(1266)Abu variant, (b) inhibition of cell adhesion and spreading assays, and (c) triple-helical alpha1(IV)1263-1277 affinity chromatography with whole cell lysates and glycosaminoglycans. Triple-helical alpha1(IV)1263-1277 was bound by melanoma cell CD44/chondroitin sulfate proteoglycan receptors and not by the collagen-binding integrins or melanoma-associated proteoglycan. Melanoma cell adhesion to and spreading on the triple-helical alpha1(IV)1263-1277 sequence was then compared for glycosylated (replacement of Lys(1265) with Hyl(O-beta-d-galactopyranosyl)) versus non-glycosylated ligand. Glycosylation was found to strongly modulate both activities, as adhesion and spreading were dramatically decreased due to the presence of galactose. CD44/chondroitin sulfate proteoglycan did not bind to glycosylated alpha1(IV)1263-1277. Overall, this study (a) is the first demonstration of the prophylactic effects of glycosylation on tumor cell interaction with the basement membrane, (b) provides a rare example of an apparent unfavorable interaction between carbohydrates, and (c) suggests that sugars may mask "cryptic sites" accessible to tumor cells with cell surface or secreted glycosidase activities.

Discordant expression of selectin ligands and sialyl Lewis x-related epitopes on murine myeloid cells

Murine leukocytes are thought to express alpha2-3-sialylated and alpha1-3-fucosylated selectin ligands such as sialyl Lewis x (sLe(x)), although monoclonal antibodies (mAbs) to sLe(x) or Le(x) reportedly do not bind to murine leukocytes. We observed that P- and E-selectin bound to pronase-sensitive ligands on murine monocytic WEHI-3 cells and murine neutrophils, indicating that the ligands for both selectins are glycoproteins. CSLEX-1, HECA-452, and other widely used mAbs to sLe(x) and Le(x) did not bind to WEHI-3 cells and bound at very low levels to murine neutrophils. Only the anti-sLe(x) mAbs 2H5 and KM93, which also recognize nonfucosylated glycans, bound to WEHI-3 cells. 2H5 and KM93 bound to pronase-resistant structures, indicating that the mAbs did not identify selectin ligands. Treatment of WEHI-3 cells with glycosidases or chlorate demonstrated that sialic acid modifications, alpha1-3-galactosylation, or sulfation did not mask epitopes for mAbs to sLe(x) or Le(x). Compared to human promyelocytic HL-60 cells, WEHI-3 cells and murine neutrophils expressed low alpha1-3-fucosyltransferase activities. Consistent with very low endogenous fucosylation, forced fucosylation of intact WEHI-3 cells or murine neutrophils by exogenous alpha1-3-fucosyltransferase FTVI and GDP-fucose created many new epitopes for anti-sLe(x) mAbs such as HECA-452 and CSLEX-1. Nevertheless, forced fucosylation of intact cells did not significantly augment their ability to bind to fluid-phase P- or E-selectin or to roll on immobilized P- or E-selectin under flow. These data suggest that murine myeloid leukocytes fucosylate only a few specific glycans, which interact preferentially with P- and E-selectin.

Differential expression of LacdiNAc, fucosylated LacdiNAc, and Lewis x glycan antigens in intramolluscan stages of Schistosoma mansoni

We report the expression of 3 well-characterized adult Schistosoma mansoni glycan antigens among molluscan stages of the parasite. These antigens are LacdiNAc (LDN; GalNAcbeta1-4GlcNAc-R), fucosylated LacdiNAc (LDNF; GalNAc[Fucal-3]beta1-4GlcNAc-R), and Lewis x (Le(x); Gal[Fucalpha1-3]beta1-4GlcNAc-R). The presence of the glycans was determined by both immunoblot and immunohistological methods using monoclonal antibodies that specifically recognize each glycan epitope. Immunoblot analyses reveal that LDN and LDNF epitopes are expressed on many different glycoproteins, including eggs, mother sporocysts, daughter sporocysts, and cercariae, although LDN expression among daughter sporocysts is greatly reduced. LDN and LDNF epitopes are localized on the tegument and in the intrasporocyst cell masses of both in vitro-derived and in vivo-derived mother sporocysts and in the daughter sporocysts derived on day 16 after infection. Unexpectedly, high levels of LDN and LDNF glycans were detected in the infected, but not in the uninfected, snail hemolymph, suggesting that the infecting larvae secrete LDN and LDNF glycoconjugates into the snail hosts. In contrast, the expression of Le(x) antigen among the molluscan stages is highly restricted. Le(x) is present on a few high-molecular weight glycoproteins in eggs and cercariae but is undetectable in mother and daughter sporocysts. Taken together with our earlier studies on vertebrate stages of S. mansoni, these results show that LDN and LDNF glycans are conserved during schistosome development. The study further extends the evidence that Le(x) is a developmentally regulated antigen in schistosomes.

