The critical role of the stem region as a functional domain responsible for the oligomerization and Golgi localization of N-acetylglucosaminyltransferase V. The involvement of a domain homophilic interaction

Structure and function of N-acetylglucosaminyltransferase V (GnT-V)

BACKGROUND

The β1,6-GlcNAc branch in N-glycans, produced by a glycosyltransferase N-acetylglucosaminyltransferase V (GnT-V or MGAT5), is associated with cancer and autoimmune diseases.

SCOPE

Here, we summarize the structure and activity regulation of GnT-V. We also describe the roles of the β1,6-GlcNAc branch on glycoproteins in cells and the phenotypes of Mgat5-deficient mice, focusing on cancer and the immune system.

MAJOR CONCLUSIONS

GnT-V has a unique structure for substrate recognition, and its activity and function are regulated by shedding. The glycans produced by GnT-V play pivotal roles in the differentiation of neural cells, cancer malignancy and immunotherapy, and the development of autoimmune diseases by regulating the functions and cell surface residency of glycoproteins.

GENERAL SIGNIFICANCE

Controlling the expression or activity of GnT-V could be a therapeutic option against cancer and autoimmune diseases. Future work should clarify how GnT-V selectively modifies the specific glycoproteins or N-glycosylation sites in vivo.

The stem region of group A transferase is crucial for its specificity, and its alteration promotes heterologous Forssman synthase activity

Some stem region mutants of human blood group A transferase (hAT) possess Forssman synthase (FS) activity, but very little is known about the mechanisms responsible for this enzymatic crosstalk. We performed confocal microscopy and image analysis to determine whether different intra-Golgi localization was accountable for this acquired activity. We also performed structural modeling and mutational and normal mode analyses. We introduced new mutations in the stem region and tested its FS and AT activities. No differences in subcellular localization were found between hAT and FS-positive mutants. AlphaFold models of hAT and mFS (mouse Forssman synthase) showed that the hAT stem region has a tether-like stem region, while in mFS, it encircles its catalytic domain. In silico analysis of FS-positive mutants indicated that stem region mutations induced structural changes, decreasing interatomic interactions and mobility of hAT that correlated with FS activity. Several additional mutations introduced in that region also bestowed FS activity without altering the AT activity: hAT 37-55 aa substitution by mFS 34-52, 37-55 aa deletion, and missense mutations: S46P, Q278Y, and Q286M. Stem region structure, mobility, and interactions are crucial for hAT specificity. Moreover, stem region mutations can lead to heterologous Forssman activity without changes in the catalytic machinery.

Structural basis for matriglycan synthesis by the LARGE1 dual glycosyltransferase

LARGE1 is a bifunctional glycosyltransferase responsible for generating a long linear polysaccharide termed matriglycan that links the cytoskeleton and the extracellular matrix and is required for proper muscle function. This matriglycan polymer is made with an alternating pattern of xylose and glucuronic acid monomers. Mutations in the LARGE1 gene have been shown to cause life-threatening dystroglycanopathies through the inhibition of matriglycan synthesis. Despite its major role in muscle maintenance, the structure of the LARGE1 enzyme and how it assembles in the Golgi are unknown. Here we present the structure of LARGE1, obtained by a combination of X-ray crystallography and single-particle cryo-EM. We found that LARGE1 homo-dimerizes in a configuration that is dictated by its coiled-coil stem domain. The structure shows that this enzyme has two canonical GT-A folds within each of its catalytic domains. In the context of its dimeric structure, the two types of catalytic domains are brought into close proximity from opposing monomers to allow efficient shuttling of the substrates between the two domains. Together, with putative retention of matriglycan by electrostatic interactions, this dimeric organization offers a possible mechanism for the ability of LARGE1 to synthesize long matriglycan chains. The structural information further reveals the mechanisms in which disease-causing mutations disrupt the activity of LARGE1. Collectively, these data shed light on how matriglycan is synthesized alongside the functional significance of glycosyltransferase oligomerization.

N-acetylglucosaminyltransferase-V (GnT-V)-enriched small extracellular vesicles mediate N-glycan remodeling in recipient cells

Small extracellular vesicles (sEVs) secreted from cancer cells play pivotal roles in cancer metastasis and malignancy by transferring biomolecules and conditioning future metastatic sites. Studies have elucidated structures and functions of glycans on sEVs; however, whether sEVs remodel glycans in recipient cells remains poorly understood. Here, we examined the enzyme activity of glycosyltransferases for complex N-glycan biosynthesis in cancer-derived sEVs and discovered that cancer-related glycosyltransferase, N-acetylglucosaminyltransferase-V (GnT-V, a.k.a. MGAT5), is selectively enriched in sEVs among various glycosyltransferases. GnT-V in sEVs is a cleaved form, and cleavage by SPPL3 protease is necessary for loading GnT-V in sEVs. Fractionation experiments and single-particle imaging further revealed that GnT-V was enriched in non-exosomal sEVs. Strikingly, we found that enzymatically active GnT-V in sEVs was transferred to recipient cells and the N-glycan structures of recipient cells were remodeled to express GnT-V-produced glycans. Our results suggest GnT-V-enriched sEVs' role in glycan remodeling in cancer metastasis.

Shedding of N-acetylglucosaminyltransferase-V is regulated by maturity of cellular N-glycan

The number of N-glycan branches on glycoproteins is closely related to the development and aggravation of various diseases. Dysregulated formation of the branch produced by N-acetylglucosaminyltransferase-V (GnT-V, also called as MGAT5) promotes cancer growth and malignancy. However, it is largely unknown how the activity of GnT-V in cells is regulated. Here, we discover that the activity of GnT-V in cells is selectively upregulated by changing cellular N-glycans from mature to immature forms. Our glycomic analysis further shows that loss of terminal modifications of N-glycans resulted in an increase in the amount of the GnT-V-produced branch. Mechanistically, shedding (cleavage and extracellular secretion) of GnT-V mediated by signal peptide peptidase-like 3 (SPPL3) protease is greatly inhibited by blocking maturation of cellular N-glycans, resulting in an increased level of GnT-V protein in cells. Alteration of cellular N-glycans hardly impairs expression or localization of SPPL3; instead, SPPL3-mediated shedding of GnT-V is shown to be regulated by N-glycans on GnT-V, suggesting that the level of GnT-V cleavage is regulated by its own N-glycan structures. These findings shed light on a mechanism of secretion-based regulation of GnT-V activity.

