Evidence for a molecular distinction between Golgi and cell surface forms of beta 1,4-galactosyltransferase

Altered N-linked glycosylation in depression: A pre-clinical study

BACKGROUND

Neuroimmune plays an important role in major depressive disorders (MDD). N-linked protein glycosylation (NLG) might contribute to depression by regulating the neuroinflammatory response. As microglia is the main executor of neuroimmune function in the central neural system (CNS), targeting the process of N-linked protein glycosylation of microglia in the mice used for studying depression might potentially offer new avenues for the strategy for MDD.

METHODS

The chronic unpredictable mild stress (CUMS) mouse model was established for the whole brain microglia isolating. Then, RNA samples of microglia were extracted for transcriptome sequencing and mRNA analysis. Immunofluorescence (IF) was used to identify the expression level of NLG-related enzyme, B4galt1, in microglia.

RESULTS

The data showed that NLG was positively related to depression. Moreover, the NLG-related gene, B4galt1 increased expression in the microglia of CUMS mice. Then, the inhibition of NLG reversed the depressive behavior in CUMS mice. The expression level of B4galt1 in CUMS mice was upregulating following the NLG-inhibitor treatment. Similar results haven't been observed in neurons. Information obtained from these experiments showed increasing expression of B4galt1 in microglia following depressive-like behaviors.

CONCLUSIONS

These findings indicate that NLG in microglia is associated with MDD, and suggest that therapeutically targeting NLG might be an effective strategy for depression.

LIMITATIONS

How to modulate the B4galt1 or NLG pathways in microglia efficiently and economically request new technologies.

Applying transcriptomics to studyglycosylation at the cell type level

The complex multi-step process of glycosylation occurs in a single cell, yet current analytics generally cannot measure the output (the glycome) of a single cell. Here, we addressed this discordance by investigating how single cell RNA-seq data can be used to characterize the state of the glycosylation machinery and metabolic network in a single cell. The metabolic network involves 214 glycosylation and modification enzymes outlined in our previously built atlas of cellular glycosylation pathways. We studied differential mRNA regulation of enzymes at the organ and single cell level, finding that most of the general protein and lipid oligosaccharide scaffolds are produced by enzymes exhibiting limited transcriptional regulation among cells. We predict key enzymes within different glycosylation pathways to be highly transcriptionally regulated as regulatable hotspots of the cellular glycome. We designed the Glycopacity software that enables investigators to extract and interpret glycosylation information from transcriptome data and define hotspots of regulation.

•

RNA-seq can provide information on the glycosylation metabolic network state

•

It is possible to readout glycosylation capacity from single cell RNA-seq data

•

Genes regulating the biosynthesis of common glycan scaffolds show little regulation

•

Key enzymes in the glycosylation network are predicted to be regulatable hotspots

N-glycosylation of the human β1,4-galactosyltransferase 4 is crucial for its activity and Golgi localization

β1,4-galactosyltransferase 4 (B4GalT4) is one of seven B4GalTs that belong to CAZy glycosyltransferase family 7 and transfer galactose to growing sugar moieties of proteins, glycolipids, glycosaminoglycans as well as single sugar for lactose synthesis. Herein, we identify two asparagine-linked glycosylation sites in B4GalT4. We found that mutation of one site (Asn220) had greater impact on enzymatic activity while another (Asn335) on Golgi localization and presence of N-glycans at both sites is required for production of stable and enzymatically active protein and its secretion. Additionally, we confirm B4GalT4 involvement in synthesis of keratan sulfate (KS) by generating A375 B4GalT4 knock-out cell lines that show drastic decrease in the amount of KS proteoglycans and no significant structural changes in N- and O-glycans. We show that KS decrease in A375 cells deficient in B4GalT4 activity can be rescued by overproduction of either partially or fully glycosylated B4GalT4 but not with N-glycan-depleted B4GalT4 version.

Expression and Characterization of Human β-1, 4-Galactosyltransferase 1 (β4GalT1) Using Silkworm-Baculovirus Expression System

Baculovirus expression vector system (BEVS) is widely known as a mass-production tool to produce functional recombinant glycoproteins except that it may not be always suitable for medical practice due to the differences in the structure of N-linked glycans between insects and mammalian. Currently, various approaches have been reported to alter N-linked glycan structures of glycoproteins derived from insects into terminally sialylated complex-type N-glycans. In the light of those studies, we also proposed in vitro maturation of N-glycan with mass-produced and purified glycosyltransferases by silkworm-BEVS. β-1,4-Galactosyltransferase 1 (β4GalT1) is known as one of type II transmembrane enzymes that transfer galactose in a β-1, 4 linkage to accepter sugars, and a key enzyme for further sialylation of N-glycans. In this study, we developed a large-scale production of recombinant human β4GalT1 (rhβ4GalT1) with N- or C-terminal tags in silkworm-BEVS. We demonstrated that rhβ4GalT1 is N-glycosylated and without mucin-type glycosylation. Interestingly, we found that purified rhβ4GalT1 from silkworm serum presented higher galactosyltransferase activity than that expressed from cultured mammalian cells. We also validated the UDP-galactose transferase activity of produced rhβ4GalT1 proteins by using protein subtracts from silkworm silk gland. Taken together, rhβ4GalT1 from silkworms can become a valuable tool for producing high-quality recombinant glycoproteins with mammalian-like N-glycans.

Mouse fibroblasts null for the long isoform of β1,4-galactosyltransferase-I show defective cell-matrix interactions

β1,4 Galactosyltransferase-I (GalT-I) is expressed as two nearly identical polypeptides that differ only in the length of their cytoplasmic domains. The longer isoform has been implicated as a cell surface receptor for extracellular glycoside ligands, such as laminin. To more stringently test the function of the long GalT-I isoform during cell interactions with laminin, we created multiple independent fibroblastic cell lines that fail to express the long isoform, but which express the short GalT-I isoform normally and appear to have normal intracellular galactosylation. Cells devoid of the long GalT-I isoform are unable to adhere and spread on laminin substrates as well as control cells, but retain near normal interactions with fibronectin, which do not rely upon surface GalT-I function. The loss of the long GalT-I isoform also leads to a loss of actin stress fibers, focal adhesions and rac GTPase activation.

Multifaceted roles of 5'-regulatory region of the cancer associated gene B4GALT1 and its comparison with the gene family

β1,4-Galactosylransferases are a family of enzymes encoded by seven B4GALT genes and are involved in the development of anticancer drug resistance and metastasis. Among these genes, the B4GALT1 shows significant variations in the transcript origination sites in different cell types/tissues and encodes an interesting dually partitioning β-1, 4-galactosyltransferase protein. We identified at 5'-end of B4GALT1 a 1.454 kb sequence forming a transcription regulatory region, referred to by us as the TR1-PE1, had all characteristics of a bidirectional promoter directing the transcription of B4GALT1 in a divergent manner along with its long non-coding RNA (lncRNA) antisense counterpart B4GALT1-AS1. The TR1-PE1 showed unique dinucleotide base-stacking energy values specific to transcription factor binding sites (TFBSs), INR and BRE, and harbored CpG Island (CGI) that showed GC skew with potential for R-loop formation at the transcription starting sites (TSSs). The 5'-regulatory axis of B4GALT1 also included five more novel TFBSs for CTCF, GLI1, TCF7L2, GATA3 and SOX5, in addition to unique (TG)18 repeats in conjunction with 22 nucleotide TG-associated sequence (TGAS). The five lncRNA B4GALT1-AS1 transcripts showed significant complementarity with B4GALT1 mRNA. In contrast, the rest of B4GALT genes showed fewer lncRNAs, and all lacked the (TG)18 and TGAS. Our results are strongly supported by the FANTOM5 study which showed tissue-specific variations in transcript origination sites for this gene. We suggest that the unique expression patterns for the B4GALT1 in normal and malignant tissues are controlled by a differential usage of 5'-B4GALT1 regulatory units along with a post-transcriptional regulation by the antisense RNA, which in turn govern the cell-matrix interactions, neoplastic progression, anticancer drug sensitivity, and could be utilized in personalized therapy.