Viral vector-mediated delivery of competing glycosyltransferases modifies epitope expression cell specifically

The glycoconjugate epitopes 3-fucosyl-N-acetyllactosamine (CD15) and sulfoglucuronylcarbohydrate (SGC) mediate cell adhesion events in several systems, and are regulated both spatially and temporally during cerebellar development. In cotransfection studies using COS-1 cells, competition between glycosyltransferases that utilize a common precursor involved in the final synthetic steps of these epitopes, can modulate epitope expression. For example, cotransfection of rat alpha1,3-fucosyltransferase IV (Fuc-TIV) and either rat glucuronic acid transferase P (GlcAT) or pig alpha1,3-galactosyltransferase (GalT) resulted in the dominance of either SGC or GalalphaGal epitope expression, respectively, with blockage of CD15 epitope expression. Viral vectors expressing these glycosyltransferases were used to determine whether competition plays a role in establishing epitope dominance in cerebellar cells, and whether overexpression of competing glycosyltransferases could be used to block epitope expression. Infection of cerebellar astrocytes with viral vectors expressing either Fuc-TIV, or Fuc-TIX, caused dramatic increases in CD15 expression in the presence of continued endogenous SGC epitope expression. Likewise, viral transduction with GalT resulted in GalalphaGal expression without affecting endogenous CD15 or SGC expression. Thus, competition between these enzymes does not appear to play a role in establishing epitope expression in astrocytes, and transduction of these enzymes does not provide a method of blocking the expression of endogenous epitopes. In contrast to what was observed for astrocytes, infection with viral vectors expressing either Fuc-T, GlcAT, or GalT did not result in significant expression of the relevant epitopes (CD15, SGC or GalalphaGal, respectively) on granule neurons. These results suggest a different complement of precursors are present in granule neurons and astrocytes, presumably due to the presence of different complements of glycosyltransferases in these cells.

Expression of alpha-gal epitopes on HeLa cells transduced with adenovirus containing alpha1,3galactosyltransferase cDNA

Alpha1,3galactosyltransferase (alpha1,3GT) synthesizes alpha-gal epitopes (Gal(alpha)1-3Galbeta1-4GlcNAc-R) on glycoconjugates in nonprimate mammals but not in humans. Transduction of alpha1,3GT gene into human HeLa cells by an adenovirus vector allowed for accurate kinetics studies on the appearance of alpha1,3GT and of its product, the alpha-gal epitope, in the transduced cells. Mouse alpha1,3GT cDNA was inserted into a replication-defective adenovirus vector. This viral vector, designated Ad(alpha)GT, could be propagated in human 293 cells that have the viral E1 complementing gene. Transduction of HeLa cells resulted in immediate penetration of approximately 20 Ad(alpha)GT copies into each cell and the appearance of alpha1,3GT mRNA after 4h. Catalytic activity of alpha1,3GT was first detected in the cells after 6 h. The initial appearance of alpha-gal epitopes (approximately 6 x 10(4)/cell) on cell surface glycoconjugates was detected 10 h posttransduction, whereas 24 h posttransduction each cell expressed 2 x 10(6) epitopes. The activity of alpha1,3GT in cells transduced with approximately two copies of Ad(alpha)GT was eightfold lower than that in cells transduced with approximately 20 Ad(alpha)GT copies; however, the number of alpha-gal epitopes/cell remained closely similar. This implies that increased alpha1,3GT activity above a certain saturation level does not result in a corresponding increase in the carbohydrate product, possibly because of competing glycosyltransferases.

Distinctive tissue distribution and phosphorylation of IRSp53 isoforms

An insulin-receptor substrate of 53-kDa protein (IRSp53) is an adapter protein, which interacts with the Rho-family of GTPases and mediates neurite outgrowth. It also binds to DRPLA protein, a product of the gene responsible for a polyglutamine disease, dentatorubral-pallidoluysian atrophy (DRPLA). Isoforms of human IRSp53 have been reported, each with a unique amino acid sequence at the C-terminal end. Here we report the distinctive tissue distribution and phosphorylation of three isoforms (L, S, and T-forms). Western blotting analyses with isoform-specific antibodies demonstrated that the L and S-forms were expressed in the brain, whereas the T-form was not present in any tissues examined, but was found in a cancer cell line. The L and S-forms were phosphorylated upon stimulation with insulin, and the T-form with IGF-I. Since phospho-acceptor sites were localized to the common portion, the difference in phosphorylation seems to be due to the unique C-terminal sequence.

Parietal endoderm cell line from a rat blastocyst

We have derived a cell line, RE1, from a pre-implantation rat blastocyst, resembling morphologically the L2 cell line from a parietal yolk sac carcinoma of the rat, as well as parietal endoderm cell lines of the mouse. The sub-cellular organization and epithelial characteristics of RE1 cells are described. The cells express cytokeratins of simple epithelia, and vimentin; and demonstrate synthesis of proteins of the extracellular matrix, such as laminin and collagen IV. Extensive Reichert's-like basement membrane is formed by RE1 cells when grown in suspension as aggregates. Cells have a microvillous surface morphology and abundant, rough endoplasmic reticulum which is swollen with apparent secretory material. These morphological and cytochemical features are characteristic of parietal endoderm cells in vivo, and the RE1 cell line is deduced to be rat parietal endoderm. In addition, RE1 cells were examined for expression of stage-specific embryonic antigens: cells reacted with antibody against SSEA-1/TEC-1 and EMA-1, constituting the first observation of parietal endoderm cells expressing the respective epitopes. RE1-cell monolayers did not generate transepithelial resistances or potential differences in vitro, consistent with their formation of leaky epithelia. Our observations on RE1-cell morphology and ultrastructure are consistent with the occurrence of epithelial-mesenchyme transitions in culture.

