Hypoglycosylation of a brain glycoprotein (beta-trace protein) in CDG syndromes due to phosphomannomutase deficiency and N-acetylglucosaminyl-transferase II deficiency

CDG Therapies: From Bench to Bedside

Congenital disorders of glycosylation (CDG) are a group of genetic disorders that affect protein and lipid glycosylation and glycosylphosphatidylinositol synthesis. More than 100 different disorders have been reported and the number is rapidly increasing. Since glycosylation is an essential post-translational process, patients present a large range of symptoms and variable phenotypes, from very mild to extremely severe. Only for few CDG, potentially curative therapies are being used, including dietary supplementation (e.g., galactose for PGM1-CDG, fucose for SLC35C1-CDG, Mn²⁺ for TMEM165-CDG or mannose for MPI-CDG) and organ transplantation (e.g., liver for MPI-CDG and heart for DOLK-CDG). However, for the majority of patients, only symptomatic and preventive treatments are in use. This constitutes a burden for patients, care-givers and ultimately the healthcare system. Innovative diagnostic approaches, in vitro and in vivo models and novel biomarkers have been developed that can lead to novel therapeutic avenues aiming to ameliorate the patients’ symptoms and lives. This review summarizes the advances in therapeutic approaches for CDG.

Exon-skipping antisense oligonucleotides to correct missplicing in neurogenetic diseases

Alternative splicing is an important regulator of the transcriptome. However, mutations may cause alteration of splicing patterns, which in turn leads to disease. During the past 10 years, exon skipping has been looked upon as a powerful tool for correction of missplicing in disease and progress has been made towards clinical trials. In this review, we discuss the use of antisense oligonucleotides to correct splicing defects through exon skipping, with a special focus on diseases affecting the nervous system, and the latest stage achieved in its progress.

N-glycosylation analysis by HPAEC-PAD and mass spectrometry

Changes in protein glycosylation are a hallmark of most types of cancer including ovarian carcinoma. The structural elucidation of glycans is technically challenging and it requires complementary chromatographic and spectroscopic techniques among others. Here, we describe the profiling of N-glycans from glycoproteins of SKOV3 ovarian carcinoma cells by high-performance anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD) and matrix-assisted laser desorption/ionization with time-of-flight mass spectrometry (MALDI-TOF MS). Mass spectrometry as a complementary method enables precise mass determination of N-glycan mixtures thus corroborating data obtained from HPAEC-PAD mapping in conjunction with reference oligosaccharide structures.

Primary immunodeficiency diseases associated with neurologic manifestations

Primary immunodeficiency diseases (PID) are a heterogeneous group of inherited disorders of the immune system, predisposing individuals to recurrent infections, allergy, autoimmunity, and malignancies. A considerable number of these conditions have been found to be also associated with neurologic signs and symptoms. These manifestations are considered core features of some immunodeficiency syndromes, such as ataxia-telangiectasia and purine nucleoside phosphorylase deficiency, or occur less prominently in some others. Diverse pathological mechanisms including defective responses to DNA damage, metabolic errors, and autoimmune phenomena have been associated with neurologic abnormalities; however, several issues remain to be elucidated. Greater awareness of these associated features and gaining a better understanding of the contributing mechanisms will lead to prompt diagnosis and treatment and possibly development of novel preventive and therapeutic strategies. In this review, we aim to provide a brief description of the clinical and genetic characteristics of PID associated with neurologic complications.

The impact of mass spectrometry in the diagnosis of congenital disorders of glycosylation

Contribution of mass spectrometry (MS) in the diagnosis and characterization of congenital disorders of glycosylation (CDG) has long been known. CDG type I diseases are characterized by the under-occupancy of protein N-glycosylation sites. Electrospray (ESI) MS and matrix assisted laser desorption ionization (MALDI) MS are effective for underglycosylation analyses of intact serum Transferrin (Tf) in CDG-I patients by mass determination of individual component glycoforms. Thus, high-throughput methods developed to speed-up analytical times found increasing application in clinical testing for CDG detection. ESI MS recognizable glycoform profiles of serum Tf have been reported in CDG-I different from PMM2-CDG and in individual CDG-II defects. MALDI MS analysis of acidic and neutral N-linked glycans released from total plasma or targeted glycoproteins, is the mainstream tool to explore abnormal oligosaccharide structure and changes in the relative amount of individual oligosaccharides in CDG-II patients. Here we briefly review state-of-the-art and updates of MS-based applications for the diagnosis of CDG with special emphasis to detectable glycosylation profiles reported in different CDG types.