Cleavage of the glycan-branching enzyme N-acetylglucosaminyltransferase-V (GnT-V) by signal peptide peptidase-like 3 (SPPL3) protease and extracellular secretion of active glycan GnT-V depend on GnT-V’s own glycosylation state.

Glyco-Engineering Plants to Produce Helminth Glycoproteins as Prospective Biopharmaceuticals: Recent Advances, Challenges and Future Prospects

Glycoproteins are the dominant category among approved biopharmaceuticals, indicating their importance as therapeutic proteins. Glycoproteins are decorated with carbohydrate structures (or glycans) in a process called glycosylation. Glycosylation is a post-translational modification that is present in all kingdoms of life, albeit with differences in core modifications, terminal glycan structures, and incorporation of different sugar residues. Glycans play pivotal roles in many biological processes and can impact the efficacy of therapeutic glycoproteins. The majority of biopharmaceuticals are based on human glycoproteins, but non-human glycoproteins, originating from for instance parasitic worms (helminths), form an untapped pool of potential therapeutics for immune-related diseases and vaccine candidates. The production of sufficient quantities of correctly glycosylated putative therapeutic helminth proteins is often challenging and requires extensive engineering of the glycosylation pathway. Therefore, a flexible glycoprotein production system is required that allows straightforward introduction of heterologous glycosylation machinery composed of glycosyltransferases and glycosidases to obtain desired glycan structures. The glycome of plants creates an ideal starting point for N- and O-glyco-engineering of helminth glycans. Plants are also tolerant toward the introduction of heterologous glycosylation enzymes as well as the obtained glycans. Thus, a potent production platform emerges that enables the production of recombinant helminth proteins with unusual glycans. In this review, we discuss recent advances in plant glyco-engineering of potentially therapeutic helminth glycoproteins, challenges and their future prospects.

N-acetylglucosaminyltransferase-V requires a specific noncatalytic luminal domain for its activity toward glycoprotein substrates

N-acetylglucosaminyltransferase-V (GnT-V or MGAT5) catalyzes the formation of an N-glycan β1,6-GlcNAc branch on selective target proteins in the Golgi apparatus and is involved in cancer malignancy and autoimmune disease etiology. Several three-dimensional structures of GnT-V were recently solved, and the recognition mechanism of the oligosaccharide substrate was clarified. However, it is still unclear how GnT-V selectively acts on glycoprotein substrates. In this study, we focused on an uncharacterized domain at the N-terminal side of the luminal region (N domain) of GnT-V, which was previously identified in a crystal structure, and aimed to reveal its role in GnT-V action. Using lectin blotting and fluorescence assisted cell sorting analysis, we found that a GnT-VΔN mutant lacking the N domain showed impaired biosynthetic activity in cells, indicating that the N domain is required for efficient glycosylation. To clarify this mechanism, we measured the in vitro activity of purified GnT-VΔN toward various kinds of substrates (oligosaccharide, glycohexapeptide, and glycoprotein) using HPLC and a UDP-Glo assay. Surprisingly, GnT-VΔN showed substantially reduced activity toward the glycoprotein substrates, whereas it almost fully maintained its activity toward the oligosaccharides and the glycopeptide substrates. Finally, docking models of GnT-V with substrate glycoproteins suggested that the N domain could interact with the substrate polypeptide directly. Our findings suggest that the N domain of GnT-V plays a critical role in the recognition of glycoprotein substrates, providing new insights into the mechanism of substrate-selective biosynthesis of N-glycans.

Structures and mechanism of human glycosyltransferase β1,3-N-acetylglucosaminyltransferase 2 (B3GNT2), an important player in immune homeostasis

β1,3-N-acetylglucosaminyltransferases (B3GNTs) are Golgi-resident glycosyltransferases involved in the biosynthesis of poly-N-acetyl-lactosamine chains. They catalyze the addition of the N-acetylglucosamine to the N-acetyl-lactosamine repeat as a key step of the chain elongation process. Poly-N-acetyl-lactosamine is involved in the immune system in many ways. Particularly, its long chain has been demonstrated to suppress excessive immune responses. Among the characterized B3GNTs, B3GNT2 is the major poly-N-acetyl-lactosamine synthase, and deletion of its coding gene dramatically reduced the cell surface poly-N-acetyl-lactosamine and led to hypersensitive and hyperresponsive immunocytes. Despite the extensive functional studies, no structural information is available to understand the molecular mechanism of B3GNT2, as well as other B3GNTs. Here we present the structural and kinetic studies of the human B3GNT2. Five crystal structures of B3GNT2 have been determined in the unliganded, donor substrate-bound, acceptor substrate-bound, and product(s)-bound states at resolutions ranging from 1.85 to 2.35 Å. Kinetic study shows that the transglycosylation reaction follows a sequential mechanism. Critical residues involved in recognition of both donor and acceptor substrates as well as catalysis are identified. Mutations of these invariant residues impair B3GNT2 activity in cell assays. Structural comparison with other glycosyltransferases such as mouse Fringe reveals a novel N-terminal helical domain of B3GNTs that may stabilize the catalytic domain and distinguish among different acceptor substrates.

Recognition of glycan and protein substrates by N-acetylglucosaminyltransferase-V

BACKGROUND

N-Glycosylation is crucial for protein folding, trafficking, and functions. N-Glycans have a different number of N-acetylglucosamine (GlcNAc) branches in a protein-selective manner, and the β1,6-linked GlcNAc branch on specific proteins produced by N-acetylglucosaminyltransferase-V (GnT-V or MGAT5) promotes cancer malignancy. However, little is known about how GnT-V acts on specific target proteins.

METHODS

Based on our structural model, we hypothesized that GnT-V interacts with the N-glycan core or polypeptide moiety as well as the accepter site of N-glycan. To explore this possibility, we selected four candidate residues involved in the interaction with the glycan core or surrounding amino acids, created point mutants of these residues, and examined the in vitro and in vivo activities of the mutants.

RESULTS

Our in vitro enzyme assays using various types of substrates including oligosaccharides and glycoproteins revealed that the V354N mutant had dramatically reduced activity for all tested substrates with an altered substrate preference and that K361A had reduced activity for an oligosaccharide with asparagine (Asn), but not a shorter oligosaccharide without the reducing end of GlcNAc and Asn. These results suggest that V354 and K361 are involved in the recognition of N-glycan core and surrounding amino acids. We further performed rescue experiments using GnT-V knockout HeLa cells and confirmed the importance of these residues for modifications of glycoproteins in cells.