Over-expression of nucleophosmin/B23 decreases the susceptibility of human leukemia HL-60 cells to retinoic acid-induced differentiation and apoptosis

Stable clones of HL-60 cells in which nucleophosmin/B23 was over-expressed were established. Less percentages (4-20%) of nucleophosmin/B23 over-expressed (pCR3-B23) cells exhibited the morphological characteristic of apoptosis as compared with control vector-transfected (pCR3) cells (6-53%) during the 10 microM RA treatment for 1-4 days. In flow cytometry analysis, a block in the G1 phase was noted in all the pCR3-B23 and pCR3 cells after 2 days of 10 microM RA treatment and continued to be observed at all times measured up to 6 days. Smaller peaks of apoptotic cells with less than G1 DNA content were observed in pCR3-B23 as compared with pCR3 cells after 4-6 days of 10 microM RA treatment. As measured by expressions of differentiation markers and the functional assessment of the ability to reduce nitroblue-tetrazolium, our results further showed that over-expression of nucleophosmin/B23 decreased the response of the cells to RA-induced differentiation. Less cleavage of PARP and in vitro caspase-3 activity were observed in PCR3-B23 cells as compared with pCR3 cells treated with 10 microM RA for 3-4 days. IRF-1 was induced after 6 hr of 10 microM RA treatment in the pCR3-B23 and pCR3 cells. Significantly more nucleophosmin/B23 was co-immunoprecipitated with IRF-1 from pCR3-B23 cells than from pCR3 cells during RA treatment (10 microM; 24 hr, 96 hr). The IRF-1 transcriptional activity was found to be attenuated in pCR3-B23 cells as compared with pCR3 cells during the treatment of cells with RA. Nucleophosmin/B23, through interacting with IRF-1, plays an important role in the control of the susceptibility of cells to RA-induced differentiation and apoptosis.

Restriction of high CD15 expression to a subset of rat cerebellar astroglial cells can be overcome by transduction with adenoviral vectors expressing the rat alpha 1,3-fucosyltransferase IV gene

Glycoconjugates bearing the epitope 3-fucosyl-N-acetyllactosamine (CD15) are believed to be involved in cell-cell interactions and are temporally and spatially regulated in the brain. In the rat postnatal cerebellum, CD15 is predominantly expressed in the molecular layer by Bergmann glial cells, but little CD15 expression is seen in other astroglia, and the basis for this restricted expression is not known. Adenoviral vectors were shown to efficiently deliver transgenes to cerebellar glial cells and were used to determine whether manipulation of glycosyltransferase activities could enhance the expression of CD15 in these cells. In dissociated cerebellar cell cultures, few glial cells normally express CD15. However, transduction of these cells with an adenoviral vector (AdGFPCMVFucT) that expressed both green fluorescent protein (GFP) and FLAG-tagged rat alpha 1, 3-fucosyltransferase IV (rFuc-TIV) resulted in high CD15 expression on the surface of all transduced glial cells. Likewise, infection of cerebellar slice cultures caused the appearance of CD15-positive transduced cells of glial cell morphology in the internal granule cell layer. Thus, enhancement of Fuc-T activity caused robust CD15 expression in cerebellar glial cells that normally show little expression of CD15, suggesting a role for Fuc-T levels in regulating CD15 expression in this cell type. The manipulation of levels of glycosyltransferases using adenoviral vectors may prove a useful tool to investigate questions of glycoconjugate regulation in glial cells in the developing rodent cerebellum.

Characterization of insulin receptor substrate 3 in rat liver derived cells

In rat liver derived HTC cells transfected with and expressing human insulin receptors, there are multiple p60-70 proteins that are tyrosine phosphorylated following insulin treatment of cells. Employing antibodies to insulin receptor substrate 3 (alpha-IRS-3), we found that IRS-3 is a major p60 phosphoprotein that is tyrosine phosphorylated following insulin treatment of cells and interacts with phosphatidylinositol-3-kinase (PI3K). Majority of IRS-3 when phosphorylated appears to interact with PI3K. Tyrosine phosphorylation of IRS-3 is robust at 2 min and steadily increases up to 30-90 min of insulin treatment. Following insulin treatment of cells, some high molecular weight phosphoproteins are coimmunoprecipitated with alpha-IRS-3. In summary, IRS-3 is the major p60 protein that is tyrosine phosphorylated and interacts with PI3K in HTC rat liver derived cells following insulin treatment of cells. Unlike related IRS-1/2 that is transiently phosphorylated, IRS-3 shows robust and prolonged tyrosine phosphorylation upon insulin treatment of cells and may play a role in delayed and/or prolonged insulin actions.