L-PGDS (betatrace protein) inhibits astrocyte proliferation and mitochondrial ATP production in vitro

L-PGDS is the most abundant protein present in the cerebrospinal fluid (CSF). Although CSF was believed to be homogenous in content, a previous study has showed that a marked concentration gradient of L-PGDS exists between the spinal CSF and the CSF in the subarachnoid space of patients with optic nerve disease (papilledema and normal-tension glaucoma). Astrocytes play a critical role in maintaining the integrity of axon function in the central nervous system and specifically in the optic nerve, and we therefore investigated the biochemical effects of L-PGDS on the proliferation of astrocytes and on the production of adenosine triphosphate (ATP) by astrocyte mitochondria. We found an inhibitory effect of L-PGDS on both proliferation of astrocytes and production of astrocyte ATP. The concentrations that inhibited astrocyte proliferation and ATP production were in the range measured in patients with idiopathic intracranial hypertension and in patients with normal-tension glaucoma. As the CSF is in contact with axons and mitochondria of the optic nerve (Bristow et al. Archives of Ophthalmology, 120, 791-796, 2002), we postulate that a change in the concentration of CSF protein such as L-PGDS could exercise a harmful effect on these structures.

Screening and diagnosis of congenital disorders of glycosylation

The aim of this paper is to review the diagnostics of congenital disorders of glycosylation (CDG), an ever expanding group of diseases. Development delay, neurological, and other clinical abnormalities as well as various non-specific laboratory changes can lead to the first suspicion of the disease. Still common screening test for most CDG types, including CDG Ia, is isoelectric focusing/polyacrylamide gel electrophoresis (IEF). IEF demonstrates the hypoglycosylation of various glycoproteins, usually serum transferrin. Other methods, such as agarose electrophoresis, capillary electrophoresis, high-performance liquid chromatography, micro-column separation combined with turbidimetry, enzyme-(EIA) and radioimmunoassay (RIA) have also been used for screening. However, these methods do not recognize all CDG defects, so other approaches including analysis of membrane-linked markers and urine oligosaccharides should be taken. Confirmation of diagnosis and detailed CDG subtyping starts with thorough structure analysis of the affected lipid-linked oligosaccharide or protein-(peptide)-linked-glycan using metabolic labeling and various (possibly mass-spectrometry combined) techniques. Decreased enzyme activity in peripheral leukocytes/cultured fibroblasts or analysis of affected transporters and other functional proteins combined with identification of specific gene mutations confirm the diagnosis. Prenatal diagnosis, based on enzyme assay or mutation analysis, is also available. Peri-/post-mortem investigations of fatal cases are important for genetic counseling. Evaluation of various analytical approaches and proposed algorithms for investigation complete the review.

The normal phenotype of Pmm1-deficient mice suggests that Pmm1 is not essential for normal mouse development

Phosphomannomutases (PMMs) are crucial for the glycosylation of glycoproteins. In humans, two highly conserved PMMs exist: PMM1 and PMM2. In vitro both enzymes are able to convert mannose-6-phosphate (mannose-6-P) into mannose-1-P, the key starting compound for glycan biosynthesis. However, only mutations causing a deficiency in PMM2 cause hypoglycosylation, leading to the most frequent type of the congenital disorders of glycosylation (CDG): CDG-Ia. PMM1 is as yet not associated with any disease, and its physiological role has remained unclear. We generated a mouse deficient in Pmm1 activity and documented the expression pattern of murine Pmm1 to unravel its biological role. The expression pattern suggested an involvement of Pmm1 in (neural) development and endocrine regulation. Surprisingly, Pmm1 knockout mice were viable, developed normally, and did not reveal any obvious phenotypic alteration up to adulthood. The macroscopic and microscopic anatomy of all major organs, as well as animal behavior, appeared to be normal. Likewise, lectin histochemistry did not demonstrate an altered glycosylation pattern in tissues. It is especially striking that Pmm1, despite an almost complete overlap of its expression with Pmm2, e.g., in the developing brain, is apparently unable to compensate for deficient Pmm2 activity in CDG-Ia patients. Together, these data point to a (developmental) function independent of mannose-1-P synthesis, whereby the normal knockout phenotype, despite the stringent conservation in phylogeny, could be explained by a critical function under as-yet-unidentified challenge conditions.