Interleukin-1-induced changes in the glioblastoma secretome suggest its role in tumor progression

The tumor microenvironment including glial cells and their inflammatory products regulates brain tumor development and progression. We have previously established that human glioma cells are exquisitely sensitive to IL-1 stimulation leading us to undertake a comparative analysis of the secretome of unstimulated and cytokine (IL-1)-stimulated glioblastoma cells. We performed label-free quantitative proteomic analysis and detected 190 proteins which included cytokines, chemokines, growth factors, proteases, cell adhesion molecules, extracellular matrix (ECM) and related proteins. Measuring area under the curve (AUC) of peptides for quantitation, the IL-1-induced secretome contained 13 upregulated and 5 downregulated extracellular proteins (p<0.05) compared to controls. Of these, IL-8, CCL2, TNC, Gal-1 and PTX3 were validated as upregulated and SERPINE1, STC2, CTGF and COL4A2 were validated as downregulated factors by immunochemical methods. A major representation of the ECM and related proteins in the glioblastoma secretome and their modulation by IL-1 suggested that IL-1 induces its effect in part by altering TGFβ expression, activity and signaling. These findings enhance our understanding of IL-1-induced modulation of glioma microenvironment, with implications for increased tumor invasion, migration and angiogenesis. They further provide novel targets for the glioblastoma intervention.

BIOLOGICAL SIGNIFICANCE

Present study is on an unbiased screening of the glioblastoma secretome stimulated by IL-1 which triggers neuroinflammatory cascades in the central nervous system. Network of secreted proteins were shown to be regulated revealing their possible contribution to glioma progression. Label free quantitative proteomics has provided unique novel targets for potential glioblastoma intervention.

Expression of long-form N-acetylglucosamine-6-O-sulfotransferase 1 in human high endothelial venules

Two members of the N-acetylglucosamine-6-O-sulfotransferase (GlcNAc6ST) family, GlcNAc6ST-1 and GlcNAc6ST-2, function in the biosynthesis of 6-sulfo sialyl Lewis X-capped glycoproteins expressed on high endothelial venules (HEVs) in secondary lymphoid organs. Thus, both enzymes play a critical role in L-selectin-expressing lymphocyte homing. Human GlcNAc6ST-1 is encoded by a 1593-bp open reading frame exhibiting two 5' in-frame methionine codons spaced 141 bp apart. Both resemble the consensus sequence for translation initiation. Thus, it has been hypothesized that both long and short forms of GlcNAc6ST-1 may be present, although endogenous expression of either form has not been confirmed in humans. Here, the authors developed an antibody recognizing amino acid residues between the first two human GlcNAc6ST-1 methionines. This antibody specifically recognizes the long form of the enzyme, a finding validated by Western blot analysis and immunofluorescence cytochemistry of HeLa cells misexpressing long and/or short forms of human GlcNAc6ST-1. Using this antibody, the authors carried out immunofluorescence histochemistry of human lymph node tissue sections and found endogenous expression of the long form of the enzyme in human tissue, predominantly in the trans-Golgi network of endothelial cells that form HEVs.

Synthesis and expression of CDw75 antigen in human colorectal cancer

Background

Increased ST6Gal I activity has been associated with the α(2,6)sialylation enhancement of membrane glycoconjugates observed in metastatic colorectal carcinomas (CRC). Siaα(2,6)Galβ(1,4)GlcNAc sequence, known as CDw75, is a sialylated carbohydrate determinant generated by the ST6Gal I. This epitope has been reported to be associated with the progression of gastric and colorectal tumours, hence there are only a few conclusive studies to date.

Methods

By radioisotopic techniques we evaluated the ST6Gal I activity in healthy, transitional and tumour tissues from 43 patients with CRC. By immunohistochemistry we assessed the CDw75 expression in 25 colorectal adenomas, 43 tumours, 13 transitional and 28 healthy tissues of CRC patients.

Results

ST6Gal I activity was likewise found to be statistically higher in tumour tissue respect to healthy tissue from CRC patients. CDw75 expression was positive in 20% of colorectal adenomas. Furthermore, 70% of tumour specimens and 8.3% of transitional specimens were positive for CDw75 expression, whereas none of the healthy ones showed the presence of the epitope.

Conclusion

The major contribution of this study is the inclusion of data from transitional tissue and the analysis of CDw75 antigen expression in CRC and in colorectal adenomas, little known so far. ST6Gal I activity and CDw75 antigen expression were increased in CRC. Although their comparison did not reach the statistical significance, a great extent of patients showed both, an enhanced tumour ST6Gal I activity and an increased CDw75 expression in the tumour tissue. So, these two variables may play a role in malignant transformation. The expression of CDw75 in colorectal adenomas suggests that this antigen may be a tumour marker in CRC.

Cell surface beta 1, 4-galactosyltransferase 1 promotes apoptosis by inhibiting epidermal growth factor receptor pathway

Our previous studies have shown that overexpression of beta1,4-galactosyltransferase1 (beta1,4GT1) leads to increased apoptosis induced by cycloheximide (CHX) in SMMC-7721 human hepatocarcinoma cells. However, the role of beta1,4GT1 in apoptosis remains unclear. Here we demonstrated that cell surface beta1,4GT1 inhibited the autophosphorylation of epidermal growth factor receptor (EGFR) especially at Try 1068. The phosphorylation of protein kinase B (PKB/Akt) and extracellular signal-regulated protein kinase1/2 (ERK1/2), which are downstream molecules of EGFR, were also reduced in cell surface beta1,4GT1-overexpressing cells. Furthermore, the translocations of Bad and Bax that are regulated by PKB/Akt and ERK1/2 were also increased in these cells. As a result, the release of cytochrome c from mitochondria to cytosol was increased and caspase-3 was activated. In contrast, RNAi-mediated knockdown of beta1,4GT1 increased the autophosphorylation of EGFR. These results demonstrated that cell surface beta1,4GT1 may negatively regulate cell survival possibly through inhibiting and modulating EGFR signaling pathway.

An antibody produced in tobacco expressing a hybrid beta-1,4-galactosyltransferase is essentially devoid of plant carbohydrate epitopes

N-glycosylation of a mAb may have a major impact on its therapeutic merits. Here, we demonstrate that expression of a hybrid enzyme (called xylGalT), consisting of the N-terminal domain of Arabidopsis thaliana xylosyltransferase and the catalytic domain of human beta-1,4-galactosyltransferase I (GalT), in tobacco causes a sharp reduction of N-glycans with potentially immunogenic core-bound xylose (Xyl) and fucose (Fuc) residues as shown by Western blot and MALDI-TOF MS analysis. A radioallergosorbent test inhibition assay with proteins purified from leaves of WT and these transgenic tobacco plants using sera from allergic patients suggests a significant reduction of potential immunogenicity of xylGalT proteins. A mAb purified from leaves of plants expressing xylGalT displayed an N-glycan profile that featured high levels of galactose, undetectable xylose, and a trace of fucose. Hence, a transgenic plant expressing the hybrid GalT might yield more effective and safer monoclonals for therapeutic purposes than WT plants and even transgenic plants expressing the unchanged GalT.

