Cloning and expression of human sialic acid pathway genes to generate CMP-sialic acids in insect cells

Strategies for Glycoengineering Therapeutic Proteins

Almost all therapeutic proteins are glycosylated, with the carbohydrate component playing a long-established, substantial role in the safety and pharmacokinetic properties of this dominant category of drugs. In the past few years and moving forward, glycosylation is increasingly being implicated in the pharmacodynamics and therapeutic efficacy of therapeutic proteins. This article provides illustrative examples of drugs that have already been improved through glycoengineering including cytokines exemplified by erythropoietin (EPO), enzymes (ectonucleotide pyrophosphatase 1, ENPP1), and IgG antibodies (e.g., afucosylated Gazyva^®, Poteligeo^®, Fasenra™, and Uplizna^®). In the future, the deliberate modification of therapeutic protein glycosylation will become more prevalent as glycoengineering strategies, including sophisticated computer-aided tools for “building in” glycans sites, acceptance of a broad range of production systems with various glycosylation capabilities, and supplementation methods for introducing non-natural metabolites into glycosylation pathways further develop and become more accessible.

Evolutionary conservation of human ketodeoxynonulosonic acid production is independent of sialoglycan biosynthesis

Human metabolic incorporation of nonhuman sialic acid (Sia) N-glycolylneuraminic acid into endogenous glycans generates inflammation via preexisting antibodies, which likely contributes to red meat-induced atherosclerosis acceleration. Exploring whether this mechanism affects atherosclerosis in end-stage renal disease (ESRD), we instead found serum accumulation of 2-keto-3-deoxy-d-glycero-d-galacto-2-nonulosonic acid (Kdn), a Sia prominently expressed in cold-blooded vertebrates. In patients with ESRD, levels of the Kdn precursor mannose also increased, but within a normal range. Mannose ingestion by healthy volunteers raised the levels of urinary mannose and Kdn. Kdn production pathways remained conserved in mammals but were diminished by an M42T substitution in a key biosynthetic enzyme, N-acetylneuraminate synthase. Remarkably, reversion to the ancestral methionine then occurred independently in 2 lineages, including humans. However, mammalian glycan databases contain no Kdn-glycans. We hypothesize that the potential toxicity of excess mannose in mammals is partly buffered by conversion to free Kdn. Thus, mammals probably conserve Kdn biosynthesis and modulate it in a lineage-specific manner, not for glycosylation, but to control physiological mannose intermediates and metabolites. However, human cells can be forced to express Kdn-glycans via genetic mutations enhancing Kdn utilization, or by transfection with fish enzymes producing cytidine monophosphate-Kdn (CMP-Kdn). Antibodies against Kdn-glycans occur in pooled human immunoglobulins. Pathological conditions that elevate Kdn levels could therefore result in antibody-mediated inflammatory pathologies.

Defects in Galactose Metabolism and Glycoconjugate Biosynthesis in a UDP-Glucose Pyrophosphorylase-Deficient Cell Line Are Reversed by Adding Galactose to the Growth Medium

UDP-glucose (UDP-Glc) is synthesized by UGP2-encoded UDP-Glc pyrophosphorylase (UGP) and is required for glycoconjugate biosynthesis and galactose metabolism because it is a uridyl donor for galactose-1-P (Gal1P) uridyltransferase. Chinese hamster lung fibroblasts harboring a hypomrphic UGP(G116D) variant display reduced UDP-Glc levels and cannot grow if galactose is the sole carbon source. Here, these cells were cultivated with glucose in either the absence or presence of galactose in order to investigate glycoconjugate biosynthesis and galactose metabolism. The UGP-deficient cells display < 5% control levels of UDP-Glc/UDP-Gal and > 100-fold reduction of [6-³H]galactose incorporation into UDP-[6-³H]galactose, as well as multiple deficits in glycoconjugate biosynthesis. Cultivation of these cells in the presence of galactose leads to partial restoration of UDP-Glc levels, galactose metabolism and glycoconjugate biosynthesis. The V_max for recombinant human UGP(G116D) with Glc1P is 2000-fold less than that of the wild-type protein, and UGP(G116D) displayed a mildly elevated K_m for Glc1P, but no activity of the mutant enzyme towards Gal1P was detectable. To conclude, although the mechanism behind UDP-Glc/Gal production in the UGP-deficient cells remains to be determined, the capacity of this cell line to change its glycosylation status as a function of extracellular galactose makes it a useful, reversible model with which to study different aspects of galactose metabolism and glycoconjugate biosynthesis.

A homozygous mutation in CMAS causes autosomal recessive intellectual disability in a Kazakh family

Intellectual disability (ID) describes a wide range of serious human diseases caused by defects in central nervous system development and function. Some mutant genes have been found to be associated with these diseases, but not all cases can be explained, thus suggesting that other disease-causing genes have not yet been discovered. Sialic acid is involved in a number of key biological processes, including embryo formation, nerve cell growth, and cancer cell metastasis, and very recently it has been suggested that N-acetylneuraminic acid synthase-mediated synthesis of sialic acid is required for brain and skeletal development. CMP-sialic acid synthetase (CMAS) is one of four enzymes involved in NeuNAc metabolism, as it catalyzes the formation of CMP-NeuNAc. Before the present study, no links between mutations in CMAS and incidences of human ID had been reported. In the current study, we recruited a recessive nonsyndromic ID pedigree with consanguineous marriage in which all patients have typical clinical manifestations of ID. We identified the NM_018686.3:c.563G > A (p.Arg188His) substitution in CMAS as being responsible for the disease in this family. Conservation analysis, structural prediction, and enzyme activity experiments demonstrated that (p.Arg188His) influences protein dimerization and alters CMAS enzyme activity. Our results offer a new orientation for future research and clinical diagnosis.

Structural and functional characterization of CMP-N-acetylneuraminate synthetase from Vibrio cholerae

CMP-N-acetylneuraminate synthetase (CMAS) is a key enzyme in the sialic acid incorporation pathway and plays a crucial role in the virulence and survival of several pathogenic bacteria. Here, the structural and functional properties of CMAS from the pathogenic bacterium Vibrio cholerae are reported. Upon CDP binding, a partial domain closure is observed that was previously unreported in homologous structures. Kinetic studies reveal that the enzyme shows substrate promiscuity and can activate both Neu5Ac and Neu5Gc.

