Fucosyltransferase 8 (FUT8) and β-galactoside α-2,6-sialyltransferase 1 (ST6GAL1) are glycosyltransferases that catalyze α1,6-fucosylation and α2,6-sialylation, respectively, in the mammalian N-glycosylation pathway. They are aberrantly expressed in various human diseases. FUT8 is non-glycosylated but is responsible for the fucosylation of ST6GAL1. However, the mechanism for the interaction between these two enzymes is unknown. In this study, we show that serum levels of α2,6-sialylated N-glycans are increased in Fut8−/− mice, whereas the mRNA and protein levels of ST6GAL1 are unchanged in mouse live tissues. The level of α2,6-sialylation on IgG was also enhanced in Fut8−/− mice along with ST6GAL1 catalytic activity increase in both serum and liver. Moreover, it was observed that ST6GAL1 prefers non-fucosylated substrates. Interestingly, increased core fucosylation accompanied by a reduction in α2,6-sialylation, was detected in rheumatoid arthritis patient serum. These findings provide new insight into the interactions between FUT8 and ST6GAL1.

Introduction

N-linked glycosylation is an important posttranslational process that regulates protein expression, folding, and bioactivity. Glycosylation reactions are catalyzed by various glycosyltransferases (GTases) such as α1,6-fucosyltransferase (FUT8), the only enzyme known to catalyze the transfer of a fucose residue from guanosine diphosphate (GDP)-fucose to position 6 of the innermost N-acetylglucosamine (GlcNAc) residue of an N-linked glycan to form α1,6-fucose in mammals [1]. The resultant core fucosylated N-glycans are found in a variety of glycoproteins. In the blood, α-fetoprotein, immunoglobulin (Ig)G, and haptoglobin are fucosylated glycoproteins whose various biological functions are regulated by this modification [2–4]. The physiological functions of core fucosylation have been investigated using Fut8-deficient (Fut8−/−) mice, which fail to thrive after birth and exhibit growth retardation and emphysema-like changes [5]. Lack of fucosylation of transforming growth factor-β1 and/or epidermal growth factor receptor leads to weak recognition by their respective ligands and a consequent down-regulation of signaling [5,6]. However, fucose deficiency increases the binding affinity of IgG1 antibody to Fc gamma receptor IIIa, resulting in natural killer cell-induced antibody-dependent cell-mediated cytotoxicity [2].

Sialic acid is the most abundant terminal monosaccharide of glycoconjugates on the surface of mammalian cells, and normally exists in α2,3, α2,6, and α2,8 configurations that are generated by a group of sialyltransferases [7]. β-galactoside α2,6-sialyltransferase (ST6GAL)1, which catalyzes the formation of α2,6-linked sialic acid in N-linked glycans, has 403 amino acids and a molecular mass of 47 kDa, and is a type II transmembrane glycoprotein with two N-glycosylation sites (N146 and N158), which appear to some core fucosylated N-glycans [8,9]. It can be hydrolyzed to a secreted form (molecular mass: 41 kDa) with a lumen-facing active domain by β-site APP-cleaving enzyme 1, with which it co-localizes in the Golgi apparatus [9,10]. N-glycosylation is essential for the catalytic activity of ST6GAL1, which is lost upon complete removal of all N-linked oligosaccharides [11].

RA is a prevalent chronic, systemic, and complex inflammatory autoimmune disorder characterized by the destruction of cartilage, synovitis, and hyperplastic pannus tissue, leading to severe joint destruction [12]. Currently, the presence of autoantibodies, such as anti-citrullinated protein antibody and rheumatoid factor, is widely used in the clinical diagnosis of RA. Among autoantibodies, IgG glycosylation has gained significant attention. In healthy individuals, ∼10% of IgG antibodies are sialylated; however, this fraction is markedly reduced in RA patients [13], who also exhibit elevated levels of core fucosylated native IgG [14]. Female RA patients express higher level of α2,6-sialylated IgG during pregnancy than postpartum, which is associated with relief of symptoms [15,16]. These results suggest that ST6GAL1 and FUT8 play a critical regulatory role in the immunoreaction responses in RA.

Although FUT8 is non-glycosylated, it mediates the core fucosylation of ST6GAL1 protein [8,17]. Both enzymes have been extensively studied in the context of abnormally expressed core fucose and α2,6-linked sialic acids as modifiers/regulators of N-glycan in certain diseases. Indeed, FUT8 is up-regulated in various cancers [18–20], whereas high levels of ST6GAL1 have been detected in tissues of patients with colorectal, breast, gastric, and cervical cancers [21–24]. In addition, the level of sialylated glycans was shown to be positively correlated with tumor invasion and metastasis [25]. However, the relationship between the expression/function of core fucose and α2,6-linked sialic acid is not well understood.

To address this question, we used a glycomic approach to investigate differences in serum N-glycan structure between Fut8+/+ and Fut8−/− mice. We found that overall N-glycosylation was altered in the absence of Fut8; in particular, α2,6-sialylated N-glycans were up-regulated in Fut8−/− mice. Although Fut8 deficiency had no effect on the expression level of ST6GAL1, its catalytic activity was increased in both the serum and liver. Moreover, ST6GAL1 was demonstrated that it prefers non-core fucosylated substrates in vitro. On the other hand, we also found that core fucosylation and FUT8 expression were increased, whereas α2,6-sialylation and ST6GAL1 catalytic activity were reduced in RA patients relative to healthy individuals. These results indicate that FUT8 modulates α2,6-sialylation of glycoproteins by regulating both the activity and it's preferred substrate concentration of ST6GAL1.