Tumor beta-1,4-galactosyltransferase IV overexpression is closely associated with colorectal cancer metastasis and poor prognosis

PURPOSE

To elucidate the significance of beta-1,4-galactosyltransferase IV (beta-1,4-GT-IV) in the clinical presentation and prognostication of colorectal cancer.

EXPERIMENTAL DESIGN

Tissue lysates from paired tumor and nontumor tissues of a colon cancer patient were labeled separately with fluorescent dyes Cy5 and Cy3 for two-dimensional difference in-gel electrophoresis. Subsequent matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and immunoblot analyses identified a down-regulated level of beta-1,4-GT-IV in the tumor tissue. In the follow-up study, paired tissue lysates were obtained from 100 colorectal cancer patients with immunoblot analyses done to compare the levels of beta-1,4-GT-IV expression in these patients.

RESULTS

Of 100 colorectal patients studied, 48% had down-regulated expression of beta-1,4-GT-IV in the tumor tissue but 28% of patients exhibited elevated beta-1,4-GT-IV levels. Increased beta-1,4-GT-IV in the tumor tissue was significantly coexistent with raised serum level of CA-199 and the presence of tumor metastasis (P=0.006 and P<0.001, respectively) but was independent of age and gender of patient, tumor site, tumor size, serum level of carcinoembryonic antigen, grade of tumor cell differentiation, and depth of tumor invasion. The results of logistic regression analyses suggested that tumor beta-1,4-GT-IV overexpression and tumor invasion, but not other patient variables such as tumor size and serum levels of carcinoembryonic antigen and CA19-9, were significantly correlated with the occurrence of metastases (P<0.05). In a multivariate regression analysis, the patient group with tumor beta-1,4-GT-IV overexpression strongly predicted for tumor metastasis (odds ratio, 10.009; 95% confidence interval, 2.992-33.484; P<0.001). Likewise, tumor beta-1,4-GT-IV overexpression was significantly associated with poor overall survival (P<0.01). By Cox regression analysis, this association remained significant even after adjustment for tumor metastasis (P=0.048).

CONCLUSION

Increased beta-1,4-GT-IV expression in tumor tissue was strongly associated with tumor metastases and poor prognosis in colorectal cancer.

Multiple signals are required for alpha2,6-sialyltransferase (ST6Gal I) oligomerization and Golgi localization

A single amino acid difference in the catalytic domain of two isoforms of the alpha2,6-sialyltransferase (ST6Gal I) leads to differences in their trafficking, processing, and oligomerization. The STtyr isoform is transiently localized in the Golgi and is ultimately cleaved and secreted, whereas the STcys isoform is stably localized in the Golgi and is not cleaved and secreted. The stable localization of STcys is correlated with its enhanced ability to oligomerize. To test the hypothesis that multiple signals can mediate Golgi localization and further evaluate the role of oligomerization in the localization process, we evaluated the effects of individually and simultaneously altering the cytosolic tail and transmembrane region of the STcys isoform. We found that the localization, processing, and oligomerization of STcys were not substantially changed when either the core amino acids of the cytosolic tail were deleted or the sequence and length of the transmembrane region were altered. In contrast, when these changes were made simultaneously, the STcys isoform was converted into a form that was processed, secreted, and weakly oligomerized like STtyr. We propose that STcys oligomerization is a secondary event resulting from its concentration in the Golgi via mechanisms independently mediated by its cytosolic tail and transmembrane region.

Correlation analysis between tumor-associated antigen sialyl-Tn expression and ST6GalNAc I activity in human colon adenocarcinoma

OBJECTIVES

Sialyl-Tn (sTn) is a mucin carbohydrate-associated antigen that is strongly expressed in a large number of colorectal carcinomas. In this study, we combined immunohistochemical and enzymatic techniques in order to find the correlation between sTn tissue expression and the sialyltransferase activity (ST6GalNAc I) responsible for its synthesis in colorectal cancer (CRC) patients.

METHODS

We compared sTn expression in healthy (n = 46), tumorous (n = 60) and transitional tissue (n = 46) from CRC patients, and correlated sTn altered expression with clinicopathologic variables of the patient. Furthermore, we determined ST6GalNAc I tissue activity employing asialo-ovine submaxillary mucin (asialo-OSM) as glycoprotein acceptor (n = 27).

RESULTS

The rates of sTn positive expression obtained for healthy, tumorous and transitional tissues were 15, 67 and 63%, respectively. These rates led to statistically significant differences between healthy and tumorous or transitional tissue (p = 0.001); sTn expression was related to the first stages of the tumor invasion in transitional tissue. As regards ST6GalNAc I activity, we found an enhancement in transitional tissue. Statistical correlation analysis did not reveal association between sTn expression and ST6GalNAc I activity.

CONCLUSIONS

Our findings indicated that sTn antigen tissue expression and ST6GalNAc I activity levels were not correlated in CRC, in spite of the overexpression of the antigen in tumorous and transitional tissue.

Apoptosis and involution in the mammary gland are altered in mice lacking a novel receptor, beta1,4-Galactosyltransferase I

Receptor-mediated cell-extracellular matrix (ECM) interactions are critical regulators of cell survival, and perturbing these signaling pathways can disrupt cellular differentiation and function in a variety of tissues, including the mammary gland. One such receptor is the cell surface-associated, long isoform of beta1,4-galactosyltransferase I (GalT I). Deletion of long GalT I leads to increased mammary ductal branching morphogenesis [Dev. Biol., 244 (2002) 114]. Here, we show that this expansion in the mammary epithelial (ME) cell compartment is accomplished through decreased apoptosis during pregnancy and involution. Decreased apoptosis during involution is concomitant with delayed alveolar collapse, persistent expression of the milk protein gene alpha-lactalbumin and delayed expression of genes associated with the tissue-remodeling phase of involution. Using 3-dimensional in vitro cultures, we show that the decrease in apoptosis is dependent on laminin 1, a ligand for surface GalT I, suggesting that surface GalT I negatively influences ECM-dependent cell survival, a novel function for an ECM receptor. In the best-studied examples, ECM promotes survival through integrin receptor-mediated activation of focal adhesion kinase (FAK). Aggregation of surface GalT I also activates FAK, therefore, we asked if FAK activation was altered in ME from long GalT I null mice. Activated FAK was appropriately localized to focal adhesions in long GalT I null ME. However, FAK activation was constitutively reduced 4.5-fold in long GalT I nulls relative to wild type. Expression of the integrin beta1 subunit was not affected by loss of long GalT I. Collectively, these results suggest that surface GalT I might negatively regulate ME cell survival by linking integrin-independent FAK activation to apoptotic rather than survival signaling events.