CONCLUSIONS

We identified several residues involved in the action of GnT-V toward N-glycan cores and surrounding amino acids.

GENERAL SIGNIFICANCE

Our data provide new insights into how GnT-V recognizes glycoproteins.

Structure and mechanism of cancer-associated N-acetylglucosaminyltransferase-V

N-acetylglucosaminyltransferase-V (GnT-V) alters the structure of specific N-glycans by modifying α1-6-linked mannose with a β1-6-linked N-acetylglucosamine branch. β1-6 branch formation on cell surface receptors accelerates cancer metastasis, making GnT-V a promising target for drug development. However, the molecular basis of GnT-V's catalytic mechanism and substrate specificity are not fully understood. Here, we report crystal structures of human GnT-V luminal domain with a substrate analog. GnT-V luminal domain is composed of a GT-B fold and two accessary domains. Interestingly, two aromatic rings sandwich the α1-6 branch of the acceptor N-glycan and restrain the global conformation, partly explaining the fine branch specificity of GnT-V. In addition, interaction of the substrate N-glycoprotein with GnT-V likely contributes to protein-selective and site-specific glycan modification. In summary, the acceptor-GnT-V complex structure suggests a catalytic mechanism, explains the previously observed inhibition of GnT-V by branching enzyme GnT-III, and provides a basis for the rational design of drugs targeting N-glycan branching.

Ganglioside glycosyltransferases are S-acylated at conserved cysteine residues involved in homodimerisation

Ganglioside glycosyltransferases (GGTs) are type II membrane proteins bearing a short N-terminal cytoplasmic tail, a transmembrane domain (TMD), and a lumenal catalytic domain. The expression and activity of these enzymes largely determine the quality of the glycolipids that decorate mammalian cell membranes. Many glycosyltransferases (GTs) are themselves glycosylated, and this is important for their proper localisation, but few if any other post-translational modifications of these proteins have been reported. Here, we show that the GGTs, ST3Gal-V, ST8Sia-I, and β4GalNAcT-I are S-acylated at conserved cysteine residues located close to the cytoplasmic border of their TMDs. ST3Gal-II, a GT that sialylates glycolipids and glycoproteins, is also S-acylated at a conserved cysteine located in the N-terminal cytoplasmic tail. Many other GTs also possess cysteine residues in their cytoplasmic regions, suggesting that this modification occurs also on these GTs. S-acylation, commonly known as palmitoylation, is catalysed by a family of palmitoyltransferases (PATs) that are mostly localised at the Golgi complex but also at the endoplasmic reticulum (ER) and the plasma membrane. Using GT ER retention mutants, we found that S-acylation of β4GalNAcT-I and ST3Gal-II takes place at different compartments, suggesting that these enzymes are not substrates of the same PAT. Finally, we found that cysteines that are the target of S-acylation on β4GalNAcT-I and ST3Gal-II are involved in the formation of homodimers through disulphide bonds. We observed an increase in ST3Gal-II dimers in the presence of the PAT inhibitor 2-bromopalmitate, suggesting that GT homodimerisation may be regulating S-acylation

Glycosyltransferase complexes in eukaryotes: long-known, prevalent but still unrecognized

Glycosylation is the most common and complex cellular modification of proteins and lipids. It is critical for multicellular life and its abrogation often leads to a devastating disease. Yet, the underlying mechanistic details of glycosylation in both health and disease remain unclear. Partly, this is due to the complexity and dynamicity of glycan modifications, and the fact that not all the players are taken into account. Since late 1960s, a vast number of studies have demonstrated that glycosyltransferases typically form homomeric and heteromeric complexes with each other in yeast, plant and animal cells. To propagate their acceptance, we will summarize here accumulated data for their prevalence and potential functional importance for glycosylation focusing mainly on their mutual interactions, the protein domains mediating these interactions, and enzymatic activity changes that occur upon complex formation. Finally, we will highlight the few existing 3D structures of these enzyme complexes to pinpoint their individual nature and to emphasize that their lack is the main obstacle for more detailed understanding of how these enzyme complexes interact and function in a eukaryotic cell.

In situ structural analysis of Golgi intracisternal protein arrays

We acquired molecular-resolution structures of the Golgi within its native cellular environment. Vitreous Chlamydomonas cells were thinned by cryo-focused ion beam milling and then visualized by cryo-electron tomography. These tomograms revealed structures within the Golgi cisternae that have not been seen before. Narrow trans-Golgi lumina were spanned by asymmetric membrane-associated protein arrays that had ∼6-nm lateral periodicity. Subtomogram averaging showed that the arrays may determine the narrow central spacing of the trans-Golgi cisternae through zipper-like interactions, thereby forcing cargo to the trans-Golgi periphery. Additionally, we observed dense granular aggregates within cisternae and intracisternal filament bundles associated with trans-Golgi buds. These native in situ structures provide new molecular insights into Golgi architecture and function.

UDP-galactose (SLC35A2) and UDP-N-acetylglucosamine (SLC35A3) Transporters Form Glycosylation-related Complexes with Mannoside Acetylglucosaminyltransferases (Mgats)

UDP-galactose transporter (UGT; SLC35A2) and UDP-N-acetylglucosamine transporter (NGT; SLC35A3) form heterologous complexes in the Golgi membrane. NGT occurs in close proximity to mannosyl (α-1,6-)-glycoprotein β-1,6-N-acetylglucosaminyltransferase (Mgat5). In this study we analyzed whether NGT and both splice variants of UGT (UGT1 and UGT2) are able to interact with four different mannoside acetylglucosaminyltransferases (Mgat1, Mgat2, Mgat4B, and Mgat5). Using an in situ proximity ligation assay, we found that all examined glycosyltransferases are in the vicinity of these UDP-sugar transporters both at the endogenous level and upon overexpression. This observation was confirmed via the FLIM-FRET approach for both NGT and UGT1 complexes with Mgats. This study reports for the first time close proximity between endogenous nucleotide sugar transporters and glycosyltransferases. We also observed that among all analyzed Mgats, only Mgat4B occurs in close proximity to UGT2, whereas the other three Mgats are more distant from UGT2, and it was only possible to visualize their vicinity using proximity ligation assay. This strongly suggests that the distance between these protein pairs is longer than 10 nm but at the same time shorter than 40 nm. This study adds to the understanding of glycosylation, one of the most important post-translational modifications, which affects the majority of macromolecules. Our research shows that complex formation between nucleotide sugar transporters and glycosyltransferases might be a more common phenomenon than previously thought.

