NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Varki A, Cummings RD, Esko JD, et al., editors. Essentials of Glycobiology. 2nd edition. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 2009.

Cover of Essentials of Glycobiology

Essentials of Glycobiology. 2nd edition.

Show details

Chapter 17Nucleocytoplasmic Glycosylation

and .

This chapter reviews the evidence for complex glycoconjugates in the nucleoplasmic and cytoplasmic compartments of the cell. The modification of nuclear and cytoplasmic proteins by a single β-linked N-acetylglucosamine (O-β-GlcNAc) is described in Chapter 18. Here the focus is on resident glycoproteins of the cytoplasm/nucleus with glycans of length two units or more. Current views regarding the putative functions of nuclear lectins and the possible roles of glycans as nuclear localization signals are also presented.


Most general biochemistry and cell biology texts still persist in the view that glycosylation is largely restricted to the addition of complex glycans to proteins or lipids localized on the cell surface or within lumenal compartments of subcellular organelles, such as the endoplasmic reticulum (ER), Golgi, endosomes, or lysosomes. In fact, a large body of data documents the presence of both simple and complex glycoconjugates within the nucleus and cytoplasm. Many of these studies have used plant lectins, anticarbohydrate antibodies, or natural polysaccharide (e.g., hyaluronan)-binding proteins as tools to detect glycosylated molecules within the nucleoplasmic or cytoplasmic compartments. Even though there have been more than 50 papers that have used the binding of plant lectins to detect complex glycans within the nucleus of cytoplasm, these binding studies have largely been ignored for several reasons: (1) Most glycosyltransferases involved in complex glycan biosynthesis are type II membrane proteins with their active sites within the lumen of the ER or Golgi. Thus, the “enzymatic machinery” is on the “wrong” side of the membrane for the biosynthesis of complex nuclear or cytoplasmic glycoconjugates. (2) Most of the lectin-binding studies that make claims of cytoplasmic or nuclear glycosylation present no supporting structural data to establish firmly the identity of these glycans. (3) Under some circumstances, plant lectins, such as concanavalin A, can bind to molecules by nonspecific hydrophobic interactions. Also, it is possible that the competing saccharide ligand can alter the conformation of such lectins to induce a loss of this hydrophobic binding. Therefore, although lectins are important tools in glycobiology, conclusions based exclusively on their use, particularly those not compatible with known pathways and established concepts of biochemistry and cell biology, must be followed up by rigorous structural analyses. Nevertheless, these many lectin-binding studies indicate that N-acetylglucosamine-, mannose-, and fucose-containing glycans are abundant within the nucleus and cytoplasm, suggesting that further, more rigorous analyses are warranted.


There are several possible explanations for the potential origins of complex nucleocytoplasmic glycans suggested by cytological studies, as summarized in Figure 17.1. Most simply, nonconventional soluble glycosyltransferases may reside in the cytoplasm or nucleus and directly modify proteins there (Figure 17.1a). Nucleotide sugar precursors are available, because the enzymes that synthesize them reside within these compartments. As discussed below, this is the mechanism by which glycogen synthesis is primed and by which Skp1 is modified by multiple sugar types in the cytoplasm of Dictyostelium. Glycosylated proteins might then be transferred to the nucleus through nuclear pores via conventional nuclear transport pathways.

FIGURE 17.1. Potential mechanisms for the glycosylation of cytoplasmic proteins.


Potential mechanisms for the glycosylation of cytoplasmic proteins. (a) In the simplest model, the glycosyltransferases are present in the cytoplasm and transfer sugars from nucleotide sugar donors to the acceptor protein in the cytoplasm. This is the (more...)

A second potential source of cytoplasmic complex glycans is the secretory pathway itself. An emerging area of cell biology has identified numerous proteins that return to the cytoplasm after entry into the ER or after secretion (Figure 17.1b). The Sec61/63 translocator that transports nascent proteins into the ER is also thought to support the transfer of mis-folded nascent proteins back into the cytoplasm where they are subsequently degraded by the 26S proteasome. In addition, there is evidence that endocytosed proteins, such as cholera toxin, can gain access to the cytoplasm via this pathway. Therefore, retrotranslocation is also a potential source of cytoplasmic and ultimately nuclear glycoproteins. Furthermore, there is a growing body of literature that supports the concept that cytokines, growth factors, and even their transmembrane receptors, many of which are glycosylated, can translocate to the nucleus to exert direct effects on transcription. These novel translocation pathway(s) are obscure but may involve lipid rafts, endocytosis, and conventional nuclear localization signals (Figure 17.1b). Although the existence of these transport pathways is controversial, they might provide a source of cytoplasmic and nuclear glycoproteins with conventional glycan modifications. Finally, occasional reports of conventional O-glycosylation of cytoplasmic proteins in pathological tissues raises the possibility that compartmental breakdowns result in the exposure of latent cytoplasmic proteins to Golgi glycosyltransferases.