Several pathogenic bacteria utilize sialic acid, including host-derived N-acetylneuraminic acid (Neu5Ac), in at least two ways: they use it as a nutrient source and as a host-evasion strategy by coating themselves with Neu5Ac. Given the significant role of sialic acid in pathogenesis and host-gut colonization by various pathogenic bacteria, including Neisseria meningitidis, Haemophilus influenzae, Pasteurella multocida and Vibrio cholerae, several enzymes of the sialic acid catabolic, biosynthetic and incorporation pathways are considered to be potential drug targets. In this work, findings on the structural and functional characterization of CMP-N-acetylneuraminate synthetase (CMAS), a key enzyme in the incorporation pathway, from Vibrio cholerae are reported. CMAS catalyzes the synthesis of CMP-sialic acid by utilizing CTP and sialic acid. Crystal structures of the apo and the CDP-bound forms of the enzyme were determined, which allowed the identification of the metal cofactor Mg²⁺ in the active site interacting with CDP and the invariant Asp215 residue. While open and closed structural forms of the enzyme from eukaryotic and other bacterial species have already been characterized, a partially closed structure of V. cholerae CMAS (VcCMAS) observed upon CDP binding, representing an intermediate state, is reported here. The kinetic data suggest that VcCMAS is capable of activating the two most common sialic acid derivatives, Neu5Ac and Neu5Gc. Amino-acid sequence and structural comparison of the active site of VcCMAS with those of eukaryotic and other bacterial counterparts reveal a diverse hydrophobic pocket that interacts with the C5 substituents of sialic acid. Analyses of the thermodynamic signatures obtained from the binding of the nucleotide (CTP) and the product (CMP-sialic acid) to VcCMAS provide fundamental information on the energetics of the binding process.

Improving Immunotherapy Through Glycodesign

Immunotherapy is revolutionizing health care, with the majority of high impact "drugs" approved in the past decade falling into this category of therapy. Despite considerable success, glycosylation-a key design parameter that ensures safety, optimizes biological response, and influences the pharmacokinetic properties of an immunotherapeutic-has slowed the development of this class of drugs in the past and remains challenging at present. This article describes how optimizing glycosylation through a variety of glycoengineering strategies provides enticing opportunities to not only avoid past pitfalls, but also to substantially improve immunotherapies including antibodies and recombinant proteins, and cell-based therapies. We cover design principles important for early stage pre-clinical development and also discuss how various glycoengineering strategies can augment the biomanufacturing process to ensure the overall effectiveness of immunotherapeutics.

Glycoengineering of CHO Cells to Improve Product Quality

Chinese hamster ovary (CHO) cells represent the predominant platform in biopharmaceutical industry for the production of recombinant biotherapeutic proteins, especially glycoproteins. These glycoproteins include oligosaccharide or glycan attachments that represent one of the principal components dictating product quality. Especially important are the N-glycan attachments present on many recombinant glycoproteins of commercial interest. Furthermore, altering the glycan composition can be used to modulate the production quality of a recombinant biotherapeutic from CHO and other mammalian hosts. This review first describes the glycosylation network in mammalian cells and compares the glycosylation patterns between CHO and human cells. Next genetic strategies used in CHO cells to modulate the sialylation patterns through overexpression of sialyltransfereases and other glycosyltransferases are summarized. In addition, other approaches to alter sialylation including manipulation of sialic acid biosynthetic pathways and inhibition of sialidases are described. Finally, this review also covers other strategies such as the glycosylation site insertion and manipulation of glycan heterogeneity to produce desired glycoforms for diverse biotechnology applications.

CMP-Sialic Acid Synthetase: The Point of Constriction in the Sialylation Pathway

Sialoglycoconjugates form the outermost layer of animal cells and play a crucial role in cellular communication processes. An essential step in the biosynthesis of sialylated glycoconjugates is the activation of sialic acid to the monophosphate diester CMP-sialic acid. Only the activated sugar is transported into the Golgi apparatus and serves as a substrate for the linkage-specific sialyltransferases. Interference with sugar activation abolishes sialylation and is embryonic lethal in mammals. In this chapter we focus on the enzyme catalyzing the activation of sialic acid, the CMP-sialic acid synthetase (CMAS), and compare the enzymatic properties of CMASs isolated from different species. Information concerning the reaction mechanism and active site architecture is included. Moreover, the unusual nuclear localization of vertebrate CMASs as well as the biotechnological application of bacterial CMAS enzymes is addressed.

Sialylation potentials of the silkworm, Bombyx mori; B. mori possesses an active α2,6-sialyltransferase

N-Glycosylation is an important post-translational modification in most secreted and membrane-bound proteins in eukaryotic cells. However, the insect N-glycosylation pathway and the potentials contributing to the N-glycan synthesis are still unclear because most of the studies on these subjects have focused on mammals and plants. Here, we identified Bombyx mori sialyltransferase (BmST), which is a Golgi-localized glycosyltransferase and which can modify N-glycans. BmST was ubiquitously expressed in different organs and in various stages of development and localized at the Golgi. Biochemical analysis using Sf9-expressed BmST revealed that BmST encoded α2,6-sialyltransferase and transferred N-acetylneuraminic acid (NeuAc) to the nonreducing terminus of Galβ1-R, but exhibited the highest activity toward GalNAcβ1,4-GlcNAc-R. Unlike human α2,6-sialyltransferase, BmST required the post-translational modification, especially N-glycosylation, for its full activity. N-Glycoprotein analysis of B. mori fifth instar larvae revealed that high-mannose-type structure was predominant and GlcNAc-linked and fucosylated structures were observed but endogenous galactosyl-, N-acetylgalactosaminyl- and sialyl-N-glycoproteins were undetectable under the standard analytical approach. These results indicate that B. mori genome encodes an α2,6-sialyltransferase, but further investigations of the sialylation potentials are necessary.