Organizational interplay of Golgi N-glycosyltransferases involves organelle microenvironment-dependent transitions between enzyme homo- and heteromers

Glycosylation of proteins and lipids takes place in the Golgi apparatus by the consecutive actions of functionally distinct glycosidases and glycosyltransferases. Current evidence indicates that they function as enzyme homomers and/or heteromers in the living cell. Here we investigate their organizational interplay and show that glycosyltransferase homomers are assembled in the endoplasmic reticulum. Upon transport to the Golgi, the majority of homomers are disassembled to allow the formation of enzyme heteromers between sequentially acting medial-Golgi enzymes GnT-I and GnT-II or trans-Golgi enzymes GalT-I and ST6Gal-I. This transition is driven by the acidic Golgi environment, as it was markedly inhibited by raising Golgi luminal pH with chloroquine. Our FRAP (fluorescence recovery after photobleaching) measurements showed that the complexes remain mobile Golgi membrane constituents that can relocate to the endoplasmic reticulum or to the scattered Golgi mini-stacks upon brefeldin A or nocodazole treatment, respectively. During this relocation, heteromers undergo a reverse transition back to enzyme homomers. These data unveil an unprecedented organizational interplay between Golgi N-glycosyltransferases that involves dynamic and organelle microenvironment-driven transitions between enzyme homomers and heteromers during their trafficking within the early secretory compartments.

Post-translational membrane insertion of tail-anchored transmembrane EF-hand Ca2+ sensor calneurons requires the TRC40/Asna1 protein chaperone

Calneuron-1 and -2 are neuronal EF-hand-type calcium sensor proteins that are prominently targeted to trans-Golgi network membranes and impose a calcium threshold at the Golgi for phosphatidylinositol 4-OH kinase IIIβ activation and the regulated local synthesis of phospholipids that are crucial for TGN-to-plasma membrane trafficking. In this study, we show that calneurons are nonclassical type II tail-anchored proteins that are post-translationally inserted into the endoplasmic reticulum membrane via an association of a 23-amino acid-long transmembrane domain (TMD) with the TRC40/Asna1 chaperone complex. Following trafficking to the Golgi, calneurons are probably retained in the TGN because of the length of the TMD and phosphatidylinositol 4-phosphate lipid binding. Both calneurons rapidly self-associate in vitro and in vivo via their TMD and EF-hand containing the N terminus. Although dimerization and potentially multimerization precludes TRC40/Asna1 binding and thereby membrane insertion, we found no evidence for a cytosolic pool of calneurons and could demonstrate that self-association of calneurons is restricted to membrane-inserted protein. The dimerization properties and the fact that they, unlike every other EF-hand calmodulin-like Ca(2+) sensor, are always associated with membranes of the secretory pathway, including vesicles and plasma membrane, suggests a high degree of spatial segregation for physiological target interactions.

Synthesis and evaluation of a radioiodinated trisaccharide derivative as a synthetic substrate for a sensitive N-acetylglucosaminyltransferase V radioassay

N-acetylglucosaminyltransferase V (GnT-V) is one of the most relevant glycosyltransferases to tumor invasion and metastasis. Based on previous findings of molecular recognition between GnT-V and synthetic substrates, we designed and synthesized a p-iodophenyl-derivatized trisaccharide, 2-(4-iodophenyl)ethyl 6-O-[2-O-(2-acetamido-2-deoxy-β-D-glucopyranosyl)-α-d-mannopyranosyl]-β-D-glucopyranoside (IPGMG, 1) and its radiolabeled form, [(125)I]IPGMG ([(125)I]1), for use in assays of GnT-V activity in vitro. The tributyltin derivative, 2-[4-(n-tributylstannyl)phenyl]ethyl 6-O-[2-O-(3,4,6-tri-O-acetyl-2-acetamido-2-deoxy-β-D-glucopyranosyl)-3,4,6-tri-O-acetyl-α-D-mannopyranosyl]-2,3,4-tri-O-acetyl-β-D-glucopyranoside (21), was synthesized as a precursor for the preparation of [(125)I]1. The iododestannylation of 21 using hydrogen peroxide as an oxidant followed by deacetylation yielded [(125)I]1. When [(125)I]1 was incubated in GnT-V-expressing cells with a UDP-GlcNAc donor, the production of β1-6GlcNAc-bearing IPGMG (IPGGMG, 2) was confirmed by radio-HPLC. In kinetic analysis, 1 was found to be a good substrate with a K(m) of 23.7 μM and a V(max) of 159 pmol/h. μg protein. [(125)I]1 would therefore be a useful synthetic substrate for the quantitative determination of GnT-V activity.

Chondroitin sulfate synthase-2/chondroitin polymerizing factor has two variants with distinct function

Chondroitin sulfate (CS) is a polysaccharide consisting of repeating disaccharide units of N-acetyl-D-galactosamine and d-glucuronic acid residues, modified with sulfated residues at various positions. To date six glycosyltransferases for chondroitin synthesis have been identified, and the complex of chondroitin sulfate synthase-1 (CSS1)/chondroitin synthase-1 (ChSy-1) and chondroitin sulfate synthase-2 (CSS2)/chondroitin polymerizing factor is assumed to play a major role in CS biosynthesis. We found an alternative splice variant of mouse CSS2 in a data base that lacks the N-terminal transmembrane domain, contrasting to the original CSS2. Here, we investigated the roles of CSS2 variants. Both the original enzyme and the splice variant, designated CSS2A and CSS2B, respectively, were expressed at different levels and ratios in tissues. Western blot analysis of cultured mouse embryonic fibroblasts confirmed that both enzymes were actually synthesized as proteins and were localized in both the endoplasmic reticulum and the Golgi apparatus. Pulldown assays revealed that either of CSS2A, CSS2B, and CSS1/ChSy-1 heterogeneously and homogeneously interacts with each other, suggesting that they form a complex of multimers. In vitro glycosyltransferase assays demonstrated a reduced glucuronyltransferase activity in CSS2B and no polymerizing activity in CSS2B co-expressed with CSS1, in contrast to CSS2A co-expressed with CSS1. Radiolabeling analysis of cultured COS-7 cells overexpressing each variant revealed that, whereas CSS2A facilitated CS biosynthesis, CSS2B inhibited it. Molecular modeling of CSS2A and CSS2B provided support for their properties. These findings, implicating regulation of CS chain polymerization by CSS2 variants, provide insight in elucidating the mechanisms of CS biosynthesis.