Finally, the biosynthesis of several classes of glycans is initiated by membrane-associated glycosyltransferases whose catalytic domains are situated at the cytoplasm-facing surface. This includes early steps in the synthesis of glucosylceramides and GPI anchors, as well as early assembly of dolichyl-linked N-glycosylation precursors that ultimately “flip” to the other side of the membrane and are transferred to proteins within the lumen of the ER in the secretory pathway (see Chapter 8). Other membrane-associated glycosyltransferases whose catalytic domains face the cytoplasm include those that polymerize hyaluronan (see Chapter 15), cellulose, chitin, and lipopolysaccharides, and these products are subsequently translocated across the plasma membrane. The glycan products of these membrane-associated glycosyltransferases normally exit the cytoplasmic space, but if this did not occur, they could serve as a source of novel, cytoplasmic, nonprotein-associated glycans. Alternatively, homologs of these enzymes might be capable of modifying protein acceptors.

Cytoplasmic N-Linked Glycans?

The secretory pathway is home to the biosynthesis of several classes of glycans, including the N- and O-linked types and the more rare glycans linked via C-mannosyl tryptophan. Numerous reports offer partial evidence to support the existence of N-linked glycoproteins within the cytoplasm. The α subunit of the sodium pump (Na+, K+-ATPase) from dog kidney, which is a transmembrane protein, was reported to contain N-linked glycans with terminal N-acetylglucosamine residues in its cytoplasmic domain. This conclusion was based on the enzymatic attachment of radioactive galactose to N-acetylglucosamine residues by galactosyltransferase labeling of permeabilized right-side-out membrane vesicles. N-Glycanase sensitivity of the radiolabeled products suggested that the acceptors are N-linked glycans (see Chapter 8). In such vesicles, extracellular domains would be protected from enzymatic probes by being inside sealed membranes. This provocative claim has been largely ignored because of the lack of any site-mapping and structural analysis of the putative cytoplasmic glycans, even though the primary sequence of the protein has been known for years.

Several studies have also strongly indicated that the so-called high-mobility-group proteins (HMGs), which are important structural components of chromatin, are bona fide “classical type” glycoproteins. Highly purified preparations of HMGs 14 and 17 were found to contain N-acetylglucosamine, mannose, galactose, glucose, fucose, and possibly xylose. A fucose-specific lectin, Ulex europeus agglutinin I, bound to both HMGs and isolated nucleosomes. Alkali resistance of the saccharides on the HMGs is consistent with attachment of the glycan via an N-linkage. In addition, these HMGs could be metabolically radiolabeled with tritiated fucose, galactose, mannose, or N-unsubstituted glucosamine. Later studies suggested that the glycans on HMGs 14 and 17 were required for their binding to the nuclear matrix, where the HMGs have a role in modulating chromatin structure at active sites of gene transcription. Even though these HMG glycosylation studies are widely cited as evidence for nuclear N-glycans, they also suffer from a lack of definitive structural data, such as physical proof of the linkage to protein and physical definition of the structure of the putative N-glycan(s). In fact, more recent studies have not been able to verify the presence of complex glycans of any type on the HMGs, but did find that the HMGs are modified by O-β-GlcNAc (see Chapter 18).

Sialic acid–containing glycoproteins have been reported on the cytoplasmic face of the nuclear envelope. The sialic acid–specific lectin Sambucus nigra agglutinin (SNA) was shown to bind to several proteins, previously shown to also contain O-GlcNAc. Two of these proteins were identified as major nucleoporins: p62 and p180. Prior sialidase treatment blocked binding of SNA to these nuclear pore proteins. On the basis of N-glycanase sensitivity, the sialic acids on p180 appeared to be on N-linked glycans, and those on p62 appeared to be on O-linked glycans. The authors also demonstrated that SNA blocked nuclear protein import in neuroblastoma cells, suggesting that the sialic acids might have functional importance on the nuclear pore proteins. Again, the significance of these provocative findings must await systematic structural proof of the nature of the putative sialylated glycans on these nucleoporins.

Weighing in against the possibility of resident N-glycoproteins in the cytoplasm is the existence of catabolic pathways that rapidly degrade them. Unfolded proteins that are translocated from the ER are deglycosylated by a cytoplasmic N-glycanase, polyubiquitinated, and degraded by the 26S proteasome. Alternatively, these proteins can be recognized directly by a ubiquitin ligase that is selective for high mannose N-glycans, also resulting in polyubiquitination and degradation. These turnover pathways suggest that cytoplasmic N-linked glycoproteins are short-lived and therefore unlikely to function in the cytoplasm.

A heat-shock-like nuclear chaperone protein with N-acetylglucosamine-binding activity (CBP70) has also been suggested to be N-glycosylated. A subpopulation of prion protein may be N-glycosylated and found in the nucleus, possibly in association with the N-acetyl-glucosamine-binding lectin. It is possible that the compartmentalization of N-glycoproteins within the nucleus or mitochondrion protects them from degradation by the above-mentioned cytoplasmic pathways.

Mitochondrial Glycosylation?