Rho small G-protein-dependent binding of mDia to an Src homology 3 domain-containing IRSp53/BAIAP2

mDia1 is a downstream effector of Rho small G protein that is implicated in stress fiber formation and cytokinesis. We isolated an mDia1-binding protein and identified it to be IRSp53/BAIAP2. IRSp53 and BAIAP2 have independently been isolated as a 58/53-kDa protein tyrosine phosphorylated in response to insulin and a BAI1-binding protein, respectively. BAI1 is a brain-specific seven-span transmembrane protein capable of inhibiting angiogenesis. The proline-rich formin homology 1 domain of mDia1 bound the Src homology 3 domain of IRSp53/BAIAP2 in a GTP-Rho-dependent manner. The results suggest that IRSp53/BAIAP2 is a downstream effector of mDia1.

Functional implications of the noradrenergic-cholinergic switch induced by retinoic acid in NB69 neuroblastoma cells

Some neuroblastoma cell lines change their neurotransmitter phenotype from noradrenergic to cholinergic under retinoic acid treatment. Such "neurotransmitter switch" seems to be a consequence of changes in the expression and activity of the biosynthetic machinery for both neurotransmitters. In this study, we have characterized this "neurotransmitter switch" induced by retinoic acid in a human neuroblastoma cell line (NB69) showing catecholaminergic characteristics. Retinoic acid treatment reduced tyrosine hydroxylase activity and noradrenaline levels in NB69 cells but did not modify the expression of this enzyme. Moreover, the calcium-dependent release of [(3)H]noradrenaline in control cells was highly reduced by retinoic acid treatment. On the other hand, NB69 cells treated with retinoic acid enhanced the expression of choline acetyltransferase and acquired the capability to release [(3)H]acetylcholine in a calcium-dependent way. In addition, we found that the expression of the vesicular monoamine transporter 2 (VMAT2) and the vesicular acetylcholine transporter (VAChT) was increased in those cells treated with retinoic acid. Immunostaining revealed that retinoic acid treatment changed the cellular distribution of both vesicular monoamine transporter 2 and vesicular acetylcholine transporter. In conclusion, retinoic acid induces a noradrenergic to cholinergic switch in NB69 cells by acting at several levels of the neurotransmitter phenotypic expression.

Transcriptional repression of the human galactocerebrosidase gene in squamous cell carcinomas of the larynx

Alterations of gene expression in squamous cell carcinoma (SCC) cell lines derived from the larynx and keratinocytes derived from adjacent normal mucosa of the larynx have been studied using the mRNA differential display technique. Lane-to-lane comparison of reverse transcribed mRNA showed a strong repression of a 148 bp fragment in SCC cells. The fragment was reamplified and cloned. Sequencing revealed a 99.3% homology with a region in exon 17 of the human galactocerebrosidase (GALC) gene. Northern blot analysis confirmed the differential expression of this gene in both carcinoma cell lines and laryngeal SCC biopsies in contrast with corresponding normal mucosa. To provide further evidence for the differential expression rate, both types of cells were transiently transfected with a 152 bp (-176 to -24) high regulatory promoter element of the 5' flanking region of the GALC gene. Results of 3 independent transfection experiments indicated a 16-fold repression of the GALC gene expression in SCC cells compared with benign keratinocytes. However, neither mutation nor other alterations of the promoter sequence were detected. Expression of the GALC gene is thus greatly affected in SCCs of the larynx.

Morphological domains of Lewis-X/FORSE-1 immunolabeling in the embryonic neural tube are due to developmental regulation of cell surface carbohydrate expression

The Lewis-X (LeX) carbohydrate epitope, recognized by the FORSE-1 monoclonal antibody (mAb), shares expression boundaries with neural regulatory genes and may be involved in patterning the neural tube by creating domains of differential cell adhesion. The present experiments focus on the question of what determines the expression pattern of LeX in embryonic rat brain. Comparisons of FORSE-1-positive glycolipid and protein antigens in embryonic, early postnatal, and adult tissues show that the LeX epitope is carried primarily by glycolipids during embryonic development and by a proteoglycan and glycoproteins in postnatal and adult tissue. Immunohistochemistry using FORSE-1 and an antibody to the proteoglycan phosphacan, which carries LeX, shows that the distribution of LeX is more restricted than phosphacan. These observations suggest that the precise spatial regulation of FORSE-1 binding in the embryonic forebrain is due to the expression pattern of the LeX carbohydrate on glycolipids, rather than to the transcriptional regulation of a carrier protein.