Mutational analysis of the cytoplasmic domain of beta1,4-galactosyltransferase I: influence of phosphorylation on cell surface expression

Beta1,4-galactosyltransferase I (GalT I) exists in two subcellular compartments where it performs two distinct functions. The majority of GalT I is localized in the Golgi complex where it participates in glycoprotein biosynthesis; however, a small portion of GalT I is expressed on the cell surface where it functions as a matrix receptor by binding terminal N-acetylglucosamine residues on extracellular glycoside ligands. The GalT I polypeptide occurs in two alternate forms that differ only in the length of their cytoplasmic domains. It is thought that the longer cytoplasmic domain is responsible for GalT I function as a cell surface receptor because of its ability to associate with the detergent-insoluble cytoskeleton. In this study, we demonstrate that the long GalT I cytoplasmic and transmembrane domains are capable of targeting a reporter protein to the plasma membrane, whereas the short cytoplasmic and transmembrane domains do not have this property. The surface-localized GalT I reporter protein partitions with the detergent-insoluble pool, a portion of which co-fractionates with caveolin-containing lipid rafts. Site-directed mutagenesis of the cytoplasmic domain identified a requirement for serine and threonine residues for cell surface expression and function. Replacing either the serine or threonine with aspartic acid reduces surface expression and function, whereas substitution with neutral alanine has no effect on surface expression or function. These results suggest that phosphorylation negatively regulates GalT I function as a surface receptor. Consistent with this, phosphorylation of the endogenous, full-length GalT I inhibits its stable expression on the cell surface. Thus, the 13 amino acid extension unique to the long GalT I isoform is required for GalT I expression on the cell surface, the function of which is regulated by phosphorylation.

Ganglioside expression in tissues of mice lacking beta2-microglobulin

This study presents a comparative analysis of gangliosides from lymphoid (spleen and thymus) and other (brain, liver, lungs and muscle) tissues of C57BL/6 mice lacking the gene for beta2-microglobulin (beta2M), a constitutive component of the MHC class I molecule. Ganglioside fractions in the tissues of mice homozygous (beta2M-/-) and heterozygous (beta2M-/+) for the gene deletion were determined by high performance thin-layer chromatography (HPTLC), followed by immunostaining with specific polyclonal antibodies. Ubiquitous gangliosides GM3(Neu5Ac) and GM3(Neu5Gc) were the dominant gangliosides in the lungs of the control beta2M-/+ mice, whereas the homozygous knockout mice had substantially decreased expression of these structures. The lungs of the beta2M-/- mice also had reduced expression of T-lymphocyte-specific GM1b-type gangliosides (GM1b and GalNAc-GM1b). beta2M-deficient mice also had more GM1a and GD1a gangliosides in the liver, and several neolacto-series gangliosides were increased in the brain and lungs. This study provides in vivo evidence that the beta2M molecule can influence the acquisition of a distinct ganglioside assembly in different mouse organs, implicating its non-immunological functions.

Enhanced branching morphogenesis in mammary glands of mice lacking cell surface beta1,4-galactosyltransferase

Development of the mammary gland is influenced both by the systemic hormonal environment and locally through cell-cell and cell-extracellular matrix (ECM) interactions. We have previously demonstrated aberrant mammary gland morphogenesis in transgenic mice with elevated levels of the long isoform of beta1,4-galactosyltransferase 1 (GalT), a proportion of which is targeted to the plasma membrane, where it plays a role in cell-ECM interactions. Here, we show that mammary glands of mice lacking the long GalT isoform exhibit a complementary phenotype. Cell-surface GalT activity was reduced by over 60%, but because the short GalT isoform is intact, total GalT activity was reduced only slightly relative to wild type. Mammary glands from long GalT-null mice were characterized by excess branching, and this phenotype was accompanied by altered expression of laminin chains. Laminin alpha1 and alpha3 were reduced 2.4- and 3.0-fold, respectively, while expression of laminin gamma2 was elevated 2.3-fold. The expression and cleavage of laminin gamma2 have been correlated with branching and cell migration, and Western blotting revealed an altered pattern in gamma2 cleavage products in long GalT-null mammary glands. We then examined the expression of metalloproteases that cleave laminins or that have been shown to play a role in mammary gland morphogenesis. Expression of MT1-MMP, a membrane-bound protease that can cleave laminin gamma2, was elevated 5.5-fold in the long GalT-nulls. MMP 7 was also elevated 5.1-fold. Our results suggest that expression of surface GalT is important for the proper regulation of matrix expression and deposition, which in turn regulates the proper branching morphogenesis of the mammary epithelial ductal system.

Functional interaction between the SSeCKS scaffolding protein and the cytoplasmic domain of β1,4-galactosyltransferase

The β1,4-galactosyltransferase family contains at least seven unique gene products, of which β1,4-galactosyltransferase I (GalT) is the most exhaustively studied. GalT exists in the Golgi complex, similar to many other glycosyltransferases, as well as on the cell surface, where it functions as a signaling receptor for extracellular glycoside ligands. When expressed on the surface, GalT associates with the cytoskeleton and, upon ligand-induced aggregation, induces cell-type specific intracellular signal cascades. In an effort to define the mechanisms by which surface GalT exerts these intracellular effects, we used the yeast two-hybrid system to identify proteins that specifically interact with the GalT cytoplasmic domain.

The yeast two-hybrid screen identified two distinct clones (1.12 and 2.52) that showed identity to portions of SSeCKS (Src Suppressed CKinase Substrate). SSeCKS is a previously defined kinase and cytoskeleton scaffolding protein whose subcellular distribution and functions are remarkably similar to those attributed to GalT. Both SSeCKS and GalT have been localized to the perinuclear/Golgi region as well as to filopodia/lamellipodia. SSeCKS and GalT have been implicated in regulating cell growth, actin filament dynamics, and cell spreading. Interestingly, 1.12 and 2.52-GFP constructs were localized to subcellular domains that correlated with the two purported subcellular distributions for GalT; 2.52 being confined to the Golgi, whereas 1.12 localized primarily to filopodia. Coimmunoprecipitation assays demonstrate stable binding between the GalT cytoplasmic domain and the 1.12 and 2.52 domains of SSeCKS in appropriately transfected cells. Similar assays demonstrate binding between the endogenous GalT and SSeCKS proteins also. Coimmunoprecipitation assays were performed in both directions and produced similar results (i.e. using either anti-GalT domain or anti-SSeCKS domain antibodies as the precipitating reagent). A functional interaction between the GalT cytoplasmic domain and SSeCKS was illustrated by the ability of either the 1.12 or 2.52 SSeCKS domain to restore a normal adhesive phenotype in cells overexpressing the TL-GFP dominant negative construct. TL-GFP is composed of the GalT cytoplasmic and transmembrane domains fused to GFP, and leads to a loss of cell adhesion on laminin by displacing the endogenous GalT from its cytoskeleton binding sites. This is the first reported interaction between a glycosyltransferase and a scaffolding protein, and suggests that SSeCKS serve to integrate the various functions ascribed to the GalT cytoplasmic domain.