Activation of beta1,3-N-acetylglucosaminyltransferase-2 (beta3Gn-T2) by beta3Gn-T8. Possible involvement of beta3Gn-T8 in increasing poly-N-acetyllactosamine chains in differentiated HL-60 cells

Enzymatic activities of some glycosyltransferases are markedly increased via complex formation with other transferases or cofactor proteins. We previously showed that beta1,3-N-acetylglucosaminyltransferase-2 (beta3Gn-T2) and beta3Gn-T8 can form a heterodimer in vitro and that the complex exhibits much higher enzymatic activity than either enzyme alone (Seko, A., and Yamashita, K. (2005) Glycobiology 15, 943-951). Here we examined this activation and the biological significance of complex formation in differentiated HL-60 cells. beta3Gn-T2 and -T8 were co-immunoprecipitated from the lysates of both-transfected COS-7 cells, indicating their association in vivo. We prepared inactive mutants of both enzymes by destroying the DXD motifs. The mixture of mutated beta3Gn-T2 and intact beta3Gn-T8 did not exhibit any activation, whereas the mixture of intact beta3Gn-T2 and mutated beta3Gn-T8 had increased activity, indicating the activation of beta3Gn-T2 via complex formation. Next, we compared expression levels of beta3Gn-T1-T8 in HL-60 cells and DMSO-treated differentiated HL-60 cells, which produce larger poly-N-acetyllactosamine chains. The expression level of beta3Gn-T8 in the differentiated cells was 2.6-fold higher than in the untreated cells. Overexpression of beta3Gn-T8, but not beta3Gn-T2, induced an increase in poly-N-acetyllactosamine chains in HL-60 cells. These results raise a possibility that up-regulation of beta3Gn-T8 in differentiated HL-60 cells increases poly-N-acetyllactosamine chains by activating intrinsic beta3Gn-T2.

Glycomic analysis of alpha-fetoprotein L3 in hepatoma cell lines and hepatocellular carcinoma patients

The N-glycan structures of the Lens culinaris agglutinin (LCA)-reactive fraction of alpha-fetoprotein (AFP-L3), a tumor marker of hepatocellular carcinomas (HCC), were analyzed in relationship to glycosyltransferases and LCA-affinity electrophoresis. Using HPLC and MALDI-TOF MS, we determined the N-glycan structures of AFP from HCC cell lines, and demonstrated they were affected by N-acetylglucosaminyltransferase III and fucosyltransferase VIII, but not by N-acetylglucosaminyltransferase V. Moreover, we identified the N-glycan structures of AFP in HCC patients.

Cerebroside Sulfotransferase Forms Homodimers in Living Cells

Cerebroside sulfotransferase (CST) catalyzes the 3'-sulfation of galactose residues in several glycolipids. Its major product in the mammalian brain is sulfatide, which is an essential myelin component. Using epitope-tagged variants, murine CST was found to localize to the Golgi apparatus, but in contrast to previous assumptions, not to the trans-Golgi network. An examination of enhanced green fluorescent protein (EGFP)-tagged CST suggests that CST forms homodimers and that dimerization is mediated by the lumenal domain of the enzyme, as shown by immunoprecipitation and density gradient centrifugation. In order to verify that dimerization of CST observed by biochemical methods reflects the behavior of the native protein within living cells, the mobility of CST-EGFP was examined using fluorescence correlation spectroscopy. These experiments confirmed the homodimerization of CST-EGFP fusion proteins in vivo. In contrast to full-length CST, a fusion protein of the amino-terminal 36 amino acids of CST fused to EGFP was exclusively found as a monomer but nevertheless showed Golgi localization.

Complex formation regulates the glycosylation of the reversibly glycosylated polypeptide

Reversible glycosylated polypeptides (RGPs) are highly conserved plant-specific proteins, which can perform self-glycosylation. These proteins have been shown essential in plants yet its precise function remains unknown. In order to understand the function of this self-glycosylating polypeptide, it is important to establish what factors are involved in the regulation of the RGP activity. Here we show that incubation at high ionic strength produced a high self-glycosylation level and a high glycosylation reversibility of RGP from Solanum tuberosum L. In contrast, incubation at low ionic strength led to a low level of glycosylation and a low glycosylation reversibility of RGP. The incubation at low ionic strength favored the formation of high molecular weight RGP-containing forms, whereas incubation at high ionic strength produced active RGP with a molecular weight similar to the one expected for the monomer. Our data also showed that glycosylation of RGP, in its monomeric form, was highly reversible, whereas, a low reversibility of the protein glycosylation was observed when RGP was part of high molecular weight structures. In addition, glycosylation of RGP increased the occurrence of non-monomeric RGP-containing forms, suggesting that glycosylation may favor multimer formation. Finally, our results indicated that RGP from Arabidopsis thaliana and Pisum sativum are associated to golgi membranes, as part of protein complexes. A model for the regulation of the RGP activity and its binding to golgi membranes based on the glycosylation of the protein is proposed where the sugars linked to oligomeric form of RGP in the golgi may be transferred to acceptors involved in polysaccharide biosynthesis.

Cloning and recombinant expression of active full-length xylosyltransferase I (XT-I) and characterization of subcellular localization of XT-I and XT-II

Xylosyltransferase I (XT-I) catalyzes the transfer of xylose from UDP-xylose to serine residues in proteoglycan core proteins. This is the first and apparently rate-limiting step in the biosynthesis of the tetrasaccharide linkage region in glycosaminoglycan-containing proteoglycans. The XYLT-II gene codes for a highly homologous protein, but its physiological function is not yet known. Here we present for the first time the construction of a vector encoding the full-length GFP-tagged human XT-I and the recombinant expression of the active enzyme in mammalian cells. We expressed XT-I-GFP and various GFP-tagged XT-I and XT-II mutants with C-terminal truncations and deletions in HEK-293 and SaOS-2 cells in order to investigate the intracellular localization of XT-I and XT-II. Immunofluorescence analysis showed a distinct perinuclear pattern of XT-I-GFP and XT-II-GFP similar to that of alpha-mannosidase II, which is a known enzyme of the Golgi cisternae. Furthermore, a co-localization of native human XT-I and alpha-mannosidase II could also be demonstrated in untransfected cells. Using brefeldin A, we could also show that both xylosyltransferases are resident in the early cisternae of the Golgi apparatus. For its complete Golgi retention, XT-I requires the N-terminal 214 amino acids. Unlike XT-I, for XT-II, the first 45 amino acids are sufficient to target and retain the GFP reporter in the Golgi compartment. Here we show evidence that the stem regions were indispensable for Golgi localization of XT-I and XT-II.