On the basis of lectin-binding studies, mitochondria have been suggested to contain complex glycoconjugates. Most mitochondrial proteins are synthesized within the cytoplasm under the direction of nuclear genes, suggesting that these glycoconjugates might originate in the cytoplasm. Recently, two glycoproteins located on the inner mitochondrial membrane have been identified, and they appear to be conventionally N-glycosylated in the rER based on pulse-chase labeling studies and susceptibility to N-glycanase. These glycoproteins might transit to mitochondria via a novel vesicle transport pathway or by retro-translocation to the cytoplasm followed by conventional uptake into mitochondria. High-resolution electron microscopy shows that some mitochondrial membranes are in contact with portions of the ER, offering another option for direct membrane transfer into the mitochondria. In plants, a galactolipid is synthesized in mitochondrial membranes, which would provide an alternative explanation for mitochondrial lectin labeling.

Nucleocytoplasmic N-Glycan Glycosyltransferases?

In support of the existence of novel nucleocytoplasmic glycosyltransferases, there are several reports in the literature of glycosyltransferase activities in highly purified preparations of rat liver nuclei judged to be >99% pure by marker enzyme analysis. These studies document the transfer of N-acetylglucosamine from UDP-GlcNAc to endogenous acceptors and show that at least 80% of the activity is blocked by low concentrations of the antibiotic tunicamycin, suggesting the involvement of N-linked biosynthetic intermediates, such as N-acetylglucosaminyl-pyrophosphoryldolichol (GlcNAc-PP-dolichol). Later studies demonstrated the direct transfer of chitobiose (GlcNAcβ1-4GlcNAc) from chitobiosyl-dolichol to endogenous nuclear acceptors by these nuclear preparations, suggesting a novel pathway of N-glycosylation. The products of these in vitro reactions were found to be N-linked chitobiosyl moieties, based on their sensitivity to peptide N-glycosidase F (N-glycanase) and hydrazinolysis, but also on their insensitivity to alkali-induced β-elimination (see Chapter 47). Similar studies have documented the presence of nuclear mannosyl-transferases. These studies are provocative, but they must also be interpreted with caution. The ER, which is the widely accepted site of N-glycosylation (see Chapter 8), is functionally contiguous with the outer nuclear envelope, and cell biologists have detected infoldings of this membrane into the nuclear interior in some cells. Even a minor contamination of nuclear envelope with ER could lead to misinterpretation of these findings. In addition, it is very difficult to purify nuclei such that other cellular components do not adhere non-specifically to the otherwise “pure” nuclei during their preparation. Given these potential problems, widespread acceptance of the existence of these nuclear glycosyltransferases must await independent confirmation by more direct criteria.

Recently, an unusual N-linked glycan has been detected crystallographically on a capsid protein of a virus, PBCV-1. This capsid protein (VP54) is assembled, posttranslationally modified, and incorporated into viral structures within the cytoplasm of Chlorella algae. The PBCV-1 genome encodes multiple sequences predicted to encode cytoplasmically localized glycosyltransferases that lack targeting sequences for the secretory pathway. N-glycosylation of VP54 is unusual in that the asparagine attachment site is not associated with the canonical N-glycosylation sequon used for N-glycosylation in the secretory pathway. This example raises the possibility that a variant form of N-glycosylation has evolved within the cytoplasm, but, because PBCV-1 eventually exits the host after cell lysis, the product glycoprotein appears to ultimately function outside of the cell.


The hydroxyl side chains of threonine, serine, tyrosine, hydroxyproline, and hydroxylysine constitute potential sites of attachment of so-called O-linked glycans. Threonine and serine are the predominant linkage residues in the secretory pathway, but interestingly, the best-characterized examples of cytoplasmic O-glycosylation (other than O-GlcNAc; see Chapter 18) occur on tyrosine and hydroxyproline. Unlike secretory pathway glycosylation, cytoplasmic glycosylation pathways appear to target only single proteins. The following sections review evidence for these and other kinds of cytoplasmic O-glycosylation.


Glycogen is a large homopolysaccharide of glucose. It serves as a short-term storage form of glucose and can be broken down to provide glucose to liver, muscle, and blood during stress. Several reports have unequivocally established that glycogen is covalently attached at its reducing terminus to a polypeptide and thus is a glycoprotein. In 1975, liver extracts were shown to synthesize from UDP-glucose a glycogen-like product that was precipitated by trichloroacetic acid, indicating that it had a protein component. In vitro, maltotetraose [(Glcα 1-4)4] is the smallest saccharide that primes glycogen elongation. Therefore, it was proposed that protein, not carbohydrate, was the original primer for glycogen synthesis. In 1985, the glycogen protein primer was named glycogenin. Glycogenin was then purified in a complex with glycogen synthase. Glycogenin has 322 amino acids and the first glucosyl residue is attached through a glycosidic linkage to the hydroxyl group of tyrosine-194 of the protein. In muscle, each molecule of glycogenin is covalently attached to one molecule of glycogen (i.e., a ratio of 1, where the molecular weight of glycogen is 107, approximately the maximum mass of a glycogen molecule in muscle). In liver, the mass ratio of glycogenin to glycogen is about 0.0025. This small ratio reflects both the enormous size of liver glycogen (the molecular weight can be as much as 5 × 109and the existence of large numbers of free glycogen chains in liver glycogen granules.