Coexpression of alpha1,2 galactosyltransferase and UDP-galactose transporter efficiently galactosylates N- and O-glycans in Saccharomyces cerevisiae

We have studied in vivo neo-galactosylation in Saccharomyces cerevisiae and analyzed the critical factors involved in this system. Two heterologous genes, gma12(+) encoding alpha1, 2-galactosyltransferase (alpha1,2 GalT) from Schizosaccharomyces pombe and UGT2 encoding UDP-galactose (UDP-Gal) transporter from human, were functionally expressed to examine the intracellular conditions required for galactosylation. Detection by fluorescence labeled alpha-galactose specific lectin revealed that 50% of the cells incorporated galactose to cell surface mannoproteins only when the gma12(+) and hUGT2 genes were coexpressed in galactose media. Integration of both genes in the Delta mnn1 background cells increased galactosylation to 80% of the cells. Correlation between cell surface galactosylation and UDP-galactose transport activity indicated that an exogenous supply of UDP-Gal transporter rather than alpha1,2 GalT played a key role for efficient galactosylation in S.cerevisiae. In addition, this heterologous system enabled us to study the in vivo function of S. pombe alpha1,2 GalT to prove that it transfers galactose to both N - and O -linked oligosaccharides. Structural analysis indicated that this enzyme transfers galactose to O -mannosyl residue attached to polypeptides and produces Galalpha1,2-Man1-O-Ser/Thr structure. Thus, we have successfully generated a system for efficient galactose incorporation which is originally absent in S. cerevisiae, suggesting further possibilities for in vivo glycan remodeling toward therapeutically useful galactose containing heterologous proteins in S. cerevisiae.

Mouse C127 cells transfected with fucosyltransferase fuc-TIII express masked Lewisx but not Lewisx antigen

To study human alpha1,3/1,4fucosyltransferase (Fuc-TIII) as an alpha1,3 fucosyltransferase, we constructed two cell clones, C127-FT and C127-T-FT, by transfecting cDNA in parental (C127) or Polyoma T antigen expressing (C127-T) mouse cells, respectively. Both C127-FT and C127-T-FT clones express high levels of a fucosyltransferase activity kinetically similar to Fuc-TIII and an RNA that is amplified by a Fuc-TIII-specific oligonucleotide primer pair after reverse transcription. Clone C127-FT is Lewisxpositive, by flow cytometry, only after alpha-galactosidase or sialidase treatment, and releases [3H]Fuc N-glycans which efficiently bind to immobilized Griffonia simplicifolia I and Sambucus nigra lectins. Immunoblotting confirms that C127-FT glycoproteins acquire Lewisxreactivity only after specific deglycosylation, and shows that a small subset of Griffonia simplicifolia I isolectin B4reactive glycoproteins bears masked Lewisx, suggesting fine substrate recognition by Fuc-TIII. Moreover, transient transfection of H type alpha1, 2fucosyltransferase in clone C127-T-FT directs synthesis of Lewisyantigen, as detected by flow cytometry. Results indicate that Fuc-TIII expressed in C127 cells synthesizes masked Lewisxantigen while Lewisxantigen is not detectable.

Embryoglycan ectodomains regulate biological activity of FGF-2 to embryonic stem cells

Basic fibroblast growth factor (FGF-2) functions as a natural inducer of mesoderm, regulator of cell differentiation and autocrine modulator of cell growth and transformation. The FGF-2 signals are transduced through receptors with intrinsic protein tyrosine kinase activity. However, receptor binding and activation is governed by extracellular matrix, cell surface or soluble proteoglycans. This paper focuses on the role of proteoglycans synthesized by embryonic cells, embryoglycans, in FGF-2 signaling via FGF receptor-1 (FGFR-1). We found that embryoglycan ectodomain Lewis X, analog of developmentally regulated embryonic cell surface epitope TEC 1, promotes oligomerization of FGF-2 in the cell free chemical crosslinking. In vitro assays show that a large molar excess of extracellular Lewis X does not inhibit binding of FGF-2 to embryonic stem (ES) cells, but prevents the mitogenic effect of FGF-2. Western blot analysis of ES cells revealed the presence of abundant 52 kDa and trace amounts of 67 and 125 kDa isoforms of FGFR-1. However, none of these isoforms undergo any detectable changes in tyrosine phosphorylation under the conditions that modulate the mitogenic effect of FGF-2. Rather, a primary substrate of all receptor tyrosine kinases, phospholipase C gamma (PLC gamma), is activated by both FGF-2 and Lewis X. The combination, FGF-2 plus Lewis X, leads to weak inhibition, when compared with the effects of FGF-2 and Lewis X, respectively. In accordance, the level of phosphorylation of non-receptor tyrosine kinase c-Src is reduced in a reversed pattern to PLC(gamma). Furthermore, in this particular cell type we show the presence of activated forms of extracellular signal-related kinase (ERK) in all nontreated and treated cells. These findings demonstrate that embryoglycan ectodomains may act as negative regulators of FGF-2-induced ES cell proliferation, most likely through the FGFR-1-independent signaling pathway.

Increased elongation of N-acetyllactosamine repeats in doubly glycosylated lysozyme with a particular spacing of the glycosylation sites

Lysozyme is an example of an extensively studied secretory enzyme. Glycosylated mutant human lysozyme has been used as a model in studies on the biosynthesis of N-acetyllactosamine repeats in N-linked oligosaccharides. We examined the biosynthesis of the repeats in two doubly glycosylated mutants and describe here a rapid purification and separation of singly and doubly glycosylated molecules. In one of the mutants, the elongation of the repeats is enhanced if the molecules are doubly glycosylated, but not if the carbohydrate is attached to either site individually. This enhancement is not seen in the other doubly glycosylated mutant. Since lysozyme is not structurally related to glycoproteins bearing carbohydrate with N-acetyllactosamine repeats, we propose that in multivalent substrates the synthesis of the repeats can be promoted by a proper spacing of the elongated carbohydrate antennae in addition to any role of the protein backbone.