Materials and methods

Animals

The establishment of Fut8−/− mice has been described previously [5]. The Fut8+/+ and Fut8−/− mice on an Institute of Cancer Research genetic background were used and maintained in a room illuminated for 12 h (08:00–20:00) and kept at 24 ± 1°C with free access to food and water in specific pathogen-free laboratory animal facility. All animal experiments were approved by the Animal Care and Use Committee of Dalian Medical University and completed in Dalian Medical University Laboratory Animal Center. The approval number was AEE 17013. Blood samples were collected through retro-orbital puncture from the mice anesthetized by tribromoethanol (0.2 ml/10 g, i.p.). Cardiac perfusion was performed with cold phosphate-buffered saline (PBS).

Serum samples

Female patients (aged 39–85, mean age 56 ± 13) were diagnosed with RA according to the American College of Rheumatology 2010 criteria [26]. Serum samples from RA, healthy volunteers, and osteoarthritis (OA) patients (n = 3, respectively), as well as synovial fluid from RA and OA (n = 3, respectively) patients were collected at the Dalian Municipal Central Hospital affiliated with the Dalian Medical University, China. The clinicopathological factors of RA are summarized in Table 1. All samples were stored at −80°C. Each aliquot was thawed no more than two times before use. Informed consents were obtained from all patients and this study and all research protocols were approved by the Institutional Review Board of Dalian Medical University, in accordance with the established guidelines for the use of patient's information and samples.

Table 1.
Summary of information about RA patients
No.SexAgeAnti-CCP1 (U/ml)RF2 (IU/ml)CRP3 (mg/dl)ESR4 (mm/h)
Female 60 119.1 43.7 0.18 40 
Female 60 127.5 86.3 2.89 49 
Female 39 >200 1150 1.09 42 
Female 39 >200 271 2.74 49 
Female 47 102 68.3 1.87 98 
Female 52 110.5 22.3 0.95 21 
Female 61 >200 2800 1.42 72 
Female 53 52.7 211 3.6 63 
Female 54 79 31 1.43 — 
10 Female 73 >200 183 1.14 58 
11 Female 55 151.6 27 8.7 50 
12 Female 85 >200 512 6.2 56 
No.SexAgeAnti-CCP1 (U/ml)RF2 (IU/ml)CRP3 (mg/dl)ESR4 (mm/h)
Female 60 119.1 43.7 0.18 40 
Female 60 127.5 86.3 2.89 49 
Female 39 >200 1150 1.09 42 
Female 39 >200 271 2.74 49 
Female 47 102 68.3 1.87 98 
Female 52 110.5 22.3 0.95 21 
Female 61 >200 2800 1.42 72 
Female 53 52.7 211 3.6 63 
Female 54 79 31 1.43 — 
10 Female 73 >200 183 1.14 58 
11 Female 55 151.6 27 8.7 50 
12 Female 85 >200 512 6.2 56 
1

Anti-CCP: anti-cyclic citrullinated peptide;

2

RF: rheumatoid factor;

3

CRP: C-reactive protein;

4

ESR: erythrocyte sedimentation rate.

Reagents

Biotinylated lectins, Aleuria aurantia lectin (AAL), Lens culinaris lectin (LCA), Sambucus nigra lectin (SNA), Phaseolus vulgaris erythroagglutinin (PHA-E), Maackia amurensis lectin II (MAL II), Phaseolus vulgaris leucoagglutinin (PHA-L) (Supplementary Table S1) were purchased from Vector Laboratories Inc. (Burlingame, CA, U.S.A.). Rabbit anti-FUT8 polyclonal antibody (pAb) was purchased from Abcam (Cambrige, MA). Mouse anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) monoclonal antibody (mAb) was purchased from Proteintech Group (Chicago, IL). Anti-ST6GAL1 (M2 rabbit pAb) was purchased from Immuno-Biological Laboratories (Co., Ltd., Fujioka, Gunma, Japan). Streptavidin and goat anti-rabbit/mouse IgG conjugated with horseradish peroxidase (HRP) was purchased from Beyotime (Shanghai, China). Cytidine-5′-monophospho-N-acetylneuraminic acid (CMP-NeuAc) sodium salt, Guanosine 5′-diphospho-β-l-fucose (GDP-l-fucose) sodium salt, 2-(N-morpholino) ethanesulfonic acid (MES), ammonium acetate, 1-butanol and Direct Blue (DB)-71 were purchased from Sigma–Aldrich (St. Louis, MO, U.S.A.). Sodium cacodylate trihydrate was purchased from Acros Organics (Belgium, U.S.A.). The standard 2-aminopyridine (PA)-digalactosylated, -monosialylated and -disialylated biantennary oligosaccharides were purchased from Takara Bio (Dalian, China). Mouse CD45R (B220) MicroBeads was purchased from Miltenyi Biotec (Shanghai, China). Protein G Sefinose (TM) Resin (Settled Resin) was purchased from Sangon Biotech (Shanghai, China). Tert-amyl alcohol and 2,2,2-Tribromoethanol were purchased from Macklin Biochemical (Shanghai, China).