Glycogenin is autocatalytic, attaching glucose to itself from the donor (UDP-Glc) and is classified as a retaining hexosyltransferase (see Chapter 5). Glycogenin catalyzes two chemically distinct reactions: formation of the Glcα1-Tyr linkage and the subsequent Glcα1-4Glc linkages. The first reaction was established using recombinant glycogenin from an Escherichia coli strain unable to synthesize UDP-Glc. A mutant glycogenin lacking Tyr-194 can glucosylate a catalytically inactive glycogenin, showing interpolypeptide transfer. Glycogenin occurs as a dimer in solution and in crystals, and elongation of the glucan chain is essentially unaffected by dilution. Therefore, glycogenin appears to extend the glucan chain on the other subunit of a functional dimer, as depicted in Figure 17.2. This self-glucosylation yields a malto-octaose (eight glucose units), and both p-nitro-phenyl α-glucoside and α-malto-oligosaccharides are competitive substrates. However, the distances between UDP-Glc and Tyr-194 (of either subunit) are too great in the known crystal structures of glycogenin for this mechanism to explain initial priming, which warrants further analysis. To date, glycogenin is the only known example of the glycosylation of tyrosine residues in animals.

FIGURE 17.2. Opposing models for the self-glucosylation of glycogenin.


Opposing models for the self-glucosylation of glycogenin. Structural data support the hypothesis of intersubunit elongation rather than intrasubunit elongation. (Top) Active site of one subunit of the glycogenin dimer adding α1-4Glc residues to (more...)

Glycogenin-like proteins are found in a wide range of plants, animals, and free-living lower eukaryotes. E. coli also appears to have a glycogenin-like molecule, but it is incapable of glucosylating recombinant mammalian glycogenin. Based on its primary sequence, glycogenin has been classified in the CAZy database (see Chapter 7) as a member of the family GT-8 glycosyltransfereases, which includes bacterial lipopolysaccharide glucosyl, galactosyl transferases, and galactinol synthases, all of which are apparently cytoplasmic like glycogenin. Remarkably, a novel Glcβ1-Arg linkage has been described in a plant protein associated with starch synthesis, but this protein has no apparent sequence similarity to glycogenin.

Biochemically, glycogenin may have a key role in the regulation of glucose homeostasis. Glycogenin has a Km for UDP-Glc that is about 1000 times lower than the Km of glycogen synthase for the sugar nucleotide. Production of the glycogenin primer, especially in muscle cells, can be the rate-limiting step in glycogen formation. Thus, levels of glycogenin could override the better understood, hormonally controlled mechanisms that involve protein phosphorylation/dephosphorylation and regulate glycogen elongation. Figure 17.3 illustrates a model of glycogen synthesis by glycogenin, glycogen synthase, and branching enzyme.

FIGURE 17.3. Model for glycogen biosynthesis via glycogenin (GN), glycogen synthase (GS), and branching enzyme, as proposed by Carl Smythe and Philip Cohen in the late 1980s.


Model for glycogen biosynthesis via glycogenin (GN), glycogen synthase (GS), and branching enzyme, as proposed by Carl Smythe and Philip Cohen in the late 1980s. (Step 1) GN is self-priming, attaching glucose to Tyr-194. (Step 2) GN elongates the glucose (more...)

Hydroxyproline-linked Glycans

A small fucosylated protein, Skp1, is found in the cytoplasm of a free-living soil amoeba, the cellular slime mold Dictyostelium. Skp1 has an established role in yeast, plants, and animals as an adaptor in certain E3-ubiquitin ligase complexes that are active in the cytoplasm and nucleus. E3SCF-ubiquitin ligases promote the polyubiquitination and eventual proteasomal degradation of cell-cycle regulatory proteins, transcriptional factors, and other signal-transduction proteins. Mass spectrometric structural analyses of Skp1 peptides established that it is glycosylated at Pro-143 after prior hydroxylation. The glycan is a pentasaccharide consisting of a core trisaccharide equivalent to the type 1 blood group H structure: Fucα1-2Galβ1-3GlcNAc- (Figure 17.4, top). This glycan is further modified by an α1-3-linked galactose residue on the fucose and another α-linked galactose on the α-Gal or fucose residue. In these studies, the great majority of Pro-143 is hydroxylated, and nearly all of the hydroxylated form is glycosylated. The linkage to protein, localization of this type of structure in the cytoplasm, and the structure itself are all novel, which suggests a very atypical glycosylation pathway in this organism and perhaps in other eukaryotes. Because Pro-143 is conserved in the Skp1 of plants, invertebrates, and lower eukaryotes, and sequences related to those of known Skp1 modification enzymes are present in the genomes of other protists, similar glycosylation may occur in other organisms.

FIGURE 17.4. Mechanism of glycosylation of Skp1 in the cytoplasm of Dictyostelium.


Mechanism of glycosylation of Skp1 in the cytoplasm of Dictyostelium. (Top) Skp1 Pro-143 is sequentially modified by a soluble prolyl 4-hydroxylase and five soluble, sugar nucleotide–dependent glycosyltransferase activities. (αGlcNAcT) (more...)