Glycoengineering of Chinese hamster ovary cells for enhanced erythropoietin N-glycan branching and sialylation

Sialic acid, a terminal residue on complex N-glycans, and branching or antennarity can play key roles in both the biological activity and circulatory lifetime of recombinant glycoproteins of therapeutic interest. In order to examine the impact of glycosyltransferase expression on the N-glycosylation of recombinant erythropoietin (rEPO), a human α2,6-sialyltransferase (ST6Gal1) was expressed in Chinese hamster ovary (CHO-K1) cells. Sialylation increased on both EPO and CHO cellular proteins as observed by SNA lectin analysis, and HPLC profiling revealed that the sialic acid content of total glycans on EPO increased by 26%. The increase in sialic acid content was further verified by detailed profiling of the N-glycan structures using mass spectra (MS) analysis. In order to enhance antennarity/branching, UDP-N-acetylglucosamine: α-1,3-D-mannoside β1,4-N-acetylglucosaminyltransferase (GnTIV/Mgat4) and UDP-N-acetylglucosamine:α-1,6-D-mannoside β1,6-N-acetylglucosaminyltransferase (GnTV/Mgat5), was incorporated into CHO-K1 together with ST6Gal1. Tri- and tetraantennary N-glycans represented approximately 92% of the total N-glycans on the resulting EPO as measured using MS analysis. Furthermore, sialic acid content of rEPO from these engineered cells was increased ∼45% higher with tetra-sialylation accounting for ∼10% of total sugar chains compared to ∼3% for the wild-type parental CHO-K1. In this way, coordinated overexpression of these three glycosyltransferases for the first time in model CHO-K1 cell lines provides a mean for enhancing both N-glycan branching complexity and sialylation with opportunities to generate tailored complex N-glycan structures on therapeutic glycoproteins in the future.

Strategies for Engineering Protein N-Glycosylation Pathways in Mammalian Cells

Complexity and heterogeneity of oligosaccharides present a considerable challenge to the biopharmaceutical industry to manufacture biotherapeutics with reproducible and consistent glycoform profiles. Mammalian cells, especially Chinese hamster ovary cells, are the most widely used platform for the production of biotherapeutics. The glycans produced are predominantly of the complex type, with some differences between human and nonhuman mammalian glycosylation existing. This review briefly summarizes metabolic glyco-engineering strategies used in mammalian cells in order to alter the glycosylation patterns attached to proteins applied for diverse biotechnology applications.

The transcriptome of the baculovirus Autographa californica multiple nucleopolyhedrovirus in Trichoplusia ni cells

Baculoviruses are important insect pathogens that have been developed as protein expression vectors in insect cells and as transduction vectors for mammalian cells. They have large double-stranded DNA genomes containing approximately 156 tightly spaced genes, and they present significant challenges for transcriptome analysis. In this study, we report the first comprehensive analysis of AcMNPV transcription over the course of infection in Trichoplusia ni cells, by a combination of strand-specific RNA sequencing (RNA-Seq) and deep sequencing of 5' capped transcription start sites and 3' polyadenylation sites. We identified four clusters of genes associated with distinctive patterns of mRNA accumulation through the AcMNPV infection cycle. A total of 218 transcription start sites (TSS) and 120 polyadenylation sites (PAS) were mapped. Only 29 TSS were associated with a canonical TATA box, and 14 initiated within or near the previously identified CAGT initiator motif. The majority of viral transcripts (126) initiated within the baculovirus late promoter motif (TAAG), and late transcripts initiated precisely at the second position of the motif. Analysis of 3' ends showed that 92 (77%) of the 3' PAS were located within 30 nucleotides (nt) downstream of a consensus termination signal (AAUAAA or AUUAAA). A conserved U-rich region was found approximately 2 to 10 nt downstream of the PAS for 58 transcripts. Twelve splicing events and an unexpectedly large number of antisense RNAs were identified, revealing new details of possible regulatory mechanisms controlling AcMNPV gene expression. Combined, these data provide an emerging global picture of the organization and regulation of AcMNPV transcription through the infection cycle.

Identification and biochemical characterization of two functional CMP-sialic acid synthetases in Danio rerio

Sialic acids (Sia) form the nonreducing end of the bulk of cell surface-expressed glycoconjugates. They are, therefore, major elements in intercellular communication processes. The addition of Sia to glycoconjugates requires metabolic activation to CMP-Sia, catalyzed by CMP-Sia synthetase (CMAS). This highly conserved enzyme is located in the cell nucleus in all vertebrates investigated to date, but its nuclear function remains elusive. Here, we describe the identification and characterization of two Cmas enzymes in Danio rerio (dreCmas), one of which is exclusively localized in the cytosol. We show that the two cmas genes most likely originated from the third whole genome duplication, which occurred at the base of teleost radiation. cmas paralogues were maintained in fishes of the Otocephala clade, whereas one copy got subsequently lost in Euteleostei (e.g. rainbow trout). In zebrafish, the two genes exhibited a distinct spatial expression pattern. The products of these genes (dreCmas1 and dreCmas2) diverged not only with respect to subcellular localization but also in substrate specificity. Nuclear dreCmas1 favored N-acetylneuraminic acid, whereas the cytosolic dreCmas2 showed highest affinity for 5-deamino-neuraminic acid. The subcellular localization was confirmed for the endogenous enzymes in fractionated zebrafish lysates. Nuclear entry of dreCmas1 was mediated by a bipartite nuclear localization signal, which seemed irrelevant for other enzymatic functions. With the current demonstration that in zebrafish two subfunctionalized cmas paralogues co-exist, we introduce a novel and unique model to detail the roles that CMAS has in the nucleus and in the sialylation pathways of animal cells.

Co-expression vs. co-infection using baculovirus expression vectors in insect cell culture: Benefits and drawbacks

The baculovirus expression vector system (BEVS) is a versatile and powerful platform for protein expression in insect cells. With the ability to approach similar post-translational modifications as in mammalian cells, the BEVS offers a number of advantages including high levels of expression as well as an inherent safety during manufacture and of the final product. Many BEVS products include proteins and protein complexes that require expression from more than one gene. This review examines the expression strategies that have been used to this end and focuses on the distinguishing features between those that make use of single polycistronic baculovirus (co-expression) and those that use multiple monocistronic baculoviruses (co-infection). Three major areas in which researchers have been able to take advantage of co-expression/co-infection are addressed, including compound structure-function studies, insect cell functionality augmentation, and VLP production. The core of the review discusses the parameters of interest for co-infection and co-expression with time of infection (TOI) and multiplicity of infection (MOI) highlighted for the former and the choice of promoter for the latter. In addition, an overview of modeling approaches is presented, with a suggested trajectory for future exploration. The review concludes with an examination of the gaps that still remain in co-expression/co-infection knowledge and practice.