Immunoprecipitation of ST6GAL1 and IgG

A total of 100 mg liver tissue was snap-frozen in liquid nitrogen and ground; the powder was homogenized in 1 ml of lysis buffer containing a protease inhibitor mixture and centrifuged. The supernatant was incubated overnight at 4°C with gentle inversion with anti-ST6GAL1 antibody and protein G agarose. Immunoprecipitates were subjected to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS–PAGE), and the separated proteins were transferred to polyvinylidene difluoride (PVDF) membranes that were incubated with anti-ST6GAL1 antibody for immunoblot analysis or with LCA for lectin blotting, or were stained with DB-71 for N-glycan purification and matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) analysis [27].

IgG was captured from 25 μl pooled serum (n = 4) using 25 μl of Protein G beads in 500 μl of PBS. The proteins were allowed to interact with the beads with shaking for 4 h. After washing the beads three times with PBS, the IgG was released by boiling in 25 μl of 1 × loading buffer for 5 min [28].

Lectin and western blotting

Protein concentration was determined with a bicinchoninic acid protein assay kit (Takara Bio, Otsu, Japan). Proteins samples were separated by 10% SDS–PAGE. One gel was stained with Coomassie Brilliant Blue R-250, while proteins on another gel were transferred to a 0.45 μm PVDF membrane (Millipore, Bedford, MA, U.S.A.), the fixation of the electroblotted membranes were performed as recently reported by our group [29]. For Letin blotting, the electroblotted membranes were immersed in acetone at room temperature for 30 min, accompanied by gentle shaking, and followed by sample heating at 100°C for 30 min. For immunoblotting, the electroblotted membranes were immersed in acetone at 0°C for 30 min with gentle shaking, which was followed by heating at 50°C for 30 min. The fixed membranes were briefly immersed in methanol and then in Tris-buffered saline (TBS) containing 0.05% Tween-20 (TBS-T) for several minutes before their blocking with 5% bovine serum albumin (BSA) in TBS-T for 1 h. After washing three times for 5 min each with TBS-T, the membrane was incubated overnight at 4°C with the biotinylated AAL, LCA, SNA, and PHA-E diluted 1 : 15 000; MAL II and PHA-L diluted 1 : 1000; or anti-FUT8 (1 : 1000), anti-ST6GAL1 (1 : 100) antibodies, with anti-GAPDH antibody (1 : 5000) used as a loading control. After washing three times for 5 min each with TBS-T, the membrane was incubated for 1 h at room temperature with HRP-conjugated streptavidin or goat anti-rabbit/-mouse IgG diluted with TBS-T. Antibody binding was visualized using Enhanced Chemiluminescence Plus reagent (Beyotime Institute of Biotechnology, Shanghai, China).

Immunohistochemical and lectin histochemical analysis

Whole liver tissue from Fut8+/+ and Fut8−/− mice was fixed and embedded in paraffin, and cut into 4 μm-thick sections. After deparaffinization in xylene and hydration through a graded series of ethanol to H2O, antigen retrieval was performed in citrate solution (Beyotime Institute of Biotechnology) by heating in a microwave for 20 min. Endogenous peroxidase activity was blocked by incubation in 0.3% H2O2 for 15 min. After washing with PBS, sections were blocked with 5% BSA in TBS-T and incubated overnight at 4°C with lectin (LCA, 1 : 500 or SNA, 1 : 100) or antibody (anti-FUT8 and -ST6GAL1, 1 : 200) followed by HRP-conjugated secondary streptavidin or antibody for 2 h at room temperature. Sections were reacted with 3,3-diaminobenzidine and counterstained with hematoxylin and eosin for visualization.

Real-time PCR

Total RNA was isolated from liver tissue of Fut8+/+ and Fut8−/− mice using TRIzol reagent (Invitrogen, Carlsbad, CA) and converted into cDNA using the PrimeScript RT Reagent Kit with gDNA Eraser (Takara Bio) according to the manufacturer's protocol. Real-time PCR was carried out using SYBR Green Master Mix (Takara Bio) and melting curves were generated using StepOne software (Applied Biosystems, Foster City, CA, U.S.A.) to ensure that a single product was amplified. Data were analyzed by relative quantification [30]. Primers were designed using Oligo Primer Analysis software (Molecular Biology Insights, Colorado Springs, CO, U.S.A.) and are shown in Table 2.