Pitfalls and drawbacks in screening of congenital disorders of glycosylation

Congenital disorders of glycosylation include a group of diseases, each of them caused by different protein (mostly enzyme) impairment due to a specific gene defect. The many subtypes are classified according to clinical features, enzymology and molecular genetic analyses. Problems in diagnostics arise from the great diversity in clinical presentation, usually age-related, and different severities of individual types of these, by far underdiagnosed, diseases. Also the biochemical findings tend to vary, even within a single type. No one screening test, common for all types, is available so far. Several methods of choice may be used in the first approach; other procedures must follow for detailed typing of the defect. Possible drawbacks and pitfalls in the diagnostics from the viewpoint of our 3-year studies and practical screening experience are presented.

Congenital disorders of glycosylation

Congenital disorders of glycosylation (CDG) are a rapidly growing group of genetic diseases that are due to defects in the synthesis of glycans and in the attachment of glycans to other compounds. Most CDG are multisystem diseases that include severe brain involvement. The CDG causing sialic acid deficiency of N-glycans can be diagnosed by isoelectrofocusing of serum sialotransferrins. An efficient treatment, namely oral D-mannose, is available for only one CDG (CDG-Ib). In many patients with CDG, the basic defect is unknown (CDG-x). Glycan structural analysis, yeast genetics, and knockout animal models are essential tools in the elucidation of novel CDG. Eleven primary genetic glycosylation diseases have been discovered and their basic defects identified: six in the N-glycan assembly, three in the N-glycan processing, and two in the O-glycan (glycosaminoglycan) assembly. This review summarizes their clinical, biochemical, and genetic characteristics and speculates on further developments in this field.

Defect in N-glycosylation of proteins is tissue-dependent in congenital disorders of glycosylation Ia

The biochemical hallmark of Congenital Disorders of Glycosylation (CDG) including type Ia is a defective N-glycosylation of serum glycoproteins. Hypoglycosylated forms of alpha1-antitrypsin have been detected by Western blot in serum from CDG Ia patients. In contrast we were not able to detect hypoglycosylation in alpha1-antitrypsin synthesized by fibroblasts, keratinocytes, enterocytes, and leukocytes. Similarly no hypoglycosylation was detectable in a membrane-associated N-linked glycoprotein, the facilitative glucose transporter GLUT-1 and also in serum immunoglobulin G isolated from sera of CDG Ia patients. We conclude that the phenotypic expression of CDG Ia is tissue-dependent.

Biochemical, structural, genetic, physiological, and pathophysiological features of lipocalin-type prostaglandin D synthase

Lipocalin-type prostaglandin (PG) D synthase (PGDS) catalyzes the isomerization of PGH(2), a common precursor of various prostanoids, to produce PGD(2), a potent endogenous somnogen and nociceptive modulator, in the presence of sulfhydryl compounds. PGDS is an N-glycosylated monomeric protein with an M(r) of 20000-31000 depending on the size of the glycosyl moiety. PGDS is localized in the central nervous system and male genital organs of various mammals and in the human heart and is secreted into the cerebrospinal fluid, seminal plasma, and plasma, respectively, as beta-trace. The PGDS concentrations in these body fluids are useful for the diagnosis of several neurological disorders, dysfunction of sperm formation, and cardiovascular and renal diseases. The cDNA and gene for PGDS have been isolated from several animal species, and the tissue distribution and cellular localization have also been determined. This enzyme is considered to be a dual functional protein; i.e. it acts as a PGD(2)-producing enzyme and also as a lipophilic ligand-binding protein, because the enzyme binds biliverdin, bilirubin (K(d)=30 nM), retinaldehyde, retinoic acid (K(d)=80 nM) with high affinities. X-ray crystallographic analyses revealed that PGDS possesses a beta-barrel structure with a hydrophobic pocket in which an active thiol, Cys(65), the active center for the catalytic reaction, was located facing to the inside of the pocket. Gene-knockout and transgenic mice for PGDS were generated and found to have abnormalities in the regulation of nociception and sleep.