Extracts of Dictyostelium can hydroxylate and glycosylate recombinant Skp1 in vitro. The Skp1 prolyl 4-hydroxylase is the Dictyostelium ortholog of animal cytoplasmic prolyl 4- hydroxylases that regulate the O2 and E3VHL ubiquitin-ligase-dependent degradation of the hypoxia-inducible factor-1α transcriptional factor subunit (HIFα) (Figure 17.4, bottom). The biochemical assays suggest that the sugars are added sequentially by soluble UDP-sugar-dependent enzymes that reside within the cytoplasm. The proteins that catalyze the first four glycosyltransferase reactions have been purified to homogeneity. Cloning of the gene for the N-acetylglucosaminyltransferase, which adds the first sugar, showed it to be related in sequence to the Thr/Ser polypeptide α-N-acetylgalactosaminyl- and α-N-acetylglucos-aminyltransferases that initiate mucin-type O-glycosylation in the Golgi of animals and protists, and, therefore, it is expected to form an anomeric linkage opposite to that of O-β-N-acetylglucosaminyltransferase (see Chapter 18). However, unlike all other known polypeptide α-N-acetylgalactosaminyl- and α-N-acetylglucosaminyltransferases, the Skp1 N-acetylglucosaminyltransferase does not possess an amino-terminal signal anchor, consistent with its biochemical fractionation as a soluble cytoplasmic protein (Figure 17.4, bottom). These findings suggest that the mechanism of initiation of Skp1 glycosylation is similar to that of mucin-type domains of animal secretory proteins, except that a different N-acetylhexosamine (N-acetylglucosamine vs. N-acetylgalactosamine) is attached to a distinct hydroxyamino acid (hydroxyproline, not threonine or serine) in the cytoplasm instead of the Golgi lumen. Cloning of the gene for the fucosyltransferase, which adds the third sugar, led to the discovery that both the second (β-Gal) and third (α-L-Fuc) sugars are added by separate domains of the same protein (Figure 17.4, bottom). This protein is also predicted to be soluble. The β1-3-galactosyltransferase catalytic domain is most similar to bacterial lipopolysaccharide and capsular glycosyltransferases, which are also cytoplasmic, suggesting that cytoplasmic glycosylation in Dictyostelium has its evolutionary origins in bacterial glycolipid synthesis. The β-galactosyltransferase catalytic domain belongs to the inverting GT family 2 (CAZy database), which is diagnostic for glycosyltransferases with their catalytic domains exposed to the cytoplasm. However, in both bacteria and eukaryotes, these enzyme family members are usually membrane-associated and their products, including lipopolysaccharides, dolichol phosphate sugars, and polysaccharides such as hyaluronan, cellulose, and chitin, are translocated to the lumen or cell exterior. Bifunctional diglycosyl-transferases with a similar architecture are used for cell-surface glycosaminoglycan and polysaccharide synthesis in certain bacteria and in the eukaryotic Golgi. The fourth sugar is added by another two-domain glycosyltransferase in which the α1-3-galactosyltransferase domain is fused to a β-propeller-like domain of unknown function. The catalytic domain is related to a large number of plant Golgi proteins implicated in pectin biosynthesis.

In contrast to glycogenin, Dictyostelium Skp1 possesses a hetero-oligomeric glycan that is the product of multiple enzymes that act sequentially on the acceptor protein substrate by a mechanism most similar to that of O-linked glycosylation in the eukaryotic Golgi. Like glycogenin, Skp1 is the only known protein substrate for its glycan modification. Both glycogenin and Skp1 have novel sugar–amino acid linkages. Mutagenesis of Pro-143 in Dictyostelium Skp1 or inhibition of Pro-143 hydroxylation reduces the steady-state level of Skp1 in the nucleus, suggesting an involvement of glycosylation in the subcellular compartmentalization of this protein. Disruption of the Skp1 prolyl 4-hydroxylase gene interferes with an O2-dependent step in Dictyostelium development. This reflects a function for the early part of the modification pathway, possibly involving addition of α-GlcNAc, because disruption of the subsequently acting diglycosyltransferase gene has only a minimal effect on development.

Phosphoglucosylation and Cytoplasmic O-Linked Mannose

α-Glc-1-P can be transferred from UDP-Glc to O-linked mannosyl residues situated predominantly on a single 62-kD protein called parafusin. Note that this form of cytoplasmic O-mannose appears to be distinct from the O-mannosylation that is common in the lumen of the ER, Golgi, and cell surface of many eukaryotes, particularly fungi (see Chapter 12). Parafusin responds to a secretagogue by reversibly dissociating from cellular membranes at the same time that Glc-P is released. Considerable biochemical evidence for the occurrence of Glcα1-P and Man-P linkages, which are likely to be conjoined in a single phosphodiester moiety, has been obtained from in vitro studies of the modification of parafusin in extracts from organisms as diverse as Paramecium, birds, and mammals. Parafusin is a member of the phosphoglucomutase (PGM) protein family, named for the enzyme that catalyzes the interconversion of Glc-1-P and Glc-6-P. Although parafusin itself appears not to exhibit PGM activity, there is evidence that the active PGM protein from yeast and rat liver is similarly modified in extracts. Latency studies established that both the novel glucose phosphotransferase that acts on parafusin and a Glc-1-phosphate phosphodiesterase that cleaves Glc-P from parafusin have their active sites within the cytoplasm. Cochromatography studies of hydrolyzed parafusin from cells metabolically labeled with radioactive sugar precursors suggest that PGM-like proteins are similarly modified in vivo. The modification of PGM-like proteins is highly responsive to calcium levels and appears to regulate the association of parafusin with the membrane. Unfortunately, nothing is known about the mannosyltransferases that modify the PGM-like proteins, and little is known about the underlying mannosyl glycan structures modified by Glc-1-P. Identification of the enzyme, genes, and structural confirmation of the sugar linkages (Chapter 47) is needed to confirm this modification. Nonetheless, these findings suggest that both the O-mannosylation and dynamic Glc-1-phosphoryation of PGM-like proteins may be very important regulatory forms of cytoplasmic glycosylation.