Cerebellar ataxia with elevated cerebrospinal free sialic acid (CAFSA)

In order to identify new metabolic abnormalities in patients with complex neurodegenerative disorders of unknown aetiology, we performed high resolution in vitro proton nuclear magnetic resonance spectroscopy on patient cerebrospinal fluid (CSF) samples. We identified five adult patients, including two sisters, with significantly elevated free sialic acid in the CSF compared to both the cohort of patients with diseases of unknown aetiology (n = 144; P < 0.001) and a control group of patients with well-defined diseases (n = 91; P < 0.001). All five patients displayed cerebellar ataxia, with peripheral neuropathy and cognitive decline or noteworthy behavioural changes. Cerebral MRI showed mild to moderate cerebellar atrophy (5/5) as well as white matter abnormalities in the cerebellum including the peridentate region (4/5), and at the periventricular level (3/5). Two-dimensional gel analyses revealed significant hyposialylation of transferrin in CSF of all patients compared to age-matched controls (P < 0.001)--a finding not present in the CSF of patients with Salla disease, the most common free sialic acid storage disorder. Free sialic acid content was normal in patients' urine and cultured fibroblasts as were plasma glycosylation patterns of transferrin. Analysis of the ganglioside profile in peripheral nerve biopsies of two out of five patients was also normal. Sequencing of four candidate genes in the free sialic acid biosynthetic pathway did not reveal any mutation. We therefore identified a new free sialic acid syndrome in which cerebellar ataxia is the leading symptom. The term CAFSA is suggested (cerebellar ataxia with free sialic acid).

Serum-free transient protein production system based on adenoviral vector and PER.C6 technology: high yield and preserved bioactivity

Stable E1 transformed cells, like PER.C6, are able to grow at scale and to high cell densities. E1-deleted adenoviruses replicate to high titer in PER.C6 cells whereas subsequent deletion of E2A from the vector results in absence of replication in PER.C6 cells and drastically lowers the expression of adenovirus proteins in such cells. We therefore considered the use of an DeltaE1/DeltaE2 type 5 vector (Ad5) to deliver genes to PER.C6 cells growing in suspension with the aim to achieve high protein yield. To evaluate the utility of this system we constructed DeltaE1/DeltaE2 vector carrying different classes of protein, that is, the gene coding for spike protein derived from the Coronavirus causing severe acute respiratory syndrome (SARS-CoV), a gene coding for the SARS-CoV receptor or the genes coding for an antibody shown to bind and neutralize SARS-CoV (SARS-AB). The DeltaE1/DeltaE2A-vector backbones were rescued on a PER.C6 cell line engineered to constitutively over express the Ad5 E2A protein. Exposure of PER.C6 cells to low amounts (30 vp/cell) of DeltaE1/DeltaE2 vectors resulted in highly efficient (>80%) transduction of PER.C6 cells growing in suspension. The efficient cell transduction resulted in high protein yield (up to 60 picogram/cell/day) in a 4 day batch production protocol. FACS and ELISA assays demonstrated the biological activity of the transiently produced proteins. We therefore conclude that DeltaE1/DeltaE2 vectors in combination with the PER.C6 technology may provide a viable answer to the increasing demand for high quality, high yield recombinant protein.

The Central Cavity from the (Alpha/Alpha)6 Barrel Structure of Anabaena sp. CH1 N-Acetyl-d-glucosamine 2-Epimerase Contains Two Key Histidine Residues for Reversible Conversion

N-acetyl-D-glucosamine 2-epimerase (GlcNAc 2-epimerase) catalyzes the reversible epimerization between N-acetyl-D-glucosamine (GlcNAc) and N-acetyl-D-mannosamine (ManNAc). We report here the 2.0 A resolution crystal structure of the GlcNAc 2-epimerase from Anabaena sp. CH1. The structure demonstrates an (alpha/alpha)(6) barrel fold, which shows structural homology with porcine GlcNAc 2-epimerase as well as a number of glycoside hydrolase enzymes and other sugar-metabolizing enzymes. One side of the barrel structure consists of short loops involved in dimer interactions. The other side of the barrel structure is comprised of long loops containing six short beta-sheets, which enclose a putative central active-site pocket. Site-directed mutagenesis of conserved residues near the N-terminal region of the inner alpha helices shows that R57, H239, E308, and H372 are strictly required for activity. E242 and R375 are also essential in catalysis. Based on the structure and kinetic analysis, H239 and H372 may serve as the key active site acid/base catalysts. These results suggest that the (alpha/alpha)(6) barrel represents a steady fold for presenting active-site residues in a cleft at the N-terminal ends of the inner alpha helices, thus forming a fine-tuned catalytic site in GlcNAc 2-epimerase.

Identification of the nuclear export signals that regulate the intracellular localization of the mouse CMP-sialic acid synthetase

The CMP-sialic acid synthetase (CSS) catalyzes the activation of sialic acid (Sia) to CMP-Sia which is a donor substrate of sialyltransferases. The vertebrate CSSs are usually localized in nucleus due to the nuclear localization signal (NLS) on the molecule. In this study, we first point out that a small, but significant population of the mouse CMP-sialic acid synthetase (mCSS) is also present in cytoplasm, though mostly in nucleus. As a mechanism for the localization in cytoplasm, we first identified two nuclear export signals (NESs) in mCSS, based on the localization studies of the potential NES-deleted mCSS mutants as well as the potential NES-tagged eGFP proteins. These two NESs are conserved among mammalian and fish CSSs, but not present in the bacterial or insect CSS. These results suggest that the intracellular localization of vertebrate CSSs is regulated by not only the NLS, but also the NES sequences.

Protein N-glycosylation in the baculovirus-insect cell expression system and engineering of insect cells to produce "mammalianized" recombinant glycoproteins

Baculovirus expression vectors are frequently used to express glycoproteins, a subclass of proteins that includes many products with therapeutic value. The insect cells that serve as hosts for baculovirus vector infection are capable of transferring oligosaccharide side chains (glycans) to the same sites in recombinant proteins as those that are used for native protein N-glycosylation in mammalian cells. However, while mammalian cells produce compositionally more complex N-glycans containing terminal sialic acids, insect cells mostly produce simpler N-glycans with terminal mannose residues. This structural difference between insect and mammalian N-glycans compromises the in vivo bioactivity of glycoproteins and can potentially induce allergenic reactions in humans. These features obviously compromise the biomedical value of recombinant glycoproteins produced in the baculovirus expression vector system. Thus, much effort has been expended to characterize the potential and limits of N-glycosylation in insect cell systems. Discoveries from this research have led to the engineering of insect N-glycosylation pathways for assembly of mammalian-style glycans on baculovirus-expressed glycoproteins. This chapter summarizes our knowledge of insect N-glycosylation pathways and describes efforts to engineer baculovirus vectors and insect cell lines to overcome the limits of insect cell glycosylation. In addition, we consider other possible strategies for improving glycosylation in insect cells.