The cytoplasmic, transmembrane, and stem regions of glycosyltransferases specify their in vivo functional sublocalization and stability in the Golgi

We provide evidence for the presence of targeting signals in the cytoplasmic, transmembrane, and stem (CTS) regions of Golgi glycosyltransferases that mediate sorting of their intracellular catalytic activity into different functional subcompartmental areas of the Golgi. We have constructed chimeras of human alpha1, 3-fucosyltransferase VI (FT6) by replacement of its CTS region with those of late and early acting Golgi glycosyltransferases and have stably coexpressed these constructs in BHK-21 cells together with the secretory reporter glycoprotein human beta-trace protein. The sialyl Lewis X:Lewis X ratios detected in beta-trace protein indicate that the CTS regions of the early acting GlcNAc-transferases I (GnT-I) and III (GnT-III) specify backward targeting of the FT6 catalytic domain, whereas the CTS region of the late acting human alpha1,3-fucosyltransferase VII (FT7) causes forward targeting of the FT6 in vivo activity in the biosynthetic glycosylation pathway. The analysis of the in vivo functional activity of nine different CTS chimeras toward beta-trace protein allowed for a mapping of the CTS donor glycosyltransferases within the Golgi/trans-Golgi network: GnT-I < (ST6Gal I, ST3Gal III) < GnT-III < ST8Sia IV < GalT-I < (FT3, FT6) < ST3Gal IV < FT7. The sensitivity or resistance of the donor glycosyltransferases toward intracellular proteolysis is transferred to the chimeric enzymes together with their CTS regions. Apparently, there are at least three different signals contained in the CTS regions of glycosyltransferases mediating: first, their Golgi retention; second, their targeting to specific in vivo functional areas; and third, their susceptibility toward intracellular proteolysis as a tool for the regulation of the intracellular turnover.

Matrix-assisted laser desorption/ionization mass spectrometry of carbohydrates

This review describes the application of matrix-assisted laser desorption/ionization (MALDI) mass spectrometry to carbohydrate analysis and covers the period 1991-1998. The technique is particularly valuable for carbohydrates because it enables underivatised, as well as derivatised compounds to be examined. The various MALDI matrices that have been used for carbohydrate analysis are described, and the use of derivatization for improving mass spectral detection limits is also discussed. Methods for sample preparation and for extracting carbohydrates from biological media prior to mass spectrometric analysis are compared with emphasis on highly sensitive mass spectrometric methods. Quantitative aspects of MALDI are covered with respect to the relationship between signal strength and both mass and compound structure. The value of mass measurements by MALDI to provide a carbohydrate composition is stressed, together with the ability of the technique to provide fragmentation spectra. The use of in-source and post-source decay and collision-induced fragmentation in this context is described with emphasis on ions that provide information on the linkage and branching patterns of carbohydrates. The use of MALDI mass spectrometry, linked with exoglycosidase sequencing, is described for N-linked glycans derived from glycoproteins, and methods for the analysis of O-linked glycans are also covered. The review ends with a description of various applications of the technique to carbohydrates found as constituents of glycoproteins, bacterial glycolipids, sphingolipids, and glycolipid anchors.

Carbohydrate-deficient glycoprotein syndrome type IA (phosphomannomutase-deficiency)

The carbohydrate-deficient glycoprotein or CDG syndromes (OMIM 212065) are a recently delineated group of genetic, multisystem diseases with variable dysmorphic features. The known CDG syndromes are characterized by a partial deficiency of the N-linked glycans of secretory glycoproteins, lysosomal enzymes, and probably also membranous glycoproteins. Due to the deficiency of terminal N-acetylneuraminic acid or sialic acid, the glycan changes can be observed in serum transferrin or other glycoproteins using isoelectrofocusing with immunofixation as the most widely used diagnostic technique. Most patients show a serum sialotransferrin pattern characterized by increased di- and asialotransferrin bands (type I pattern). The majority of patients with type I are phosphomannomutase deficient (type IA), while in a few other patients, deficiencies of phosphomannose isomerase (type IB) or endoplasmic reticulum glucosyltransferase (type IC) have been demonstrated. This review is an update on CDG syndrome type IA.