Molecular cloning and characterization of an alpha1,3 fucosyltransferase, CEFT-1, from Caenorhabditis elegans

We report on the identification, molecular cloning, and characterization of an alpha1,3 fucosyltransferase (alpha1,3FT) expressed by the nematode, Caenorhabditis elegans . Although C. elegans glycoconjugates do not express the Lewis x antigen Galbeta1-->4[Fucalpha1-->3]GlcNAcbeta-->R, detergent extracts of adult C.elegans contain an alpha1,3FT that can fucosylate both nonsialylated and sialylated acceptor glycans to generate the Lexand sialyl Lexantigens, as well as the lacdiNAc-containing acceptor GalNAcbeta1-->4GlcNAcbeta1-->R to generate GalNAcbeta1-->4 [Fucalpha1-->3]GlcNAcbeta1-->R. A search of the C.elegans genome database revealed the existence of a gene with 20-23% overall identity to all five cloned human alpha1,3FTs. The putative cDNA for the C.elegans alpha1,3FT (CEFT-1) was amplified by PCR from a cDNA lambdaZAP library, cloned, and sequenced. COS7 cells transiently transfected with cDNA encoding CEFT-1 express the Lex, but not sLexantigen. The CEFT-1 in the transfected cell extracts can synthesize Lex, but not sialyl Lex, using exogenous acceptors. A second fucosyltransferase activity was detected in extracts of C. elegans that transfers Fuc in alpha1,2 linkage to Gal specifically on type-1 chains. The discovery of alpha-fucosyltransferases in C. elegans opens the possibility of using this well-characterized nematode as a model system for studying the role of fucosylated glycans in the development and survival of C.elegans and possibly other helminths.

Two soluble glycosyltransferases glycosylate less efficiently in vivo than their membrane bound counterparts

Many Golgi glycosyltransferases are type II membrane proteins which are cleaved to produce soluble forms that are released from cells. Cho and Cummings recently reported that a soluble form of alpha1, 3-galactosyltransferase was comparable to its membrane bound counterpart in its ability to galactosylate newly synthesized glycoproteins (Cho,S.K. and Cummings,R.D. (1997) J. Biol. Chem., 272, 13622-13628). To test the generality of their findings, we compared the activities of the full length and soluble forms of two such glycosyltransferases, ss1,4 N-Acetylgalactosaminyltransferase (GM2/GD2/ GA2 synthase; GalNAcT) and beta galactoside alpha2,6 sialyltransferase (alpha2,6-ST; ST6Gal I), for production of their glycoconjugate products in vivo . Unlike the full length form of GalNAcT which produced ganglioside GM2 in transfected cells, soluble GalNAcT did not produce detectable GM2 in vivo even though it possessed in vitro GalNAcT activity comparable to that of full length GalNAcT. When compared with cells expressing full length alpha2,6-ST, cells expressing a soluble form of alpha2,6-ST contained 3-fold higher alpha2,6-ST mRNA levels and secreted 7-fold greater alpha2,6-ST activity as measured in vitro , but in striking contrast contained 2- to 4-fold less of the alpha2,6-linked sialic acid moiety in cellular glycoproteins in vivo . In summary these results suggest that unlike alpha1,3-galactosyltransferase the soluble forms of these two glycosyltransferases are less efficient at glycosylation of membrane proteins and lipids in vivo than their membrane bound counterparts.

Stimulation of the high-affinity IgE receptor results in the tyrosine phosphorylation of a 60 kD protein which is associated with the protein-tyrosine kinase, Csk

The protein tyrosine kinase Csk downregulates the activity of the Src family of kinases and has a negative effect on signal transduction through several Src kinase-associated receptors. Because the Src-family kinase Lyn plays a pivotal role in FcepsilonRI-mediated cellular activation, we examined whether Csk is involved in FcepsilonRI signaling events. Using anti-Csk antibodies and recombinant fusion proteins we detected a single tyrosine-phosphorylated protein of 60 kD (herein referred to as 'p60') that associates with the SH2 domain of Csk after stimulation of the FcepsilonRI. p60 phosphorylation reached a maximum within one minute and remained constant while the receptors were aggregated; disaggregation of the receptors resulted in rapid dephosphorylation of p60. The phosphorylation of p60 was only detected after activation by IgE and antigen and not by stimulation with PMA and/or ionomycin. Phosphorylated p60 was associated entirely with the membrane fraction of the cells. A considerable fraction of Csk was associated with the membrane in both unstimulated and stimulated cells, this fraction did not change upon activation. p60 coprecipitated with Csk from both unstimulated and FcepsilonRI stimulated cells and was phosphorylated by the immunocomplex. Total kinase activity of Csk immunoprecipitates increased upon FcepsilonRI stimulation. p60 did not react with antibodies to a number of known signaling molecules, including the recently cloned, GAP-associated protein, p62dok. Our data demonstrate that Csk associates with a membrane-anchored protein complex that is directly involved in FcepsilonRI signal transduction.