Isolation and analysis of a baculovirus vector that supports recombinant glycoprotein sialylation by SfSWT-1 cells cultured in serum-free medium

The inability to sialylate recombinant glycoproteins is a critical limitation of the baculovirus-insect cell expression system. This limitation is due, at least in part, to the absence of detectable sialyltransferase activities and CMP-sialic acids in the insect cell lines routinely used as hosts in this system. SfSWT-1 is a transgenic insect cell line encoding five mammalian glycosyltransferases, including sialyltransferases, which can contribute to sialylation of recombinant glycoproteins expressed by baculovirus vectors. However, sialylation of recombinant glycoproteins requires culturing SfSWT-1 cells in the presence of fetal bovine serum or another exogenous source of sialic acid. To eliminate this requirement and extend the utility of SfSWT-1 cells, we have isolated a new baculovirus vector, AcSWT-7B, designed to express two mammalian enzymes that can convert N-acetylmannosamine to CMP-sialic acid during the early phase of infection. AcSWT-7B was also designed to express a model recombinant glycoprotein during the very late phase of infection. Characterization of this new baculovirus vector showed that it induced high levels of intracellular CMP-sialic acid and sialylation of the recombinant N-glycoprotein upon infection of SfSWT-1 cells cultured in serum-free medium supplemented with N-acetylmannosamine. In addition, co-infection of SfSWT-1 cells with AcSWT-7B plus a conventional baculovirus vector encoding human tissue plasminogen activator resulted in sialylation of this recombinant N-glycoprotein under the same culture conditions. These results demonstrate that AcSWT-7B can be used in two different ways to support recombinant N-glycoprotein sialylation by SfSWT-1 cells in serum-free medium. Thus, AcSWT-7B can be used to extend the utility of this previously described transgenic insect cell line for recombinant sialoglycoprotein production.

Identification and partial characterization of soluble and membrane-bound KDN(deaminoneuraminic acid)-glycoproteins in human ovarian teratocarcinoma PA-1, and enhanced expression of free and bound KDN in cells cultured in mannose-rich media

KDN (Deaminoneuraminic acid, or deaminated neuraminic acid) is a minor but biosynthetically independent member of the sialic acid. Human occurrence of KDN has already been established, although its level is so little that it is often undetectable by conventional sialic acid analysis. Elevated expression of KDN in fetal cord blood cells and some malignant tumor cells have been reported. However, in mammalian cells and tissues KDN mostly occurs as the free sugar and little occurred conjugated to glycolipids and/or glycoproteins. A positive correlation between the ratio of free KDN/free Neu5Ac in ovarian adenocarcinomas and the stage of malignancy has been noted for diagnostic use. We hypothesized that elevated expression of KDN in mammalian systems may be closely related to elevated activities of enzymes involved in the formation of sialoglycoconjugates and/or aberrant supply of the precursor sugar, mannose, used in the biosynthesis of KDN. In this study we used human ovarian teratocarcinoma cells PA-1 to further analyze KDN expression in human cells. Major findings reported in this paper are, (i) a 30 kDa KDN-glycoprotein immunostainable with monoclonal antibody, mAb.kdn3G, (specific for the KDNalpha2 --> 3Galbeta1--> epitope) and sensitive to KDNase was identified in the membrane fraction of the cell: (ii) a 49 kDa KDN-glycoprotein that is not reactive with mAb.kdn3G but is sensitive to KDNase was identified in the soluble fraction: and (iii) PA-1 cells showed unique response to mannose added to the growth medium in that the levels of both free and bound forms of KDN are elevated. This is the first report on the identification of mammalian KDN-glycoproteins by chemical and biochemical methods.

KDN (Deaminated neuraminic acid): Dreamful past and exciting future of the newest member of the sialic acid family

KDN is an abbreviation for 2-keto-3-deoxy-D-glycero-D-galacto-nononic acid, and its natural occurrence was revealed in 1986 by a research group including the present authors. Since sialic acid was used as a synonym for N-acylneuraminic acid at that time, there was an argument if this deaminated neuraminic acid belongs to the family of sialic acids. In this review, we describe the 20 years history of studies on KDN (KDNology), through which KDN has established its position as a distinct member of the sialic acid family. These studies have clarified that: (1) KDN occurs widely among vertebrates and bacteria similar to the occurrence of the more common sialic acid, N-acetylneuraminic acid (Neu5Ac), but its abundant occurrence in animals is limited to lower vertebrates. (2) KDN is found in almost all types of glycoconjugates, including glycolipids, glycoproteins and capsular polysaccharides. (3) KDN residues are linked to almost all glycan structures in place of Neu5Ac. All linkage types known for Neu5Ac; alpha2,3-, alpha2,4-, alpha2,6-, and alpha2,8- are also found for KDN. (4) KDN is biosynthesized de novo using mannose as a precursor sugar, which is activated to CMP-KDN and transferred to acceptor sugar residues. These reactions are catalyzed by enzymes, some of which preferably recognize KDN, but many others prefer Neu5Ac to KDN. In addition to these basic findings, elevated expression of KDN was found in fetal human red blood cells compared with adult red blood cells, and ovarian tumor tissues compared with normal controls. KDNase, an enzyme which specifically cleaves KDN-linkages, was discovered in a bacterium and monoclonal antibodies that specifically recognize KDN residues in KDNalpha2,3-Gal- and KDNalpha2,8-KDN-linkages have been developed. These have been used for identification of KDN-containing molecules. Based on past basic studies and variety of findings, future perspective of KDNology is presented.

Expression of a functional Drosophila melanogaster CMP-sialic acid synthetase. Differential localization of the Drosophila and human enzymes

CMP-N-acetylneuraminic acid is a critical metabolite in the generation of glycoconjugates that play a role in development and other physiological processes. Whereas pathways for its generation are firmly established in vertebrates, the presence and function of the relevant synthetic enzyme in insects and other protostomes is unknown. In this study, we characterize the first functional CMP-sialic acid synthase (DmCSAS) from any protostome lineage expressed from a D. melanogaster cDNA clone. Homologous genes were subsequently identified in other insect species. The gene is developmentally regulated, with expression first appearing at 12-24 h of embryogenesis, low expression through larval and pupal stages, and greatly enriched expression in the adult head, suggesting a possible role in the central nervous system. Activity of the enzyme was verified by an increase in in vitro and in vivo CMP-N-acetylneuraminic acid levels when expressed in a heterologous host. Unlike all known vertebrate CMP-sialic acid synthetase (CSAS) proteins that localize to the nucleus, the D. melanogaster CSAS protein was targeted to the Golgi compartment when expressed in both heterologous mammalian and insect cell lines. Replacement of the N-terminal leader sequence of DmCSAS with the human CSAS N-terminal sequence resulted in the redirection of the chimeric CSAS protein to the nucleus but with a concomitant loss of enzymatic activity. The localization of CSAS orthologs to different intracellular organelles represents, to our knowledge, the first example of differential protein targeting of orthologs in eukaryotes and reveals how the sialylation pathway diverged during the evolution of protostomes and deuterostomes.