Demonstration of the expression and the enzymatic activity of N-acetylglucosaminyltransferase IX in the mouse brain

We recently reported on a brain-specific beta1,6-N-acetylglucosaminyltransferase IX (GnT-IX, also referred to as GnT-VB), a GnT-V homologue, which acts on alpha-linked mannose of N-glycans and O-mannosyl glycans. To distinguish functions of GnT-IX with GnT-V, we examined the distribution of GnT-IX and GnT-V transcripts in mouse tissues by Northern blot analysis. The two enzymes were differentially expressed as has previously been observed in human tissues. GnT-IX transcripts were restricted to the cerebrum, cerebellum, thymus and testis, whereas GnT-V transcripts were expressed ubiquitously in mouse tissues. To investigate the localization of these enzymes in mouse tissues in more detail, a polyclonal antibody against GnT-IX was prepared. The antibody specifically recognized GnT-IX, but not GnT-V, in the Golgi apparatus, as confirmed by the use of GnT-IX and GnT-V transfectants. In agreement with the Northern blot analysis data, an immunohistochemical study showed substantial expression of GnT-IX in the brain, while no expression was observed in the liver. Moreover, to exclude GnT-V contamination, we performed an enzymatic assay for GnT-IX using a Mgat5 (GnT-V)-null mouse brain as an enzyme source and found the enzymatic activities do, in fact, exist in mouse brain. The reaction product was confirmed by high performance liquid chromatography and mass spectrometry. These results demonstrate, for the first time, that GnT-IX protein is actually expressed and may function as a glycosyltransferase in the brain.

Commuting between Golgi cisternae—Mind the GAP!

Intracellular transport has remained central to cell biology now for more than 40 years. Despite this, we still lack an overall mechanistic framework that describes transport in different parts of the cell. In the secretory pathway, basic questions, such as how biosynthetic cargo traverses the pathway, are still debated. Historically, emphasis was first put on interpreting function from morphology at the ultrastructural level revealing membrane structures such as the transitional ER, vesicular carriers, vesicular tubular clusters, Golgi cisternae, Golgi stacks and the Golgi ribbon. This emphasis on morphology later switched to biochemistry and yeast genetics yielding many of the key molecular players and their associated functions that we know today. More recently, microscopy studies of living cells incorporating biophysics and system analysis has proven useful and is often used to readdress earlier findings, sometimes with surprising outcomes.

Identification of a region within the cytoplasmic domain of the subtype B Vpu protein of human immunodeficiency virus type 1 (HIV-1) that is responsible for retention in the golgi complex and its absence in the Vpu protein from a subtype C HIV-1

The structure of the Vpu protein of human immunodeficiency virus type 1 (HIV-1) is composed of a short Nterminal domain (NTD), a transmembrane domain (TM), and a cytoplasmic domain (CD). Previous studies have shown that the Vpu protein from subtype B HIV-1 is transported predominantly to the rough endoplasmic reticulum (RER)/Golgi complex compartments of the cell and is not incorporated into virions. Using a previously described VpuEGFP reporter system in which the Vpu protein was fused to the gene for enhanced green fluorescent protein (EGFP), we showed that the subtype B Vpu fusion protein was localized to the RER/Golgi region of the cell, similar to the native protein. In the present study, we show that fusion of the subtype C Vpu to EGFP results in a fusion protein that is transported to the cell surface. Using this reporter system, chimeric Vpu proteins in which the CD of the subtype B and C proteins were exchanged showed that the CD was sufficient for targeting the subtype B protein to the Golgi complex of the cell. Following identification of the cytoplasmic domain as being responsible for intracellular targeting, we then generated a series of mutants in which 13, 23, 31, 38, 51, and 56 amino acids were deleted from the cytoplasmic domain of subtype B Vpu. These deletion mutants were analyzed by SDS-PAGE for size, for membrane localization, and intracellular localization by confocal fluorescence microscopy. Our results indicate that the mutant with the carboxyl-terminal 13 amino acids deleted was still localized to the Golgi complex but mutants with 23, 31, 38, 51, and 56 amino acids from the carboxyl-terminus of the subtype B Vpu were transported to the cell surface. These results suggest that a signal for the retention of the subtype B Vpu within the Golgi complex resides in the second alpha-helical domain.

Stem domains of heparan sulfate 6-O-sulfotransferase are required for Golgi localization, oligomer formation and enzyme activity

Heparan sulfate O-sulfotransferases catalyze the O-sulfation of the glucosamine and uronic acid residues of heparan sulfate, thereby determining the binding sites for ligands necessary for important biological functions such as the formation of morphogen gradients and growth factor signaling. Here we investigated the localization of the three heparan sulfate 6-O-sulfotransferase (HS6ST) isoforms and the mechanism of their localization. All three GFP-tagged HS6STs localized in the Golgi apparatus. C-5 epimerase and HS2ST have been shown to form complexes that facilitate their localization in the Golgi but we found that the absence of HS2ST did not alter the localization of any of the HS6STs. Neither the forced expression of HS2ST in the rough endoplasmic reticulum (ER), the deletion of most of the lumenal domain nor increasing the length of the transmembrane domain had any effect on the localization of HS6STs. However, deletions in the stem region did affect the Golgi localization of the HS6STs and also reduced their sulfotransferase activity and oligomer formation. These findings suggest that the stem region of HS6ST plays an important role in normal functioning, including the transit of HS6ST to the Golgi apparatus and maintaining the active conformation essential for enzyme activity.