Cell surface beta-1,4-galactosyltransferase-I activates G protein-dependent exocytotic signaling

ZP3 is a protein in the mammalian egg coat (zona pellucida) that binds sperm and stimulates acrosomal exocytosis, enabling sperm to penetrate the zona pellucida. The nature of the ZP3 receptor/s on sperm is a matter of considerable debate, but most evidence suggests that ZP3 binds to beta-1,4-galactosyltransferase-I (GalTase) on the sperm surface. It has been suggested that ZP3 induces the acrosome reaction by crosslinking GalTase, activating a heterotrimeric G protein. In this regard, acrosomal exocytosis is sensitive to pertussis toxin and the GalTase cytoplasmic domain can precipitate G(i) from sperm lysates. Sperm from mice that overexpress GalTase bind more soluble ZP3 and show accelerated G protein activation, whereas sperm from mice with a targeted deletion in GalTase have markedly less ability to bind soluble ZP3, undergo the ZP3-induced acrosome reaction, and penetrate the zona pellucida. We have examined the ability of GalTase to function as a ZP3 receptor and to activate heterotrimeric G proteins using Xenopus laevis oocytes as a heterologous expression system. Oocytes that express GalTase bound ZP3 but did not bind other zona pellucida glycoproteins. After oocyte maturation, ZP3 or GalTase antibodies were able to trigger cortical granule exocytosis and activation of GalTase-expressing eggs. Pertussis toxin inhibited GalTase-induced egg activation. Consistent with G protein activation, both ZP3 and anti-GalTase antibodies increased GTP-gamma[(35)S] binding as well as GTPase activity in membranes from eggs expressing GalTase. Finally, mutagenesis of a putative G protein activation motif within the GalTase cytoplasmic domain eliminated G protein activation in response to ZP3 or anti-GalTase antibodies. These results demonstrate directly that GalTase functions as a ZP3 receptor and following aggregation, is capable of activating pertussis toxin-sensitive G proteins leading to exocytosis.

Can relative spermatozoal galactosyltransferase activity be predictive of dairy bull fertility?

The best and poorest bovine semen samples used commercially for artificial insemination in dairy cattle typically differ in pregnancy rates by 20 to 25% but are within a range that pregnancy rates cannot be predicted consistently by commonly used laboratory assays. Sperm motility and morphology are the characteristics most often evaluated. Laboratory assays that measure other functional traits of sperm may be useful as supplemental assays to increase the reliability of predicting fertility. One such functional trait is the ability of sperm to bind to the zona pellucida, a process mediated by complementary receptors on each gamete. On mouse sperm, beta1,4-galactosyltransferase acts as a receptor for the zona pellucida. Beta1,4-galactosyltransferase is expressed on sperm from many mammals, including bovine sperm, and is a candidate for a zona pellucida receptor. The ability of sperm to bind to the zona pellucida may be related to the amount of beta1,4-galactosyltransferase present on sperm. The aim of this work was to determine if bull sperm beta1,4-galactosyltransferase activity was related to fertility. Beta1,4-galactosyltransferase enzyme assays were performed on sperm from 24 bulls whose fertility was estimated by nonreturn rate and on sperm from a second group of seven bulls whose fertility was ranked by in vivo competitive fertilization. Beta1,4-galactosyltransferase activity varied between individual bulls but was not correlated to fertility as estimated by nonreturn rate or by competitive fertilization. These results demonstrate that beta1,4-galactosyltransferase activity on sperm varies between animals, but that beta1,4-galactosyltransferase activity alone is not an accurate indicator of fertility in dairy bulls.

Cloning and chromosomal mapping of human glucuronyltransferase involved in biosynthesis of the HNK-1 carbohydrate epitope

The HNK-1 carbohydrate is expressed on various cell adhesion molecules in the nervous system and is suggested to play a role in cell-cell and cell-substrate interactions. Here we describe the isolation of a cDNA encoding human glucuronyltransferase (GlcAT-P), which is a key enzyme in the biosynthesis of the HNK-1 carbohydrate. The primary structure deduced from the cDNA sequence predicted a type II transmembrane protein of 334 amino acids. Human GlcAT-P was 98.2% identical with rat GlcAT-P in amino acid sequence, the exception being the length of the cytoplasmic tail. Northern blot analysis indicated that human GlcAT-P is expressed mainly in the brain. There is a single copy of the human GlcAT-P gene (HGMW-approved symbol B3GAT1), and it was mapped to chromosome 11q25.

The effects of targeting the vaccinia virus B5R protein to the endoplasmic reticulum on virus morphogenesis and dissemination

The consequence of redirecting the vaccinia virus (VV) B5R protein to the endoplasmic reticulum (ER) has been investigated by the addition of an ER retrieval signal KKSL (K(2)X(2)) to the B5R C-terminus. This mutant B5R gene and a version of the gene with the inactive ER retrieval sequence KKSLAL (K(2)X(4)) were inserted into the thymidine kinase locus of a VV mutant lacking the B5R gene, vDeltaB5R. Similar levels of B5R protein were made by each virus, but the B5R-K(2)X(2) protein remained sensitive to endoglycosidase H and colocalised with protein disulphide isomerase in the ER. In contrast, the B5R-K(2)X(4) protein colocalised with 1, 4-galactosyltransferase in the trans-Golgi network. Electron microscopy revealed that even when the B5R protein was redirected to the ER, intracellular mature virus particles were wrapped by cellular membranes to form intracellular enveloped virus particles, although more incompletely wrapped particles were evident compared with wild type. These intracellular enveloped virus particles were, however, unable to efficiently induce the polymerisation of actin and the plaque size formed by vB5R-K(2)X(2) was small. Nevertheless, the amount and specific infectivity of EEV produced by vB5R-K(2)X(2) were similar to those of wild type, despite the dramatic reduction in the amount of B5R protein present in vB5R-K(2)X(2) EEV.

The level of cell surface beta1,4-galactosyltransferase I influences the invasive potential of murine melanoma cells

Beta1,4-Galactosyltransferase I (GalT I) is localized on the leading lamellipodia of migrating cells, where it associates with the cytoskeleton and facilitates cell spreading and migration on basal lamina matrices. It has previously been reported that a variety of highly metastatic murine and human cell lines are characterized by elevated levels of cell surface GalT I, although the intracellular biosynthetic pool is similar between cells of high and low metastatic potential. In this study, we examined whether the elevated expression of surface GalT I characteristic of metastatic cells is instructive or incidental to their metastatic behavior by altering the expression of surface GalT I and by the use of GalT I-specific perturbants. Surface GalT I levels were positively and negatively altered on murine melanoma cells by either overexpressing full-length GalT I or by homologous recombination, respectively. The consequences of altered surface GalT I expression on cell invasion in vitro and lung colonization in vivo were determined. Increasing surface GalT I expression on cells of low metastatic potential to levels characteristic of highly metastatic cells recapitulated the highly invasive phenotype in vitro. Alternatively, decreasing surface GalT I expression on highly metastatic cells to levels characteristic of low metastatic cells reduced their invasive behavior in vitro and metastatic activity in vivo. Within the physiological range of surface GalT I expression, the invasive potential of each clonal cell line correlated strongly with the level of surface GalT I expressed. As an independent means to assess the involvement of surface GalT I in metastatic behavior, cells were pretreated with two different classes of surface GalT I perturbants, a competitive oligosaccharide substrate and a substrate modifier protein. Both perturbants inhibited metastatic colonization of the lung, whereas control reagents did not. Finally, as reported by others, surface GalT I on metastatic cells selectively interacted with one glycoprotein substrate, or ligand, of approximately 100 kDa, the identity of which remains obscure. These results show that the elevated expression of surface GalT I characteristic of highly metastatic cells contributes to their invasive phenotype in vitro and to their metastatic phenotype in vivo.