Jacalin interacts with Asn-linked glycopeptides containing multi-antennary oligosaccharide structure with terminal alpha-linked galactose

The carbohydrate binding properties of jacalin lectin were examined using RAF9 cell-derived D-[6-3H]glucosamine-radiolabeled total glycopeptides containing N-linked and O-linked oligosaccharides. The binding of N-linked glycopeptides to jacalin was abolished by treatment of alpha-galactosidase whereas O-linked glycopeptides were still bound lectin after this treatment. The removal of O-linked oligosaccharides by mild alkaline/borohydride treatment completely eliminated the lectin binding of alpha-galactosidase treated glycopeptides. These results demonstrate that jacalin interacts with cellular glycopeptides containing N-linked oligosaccharides with terminal alpha-galactose residues as well as glycopeptides containing O-linked oligosaccharides.

Secretion of alpha1,3-galactosyltransferase by cultured cells and presence of enzyme in animal sera

Glycosyltransferases are normally synthesized as membrane-anchored proteins. However, we recently found that the murine enzyme UDP-Gal:Gal beta1 -->4GLcNAc (Gal to Gal) alpha1,3 galactosyltransferase (alpha1,3GT) is secreted in a soluble form into media by mouse teratocarcinoma F9 cells (Cho SK, Yeh J-C, Cho M, Cummings RD (1996) J Biol Chem 271: 3238-46). To study the biosynthesis of this enzyme and whether secretion of the soluble enzyme is a general phenomenon, a solid-phase assay was developed for the alpha1,3GT activity. A recombinant and soluble form of the murine alpha1,3GT was produced in H293 cells (H293-alpha1,3GT) to aid in optimizing the assay. Desialylated orosomucoid was used as an immobilized acceptor in coated microtiter plates. The formation of product was detected by a biotinylated human-derived anti-alpha-Gal IgG and streptavidin conjugated to either alkaline phosphatase or the recombinant bioluminescent protein aequorin. Enzyme activity was dependent on the concentrations of asialoorosomucoid, UDP-Gal, alpha1,3GT and the time of incubation. The assay was also useful in monitoring alpha1,3GT activity during enzyme enrichment procedures. Using this assay, we found that alpha1,3GT activity was present in both cell extracts and culture media of several mammalian cell lines. Enzyme activity was also present in the sera from several mammals, but activity was absent in the sera from either humans or baboons. Our results demonstrate the development of a novel assay for the alpha1,3GT and provide evidence that secretion of the enzyme is a common biological phenomenon.

Transcriptional regulation of N-acetylglucosaminyltransferase V by the src oncogene

Transformation of baby hamster kidney fibroblasts by the Rous sarcoma virus causes a significant increase in the GlcNAcbeta(1, 6)Man-branched oligosaccharides by elevating the activity and mRNA transcript levels encoding N-acetylglucosaminyltransferase V (GlcNAc-T V). Elevated activity and mRNA levels could be inhibited by blocking cell proliferation with herbimycin A, demonstrating that Src kinase activity can regulate GlcNAc-T V expression. 5' RACE analysis was used to identify a 3-kilobase 5'-untranslated region from GlcNAc-T V mRNA and locate a transcriptional start site in a 25-kilobase pair GlcNAc-T V human genomic clone. A 6-kilobase pair fragment of the 5' region of the gene contained AP-1 and PEA3/Ets binding elements and, when co-transfected with a src expression plasmid into HepG2 cells, conferred src-stimulated transcriptional enhancement upon a luciferase reporter gene. This stimulation by src could be antagonized by co-transfection with a dominant-negative mutant of the Raf kinase, suggesting the involvement of Ets transcription factors in the regulation of GlcNAc-T V gene expression. The src-responsive element was localized by 5' deletion analysis to a 250-base pair region containing two overlapping Ets sites. src stimulation of transcription from this region was inhibited by co-transfection with a dominant-negative mutant of Ets-2, demonstrating that the effects of the src kinase on GlcNAc-T V expression are dependent on Ets.

Regulation of N-acetylglucosaminyltransferase V and Asn-linked oligosaccharide beta(1,6) branching by a growth factor signaling pathway and effects on cell adhesion and metastatic potential

Recent evidence demonstrates that the changes in the size of N-linked oligosaccharides that correlate with cell transformation and tumorigenicity are due at least in part to the regulation of expression of a glycosyltransferase involved in the branching of N-linked structures, N-acetylglucosaminyltransferase V or GlcNAc-T V. Studies have shown that the increases in GlcNAc-T V expression after oncogenic transformation are most likely caused by direct effects on the GlcNAc-T V promoter by the Ets family of transcriptional activators, which are up-regulated by a cellular proliferation signaling pathway. This pathway begins with growth factor receptors that activate tyrosine kinases at the cell surface and proceeds through src, ras, and raf. Additional evidence for the association between cellular proliferation and GlcNAc-T V expression will be presented, as well as a discussion of the effects of beta(1,6) branching on several of the phenotypes of oncogenically transformed cells, including metastatic potential.