Production and N-glycan analysis of secreted human erythropoietin glycoprotein in stably transfected Drosophila S2 cells

Schneider 2 (S2) cells from Drosophila melanogaster have been used as a plasmid-based, non-lytic expression system for foreign proteins. Here, a plasmid encoding the human erythropoietin (hEPO) gene fused with a hexahistidine (His(6)) tag under the control of the Drosophila metallothionein (MT) promoter was stably transfected into Drosophila S2 cells. After copper sulfate induction, transfected S2 cells were found to secrete hEPO with a maximum expression level of 18 mg/L and a secretion efficiency near 98%. The secreted hEPO from Drosophila S2 had an apparent molecular weight of about 23-27 kDa which was significantly lower than a recombinant hEPO expressed in Chinese hamster ovary (CHO) cells (about 36 kDa). N-glycosidase F digestion almost completely eliminated the difference and resulted in the same molecular weight ( approximately 20 kDa) of de-N-glycosylated hEPO proteins. These data suggest that recombinant hEPO from S2 cells was modified with smaller N-glycans. Subsequently, the major N-glycans were identified following glycoamidase A digestion, labeling with 2-aminopyridine (PA), and two-dimensional high-performance liquid chromatography (HPLC) analysis in concert with exoglycosidase digestion. This analysis of N-glycans revealed that hEPO was modified to include paucimannosidic glycans containing two or three mannose residues with or without core fucose. A similar glycosylation pattern was observed on a recombinant human transferrin expressed in S2 cells. These results provide a detailed analysis of multiple N-glycan structures produced in a Drosophila cell line that will be useful in the subsequent application of these cells for the generation of heterologous glycoproteins.

Engineering intracellular CMP-sialic acid metabolism into insect cells and methods to enhance its generation

Previous studies have reported that insect cell lines lack the capacity to generate endogenously the nucleotide sugar, CMP-Neu5Ac, required for sialylation of glycoconjugates. In this study, the biosynthesis of this activated form of sialic acid completely from endogenous metabolites is demonstrated for the first time in insect cells by expressing the mammalian genes required for the multistep conversion of endogenous UDP-GlcNAc to CMP-Neu5Ac. The genes for UDP-GlcNAc-2-epimerase/ManNAc kinase (EK), sialic acid 9-phosphate synthase (SAS), and CMP-sialic acid synthetase (CSAS) were coexpressed in insect cells using baculovirus expression vectors, but the CMP-Neu5Ac and precursor Neu5Ac levels synthesized were found to be lower than those achieved with ManNAc supplementation due to feedback inhibition of the EK enzyme by CMP-Neu5Ac. When sialuria-like mutant EK genes, in which the site for feedback regulation has been mutated, were used, CMP-Neu5Ac was synthesized at levels more than 4 times higher than that achieved with the wild-type EK and 2.5 times higher than that achieved with ManNAc feeding. Addition of N-acetylglucosamine (GlcNAc), a precursor for UDP-GlcNAc, to the media increased the levels of CMP-Neu5Ac even more to a level 7.5 times higher than that achieved with ManNAc supplementation, creating a bottleneck in the conversion of Neu5Ac to CMP-Neu5Ac at higher levels of UDP-GlcNAc. The present study provides a useful biochemical strategy to synthesize and enhance the levels of the sialylation donor molecule, CMP-Neu5Ac, a critical limiting substrate for the generation of complex glycoproteins in insect cells and other cell culture systems.

Development of a simple and efficient method for assaying cytidine monophosphate sialic acid synthetase activity using an enzymatic reduced nicotinamide adenine dinucleotide/oxidized nicotinamide adenine dinucleotide converting system

A new reliable method to assay the activity of cytidine monophosphate sialic acid (CMP-Sia) synthetase (CSS) has been developed. The activation of sialic acids (Sia) to CMP-Sia is a prerequisite for the de novo synthesis of sialoglycoconjugates. In vertebrates, CSS has been cloned from human, mouse, and rainbow trout, and the crystal structure has been resolved for the mouse enzyme. The mouse and rainbow trout enzyme have been compared with respect to substrate specificity, demonstrating that the mouse enzyme exhibits a pronounced specificity for N-acetylneuraminic acid (Neu5Ac), while the rainbow trout CSS is equally active with either of three Sia species, Neu5Ac, N-glycolylneuraminic acid (Neu5Gc), and deaminoneuraminic acid (KDN). However, molecular details that explain the pronounced substrate specificities are unknown. Understanding the catalytic mechanisms of these enzymes is of major importance, since CSSs play crucial roles in cellular sialylation patterns and thus are potential drug targets in a number of pathophysiological situations. The availability of the cDNAs and the obtained structural data enable rational approaches; however, these efforts are limited by the lack of a reliable high-throughput assay system. Here we describe a new assay system that allows product quantification in a reduced nicotinamide adenine dinucleotide (NADH)-dependent color reaction. The activation reaction catalyzed by CSS, CTP+Sia-->CMP-Sia+pyrophosphate, was evaluated by a consumption of Sia, which corresponds to that of NADH on the following two successive reactions: (i) Sia-->pyruvate+ManNAc (or Man), catalyzed by a sialic acid lyase (SAL), and (ii) pyruvate+NADH-->lactate+oxidized nicotinamide adenine dinucleotide (NAD+), catalyzed by a lactate dehydrogenase (LDH). Consumption of NADH can be photometrically monitored on a microtiter plate reader for a number of test samples at the same time. Furthermore, based on the quantification of CSS used in the SAL/LDH assay, relative activities toward Sia derivatives have been obtained. The preference of mouse CSS toward Neu5Ac and the ability of the rainbow trout enzyme to activate both KDN and Neu5Ac were confirmed. Thus, this simple and time-saving method is suitable for a systematic comparison of enzyme activity of structurally mutated enzymes based on the relative specific activity.