Three proteins involved in Caenorhabditis elegans vulval invagination are similar to components of a glycosylation pathway

We have molecularly analyzed three genes, sqv-3, sqv-7, and sqv-8, that are required for wild-type vulval invagination in Caenorhabditis elegans. The predicted SQV-8 protein is similar in sequence to two mammalian beta(1,3)-glucuronyltransferases, one of which adds glucuronic acid to protein-linked galactose-beta(1, 4)-N-acetylglucosamine. SQV-3 is similar to a family of glycosyltransferases that includes vertebrate beta(1, 4)-galactosyltransferases, which create galactose-beta(1, 4)-N-acetylglucosamine linkages. One model is therefore that SQV-8 uses a SQV-3 product as a substrate. SQV-7 is similar to members of a family of nucleotide-sugar transporters. The sqv genes therefore are likely to encode components of a conserved glycosylation pathway that assembles a C. elegans carbohydrate moiety, the absence of which perturbs vulval invagination.

The role of glycosylation in flavonol-induced pollen germination

Flavonols are small (C15) plant-specific molecules that are required for petunia and maize pollen to germinate. They exist in two chemical forms: the aglycone or glycosyl conjugates. Flavonol-deficient pollen is biochemically complemented by flavonol aglycones but not by the glycosylated forms that accumulate in wild type (WT) pollen. Coincident with the biochemical induction of germination, the added flavonol aglycone is rapidly converted to a galactoside and then to a glucosyl galactoside (diglycoside) that is identical to the compound present in WT pollen. A flavonol 3-O-galactosyltransferase (F3GalTase) activity has been identified that controls the formation of glycosylated flavonols in pollen. Importantly, this enzyme also catalyzes the reverse reaction, i.e. the production of the flavonol aglycone from the galactoside and UDP (Fig. 1). F3GalTase/RevGalTase therefore has the potential to control the level of the bioactive flavonol species and as a result, pollen germination.

The chicken genome contains two functional nonallelic beta1,4-galactosyltransferase genes. Chromosomal assignment to syntenic regions tracks fate of the two gene lineages in the human genome

Two distinct but related groups of cDNA clones, CKbeta4GT-I and CKbeta4GT-II, have been isolated by screening a chicken hepatoma cDNA library with a bovine beta1,4-galactosyltransferase (beta4GT) cDNA clone. CKbeta4GT-I is predicted to encode a type II transmembrane glycoprotein of 41 kDa with one consensus site for N-linked glycosylation. CKbeta4GT-II is predicted to encode a type II transmembrane glycoprotein of 43 kDa with five potential N-linked glycosylation sites. At the amino acid level, the coding regions of CKbeta4GT-I and CKbeta4GT-II are 52% identical to each other and 62 and 49% identical, respectively, to bovine beta4GT. Despite this divergence in amino acid sequence, high levels of expression of each cDNA in Trichoplusia ni insect cells demonstrate that both CKbeta4GT-I and CKbeta4GT-II encode an alpha-lactalbumin-responsive, UDP-galactose:N-acetylglucosamine beta4-galactosyltransferase. An analysis of CKbeta4GT-I and CKbeta4GT-II genomic clones established that the intron positions within the coding region are conserved when compared with each other, and these positions are identical to the mouse and human beta4GT genes. Thus CKbeta4GT-I and CKbeta4GT-II are the result of the duplication of an ancestral gene and subsequent divergence. CKbeta4GT-I maps to chicken chromosome Z in a region of conserved synteny with the centromeric region of mouse chromosome 4 and human chromosome 9p, where beta4-galactosyltransferase (EC 2.4.1.38) had previously been mapped. Consequently, during the evolution of mammals, it is the CKbeta4GT-I gene lineage that has been recruited for the biosynthesis of lactose. CKbeta4GT-II maps to a region of chicken chromosome 8 that exhibits conserved synteny with human chromosome 1p. An inspection of the current human gene map of expressed sequence tags reveals that there is a gene noted to be highly similar to beta4GT located in this syntenic region on human chromosome 1p. Because both the CKbeta4GT-I and CKbeta4GT-II gene lineages are detectable in mammals, duplication of the ancestral beta4-galactosyltransferase gene occurred over 250 million years ago in an ancestral species common to both mammals and birds.

Sperm from beta 1,4-galactosyltransferase-null mice are refractory to ZP3-induced acrosome reactions and penetrate the zona pellucida poorly

A variety of sperm surface components have been suggested to mediate gamete recognition by binding to glycoside ligands on the egg coat glycoprotein ZP3. The function of each of these candidate receptors is based upon varying degrees of circumstantial and direct evidence; however, the effects on fertilization of targeted mutations in any of these candidate receptors have not yet been reported. In this paper, we describe the effects of targeted mutations in beta1,4-galactosyltransferase, the best studied of the candidate receptors for ZP3. Surprisingly, galactosyltransferase-null (gt[−/−]) males are fertile; however, sperm from gt(−/−) males bind less radiolabeled ZP3 than wild-type sperm, and are unable to undergo the acrosome reaction in response to either ZP3 or anti-galactosyltransferase antibodies, as do wild-type sperm. In contrast, gt(−/−) sperm undergo the acrosome reaction normally in response to calcium ionophore, which bypasses the requirement for ZP3 binding. The inability of gt(−/−) sperm to undergo a ZP3-induced acrosome reaction renders them physiologically inferior to wild-type sperm, as assayed by their relative inability to penetrate the egg coat and fertilize the oocyte in vitro. Thus, although ZP3 binding and subsequent induction of the acrosome reaction are dispensable for fertilization, they impart a physiological advantage to the fertilizing sperm. A second strain of mice was created that is characterized by a loss of of the long galactosyltransferase isoform responsible for ZP3-dependent signal transduction, but which maintains normal levels of Golgi galactosylation. Sperm from these mice show that the defective sperm-egg interactions in gt(−/−) mice are due directly to a loss of the long galactosyltransferase isoform from the sperm surface and are independent of the state of intracellular galactosylation during spermatogenesis.

Molecular cloning and characterization of beta-1,4-galactosyltransferase expressed in mouse testis

Using an affinity-purified antibody against beta-1,4-galactosyltransferase purified from F9 embryonal carcinoma cells, we screened a lambda gt11 expression library constructed from the same cell line. The cDNA clone obtained encoded a 46-kDa protein. Among adult mouse organs, the testis was found to express large amounts of the corresponding mRNA. An antibody against the 46-kDa protein was raised in a rabbit by immunization with a maltose-binding-protein fusion protein produced in Escherichia coli. The affinity-purified antibody against the cloned 46-kDa protein reacted with a 59-kDa protein in sperm extract on western blotting. Among the proteins in the F9 beta-1,4-galactosyltransferase preparation, which were mostly 68-kDa and 59-kDa species, 59-kDa and 50-kDa bands reacted with the antibody against the cloned 46-kDa protein. The cloned 46-kDa protein had type-II transmembrane topology and some blocks of sequence similarity with the known beta-1,4-galactosyltransferase. Furthermore, predicted secondary structures were similar over large portions of the two proteins. The histidine-tagged 46-kDa protein produced in E. coli was partially adsorbed onto a N-acetylglucosamine-Affi-Gel column and eluted by 20 mM N-acetylglucosamine. The histidine-tagged 46-kDa protein, which was purified to homogeneity, had a small, but significant amount of beta-1,4-galactosyltransferase activity. These results strongly suggest that the 46-kDa protein is a beta-1,4-galactosyltransferase.