Carbohydrate-deficient glycoprotein syndrome type II

The carbohydrate-deficient glycoprotein syndromes (CDGS) are a group of autosomal recessive multisystemic diseases characterized by defective glycosylation of N-glycans. This review describes recent findings on two patients with CDGS type II. In contrast to CDGS type I, the type II patients show a more severe psychomotor retardation, no peripheral neuropathy and a normal cerebellum. The CDGS type II serum transferrin isoelectric focusing pattern shows a large amount (95%) of disialotransferrin in which each of the two glycosylation sites is occupied by a truncated monosialo-monoantennary N-glycan. Fine structure analysis of this glycan suggested a defect in the Golgi enzyme UDP-GlcNAc:alpha-6-D-mannoside beta-1,2-N-acetylglucosaminyltransferase II (GnT II; EC 2.4.1.143) which catalyzes an essential step in the biosynthetic pathway leading from hybrid to complex N-glycans. GnT II activity is reduced by over 98% in fibroblast and mononuclear cell extracts from the CDGS type II patients. Direct sequencing of the GnT II coding region from the two patients identified two point mutations in the catalytic domain of GnT II, S290F (TCC to TTC) and H262R (CAC to CGC). Either of these mutations inactivates the enzyme and probably also causes reduced expression. The CDG syndromes and other congenital defects in glycan synthesis as well as studies of null mutations in the mouse provide strong evidence that the glycan moieties of glycoproteins play essential roles in the normal development and physiology of mammals and probably of all multicellular organisms.

beta-Trace protein in human cerebrospinal fluid: a diagnostic marker for N-glycosylation defects in brain

As carbohydrate-deficient glycoprotein syndromes (CDGS) are multisystemic disorders with impaired central nervous function in nearly all cases, we tested isoforms of beta-trace protein (beta TP), a 'brain-type' glycosylated protein in cerebrospinal fluid (CSF) of nine patients with the characteristic CDGS type I pattern of serum transferrin. Whereas the serum transferrin pattern did not discriminate between the various subtypes of CDGS type I (CDGS type Ia, type Ic, and patients with unknown defect), beta TP isoforms of CDGS type Ia patients differed from that of the other CDGS type I patients. The percentage of abnormal beta TP isoforms correlated with the severity of the neurological symptoms. Furthermore, two patients are described, who illustrate that abnormal protein N-glycosylation can occur restricted to either the 'peripheral' serum or the central nervous system compartment. This is the first report presenting evidence for an N-glycosylation defect restricted to the brain. Testing beta TP isoforms is a useful tool to detect protein N-glycosylation disorders in the central nervous system.

The sugar moiety of Tamm-Horsfall protein is affected by the carbohydrate-deficient glycoprotein type I syndrome. A case study

As the sugar moiety of Tamm-Horsfall protein (THP) is affected by many pathological conditions, the aim of this study was to examine the influence of carbohydrate-deficient glycoprotein syndrome (CDG) on THP glycans. THP was isolated from urine of one patient with CDG type I and N-glycan profiling, analysis of monosaccharide content, determination of THP reactivity with specific lectins and with anti-THP antibodies were performed. THP of the CDG patient showed markedly lower amounts of all monosaccharides. Diminished amounts of lactosamine-type chains, galactose and alpha2,3 linked sialic acid were expressed in lower reactivity with PHA-L, DSA and MAA, respectively. These modifications were reflected in altered proportions of tetrasialylated and disialylated oligosaccharide chains. THP of the CDG patient reacted slightly more with anti-THP antibodies. Our results indicate that the CDG type I affects the THP sugar moiety and slightly enhances the THP immunoreactivity.

Changes in concanavalin A-reactive proteins in neurological disorders

Changes of glycosylation of cerebrospinal fluid proteins such as alpha2-macroglobulin, and prostaglandin D synthase were studied by lectin blotting, using concanavalinA, in multiple sclerosis (n = 42) and neuropathies (n = 20) in comparison to neurological controls (n = 22). The concanavalinA-reactivity of alpha2-macroglobulin, which was increased in the neuropathies but not in multiple sclerosis compared to controls, correlated with the total concanavalinA-reactivity in controls and neuropathies but not in multiple sclerosis, indicating that the protein could be abnormally glycosylated in the latter disease. Although the concentration and the concanavalinA-reactivity of prostaglandin D synthase were not significantly different in the three groups, the two parameters correlated only in neuropathies but not in controls or multiple sclerosis, probably due to the high heterogeneity of the protein. These changes deserve to be studied in further detail in view of their potential clinical applications.