Table 2.
Real-time PCR conditions and primer sequences for analysis of gene expression
GenePrimersAmplicon (bp)
FUT1 5′-ACTATTGGCACCTTTGGCT-3′
5′-ATTGATGCCCACCCACTCAG-3′ 
144 
FUT2 5′-AATAGTGAAGCTCCAAACCC-3′
5′-TCTGTTCTTGCCGCGTT-3′ 
78 
FUT4 5′-ATGAACTTCGAATCGCCCTC-3′
5′-CCATAGGGCACGAAGACATCC-3′ 
110 
FUT7 5′-AATCGCCCAGTAATACCCAT-3′
5′-AATCACGCCGATAGCTCAG-3′ 
75 
FUT9 5′-AGCACCGTGGAGTAGTTCAGCA-3′
5′-TGAAGCAGCCCAGGATGATG-3′ 
132 
FUT10 5′-AACACCTGGAGCCGTTT-3′
5′-GACCACCACAGCACGAT-3′ 
94 
FUT11 5′-AACGATCCCATGTTGCCTA-3′
5′-CATTCAGCCGAGCCAGT-3′ 
76 
ST6GAL1 5′-CTTTCTCCTGTTTGCCATC-3′
5′-AATGTAAGAGCCTCATAGTCG-3′ 
63 
ST3GAL1 5′-CAACAACCTGAGCGACACC-3′
5′-ATCTCAGGCCCATACGAGGA-3′ 
150 
ST3GAL2 5′-CGAGAACCCCTACCGCTT-3′
5′-TTGCCTGAGTTCCCGACCA-3′ 
69 
ST3GAL3 5′-GTCCTCGCCAACAAGTCT-3′
5′-GTCTTGCTGCCCACGTC-3′ 
107 
ST3GAL4 5′-CTTCTTCTCCGGGTGCT-3′
5′-CCACAACACAGCGACGACA-3′ 
82 
ST3GAL5 5′-GCCCTCAACCAGTTCGATG-3′
5′-ACGTGTTCAGAGTAACCCT-3′ 
68 
ST3GAL6 5′-ATAATTTTCCGTTGCCCTA-3′
5′-CACCTTTTACATGGCACT-3′ 
118 
NEU1 5′-CCAGCTCCGGCATTGTCT-3′
5′-GGCTGCTTCTTTCCATCCGT-3′ 
184 
SCT 5′-CTGCCGCTTACACCGTA-3′
5′-TAGCTAACAGGCCAACAC-3′ 
123 
MGAT3 5′-CAACGCCATCAACATCAAC-3′
5′-GTGGCGGATGTACTCGAAGG-3′ 
175 
MGAT5 5′-CATCCACCACGTCCCTGT-3′
5′-TTCCTGGTGACTTTCGGCTT-3′ 
163 
GAPDH 5′-AAATGGTGAAGGTCGGTGTG-3′
5′-TGAAGGGGTCGTTGATGG-3′ 
108 
GenePrimersAmplicon (bp)
FUT1 5′-ACTATTGGCACCTTTGGCT-3′
5′-ATTGATGCCCACCCACTCAG-3′ 
144 
FUT2 5′-AATAGTGAAGCTCCAAACCC-3′
5′-TCTGTTCTTGCCGCGTT-3′ 
78 
FUT4 5′-ATGAACTTCGAATCGCCCTC-3′
5′-CCATAGGGCACGAAGACATCC-3′ 
110 
FUT7 5′-AATCGCCCAGTAATACCCAT-3′
5′-AATCACGCCGATAGCTCAG-3′ 
75 
FUT9 5′-AGCACCGTGGAGTAGTTCAGCA-3′
5′-TGAAGCAGCCCAGGATGATG-3′ 
132 
FUT10 5′-AACACCTGGAGCCGTTT-3′
5′-GACCACCACAGCACGAT-3′ 
94 
FUT11 5′-AACGATCCCATGTTGCCTA-3′
5′-CATTCAGCCGAGCCAGT-3′ 
76 
ST6GAL1 5′-CTTTCTCCTGTTTGCCATC-3′
5′-AATGTAAGAGCCTCATAGTCG-3′ 
63 
ST3GAL1 5′-CAACAACCTGAGCGACACC-3′
5′-ATCTCAGGCCCATACGAGGA-3′ 
150 
ST3GAL2 5′-CGAGAACCCCTACCGCTT-3′
5′-TTGCCTGAGTTCCCGACCA-3′ 
69 
ST3GAL3 5′-GTCCTCGCCAACAAGTCT-3′
5′-GTCTTGCTGCCCACGTC-3′ 
107 
ST3GAL4 5′-CTTCTTCTCCGGGTGCT-3′
5′-CCACAACACAGCGACGACA-3′ 
82 
ST3GAL5 5′-GCCCTCAACCAGTTCGATG-3′
5′-ACGTGTTCAGAGTAACCCT-3′ 
68 
ST3GAL6 5′-ATAATTTTCCGTTGCCCTA-3′
5′-CACCTTTTACATGGCACT-3′ 
118 
NEU1 5′-CCAGCTCCGGCATTGTCT-3′
5′-GGCTGCTTCTTTCCATCCGT-3′ 
184 
SCT 5′-CTGCCGCTTACACCGTA-3′
5′-TAGCTAACAGGCCAACAC-3′ 
123 
MGAT3 5′-CAACGCCATCAACATCAAC-3′
5′-GTGGCGGATGTACTCGAAGG-3′ 
175 
MGAT5 5′-CATCCACCACGTCCCTGT-3′
5′-TTCCTGGTGACTTTCGGCTT-3′ 
163 
GAPDH 5′-AAATGGTGAAGGTCGGTGTG-3′
5′-TGAAGGGGTCGTTGATGG-3′ 
108 

Purification of N-glycans from serum and IgG for mass spectrometry

Pooled serum samples (n = 4, 300 μg) from 3-month-old Fut8+/+ and Fut8−/− mice were dissolved in 30 μl of 80 mM ammonium bicarbonate containing 10 mM dl-dithiothreitol at 60°C for 30 min. Then, 2.2 μl of 250 mM iodoacetamide was added to the mixtures, and incubated at room temperature for 60 min in the dark. The mixtures were then treated with 0.05 mg/ml trypsin (Sigma–Aldrich) at 37°C for 18 h, followed by heat-inactivation of the trypsin at 95°C for 5 min. Finally, 3 mU peptide N-glycanase F (PNGase F) (Takara Bio Inc., Dalian, China) was added, and the mixtures were incubated at 37°C for 18 h. The mixtures were diluted in distilled water (500 μl) and passed through cation exchange cartridge OASIS MCX (1 ml; Waters), and then the cartridge was washed with 1 ml of distilled water. The pass-through fraction and washings were combined and lyophilized. For α2,3-neuraminidase digestion, the lyophilized residues were dissolved in 40 μl distilled water containing 8 μl reaction buffer B and 2 μl Sialidase S (ProZyme, Inc., Richmond, CA) and incubated at 37°C for 20 h. Sample mixtures containing whole serum N-glycans with or without α2,3-neuraminidase digestion were treated using the BlotGlycoRglycan purification kit according to the manufacturer's protocol (Sumitomo Bakelite Co., Tokyo, Japan). Briefly, the released glycans of PNGase F digested samples were captured using BlotGlyco beads, followed by methylesterification of sialic acids with 3-methyl-1-p-tolytriazene (98%, Sigma–Aldrich). Then, the captured glycans were labeled and released with an aminooxy-functionalized peptide reagent (aoWR). Derivatized glycans were recovered from the resin by washing with 50 μl of distilled water. Finally, the excess reagent was removed using a cleanup column provided with the kit. The obtained solution containing the glycan derivatives was analyzed via MS.