Nucleocytoplasmic Glycosaminoglycans?

As early as 1964, Yamashina and colleagues proposed the presence of glycosaminoglycans (GAGs) in purified nuclei. From 1971 to 1983, a series of studies addressed nuclear heparan sulfates in developing sea urchin embryos. Heparin stimulated messenger RNA (mRNA) synthesis in pregastrula, but not in postgastrula embryos. Other glycosaminoglycans such as chondroitin sulfates (CSs) and hyaluronans were without effect. Pulse-chase studies suggested that the nuclear glycosaminoglycans appeared first in the cytoplasm and were transported to the nucleus. β-Xylosides, which are primers of CS and heparan sulfate (HS) biosynthesis (see Chapter 16), blocked development of sea urchins at the blastula stage. Addition of postgastrula (but not pregastrula) proteoglycans to the cultures prevented this block in development. Microinjection of proteoglycans into embryos blocked development at the stage from which they were isolated. Given recent advances in the tools to study proteoglycans, these potentially exciting observations warrant a critical reexamination.

The first structural data proposing nuclear GAGs, which were not easily dismissed on the basis of potential contamination with extracellular GAGs, were published in 1989. Rat hepatocytes were radiolabeled with 35 SO4, and nuclear fractions were found to contain 11% of the total cellular HS, but this fraction was greatly enriched in a relatively rare structure: GlcA2SGlcNS6S disaccharides. The existence of this unique disaccharide makes it hard to explain the nuclear GAGs as resulting from a contamination of the nuclear fraction by cell-surface HS molecules, where such structures are not present. Later pulse-chase studies suggested that cell-surface HS is taken into the nucleus and subsequently modified to form this unique species. The uptake of HS into the nucleus was not affected by chloroquine. In support of the uptake model, when unlabeled HS proteoglycan that contains no GlcA2S was incubated with cells, about 10% ended up within the nucleus and resulted in free HS chains containing the unusual disaccharide. Given the ability of heparan sulfates to influence gene transcription in vitro, these studies could eventually prove to be highly significant. How nuclei would take up and modify such large and negatively charged molecules remains unknown. Other studies using different nuclear isolation methods on rat ovarian granulosa cells have found dermatan sulfates, but not HSs, associated with the nucleus. No unique structures were found in this study.

Using another approach, immunocytochemical studies probing for a GPI-anchored HS proteoglycan, glypican, have provided strong evidence for the nuclear accumulation of this proteoglycan, in addition to its traditionally accepted presence at the cell surface. Although the mechanism of nuclear compartmentalization is not known, there is evidence that the glypican polypeptide harbors a nuclear localization sequence that is functional when grafted onto a neutral carrier expressed in the cytoplasm. In a separate study, overexpression of a CS proteoglycan core protein in cultured cells resulted in accumulation of the core polypeptide in the cytoplasm and nucleus in addition to the secretory pathway, indicating, as has been suggested for other proteins, the possibility of dual compartmentalization. However, although these studies provide an explanation for how proteoglycan core proteins might accumulate in the nucleus, it is difficult to imagine how they have been modified by the copolymerases residing in the Golgi lumen.

Cytochemical studies using naturally occurring proteins with high-affinity binding to hyaluronan (HA) have been used to document accumulation of HA in the nuclei of cultured cells. Intracellular hyaluronan might represent a pool of polysaccharide that failed to translocate following biosynthesis, because the catalytic domain of HA synthases are oriented at the cytoplasmic surface of the plasma membrane (see Chapter 15). Some of the HA-binding proteins occur naturally within the nucleus, raising the possibility of a physiological role for nuclear hyaluronan via interaction with these HA-binding proteins. However, it is not certain that HA is the natural ligand for these binding proteins. An alternative possibility is that they interact with nuclear proteins that have been cluster-modified by O-β-GlcNAc (see Chapter 18).