Glutathione-independent prostaglandin D2 synthase in ram and stallion epididymal fluids: origin and regulation

Microsequencing after two-dimensional electrophoresis revealed a major protein, glutathione-independent prostaglandin D2 synthase (PGDS) in the anterior epididymal region fluid of the ram and stallion. In this epididymal region, PGDS was a polymorphic compound with a molecular mass around 30 kDa and a range of pI from 4 to 7. PGDS represented 15% and 8% of the total luminal proteins present in this region in the ram and stallion, respectively. The secretion of the protein as judged by in vitro biosynthesis, and the presence of its mRNA as studied by Northern blot analysis, were limited to the proximal caput epididymidis. Using a specific polyclonal antibody raised against a synthetic peptide, PGDS was found throughout the epididymis, decreasing in concentration toward the cauda region. PGDS was also detected in the testicular fluid and seminal plasma by Western blotting. Castration and efferent duct ligation in the ram led to a decrease in PGDS mRNA and secretion. PGDS mRNA was not detected in the stallion 1 mo after castration, and it was restored by testosterone supplementation. This study showed that PGDS is present in the environment of spermatozoa throughout the male genital tract. Its function in the maturation and/or protection of spermatozoa is unknown.

Genetic engineering of recombinant glycoproteins and the glycosylation pathway in mammalian host cells

The analysis of many natural glycoproteins and their recombinant counterparts from mammalian hosts has revealed that the basic oligosaccharide structures and the site occupancy of glycosylated polypeptides are primarily dictated by the protein conformation. The equipment of many frequently used host cells (e.g. BHK-21 and CHO-cells) with glycosyltransferases, nucleotide-sugar synthases and transporters appears to be sufficient to guarantee complex-type glycosylation of recombinant proteins with a high degree of terminal alpha2-3 sialylation even under high expression conditions. Some human tissue-specific terminal carbohydrate motifs are not synthesized by these cells since they lack the proper sugar-transferring enzymes (e.g. alpha1-3/4 fucosyltransferases, alpha2-6 sialyltransferases). Glycosylation engineering of these hosts by stable transfection with genes encoding terminal human glycosyltransferases allows to obtain products with tailored (human tissue-specific) glycosylation in high yields. Using site-directed mutagenesis, unglycosylated polypeptides can be successfully converted in N- and/or O-glycoproteins by transferring glycosylation domains (consisting of 7-17 amino acids) from donor glycoproteins to different loop regions of acceptor proteins. The genetic engineering of glycoproteins and of host cell lines are considered to provide a versatile tool to obtain therapeutic glyco-products with novel/improved in-vivo properties, e.g. by introduction of specific tissue-targeting signals by a rational design of terminal glycosylation motifs.

In vivo specificity of human alpha1,3/4-fucosyltransferases III-VII in the biosynthesis of LewisX and Sialyl LewisX motifs on complex-type N-glycans. Coexpression studies from bhk-21 cells together with human beta-trace protein

Each of the five human alpha1,3/4-fucosyltransferases (FT3 to FT7) has been stably expressed in BHK-21 cells together with human beta-trace protein (beta-TP) as a secretory reporter glycoprotein. In order to study their in vivo properties for the transfer of peripheral Fuc onto N-linked complex-type glycans, detailed structural analysis was performed on the purified glycoprotein. All fucosyltransferases were found to peripherally fucosylate 19-52% of the diantennary beta-TP N-glycans, and all enzymes were capable of synthesizing the sialyl LewisX (sLex) motif. However, each enzyme produced its own characteristic ratio of sLex/Lex antennae as follows: FT7 (only sLex), FT3 (14:1), FT5 (3:1), FT6 (1.1:1), and FT4 (1:7). Fucose transfer onto beta-TP N-glycans was low in FT3 cells (11% of total antennae), whereas the values for FT7, FT5, FT4, and FT6 cells were 21, 25, 35, and 47%, respectively. FT3, FT4, FT5, and FT7 transfer preponderantly one Fuc per diantennary N-glycan. FT4 preferentially synthesizes di-Lex on asialo diantennary N-glycans and mono-Lex with monosialo chains. In contrast, FT6 forms mostly alpha1,3-difucosylated chains with no, one, or two NeuAc residues. FT3, FT4, and FT6 were proteolytically cleaved and released into the culture medium in significant amounts, whereas FT7 and FT5 were found to be largely resistant toward proteolysis. Studies on engineered soluble variants of FT6 indicate that these forms do not significantly contribute to the in vivo fucose transfer activity of the enzyme when expressed at activity levels comparable to those obtained for the wild-type Golgi form of FT6 in the recombinant host cells.