For glycan analysis, purified IgG was electrophoresed by 10% SDS–PAGE, and then electroblotted onto PVDF membrane. The membrane was stained with DB-71, which does not give any signals derived from the dye during MALDI MS analysis. Stained IgG spots were excised from the membrane and then transferred into a microtube. After the membrane pieces were wet with methanol, 30 μl of solution containing 2 mU PNGase F (Takara Bio, Otsu, Japan) in 50 mM ammonium bicarbonate (pH 7.8) was added to the microtubes, which were incubated at 37°C for 18 h. The solutions in the microtubes were treated using an Oasis MCX cartridge (30 mg/ml; Waters, Milford, MA) and lyophilized. The dried material was then permethylated.

Permethylated glycans were prepared according to the method described previously [31] with some minor changes. Dimethyl sulfoxide (DMSO) containing 1% v/v distilled water (50 μl) was added to the dried glycan sample under alkaline conditions with powdered sodium hydroxide. Methyl iodide (50 μl) was then added to the mixture, and the reaction was performed at room temperature for 30 min with vigorous shaking. Distilled water (1 ml) was slowly added to the reaction mixture, and then the mixture was applied to a Sep-Pak Vac 18 cartridge (50 mg/ml; Waters) to purify the permethylated glycans. After washing with water (1 ml) several times, the permethylated glycans were eluted with 80% v/v acetonitrile in distilled (750 μl) and evaporated to dryness by the centrifugal evaporator.

Mass spectrometry

Spectra were acquired using a MALDI-TOF/TOF mass spectrometer (New ultrafleXtreme; Bruker Daltonik, Bremen, Germany). Ions were generated with a pulsed 337 nm nitrogen laser and accelerated to 25 kV. All spectra were obtained in the reflectron mode with delayed extraction of 200 ns. The measurements were carried out in the positive ion mode and were repeated at least three times. For sample preparation, 0.5 μl of 2,5-dihydroxybenzoic acid (DHB) (10 mg/ml) in 30% ethanol was spotted onto a target plate (MTP 384 target plate ground steel, Bruker Daltonik). After dried, an aliquot (0.5 μl) of the glycan solution was spotted onto the DHB crystal and dried. The intensity of the isotopic peaks of each glycan was normalized to an internal standard (maltoheptaose) with a known concentration.

Measurement of GTase activity

Membrane proteins from the liver of Fut8+/+ and Fut8−/− mice were prepared as previously described [32,33]. Briefly, mice were anesthetized with 1.2% w/v tribromoethanol in tert-amyl alcohol, and the liver was excised and homogenized in a 2-ml solution of 0.5 M sucrose, 5 mM 2-mercaptoethanol, 0.15 M NaCl, and 1 mM MgCl2. After centrifugation at 11 400×g for 10 min at 4°C, the supernatant was removed, and then centrifuged at 161 700×g for 90 min at 4°C using a ST50-ST rotor (Thermo Fisher Scientific, Waltham, MA, U.S.A.), the precipitate (Membrane proteins) was then collected in 25 mM sodium cacodylate acid (pH 6.8) with 2% Triton X-100 and stored at −80°C until measurement of enzyme activity.

The specific activity of FUT8 was determined as previously described [34]. Briefly, proteins from liver and serum as enzyme source were mixed with the assay buffer composed of 200 mM MES (pH 7.0), 1% Triton X-100, 500 µM donor (GDP-l-fucose), and 50 µM acceptor [GnGn-Asn-4-(2-pyridylamine) butylamine (PABA)]. After incubation for 8 h at 37°C, the reaction was terminated by boiling for 5 min.

To measure ST6GAL1 activity, proteins from liver and serum as the enzyme source were mixed with assay buffer composed of 50 mM sodium cacodylate buffer (pH 6.5), 300 μM donor [cytidine monophosphate (CMP)-NeuAc], and 1 µM acceptor (PA-digalactobiantennary oligosaccharide) and incubated at 37°C for 0, 4, 6, or 8 h. The reaction was terminated by boiling for 5 min.

To measure the catalytic activity of ST6GAL1 to the substrate with or without FUT8, α2,6-sialyltransferase from Photobacterium damsel as enzyme source were mixed with assay buffer composed of 100 mM MES-NaOH buffer (pH 7.5), 2 mM CMP-NeuAc, 4 µM acceptor (Fmoc-digalactobiantennary oligosaccharide), and 10 mM MgCl2 and incubated at 37°C for 30 min. Finally, the reaction was terminated by boiling for 5 min.