A soluble form of alpha1,3-galactosyltransferase functions within cells to galactosylate glycoproteins

It has been assumed that membrane-bound glycosyltransferases function within the Golgi apparatus to glycosylate glycoproteins. We now report, however, that a truncated, soluble recombinant form of the murine alpha1,3-galactosyltransferase expressed in human 293 cells is highly efficient and comparable to the full-length enzyme in alpha-galactosylating both newly synthesized membrane-associated and secreted glycoproteins. Although the soluble enzyme was secreted by cells as expected, we also found that the full-length, membrane-associated form was secreted. Unexpectedly, both secreted forms are cleaved identically at two primary sites within the stem region by endogenous protease(s) at the indicated positions in the sequence 73KDWW (downward arrow) FPS (downward arrow) WFKNG. These results demonstrate that the soluble alpha1,3-galactosyltransferase is functional within the cell and that specific proteolysis occurs in the stem region. The widespread occurrence of different soluble glycosyltransferases secreted by cells suggests that normal glycoconjugate biosynthesis may involve cooperation between membrane-bound and soluble enzymes.

Molecular cloning and expression of a bovine alpha(1,3)-fucosyltransferase gene homologous to a putative ancestor gene of the human FUT3-FUT5-FUT6 cluster

Only one bovine gene, corresponding to the human cluster of genes FUT3-FUT5-FUT6, was found by Southern blot analysis. The cognate bovine alpha(1,3)-fucosyltransferase shares 67.3, 69.0, and 69.3% amino acid sequence identities with human FUC-T3, FUC-T5, and FUC-T6 enzymes, respectively. As revealed by protein sequence alignment, potential sites for asparagine-linked glycosylation and conserved cysteines, the bovine enzyme is an intermediate between FUC-T3, FUC-T5, and FUC-T6 human enzymes. Transfected into COS-7 cells, the bovine gene induced the synthesis of an alpha(1, 3)-fucosyltransferase enzyme with type 2 substrate acceptor pattern specificity and induced expression of fucosylated type 2 epitopes (Lex and sialyl-Lex), but not of type 1 structures (Lea or sialyl-Lea), suggesting that it has an acceptor specificity similar to the human plasma FUC-T6. However, no enzyme activity was detected in bovine plasma. Gene transcripts are detected on tissues such as bovine liver, kidney, lung, and brain. The type 2 sialyl-Lex epitope was found in renal macula densa and biliary ducts, and Lex and Ley epitopes were detected on the brush border of epithelial cells of small and large intestine, suggesting a tissue distribution closer to human FUC-T3, but fucosylated type 1 structures (Lea, Leb, or sialyl-Lea) were not detected at all in any bovine tissue. Analysis of genetic distances on a combined phylogenetic tree of fucosyltransferase genes suggests that the bovine gene is the orthologous homologue of the ancestor of human genes constituting the present FUT3-FUT5-FUT6 cluster.

Porcine alpha1,3-galactosyltransferase: tissue-specific and regulated expression of splicing isoforms

Expression of the Gal alpha1,3 Gal epitope on membrane glycolipids and glycoproteins is known to vary widely from one tissue to another. In the course of studying the mechanisms underlying this variability, we have isolated from pig cDNA four sequences corresponding to four isoforms of alpha1,3-galactosyltransferase (alpha1,3GT), the Golgi enzyme that links galactose in alpha1,3 on the galactose residue of N-acetyllactosamine. The isoforms differ from each other in the alternative presence of two nucleotide stretches of 36 and 63 base pairs in a segment encoding the stem region of the protein. Stable expression experiments show that all four isoenzymes can confer alpha-galactosyltransferase activity to HeLa cells, and that they are all located within the Golgi compartment, indicating that variations in length in the stem region do not affect enzyme activity or cellular localization. Analysis of RNA from different pig organs and cells shows quantitative differences between tissues in levels of alpha1,3GT, as well as qualitative differences, the four isoforms being unequally represented in different tissues.

Expression cloning of Forssman glycolipid synthetase: a novel member of the histo-blood group ABO gene family

A phenotypic cloning approach was used to isolate a canine cDNA encoding Forssman glycolipid synthetase (FS; UDP-GalNAc:globoside alpha-1,3-N-acetylgalactosaminyltransferase; EC 2.4.1.88). The deduced amino acid sequence of FS demonstrates extensive identity to three previously cloned glycosyltransferases, including the enzymes responsible for synthesis of histo-blood group A and B antigens. These three enzymes, like FS, catalyze the addition of either N-acetylgalactosamine (GalNAc) or galactose (Gal) in alpha-1,3-linkage to their respective substrates. Despite the high degree of sequence similarity among the transferases, we demonstrate that the FS cDNA encodes an enzyme capable of synthesizing Forssman glycolipid, and demonstrates no GalNAc or Gal transferase activity when closely related substrates are examined. Thus, the FS cDNA is a novel member of the histo-blood group ABO gene family that encodes glycosyltransferases with related but distinct substrate specificity. Cloning of the FS cDNA will allow a detailed dissection of the roles Forssman glycolipid plays in cellular differentiation, development, and malignant transformation.