Comparing N-glycan processing in mammalian cell lines to native and engineered lepidopteran insect cell lines

In the past decades, a large number of studies in mammalian cells have revealed that processing of glycoproteins is compartmentalized into several subcellular organelles that process N-glycans to generate complex-type oligosaccharides with terminal N -acetlyneuraminic acid. Recent studies also suggested that processing of N-glycans in insect cells appear to follow a similar initial pathway but diverge at subsequent processing steps. N-glycans from insect cell lines are not usually processed to terminally sialylated complex-type structures but are instead modified to paucimannosidic or oligomannose structures. These differences in processing between insect cells and mammalian cells are due to insufficient expression of multiple processing enzymes including glycosyltransferases responsible for generating complex-type structures and metabolic enzymes involved in generating appropriate sugar nucleotides. Recent genomics studies suggest that insects themselves may include many of these complex transferases and metabolic enzymes at certain developmental stages but expression is lost or limited in most lines derived for cell culture. In addition, insect cells include an N -acetylglucosaminidase that removes a terminal N -acetylglucosamine from the N-glycan. The innermost N -acetylglucosamine residue attached to asparagine residue is also modified with alpha(1,3)-linked fucose, a potential allergenic epitope, in some insect cells. In spite of these limitations in N-glycosylation, insect cells have been widely used to express various recombinant proteins with the baculovirus expression vector system, taking advantage of their safety, ease of use, and high productivity. Recently, genetic engineering techniques have been applied successfully to insect cells in order to enable them to produce glycoproteins which include complex-type N-glycans. Modifications to insect N-glycan processing include the expression of missing glycosyltransferases and inclusion of the metabolic enzymes responsible for generating the essential donor sugar nucleotide, CMP- N -acetylneuraminic acid, required for sialylation. Inhibition of N -acetylglucosaminidase has also been applied to alter N-glycan processing in insect cells. This review summarizes current knowledge on N-glycan processing in lepidopteran insect cell lines, and recent progress in glycoengineering lepidopteran insect cells to produce glycoproteins containing complex N-glycans.

Mammalian-like nonsialyl complex-type N-glycosylation of equine gonadotropins in Mimic™ insect cells

Recombinant equine luteinizing hormone/chorionic gonadotropin (eLH/CG) was expressed in Mimic insect cells, that are commercial stably transformed Spodoptera frugiperda (Sf9) cells expressing five mammalian genes encoding glycosyltransferases involved in the synthesis of complex-type monosialylated N-glycans. We previously showed that it exhibited no in vivo bioactivity although expressing full in vitro bioactivity, and it was suspected that this was because of insufficient sialylation of eLH/CG N-glycans. Lectin binding analyses were performed with recombinant dimeric eLH/CG or its alpha subunit, secreted in the serum-containing supernatant of infected Sf9 and Mimic cells. Two types of specific lectin affinity assays (blot analyses and enzyme-linked immunosorbent assay) were used to compare the ability or inability of natural and recombinant gonadotropins to bind to various lectins. In natural equine chorionic gonadotropin (eCG), complex-type N-glycans terminating with both Siaalpha2,3Gal (based on Maackia amurensis agglutinin [MAA] binding) and Siaalpha2,6Gal (based on Sambucus nigra agglutinin [SNA] binding) were found, but in the alpha subunit dissociated from natural eCG, we only detected Siaalpha2-6Gal. In eLH/CG and its alpha subunit produced by Sf9 cells, N-glycans were found to be terminated by mannosyl residues (based on Galanthus nivalis agglutinin [GNA] binding), whereas those produced in Mimic cells were terminated by galactoses (based on binding to Ricinus communis agglutinin I [RCA I] , but not to SNA or MAA). This is in agreement with the fact that the nucleotide donor substrate of sialic acid is not naturally synthesized in insect cells. On the basis of binding to Arachis Hypogaea agglutinin [PNA], O-glycans exhibited the Galbeta1-3GalNAc structure in recombinant-free alpha and eLH/CG from both Sf9 and Mimic cell lines. Both N- and O-linked carbohydrate side chains synthesized in Mimic cells should thus be amenable to further acellular sialylation.

Biosynthesis of human-type N-glycans in heterologous systems

Insects, yeasts and plants generate widely different N-glycans, the structures of which differ significantly from those produced by mammals. The processing of the initial Glc2Man9GlcNAc2 oligosaccharide to Man8GlcNAc2 in the endoplasmic reticulum shows significant similarities among these species and with mammals, whereas very different processing events occur in the Golgi compartments. For example, yeasts can add 50 or even more Man residues to Man(8-9)GlcNAc2, whereas insect cells typically remove most or all Man residues to generate paucimannosidic Man(3-1)GlcNAc2N-glycans. Plant cells also remove Man residues to yield Man(4-5)GlcNAc2, with occasional complex GlcNAc or Gal modifications, but often add potentially allergenic beta(1,2)-linked Xyl and, together with insect cells, core alpha(1,3)-linked Fuc residues. However, genomic efforts, such as expression of exogenous glycosyltransferases, have revealed more complex processing capabilities in these hosts that are not usually observed in native cell lines. In addition, metabolic engineering efforts undertaken to modify insect, yeast and plant N-glycan processing pathways have yielded sialylated complex-type N-glycans in insect cells, and galactosylated N-glycans in yeasts and plants, indicating that cell lines can be engineered to produce mammalian-like glycoproteins of potential therapeutic value.

Characterization of the cellular uptake and metabolic conversion of acetylated N-acetylmannosamine (ManNAc) analogues to sialic acids

"Sialic acid engineering" refers to the strategy where cell surface carbohydrates are modified by the biosynthetic incorporation of metabolic intermediates, such as non-natural N-acetylmannosamine (ManNAc) analogues, into cellular glycoconjugates. While this technology has promising research, biomedical, and biotechnological applications due to its ability to endow the cell surface with novel physical and chemical properties, its adoption on a large scale is hindered by the inefficient metabolic utilization of ManNAc analogues. We address this limitation by proposing the use of acetylated ManNAc analogues for sialic acid engineering applications. In this paper, the metabolic flux of these "second-generation" compounds into a cell, and, subsequently, into the target sialic acid biosynthetic pathway is characterized in detail. We show that acetylated ManNAc analogues are metabolized up to 900-fold more efficiently than their natural counterparts. The acetylated compounds, however, decrease cell viability under certain culture conditions. To determine if these toxic side effects can be avoided, we developed an assay to measure the cellular uptake of acetylated ManNAc from the culture medium and its subsequent flux into sialic acid biosynthetic pathway. This assay shows that the majority ( > 80%) of acetylated ManNAc is stored in a cellular "reservoir" capable of safely sequestering this analogue. These results provide conditions that, from a practical perspective, enable the acetylated analogues to be used safely and efficaciously and therefore offer a general strategy to facilitate metabolic substrate-based carbohydrate engineering efforts. In addition, these results provide fundamental new insights into the metabolic processing of non-natural monosaccharides.