Increased expression of N-acetylglucosaminyltransferase-V in human hepatoma cells by retinoic acid and 1alpha,25-dihydroxyvitamin D3

UDP-N-acetylglucosamine: alpha-6-D-mannoside beta-1,6N-acetylglucosaminyltransferase-V activities were determined in human hepatoma cell lines of Hep3B and HepG2, and also compared with those of normal liver tissues and primary hepatocytes. When GlcNAcbeta1-2Manalpha1-3(GlcNAcbeta1-2Manalpha1-4)(Manbeta1-4GlcNAc-2-amino pyridine (GlcN,GlcN-biant-PA) and UDP-GlcNAc were used as substrates, the enzymes displayed optimum temperatures of 50 degrees C, optimum pHs of 6.5 in each case, K(m) values for UDP-GlcNAc to be 5.8 (Hep3B) and 4.5 mM (HepG2) and K(m) values for GlcN,GlcN-biant-PA (mM) to be 1.28 (Hep3B) and 2.4 (HepG2). This indicates that values of Hep3B GlcNAc-transferase-V were distinguishable with HepG2 enzyme. Furthermore, Hep3B enzyme in membrane fraction showed about 1.5-fold higher specific activity (1.423 pmol/(h mg) than that (1.066 pmol/(h mg)) of HepG2. Normal hepatocytes are characterized by very low level of GlcNAc-transferase-V activity whereas hepatoma cells contained high activities. Treatment of hepatoma cells with retinoic acid and 1alpha,2,5-dihydroxyvitamin D(3) (Vit-D(3)) resulted in an increase in GlcNAc-transferase-V activity, while treatment with dimethyl sulfoxide and cytosine-arabinoside resulted in decrease in the enzyme activity. Although retinoic acid (RA) treated cells shows a changed GlcNAc-transferase-V mRNA expression, expression of marker proteins such as alpha-fetoprotein and albumin was not changed. This is the first demonstration of GlcNAc-transferase-V activity in RA and Vit-D(3)-treated hepatoma cell lines.

Molecular cloning and characterization of human GnT-IX, a novel beta1,6-N-acetylglucosaminyltransferase that is specifically expressed in the brain

A novel beta1,6-N-acetylglucosaminyltransferase (beta1, 6GnT) cDNA was identified by a BLAST search using the amino acid sequence of human GnT-V as a query. The full-length sequence was determined by a combination of 5'-rapid amplification of cDNA end analysis and a further data base search. The open reading frame encodes a 792 amino acid protein with a type II membrane protein structure typical of glycosyltransferases. The entire sequence identity to human GnT-V is 42%. When pyridylaminated (PA) agalacto biantennary N-linked oligosaccharide was used as an acceptor substrate, the recombinant enzyme generated a novel product other than the expected GnT-V product, (GlcNAcbeta1,2-Manalpha1,3-)[GlcNAcbeta1,2-(GlcNAcbeta1,6-)Manalpha1,6-]Manbeta1,4-GlcNAcbeta1,4-GlcNAc-PA. This new product was identified as [GlcNAcbeta1,2-(GlcNAcbeta1,6-)Manalpha1,3-][Glc-NAcbeta1,2-(GlcNAcbeta1,6-)Manalpha1,6-]Manbeta1,4-GlcNAcbeta1,4-GlcNAc-PA by mass spectrometry and 1H NMR. Namely, the new GnT (designated as GnT-IX) has beta1,6GnT activity not only to the alpha1,6-linked mannose arm but also to the alpha1,3-linked mannose arm of N-glycan, forming a unique structure that has not been reported to date. Northern blot analysis showed that the GnT-IX gene is exclusively expressed in the brain, whereas the GnT-V gene is expressed ubiquitously. These results suggest that GnT-IX is responsible for the synthesis of a unique oligosaccharide structure in the brain.

Chondroitin sulfate synthase-3. Molecular cloning and characterization

Recently, it has become evident that chondroitin sulfate (CS) glycosyltransferases, which transfer glucuronic acid and/or N-acetylgalactosamine residues from each UDP-sugar to the nonreducing terminus of the CS chain, form a gene family. We report here a novel human gene (GenBank trade mark accession number AB086062) that possesses a sequence homologous with the human chondroitin sulfate synthase-1 (CSS1) gene, formerly known as chondroitin synthase. The full-length open reading frame consists of 882 amino acids and encodes a typical type II membrane protein. This enzyme contains a beta 3-glycosyltransferase motif and a beta 4-glycosyltransferase motif similar to that found in CSS1. Both the enzymes were expressed in COS-7 cells as soluble proteins, and their enzymatic natures were characterized. Both glucuronyltransferase and N-acetylgalactosaminyltransferase activities were observed when chondroitin, CS polymer, and their corresponding oligosaccharides were used as the acceptor substrates, but no polymerization reaction was observed as in the case of CSS1. The new enzyme was thus designated chondroitin sulfate synthase-3 (CSS3). However, the specific activity of CSS3 was much lower than that of CSS1. The reaction products were shown to have a GlcUA beta 1-3GalNAc linkage and a GalNAc beta 1-4GlcUA linkage in the nonreducing terminus of chondroitin resulting from glucuronyltransferase activity and N-acetylgalactosaminyltransferase activity, respectively. Quantitative real time PCR analysis revealed that the transcript level of CSS3 was much lower than that of CSS1, although it was ubiquitously expressed in various human tissues. These results indicate that CSS3 is a glycosyltransferase having both glucuronyltransferase and N-acetylgalactosaminyltransferase activities. It may make a contribution to CS biosynthesis that differs from that of CSS1.

Intracellular sorting and transport of proteins

The secretory and endocytic pathways of eukaryotic organelles consist of multiple compartments, each with a unique set of proteins and lipids. Specific transport mechanisms are required to direct molecules to defined locations and to ensure that the identity, and hence function, of individual compartments are maintained. The localisation of proteins to specific membranes is complex and involves multiple interactions. The recent dramatic advances in understanding the molecular mechanisms of membrane transport has been due to the application of a multi-disciplinary approach, integrating membrane biology, genetics, imaging, protein and lipid biochemistry and structural biology. The aim of this review is to summarise the general principles of protein sorting in the secretory and endocytic pathways and to highlight the dynamic nature of these processes. The molecular mechanisms involved in this transport along the secretory and endocytic pathways are discussed along with the signals responsible for targeting proteins to different intracellular locations.