Alpha Toxins of Clostridium Species

Small cytoplasmic G proteins (GTP-binding proteins) of the Rho family are involved in regulating the cytoskeleton. As shown in Figure 17.5, Cdc42 is involved in the formation of filopodia, Rac regulates membrane ruffling, and Rho regulates focal adhesions and stress fibers. Certain toxins from anaerobic bacteria were found to possess glycosyltransferase activities that attach a glycosyl moiety to a threonine residue in the GTP-binding site of these proteins (e.g., on Rho, glycosylation occurs at Thr-37; Figure 17.6). These secreted toxins exhibit the remarkable ability of translocating across the surface membrane into the cytoplasm of mammalian target cells. The enterotoxins from Clostridium difficile (ToxA) and Clostridium sordellii are α-glucosyltransferases that use UDP-Glc as a donor. In contrast, a similar toxin from Clostridium novyi is a UDP-GlcNAc-dependent O-α-N-acetylglucosaminyltransferase. The C. novyi toxin has no primary sequence relationship to the endogenous inverting animal O-β-GlcNAc (see Chapter 18) or Skp1 α-N-acetylglucosaminyltransferases that have been cloned. These glycosyltransferase toxins might have mammalian homologs and should prove valuable as tools to study the functions of these small G proteins; they also represent a unique example of abnormal cytoplasmic protein glycosylation.

FIGURE 17.5. Rho family proteins regulate receptor-mediated actin cytoskeleton perturbations and are targets for bacterial toxins.


Rho family proteins regulate receptor-mediated actin cytoskeleton perturbations and are targets for bacterial toxins. Note how the different Rho family members are involved in regulating different cytoskeletal structures. The figure depicts transmembrane (more...)

FIGURE 17.6. Glycosylation of Rho proteins by toxins occurs at Thr-37 and this modification has a key role in GTP/GDP nucleotide exchange by these G proteins.


Glycosylation of Rho proteins by toxins occurs at Thr-37 and this modification has a key role in GTP/GDP nucleotide exchange by these G proteins. The cartoon depicts the GTP-binding site of the Rho protein. Note the essential role of Thr-37 in the active (more...)

Other Possible Examples of Cytoplasmic O-Glycosylation

A potential example of a novel O-linked glycan has been reported on a plant nuclear-pore-associated protein. This protein is recognized by the lectin wheat-germ agglutinin and can be labeled using tritiated UDP-Gal and β1-4 galactosyltransferase, which is specific for nonreducing terminal N-acetylglucosamine. The glycan, with an approximate size of five sugars, can be released by mild alkaline degradation consistent with β-elimination from a threonine or serine residue. Confirmation by mass spectrometry is needed to establish more firmly the existence of this modification. In another study, purified mammalian cytokeratin was found to bind lectins that recognize α1,3-linked N-acetylgalactosamine, a finding that was supported by compositional studies. Lectin binding was sensitive to pretreatment with N-acetylgalactosaminidase, but this does not rule out the possibility of a contaminant protein until direct structural evidence for glycosylation is obtained. Additional potential examples are oxygen-regulated protein 150, a putative cytoplasmic glycoprotein whose nuclear uptake is inhibited by a plant lectin thought to be internalized by epithelial cells, and a cytoplasmic isoform of carbonic anhydrase II secreted from male coagulating gland epithelial cells by ectocytosis. A final example is the VP54 PBCV-1 capsid protein described above. The X-ray-diffraction analysis demonstrates that in addition to four sites for unconventional N-glycosylation, there are two additional sites of serine glycosylation that appear to be generated on the cytoplasmic precursor of the mature capsid protein.


Many studies using labeled neoglycoproteins suggest the existence of multiple nuclear lectins. For example, bovine serum albumin (BSA) derivatized with L-rhamnose, D-N-acetylglucosamine, D-glucose, lactose, Man-6-P, and L-fucose all bind to nuclei at threefold higher affinity than underivatized BSA. BSA-lactose binds to cryostat sections of nuclei. However, little is known about the molecular nature of these putative carbohydrate-binding proteins except for galectins-1 and -3 (see below) and a recently identified heat-shock chaperone protein CBP70 (described above). CBP70 has GlcNAc-binding activity that might recognize O-β-GlcNAc–modified proteins.

The galectin family of lectins in animals and the discoidin family of lectins in Dictyostel-ium are high-abundance cytoplasmic proteins that have carbohydrate-binding specificity generally directed toward β-linked galactose and N-acetylgalactosamine (see Chapter 33). Large pools of these proteins are soluble in the cytoplasm, but cell biological studies have shown that these proteins also reside at the cell surface and in the pericellular matrix. These proteins lack typical amino-terminal signal peptides and exit the cytoplasm by a poorly characterized posttranslational pathway. Biochemical and genetic studies have uncovered primarily extracellular functions for these lectins (Chapter 33), suggesting that the cytoplasmic pools are simply precursors of functional extracellular pools. Nevertheless, CBP35 (now called galectin-3) and possibly galectin-1 are present in the nucleus as part of heterogeneous ribonucleoprotein complexes (HnRnP), where they appear to be required for normal mRNA splicing in extracts. Interestingly, a neoglycoprotein comprising the blood group A tetrasaccharide attached to BSA was found to inhibit specifically mRNA splicing in vitro. In addition, a cytoplasmic function in regulating apoptosis via Bcl-2 binding has been implicated for galectin-3. These intracellular activities may be mediated by protein–protein-binding activities ascribed to these proteins, because roles for their glycan-binding activities have not been established. These findings certainly leave open the possibility that galectins have important untested functions via cytoplasmic glycoconjugates.