A 25-µl volume of the reaction mixture was analyzed by high-performance liquid chromatography (HPLC) with a fluorescent detector (excitation/emission: 320/400 nm) using a Nova-Pak C18 column (3.9 × 150 mm; Waters, Milford, MA, U.S.A.). Elution was performed at a flow rate of 0.8 ml/min at 55°C with 20 mM ammonium acetate (pH 4.0) for A, and the same buffer containing 1-butanol for B. The concentrations of 1-butanol for detection of FUT8 and ST6GAL1 were 0.1% and 0.05%, respectively [35].

Magnetic-activated cell sorting (MACS)

To collect mouse spleen B lymphocytes, spleens were crushed in PBS and red blood cells were removed with Tris-ammonium chloride lysis buffer (0.83% NH4Cl and 0.17 M Tris); the homogenate was passed through a strainer [36]. Then cluster of differentiation 45 receptor (CD45R+) cells from mouse spleen were isolated by MACS using CD45R (B220) MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany; 130-049-501) according to the manufacturer's instructions. Briefly, isolated splenocytes were incubated at 4°C for 15 min with MicroBeads in MACS buffer composed of 0.05% BSA, 2 mM EDTA, and PBS. Isolated CD45R+ cells were eluted in 1 ml MACS buffer for fluorescence-activated cell sorting (FACS).

Fluorescence-activated cell sorting

Mouse spleen B lymphocytes of Fut8+/+ and Fut8−/− mice were prepared. Single-cell suspensions were blocked in FACS buffer (0.5% BSA in PBS) on ice for 30 min, and incubated on ice with biotin-labeled LCA and SNA for 30 min. After washing, cells were incubated with Andy Fluor 488 streptavidin (GeneCopoeia, Rockville, MD, U.S.A.) at 4°C for 30 min, washed, and sorted on an Accuri C6 flow cytometer (BD Biosciences, Franklin Lakes, NJ, U.S.A.). Data were analyzed with FlowJo software (Treestar, San Carlos, CA, U.S.A.).

Statistical analysis

Band intensities were analyzed and compared using Image Lab software (Bio-Rad Laboratories) and GraphPad Prism version 6. All experiments were performed at least three times. Data are expressed as mean ± S.D. and were analyzed with the Student's t-test. The significance of differences was considered significant when p < 0.05.

Results

N-linked glycan levels of serum glycoproteins are altered in Fut8−/− mice

To investigate the impact of the loss of core fucosylation on overall N-glycan expression, serum glycoproteins from Fut8+/+ and Fut8−/− mice were analyzed by MALDI-TOF MS. Total N-linked glycan mixtures were released from pooled Fut8+/+ or Fut8−/− mouse sera (n = 4) with PNGase F, captured with BlotGlyco beads, and labeled with an aoWR that enhances glycan signals in MALDI-TOF MS [37]. The derivatized glycans were analyzed by MALDI-TOF MS (Figure 1A); all glycan signals in the MS spectra were assigned and are summarized in Table 3. To simplify the comparison of the profiles of the released glycans from Fut8+/+ or Fut8−/− mouse sera, the relative intensity of each glycan signal (1–31) to the intensity of internal standard glycan in the MS spectra was represented as a histogram (Figure 1B) along with estimated monosaccharide composition. In total, 31 possible N-glycan structures were speculated in Fut8+/+ and Fut8−/− mouse sera; nine of them (peak 4, 7, 10, 16, 21, 22, 23, 28, and 31) — corresponding to mono-fucosylated glycans present in Fut8+/+ samples — were completely absent from Fut8−/− mouse sera. However, signals 2, 5, 8, and 26 from non-fucosylated glycans were only detected in Fut8−/− samples. Signals 2, 5, and 8 lacking a core fucose unit were comparable to signals 4, 7, and 10. A total of 7 signals were up-regulated in Fut8−/− mouse sera (peak 6, 11, 14, 18, 19, 25, and 30) (Figure 1B and Table 3). Unexpectedly, all of these corresponded to sialylated glycans. N-glycan profiles revealed differences in sialylation levels between Fut8+/+ and Fut8−/− mice, with a higher level of sialylated oligosaccharides in the mutants. To identify the type of linkage by which sialic acid is attached to glycoproteins observed to be up-regulated in Fut8−/− mouse sera, we performed MS analysis of N-glycans treated with α2,3-neuraminidases following their release by PNGase F in Fut8+/+ and Fut8−/− mouse sera (Figure 1C). Nine glycans with α2,6-linked sialic acid (peaks 11, 17, 19, 20, 24, 25, 27, 29, and 30) were markedly up-regulated in Fut8−/− mouse sera after α2,3-neuraminidase digestion (Figure 1C,D and Supplementary Table S2).