Engineering sialic acid synthetic ability into insect cells: identifying metabolic bottlenecks and devising strategies to overcome them

Previous studies have indicated negligible levels of both sialylation and the precursor N-acetylneuraminic acid (Neu5Ac) in a number of insect cell lines grown in serum-free medium. The overexpression of the human sialic acid 9-phosphate synthase (SAS) in combination with N-acetylmannosamine (ManNAc) feeding has been shown to overcome this limitation. In this study we evaluated the potential bottlenecks in the sialic acid synthesis pathway in a Spodoptera frugiperda (Sf9) insect cell line and devised strategies to overcome them by overexpression of the enzymatic pathway enzymes combined with appropriate substrate feeding. Coexpression of SAS and UDP-GlcNAc 2-epimerase/ManNAc kinase, the bifunctional enzyme initiating sialic acid biosynthesis in mammals, resulted in Neu5Ac synthesis without use of any external media supplementation to demonstrate that Neu5Ac could be generated intracellularly in Sf9 cells using natural metabolic precursors. N-Acetylglucosamine (GlcNAc) feeding in combination with this coexpression resulted in much higher levels of Neu5Ac compared to levels obtained with ManNAc feeding with SAS expression alone. The lower Neu5Ac levels obtained with ManNAc feeding suggested limitations in the transport and phosphorylation of ManNAc. The bottleneck in phosphorylation was likely due to utilization of GlcNAc kinase for phosphorylation of ManNAc in insect cells and was overcome by expression of ManNAc kinase. The transport limitation was addressed by the addition of tetra-O-acetylated ManNAc, which is easily taken up by the cells. An alternative sialic acid, 2-keto-3-deoxy-D-glycero-D-galacto-nononic acid (KDN), could also be generated in insect cells, suggesting the potential for controlling not only the production of sialic acids but also the type of sialic acid generated. The levels of KDN could be increased with virtually no Neu5Ac generation when Sf9 cells were fed excess GlcNAc. The results of these studies may be used to enhance the sialylation of target glycoproteins in insect and other eukaryotic expression systems.

Humanization of lepidopteran insect-cell-produced glycoproteins

The insect cell-baculovirus expression vector system, widely used for glycoprotein production, is not ideal for pharmaceutical glycoprotein production due to the characteristics of the N-glycans in the expressed products. Insect cells lack several enzymes required for mammalian-type N-glycan synthesis and contain a specific N-acetylglucosaminidase that stunts the growth of chains and a core alpha-1,3-fucosyltransferase that yields potentially allergenic glycoforms. Current knowledge on N-glycan processing in lepidopteran insect cells is summarized, and strategies to develop better glycoprotein expression systems suitable for pharmaceutical glycoprotein production are discussed.

N-glycan structures of human transferrin produced by Lymantria dispar (gypsy moth) cells using the LdMNPV expression system

N-glycan structures of recombinant human serum transferrin (hTf) expressed by Lymantria dispar (gypsy moth) 652Y cells were determined. The gene encoding hTf was incorporated into a Lymantria dispar nucleopolyhedrovirus (LdMNPV) under the control of the polyhedrin promoter. This virus was then used to infect Ld652Y cells, and the recombinant protein was harvested at 120 h postinfection. N-glycans were released from the purified recombinant human serum transferrin and derivatized with 2-aminopyridine; the glycan structures were analyzed by a two-dimensional HPLC and MALDI-TOF MS. Structures of 11 glycans (88.8% of total N-glycans) were elucidated. The glycan analysis revealed that the most abundant glycans were Man1-3(+/-Fucalpha6)GlcNAc2 (75.5%) and GlcNAcMan3(+/-Fucalpha6)GlcNAc2 (7.4%). There was only approximately 6% of high-mannose type glycans identified. Nearly half (49.8%) of the total N-glycans contained alpha(1,6)-fucosylation on the Asn-linked GlcNAc residue. However alpha(1,3)-fucosylation on the same GlcNAc, often found in N-glycans produced by other insects and insect cells, was not detected. Inclusion of fetal bovine serum in culture media had little effect on the N-glycan structures of the recombinant human serum transferrin obtained.

Developing baculovirus-insect cell expression systems for humanized recombinant glycoprotein production

The baculovirus-insect cell expression system is widely used to produce recombinant glycoproteins for many different biomedical applications. However, due to the fundamental nature of insect glycoprotein processing pathways, this system is typically unable to produce recombinant mammalian glycoproteins with authentic oligosaccharide side chains. This minireview summarizes our current understanding of insect protein glycosylation pathways and our recent efforts to address this problem. These efforts have yielded new insect cell lines and baculoviral vectors that can produce recombinant glycoproteins with humanized oligosaccharide side chains.

GlcNAc 2-epimerase can serve a catabolic role in sialic acid metabolism

Sialic acid is a major determinant of carbohydrate-receptor interactions in many systems pertinent to human health and disease. N-Acetylmannosamine (ManNAc) is the first committed intermediate in the sialic acid biosynthetic pathway; thus, the mechanisms that control intracellular ManNAc levels are important regulators of sialic acid production. UDP-GlcNAc 2-epimerase and GlcNAc 2-epimerase are two enzymes capable of generating ManNAc from UDP-GlcNAc and GlcNAc, respectively. Whereas the former enzyme has been shown to direct metabolic flux toward sialic acid in vivo, the function of the latter enzyme is unclear. Here we study the effects of GlcNAc 2-epimerase expression on sialic acid production in cells. A key tool we developed for this study is a cell-permeable, small molecule inhibitor of GlcNAc 2-epimerase designed based on mechanistic principles. Our results indicate that, unlike UDP-GlcNAc 2-epimerase, which promotes biosynthesis of sialic acid, GlcNAc 2-epimerase can serve a catabolic role, diverting metabolic flux away from the sialic acid pathway.