Biochemical characterization and intracellular localization of the Menkes disease protein

Menkes disease is a fatal neurodegenerative disorder of childhood due to the absence or dysfunction of a putative copper-transporting P-type ATPase encoded on the X chromosome. To elucidate the biosynthesis and subcellular localization of this protein, polyclonal antisera were generated against a bacterial fusion protein encoding the 4th to 6th copper-binding domains in the amino terminus of the human Menkes protein. RNA blot analysis revealed abundant Menkes gene expression in several cell lines, and immunoblotting studies utilizing this antiserum readily detected a 178-kDa protein in lysates from these cells. Pulse-chase studies indicate that this protein is synthesized as a single-chain polypeptide which is modified by N-linked glycosylation to a mature endoglycosidase H-resistant form. Sucrose gradient fractionation of HeLa cell lysates followed by immunoblotting of individual fractions with antibodies to proteins of known intracellular location identified the Menkes ATPase in fractions similar to those containing the cation-independent mannose-6-phosphate receptor. Consistent with this observation, confocal immunofluorescence studies of these same cells localized this protein to the trans-Golgi network and a vesicular compartment with no expression in the nucleus or on the plasma membrane. Taken together, these data provide a unique model of copper transport into the secretory pathway of mammalian cells which is compatible with clinical observations in affected patients and with recent data on homologous proteins identified in prokaryotes and yeast.

Mammary gland morphogenesis is inhibited in transgenic mice that overexpress cell surface beta1,4-galactosyltransferase

Mammary gland morphogenesis is facilitated by a precise sequence of cell-cell and cell-matrix interactions, which are mediated in part through a variety of cell surface receptors and their ligands (Boudreau, N., Myers, C. and Bissell, M. J. (1995). Trends in Cell Biology 5, 1–4). Cell surface beta1,4-galactosyltransferase (GalTase) is one receptor that participates in a variety of cell-cell and cell-matrix interactions during fertilization and development, including mammary epithelial cell-matrix interactions (Barcellos-Hoff, M. H. (1992). Exp. Cell Res. 201, 225–234). To analyze GalTase function during mammary gland morphogenesis in vivo, we created transgenic animals that overexpress the long isoform of GalTase under the control of a heterologous promoter. As expected, mammary epithelial cells from transgenic animals had 2.3 times more GalTase activity on their cell surface than did wild-type cells. Homozygous transgenic females from multiple independent lines failed to lactate, whereas transgenic mice overexpressing the Golgi-localized short isoform of GalTase lactated normally. Glands from transgenic females overexpressing surface GalTase were characterized by abnormal and reduced ductal development with a concomitant reduction in alveolar expansion during pregnancy. The phenotype was not due to a defect in proliferation, since the mitotic index for transgenic and wild-type glands was similar. Morphological changes were accompanied by a dramatic reduction in the expression of milk-specific proteins. Immunohistochemical markers for epithelia and myoepithelia demonstrated that both cell types were present. To better understand how overexpression of surface GalTase impairs ductal morphogenesis, primary mammary epithelial cultures were established on basement membranes. Cultures derived from transgenic mammary glands were unable to form anastomosing networks of epithelial cells and failed to express milk-specific proteins, unlike wild-type mammary cultures that formed epithelial tubules and expressed milk proteins. Our results suggest that cell surface GalTase is an important mediator of mammary cell interaction with the extracellular matrix. Furthermore, perturbing surface GalTase levels inhibits the expression of mammary-specific gene products, implicating GalTase as a component of a receptor-mediated signal transduction pathway required for normal mammary gland differentiation.

Activation of a G protein complex by aggregation of beta-1,4-galactosyltransferase on the surface of sperm

Fertilization is initiated by the species-specific binding of sperm to the extracellular coat of the egg. One sperm receptor for the mouse egg is beta-1,4-galactosyltransferase (GalTase), which binds O-linked oligosaccharides on the egg coat glycoprotein ZP3. ZP3 binding induces acrosomal exocytosis through the activation of a pertussis toxin-sensitive heterotrimeric guanine nucleotide-binding protein (G protein). The cytoplasmic domain of sperm surface GalTase bound to and activated a heterotrimeric G protein complex that contained the Gi alpha subunit. Aggregation of GalTase by multivalent ligands elicited G protein activation. Sperm from transgenic mice that overexpressed GalTase had higher rates of G protein activation than did wild-type sperm, which rendered transgenic sperm hypersensitive to their ZP3 ligand. Thus, the cytoplasmic domain of cell surface GalTase appears to enable it to function as a signal-transducing receptor for extracellular oligosaccharide ligands.

Biological consequences of targeting beta 1,4-galactosyltransferase to two different subcellular compartments

beta 1,4-galactosyltransferase is unusual among the glycosyltransferases in that it is found in two subcellular compartments where it performs two distinct functions. In the trans-Golgi complex, galactosyltransferase participates in oligosaccharide biosynthesis, as do the other glycosyltransferases. On the cell surface, however, galactosyltransferase associates with the cytoskeleton and functions as a receptor for extracellular oligosaccharide ligands. Although we now know much regarding galactosyltransferase function in these two compartments, little is known about how it is targeted to these different sites. By cloning the galactosyltransferase gene products, certain features of the protein have been identified that may be critical for its expression on the cell surface or retention within the Golgi complex. This article discusses recent studies which suggest that a cytoplasmic sequence unique to one galactosyltransferase isoform is required for targeting a portion of this protein to the plasma membrane, enabling it to function as a cell adhesion molecule. These findings allow one to manipulate surface galactosyltransferase expression, either positively or negatively, and perturb galactosyltransferase-dependent cellular interactions during fertilization and development.

Overexpressing cell surface beta 1.4-galactosyltransferase in PC12 cells increases neurite outgrowth on laminin

Neurite outgrowth on cellular and extracellular matrices is mediated by a variety of cell surface receptors. Some of these receptors recognize peptide determinants, whereas others bind oligosaccharide ligands. Previous studies have suggested that cell surface beta 1.4-galactosyltransferase functions as one of these receptors during neurite outgrowth on basal lamina by binding to N-linked oligosaccharides in the E8 domain of laminin. However, these previous investigations have been limited to the use of galactosyltransferase inhibitory reagents to block neurite formation. Therefore, in this study, we investigated whether the level of surface galactosyltransferase directly affects the efficiency of neurite outgrowth, or rather, is incidental to neurite formation. Northern blot analysis and cell surface galactosyltransferase assays were used to select two stable PC12 transfectants that overexpress surface galactosyltransferase by approximately four-fold. Radiolabeled antibody binding to intact cells and indirect immunofluorescence confirmed the higher expression of surface galactosyltransferase on transfected cells, compared to controls. Both galactosyltransferase transfected cell lines exhibited markedly enhanced neurite initiation, neurite formation, and rates of neurite elongation by two- to three-fold. These studies demonstrate that the expression of laminin receptors can be rate-limiting during neurite outgrowth, and that the level of surface galactosyltransferase can modulate the frequency and rate of neurite formation from PC12 cells on laminin.