Table 3.
Summary of N-linked glycans released from the serum of Fut8+/+ and Fut8−/− mice in MALDI-TOF MS
Peak no.Observed m/zCalculatedm/z1Chemical composition2Structure
Fut8+/+Fut8−/−
1649.81 1649.90 1649.67 (Hex)2 + (Man)3(GlcNAc)2  
— 1731.97 1731.72 (HexNAc)2 + (Man)3 (GlcNAc)2  
1811.85 1811.99 1811.72 (Hex)3 + (Man)3(GICNAc)2  
1877.92 — 1877.78 (HexNAc)2 (Fuc) + (Man)2 (GlcNAc)2  
— 1894.03 1893.77 (Hex) (HexNAc)2 + (Man)3 (GlcNAc)2  
2011.95 2012.11 2011.78 (Hex) (HexNAc) (NenGc) + (Man)3 (GlcNAc)2  
2039.98 — 2039.83 (Hex) (HexNAc)2(Fuc) + (Man)3 (GlcNAc)2  
— 2056.14 2055.83 (Hex)2 (HexNAc)2 + (Man)3 (GlcNAc)2  
2174.01 2174.18 2173.83 (Hex)2 (HcxNAc) (NcuGc) + (Man)3(GlcNAc)2  
10 2202.03 — 2201.88 (Hex)2 (HexNAc)2 (Fue) + (Man)3 (GlcNAc)2  
11 2215.05 2215.22 2214.86 (Hex) (HexNAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
12 2336.05 2336.26 2335.88 (Hex)3 (HexNAc) (NcuGc) + (Man)3 (GlcNAc)2  
13 2361.11 2361.29 2360.91 (Hex)2 (HexNAc)2 (NenAc) + (Man)3 (GlcNAc)2  
14 2377.12 2377.30 2376.91 (Hex)2 (HexNAc)2 (NeuGc) + (Man)3(GlcNAc)2  
15 2420.12 2420.29 2420.96 (Hex)3 (HexNAc)3 + (Man)3(ClcNAc)2  
16 2523.17 — 2522.97 (Hex)2 (HexNAc)2 (Fuc) (NeuGc) + (Man)3 (GlcNAc)2  
17 2666.21 2666.42 2666.01 (Hex)2 (HexNAc)2 (NeuAc)2 + (Man)3(GlcNAc)2  
18 2682.23 2682.45 2682.00 (Hex)2 (HexNAc)2 (NcuAc) (NeuGc) + (Man)3 (GlcNAc)2  
19 2698.24 2698.45 2698.00 (Hex)2 (HexNAc)2 (NeuGc)2 + (Man)3 (GlcNAc)2  
20 2740.26 2740.47 2740.02 (Hex) (HexNAc)3 (NeuGc)2 + (Man)3 (GlcNAc)2  
21 2812.31 — 2812.07 (Hex)2 (HexNAc)2 (Fuc) (NeuAc)2 + (Man)3 (GlcNAc)2  
22 2844.30 — 2844.06 (Hex)2 (HexNAc)2 (Fuc) (NeuGc)2 + (Man)3 (GlcNAc)2  
23 2886.33 — 2886.09 (Hex) (HexNAc)3 (Fuc) (NeuGc)2 + (Man)3 (GlcNAc)2  
24 2987.36 2987.58 2987.10 (Hex)2 (HexNAc)2 (NeuAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
25 3019.38 3019.60 3019.09 (Hex)2 (HexNAc)2 (NeuGc)3 +  (Man)3 (GlcNAc)2  
26 — 3031.59 3031.14 (Hex)3 (HexNAc)3 (NeuAc)2 + (Man)3 (GlcNAc)2  
27 3063.40 3063.65 3063.13 (Hex)3 (HexNAc)3 (NeuGc)2 + (Man)3(GlcNAc)2  
28 3165.44 — 3165.15 (Hex)2 (HexNAc)2 (Fuc) (NeuGc)3 + (Man)3 (GlcNAc)2  
29 3352.48 3352.76 3352.23 (Hex)3 (HexNAc)3 (NeuAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
30 3384.53 3384.80 3384.22 (Hex)3 (HexNAc)3 (NeuGc)3 + (Man)3(GlcNAc)2  
31 3530.60 — 3530.28 (Hex)3 (HexNAc)3 (Fuc) (NeuGc)3 + (Man)2 (GlcNAc)2  
Peak no.Observed m/zCalculatedm/z1Chemical composition2Structure
Fut8+/+Fut8−/−
1649.81 1649.90 1649.67 (Hex)2 + (Man)3(GlcNAc)2  
— 1731.97 1731.72 (HexNAc)2 + (Man)3 (GlcNAc)2  
1811.85 1811.99 1811.72 (Hex)3 + (Man)3(GICNAc)2  
1877.92 — 1877.78 (HexNAc)2 (Fuc) + (Man)2 (GlcNAc)2  
— 1894.03 1893.77 (Hex) (HexNAc)2 + (Man)3 (GlcNAc)2  
2011.95 2012.11 2011.78 (Hex) (HexNAc) (NenGc) + (Man)3 (GlcNAc)2  
2039.98 — 2039.83 (Hex) (HexNAc)2(Fuc) + (Man)3 (GlcNAc)2  
— 2056.14 2055.83 (Hex)2 (HexNAc)2 + (Man)3 (GlcNAc)2  
2174.01 2174.18 2173.83 (Hex)2 (HcxNAc) (NcuGc) + (Man)3(GlcNAc)2  
10 2202.03 — 2201.88 (Hex)2 (HexNAc)2 (Fue) + (Man)3 (GlcNAc)2  
11 2215.05 2215.22 2214.86 (Hex) (HexNAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
12 2336.05 2336.26 2335.88 (Hex)3 (HexNAc) (NcuGc) + (Man)3 (GlcNAc)2  
13 2361.11 2361.29 2360.91 (Hex)2 (HexNAc)2 (NenAc) + (Man)3 (GlcNAc)2  
14 2377.12 2377.30 2376.91 (Hex)2 (HexNAc)2 (NeuGc) + (Man)3(GlcNAc)2  
15 2420.12 2420.29 2420.96 (Hex)3 (HexNAc)3 + (Man)3(ClcNAc)2  
16 2523.17 — 2522.97 (Hex)2 (HexNAc)2 (Fuc) (NeuGc) + (Man)3 (GlcNAc)2  
17 2666.21 2666.42 2666.01 (Hex)2 (HexNAc)2 (NeuAc)2 + (Man)3(GlcNAc)2  
18 2682.23 2682.45 2682.00 (Hex)2 (HexNAc)2 (NcuAc) (NeuGc) + (Man)3 (GlcNAc)2  
19 2698.24 2698.45 2698.00 (Hex)2 (HexNAc)2 (NeuGc)2 + (Man)3 (GlcNAc)2  
20 2740.26 2740.47 2740.02 (Hex) (HexNAc)3 (NeuGc)2 + (Man)3 (GlcNAc)2  
21 2812.31 — 2812.07 (Hex)2 (HexNAc)2 (Fuc) (NeuAc)2 + (Man)3 (GlcNAc)2  
22 2844.30 — 2844.06 (Hex)2 (HexNAc)2 (Fuc) (NeuGc)2 + (Man)3 (GlcNAc)2  
23 2886.33 — 2886.09 (Hex) (HexNAc)3 (Fuc) (NeuGc)2 + (Man)3 (GlcNAc)2  
24 2987.36 2987.58 2987.10 (Hex)2 (HexNAc)2 (NeuAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
25 3019.38 3019.60 3019.09 (Hex)2 (HexNAc)2 (NeuGc)3 +  (Man)3 (GlcNAc)2  
26 — 3031.59 3031.14 (Hex)3 (HexNAc)3 (NeuAc)2 + (Man)3 (GlcNAc)2  
27 3063.40 3063.65 3063.13 (Hex)3 (HexNAc)3 (NeuGc)2 + (Man)3(GlcNAc)2  
28 3165.44 — 3165.15 (Hex)2 (HexNAc)2 (Fuc) (NeuGc)3 + (Man)3 (GlcNAc)2  
29 3352.48 3352.76 3352.23 (Hex)3 (HexNAc)3 (NeuAc)2 (NeuGc) + (Man)3 (GlcNAc)2  
30 3384.53 3384.80 3384.22 (Hex)3 (HexNAc)3 (NeuGc)3 + (Man)3(GlcNAc)2  
31 3530.60 — 3530.28 (Hex)3 (HexNAc)3 (Fuc) (NeuGc)3 + (Man)2 (GlcNAc)2  