Chondroitin sulfate synthase-2. Molecular cloning and characterization of a novel human glycosyltransferase homologous to chondroitin sulfate glucuronyltransferase, which has dual enzymatic activities

Chondroitin sulfate is found in a variety of tissues as proteoglycans and consists of repeating disaccharide units of N-acetylgalactosamine and glucuronic acid residues with sulfate residues at various places. We found a novel human gene (GenBank accession number AB086063) that possesses a sequence homologous with the human chondroitin sulfate glucuronyltransferase gene which we recently cloned and characterized. The full-length open reading frame encodes a typical type II membrane protein comprising 775 amino acids. The protein had a domain containing beta 3-glycosyltransferase motif but lacked a typical beta 4-glycosyltransferase motif, which is the same as chondroitin sulfate glucuronyltransferase, whereas chondroitin synthase had both domains. The putative catalytic domain was expressed in COS-7 cells as a soluble enzyme. Surprisingly, both glucuronyltransferase and N-acetylgalactosaminyltransferase activities were observed when chondroitin, chondroitin sulfate, and their oligosaccharides were used as the acceptor substrates. The reaction products were identified to have the linkage of GlcUA beta 1-3GalNAc and GalNAc beta 1-4GlcUA at the non-reducing terminus of chondroitin for glucuronyltransferase activity and N-acetylgalactosaminyltransferase activity, respectively. Quantitative real time PCR analysis revealed that the transcripts were ubiquitously expressed in various human tissues but highly expressed in the pancreas, ovary, placenta, small intestine, and stomach. These results indicate that this enzyme could synthesize chondroitin sulfate chains as a chondroitin sulfate synthase that has both glucuronyltransferase and N-acetylgalactosaminyltransferase activities. Sequence analysis based on three-dimensional structure revealed the presence of not typical but significant beta 4-glycosyltransferase architecture.

Caveolin-1 regulates the functional localization of N-acetylglucosaminyltransferase III within the golgi apparatus

In an investigation of the mechanism underlying the functional sublocalization of glycosyltransferases within the Golgi apparatus, caveolin-1 was identified as a possible cellular factor. Caveolin-1 appears to regulate the localization of N-acetylglucosaminyltransferase III (GnT-III) in the intra-Golgi subcompartment. Structural analyses of total cellular N-glycans indicated that the overexpression of GnT-III in human hepatoma cells, in which caveolin-1 is not expressed, failed to reduce branch formation, whereas expression of caveolin-1 led to a dramatic decrease in the extent of branching with no enhancement in GnT-III activity. Because the addition of a bisecting GlcNAc by GnT-III to the core beta-Man in N-glycans prevents the action of GnT-IV and GnT-V, both of which are involved in branch formation, this result suggests that caveolin-1 facilitates the prior action of GnT-III, relative to the other GnTs, on the nascent sugar chains in the Golgi apparatus and that GnT-III is redistributed in the earlier Golgi subcompartment by caveolin-1. Indeed, when caveolin-1 was expressed in human hepatoma cells, it was found to be co-localized with GnT-III, as evidenced by the fractionation of Triton X-100-insoluble cellular membranes by density gradient ultracentrifugation. Caveolin-1 may modify the biosynthetic pathway of sugar chains via the regulation of the intra-Golgi subcompartment localization of this key glycosyltransferase.

Importance of Cys, Gln, and Tyr from the transmembrane domain of human alpha 3/4 fucosyltransferase III for its localization and sorting in the Golgi of baby hamster kidney cells

Human fucosyltransferase III (EC ) (FT3wt) is localized in the Golgi of baby hamster kidney cells and synthesizes Lewis determinants associated with cell adhesion events. Replacement of the amino acid residues from the transmembrane domain (TM) Cys-16, Gln-23, Cys-29, and Tyr-33 by Leu (FT3np) caused a shift in enzyme localization to the plasma membrane. The mislocalization caused a dramatic decrease in the amount of biosynthetic products of FT3wt, the Lewis determinants. Determination of the expression levels on the surface with mutants of the enzyme, where one, two, or three of these residues were replaced by Leu, suggested that Cys from the TM was required for the localization of FT3 in the Golgi. Furthermore, Cys-23 and Cys-29 mediated the formation of disulfide-bonded dimers but not higher molecular weight oligomers. In vitro reconstitution of intra-Golgi transport showed that FT3wt was incorporated into coatomer protein (COP) I vesicles, contrary to FT3np. These data suggested that Cys, Gln, and Tyr residues are important for FT3wt sorting into the transport vesicles possibly due to interactions with other membrane proteins.

UDP-GlcNAc concentration is an important factor in the biosynthesis of beta1,6-branched oligosaccharides: regulation based on the kinetic properties of N-acetylglucosaminyltransferase V

Human beta1,6-N-acetylglucosaminyltransferase V (GnT-V) was expressed by baculovirus-insect cell system, and the purified recombinant enzyme was kinetically characterized. The data obtained were used to establish the kinetic basis of the substrate specificity toward donor nucleotide sugars, and also revealed that K(m) values for the donors are much higher compared to those of other GlcNAc transferases, the kinetic properties of which have been reported. Because this exceptionally higher K(m) suggests that GnT-V is physiologically present at far from saturated conditions, it would appear that the production of beta1,6-branched oligosaccharide, which is formed by GnT-V, could be regulated in vivo by the concentration of the donor, UDP-GlcNAc, as well as the expression levels of the enzyme. When B16 melanoma cells, which express high levels of GnT-V, were incubated with GlcNAc, the beta1,6-branched oligosaccharide levels were increased, as judged by a lectin blot analysis, in conjunction with an increase in intracellular UDP-GlcNAc. These findings suggest that the level of UDP-GlcNAc can be a critical factor in the production of beta1,6-branched oligosaccharides, for example, by tumor cells, which have been thought to be closely associated with tumor progression and metastasis.

Structural and functional features of glycosyltransferases

Most of the glycosylation reactions that generate the great diversity of oligosaccharide structures of eukaryotic cells occur in the Golgi apparatus. This review deals with the most recent data that provide insight into the functional organization of Golgi-resident glycosyltransferases. We also focus on the recent successes in X-ray crystal structure determination of glycosyltransferases. These new structures begin to shed light on the molecular bases accounting for donor and acceptor substrate specificities as well as catalysis.

Trafficking and localisation of resident Golgi glycosylation enzymes

The localisation of glycosylation enzymes within the Golgi apparatus is fundamental to the regulation of glycoprotein and glycolipid biosynthesis. Regions responsible for specifying Golgi localisation have been identified in numerous Golgi resident enzymes. The transmembrane domain of Golgi glycosyltransferases provides a dominant localisation signal and in many cases there are also major contributions from the lumenal domain. The mechanism by which these targeting domains function in maintaining an asymmetric distribution of Golgi resident glycosylation enzymes has been intensely debated in recent years. It is now clear that the targeting of Golgi resident enzymes is intimately associated with the organisation of Golgi membranes and the control of protein and lipid traffic in both anterograde and retrograde directions. Here we discuss the recent advances into how Golgi targeting signals of glycosylation enzymes function, and propose a model for maintaining the steady-state localisation of Golgi glycosyltransferases.