Like galectins, members of the annexin family of proteins reside in the cytoplasm, at the cell surface, and in the extracellular matrix. Individual isoforms exhibit glycan-binding activity directed toward various GAGs, GPI anchors, or the bisecting N-acetylglucosamine of N-glycans. These known ligands generally reside outside of the cell and have not been specifically implicated in intracellular annexin functions, but they could be receptors for secreted annexins.

Numerous other soluble cytoplasmic proteins have been assigned sugar-binding activity that point to the potential significance of cytoplasmic glycoconjugates. A Golgi-associated actin-binding protein in Dictyostelium (comitin) has a mannose-binding activity, although the significance of this activity in vivo has not been tested. In plants, cytoplasmic mannose- and N-acetylglucosamine-binding lectins are induced by various kinds of biotic and abiotic effectors. In animals, numerous proteins have an affinity for glycogen, which exists primarily in the cytoplasm. One of these, laforin, is a protein phosphatase with a discrete glycogen-binding domain related to bacterial sugar-binding proteins. This domain is required for localization near glycogen, and evidence exists that this activity is specifically mutated in a congenital form of juvenile epilepsy.


Molecules larger than approximately 40 kD cannot diffuse freely through nuclear pores and must be specifically and actively transported into and out of the nucleus. Generally, nuclear localization sequences (NLSs) consist of one or two stretches of positive amino acids, as are found, for example, on SV40 large T antigen and nucleoplasmin. However, during the past several years, evidence has accumulated that sugars may also serve as nuclear localization signals. The neoglycoproteins BSA-glucose, BSA-fucose, and BSA-mannose are rapidly transported into the nucleus of permeabilized or microinjected living HeLa cells, whereas BSA itself is not. Sugar-mediated nuclear transport appears distinct from the basic peptide-mediated NLS pathway. Like the classical pathway, sugar-mediated nuclear transport requires energy and is blocked by the lectin, wheat-germ agglutinin. However, unlike the basic peptide system, the sugar-mediated pathway does not require cytosolic factors and is not blocked by sulfhydryl-reactive chemicals. Additional evidence shows that BSA substituted with GlcNAcβ1-4GlcNAc is rapidly localized to purified nuclei in vitro by a pathway distinct from the classically defined NLS systems. Validation of these fascinating results awaits characterization of the components involved and the identification of natural counterparts to the neoglycoproteins. A possible candidate is the Skp1 glycoprotein from Dictyostelium, although a distinct sugar specificity would be required in this species. Interference with Skp1 glycosylation inhibits nuclear accumulation in vivo, although other evidence suggests that the role of the glycan is indirect.


  1. Marchase RB, Hiller AM. Glucose phosphotransferase and intracellular trafficking. Mol Cell Biochem. 1986;72:101–107. [PubMed: 3029558]
  2. Hart GW, Haltiwanger RS, Holt GD, Kelly WG. Glycosylation in the nucleus and cytoplasm. Annu Rev Biochem. 1989;58:841–874. [PubMed: 2673024]
  3. Alonso MD, Lomako J, Lomako WM, Whelan WJ. A new look at the biogenesis of glycogen. FASEB J. 1995;9:1126–1137. [PubMed: 7672505]
  4. Busch C, Aktories K. Microbial toxins and the glycosylation of Rho family GTPases. Curr Opin Struct Biol. 2000;10:528–535. [PubMed: 11042449]
  5. Lee JY, Spicer AP. Hyaluronan: A multifunctional, megadalton, stealth molecule. Curr Opin Cell Biol. 2000;12:581–586. [PubMed: 10978893]
  6. Graves MV, Bernadt CT, Cerny R, Van Etten JL. Molecular and genetic evidence for a virus-encoded glycosyltransferase involved in protein glycosylation. Virology. 2001;285:332–345. [PubMed: 11437667]
  7. Tsai B, Ye Y, Rapaport TA. Retro-translocation of proteins from the endoplasmic reticulum into the cytosol. Nat Rev Mol Cell Biol. 2002;3:246–255. [PubMed: 11994744]
  8. Monsigny M, Rondanino C, Duverger E, Fajac I, Roche AC. Glyco-dependent nuclear import of glycoproteins, glycoplexes and glycosylated plasmids. Biochim. Biophys. Acta. 2004;1673:94–103. [PubMed: 15238252]
  9. Wang JL, Gray RM, Haudek KC, Patterson RJ. Nucleocytoplasmic lectins. Biochim. Biophys. Acta. 2004;1673:75–93. [PubMed: 15238251]
  10. West CM, van der Wel H, Sassi S, Gaucher EA. Cytoplasmic glycosylation of protein-hydrox-yproline and its relationship to other glycosylation pathways. Biochim. Biophys. Acta. 2004;1673:29–44. [PubMed: 15238247]
Copyright © 2009, The Consortium of Glycobiology Editors, La Jolla, California.
Bookshelf ID: NBK1915PMID: 20301242
PubReader format: click here to try


  • PubReader
  • Print View
  • Cite this Page
  • Disable Glossary Links

Related information

Related citations in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...