Alteration of oligosaccharide biosynthesis by genetic manipulation of glycosyltransferases

The alteration of oligosaccharide structures through genetic manipulation of glycosyltransferase activities is now a reality. It is apparent that this technique has greater consequences on oligosaccharide structure when an exogenous enzyme is introduced into cells, and in particular when this enzyme is responsible for a terminal glycosylation step. By contrast, only one study has examined the effects of overexpressing an endogenous glycosyltransferase, in which there was no detectable effect on glycosylation. However, there are still other key regulatory biosynthetic enzymes, such as GlcNAc transferase V and beta 1,3 GlcNAc transferase, whose overexpression may alter glycosylation. Both of these enzymes are required for the biosynthesis of polylactosaminoglycans (polymers of N-acetyllactosamine disaccharides), and their elevation in tumor cells correlates with increased expression of polylactosaminoglycans. Recently, the gene encoding GlcNAc transferase V has been isolated, but its transfection into cells and characterization of the resulting oligosaccharides awaits further study. Alternate strategies for modifying oligosaccharide structures could involve the introduction of more than one glycosyltransferase into cells to ensure the availability of biosynthetic intermediates. Alternatively, the disruption of specific glycosyltransferase genes by homologous recombination could be used to eliminate competing glycosyltransferases that act on a common substrate. Although oligosaccharide biosynthesis is directly dependent upon the presence or absence of specific glycosyltransferases, other factors also contribute to glycosylation. For example, the transport rate of a glycoprotein through the endoplasmic reticulum and Golgi complex, the levels of processing glycosidases, the availability of substrates, the host cell, and ultimately, the peptide backbone of the particular glycoprotein of interest are important contributors to the final outcome of oligosaccharide structure. Despite these complications, further study into the manipulation of glycosyltransferase genes may ultimately allow the controlled and predictable biosynthesis of glycoprotein sugar chains.

Control of stable lamellipodia formation by expression of cell surface beta 1,4-galactosyltransferase cytoplasmic domains

Mesenchymal cell migration on basal lamina is mediated, in part, by the binding of cell surface beta 1,4-galactosyltransferase (GalTase) to specific N-linked oligosaccharides in the E8 domain of laminin. On migrating cells, surface GalTase is anchored to the cytoskeleton; when GalTase is prevented from associating with the cytoskeleton, lamellipodia formation and subsequent migration are inhibited. To define better the involvement of GalTase-cytoskeleton interactions in cell motility, we examined the lamellipodia formation, polarity and migratory behavior of stably transfected 3T3 fibroblasts expressing increased or decreased levels of GalTase capable of interacting with the cytoskeleton. Initially, the motile behavior of individual cells was quantified in the absence of exogenous stimuli. Cells that overexpress GalTase binding sites for the cytoskeleton changed their polarity more frequently and translocated more erratically than did control cells when assayed on laminin substrata. These differences were not observed, however, when cells were plated on fibronectin, which does not contain binding sites for surface GalTase. GalTase-transfected cells were also assayed for their ability to polarize in response to a specific stimulus. In this case, the ability of a cell to reorient towards a gradient of platelet-derived growth factor was found to be directly proportional to the amount of GalTase associated with the cytoskeleton. Differences in response to platelet-derived growth factor were not due to differences in growth factor binding. Indirect immunofluorescence showed that altering the level of GalTase did not affect the ventrally distributed pool of GalTase stably associated with the cytoskeleton; however, stress fiber formation was inhibited. Thus, increasing surface GalTase binding sites for the cytoskeleton leads to erratic, multipolar behavior in the absence of any vectorial stimulus, but the ability to form a functional lamellipodium in response to a stimulus is dependent upon the amount of surface GalTase associated with the cytoskeleton. Apparently, cells are able to regulate cytoskeletal assembly and lamellipodial stability by altering the expression and/or affinity of appropriate matrix receptors, such as GalTase, and their corresponding binding sites in the cytoskeleton.

Perturbing cell surface beta-(1,4)-galactosyltransferase on F9 embryonal carcinoma cells arrests cell growth and induces laminin synthesis

Cell growth and differentiation are influenced by intercellular contact, suggesting that cell adhesion molecules may be instrumental in triggering these events. F9 embryonal carcinoma cells are an ideal system in which to examine the function of cell adhesion molecules in growth and differentiation, since the relevant cell adhesion molecules and differentiation markers are well defined. Intercellular adhesion in F9 cells is mediated by uvomorulin, or E-cadherin, and cell surface beta-(1,4)-galactosyltransferase. Since previous studies suggested that neither F9 cell growth nor differentiation is directly dependent on uvomorulin function, in this study we examined whether cell surface galactosyltransferase plays any role in F9 cell growth or differentiation. A variety of galactosyltransferase perturbants, including anti-galactosyltransferase antibodies, UDPgalactose, and the substrate modifier protein alpha-lactalbumin, inhibited the growth of F9 cells, whereas control reagents did not. To examine this in more detail, we analyzed the effects of perturbing surface galactosyltransferase on progression through the F9 cell cycle. Anti-galactosyltransferase IgG treatment inhibited ornithine decarboxylase activity and lengthened the F9 cell cycle during G1 and G2, the latter mimicking the effects of retinoic acid, a reagent known to prolong the F9 cell cycle and induce differentiation. In contrast, anti-uvomorulin antibodies had no effect on F9 cell growth, ornithine decarboxylase activity, or progression through the cell cycle. Furthermore, perturbation of surface galactosyltransferase adhesions in F9 cell aggregates induced precocious F9 cell differentiation, as assayed by increased laminin synthesis, whereas control reagents had no effect. Thus, perturbing surface galactosyltransferase adhesions in F9 cells both decreases growth and stimulates synthesis of laminin. These results imply that interactions between surface galactosyltransferase and its oligosaccharide ligand during cell adhesion may affect the normal growth-regulatory and differentiation-inducing signals, as is seen, in part, during treatment with retinoic acid.

Proteins of the Golgi apparatus. Purification to homogeneity, N-terminal sequence, and unusually large Stokes radius of the membrane-bound form of UDP-galactose:N-acetylglucosamine beta 1-4galactosyltransferase from rat liver

The Golgi marker enzyme, UDP-galactose:N-acetylglucosamine beta 1-4galactosyltransferase (beta 1-4GalT) was purified 44300-fold in its intact, membrane-bound form from rat liver membranes. The protein was isolated from detergent extracts as a high-M(r) form, having a Stokes radius approximating a globular protein of M(r) 440,000. It is comprised of a single protein component as observed on SDS/polyacrylamide gels, having an M(r) near 51,000, and does not have intermolecular disulfide cross-links. N-terminal sequencing of the enzyme demonstrated that it contains an N-terminal hydrophobic stretch deduced previously from cDNA encoding for the enzyme. Previous studies have indicated that the protein may be translated at either of two AUG sites near the 5' end of the mRNA [Russo, R. N., Shaper, N. L. & Shaper, J. H. (1990) J. Biol. Chem. 265, 3324-3331], giving rise to two polypeptides, one appended with 13 amino acids. In the work described here, evidence was only found for the sequence of the short form, missing a single methionine at the N-terminus. Mild proteolytic treatment cleaved the enzyme, giving rise to low-M(r) forms which were fully catalytically active and which, upon sequencing, were missing a 66-amino-acid stretch from the N-terminus (as compared to the mouse cDNA). Proteolytic treatment was accompanied by conversion of the form having a large Stokes radius to one approximating a globular protein with M(r) near 50,000. The N-terminal stretch appears to contribute to maintenance of the form having a large Stokes radius. This may be the result of interaction with a detergent micelle, dimerization or oligomerization, or interaction with some other large, non-protein molecule, although a detergent exchange still resulted in a form having a large Stokes radius.