m/z values are monoisotopic. ‘—' indicates ‘not detected'.

1

The glycans were calculated as aminooxy tryptophanylarginine labeled derivatives in which neuraminic acids were methylesterified, and [M+H]+;

2

Monosaccharide compositions were determined by database searching using GlycoMod (http://www.expasy.ch/tools/glycomod/). Monosaccharides are indicated as follows: Hex, hexose; HexNAc, N-acetylhexosamine; Fuc, fucose ; GlcNAc, N-acetylglucosamin ; NeuAc, N-acetylneuraminic acid ; NeuGc, N-glycolylneuraminic acid ; Man, mannose ; Gal, Galactose .

Glycan analysis by MS.

Figure 1.
Glycan analysis by MS.

(A) MALDI-TOF MS spectra of N-glycans of pooled sera (n = 4) from Fut8+/+ (upper) and Fut8−/− (lower) mice, without α2,3-neuraminidase digestion. (B) Histograms of relative glycan intensities from panel (A). (C) MALDI-TOF MS spectra of N-glycans of pooled serum (n = 4) from Fut8+/+ (upper) and Fut8−/− (lower) mice with α2,3-neuraminidase digestion. (D) Histograms of relative glycan intensities from panel (C). The relative intensities of each glycan were calculated by normalizing the signal intensity of an individual glycan to the intensity of internal standard glycan. # indicates internal standard glycan. The relative intensities of released glycan from sera of Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively. Glycan signals were shown by numerals and are summarized in Table 3 and Supplementary Table S2. *, significantly different p < 0.05, **, p < 0.01, ***, p < 0.001. Each column represents the mean ± S.D. of three different experiments.

Figure 1.
Glycan analysis by MS.

(A) MALDI-TOF MS spectra of N-glycans of pooled sera (n = 4) from Fut8+/+ (upper) and Fut8−/− (lower) mice, without α2,3-neuraminidase digestion. (B) Histograms of relative glycan intensities from panel (A). (C) MALDI-TOF MS spectra of N-glycans of pooled serum (n = 4) from Fut8+/+ (upper) and Fut8−/− (lower) mice with α2,3-neuraminidase digestion. (D) Histograms of relative glycan intensities from panel (C). The relative intensities of each glycan were calculated by normalizing the signal intensity of an individual glycan to the intensity of internal standard glycan. # indicates internal standard glycan. The relative intensities of released glycan from sera of Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively. Glycan signals were shown by numerals and are summarized in Table 3 and Supplementary Table S2. *, significantly different p < 0.05, **, p < 0.01, ***, p < 0.001. Each column represents the mean ± S.D. of three different experiments.

MALDI-TOF/TOF MS/MS defines core fucosylation status and the structural assignment of N-glycans

The structural assignment of N-glycans observed in the MS analyses was performed based on subsequent MS/MS analysis data using MALDI-TOF/TOF MS. MS/MS spectra of precursor ions with m/z 2698.20, 2698.24, 2844.38, 2056.25, and 3530.63 are shown in Figure 2. We showed that all product ions obtained from the sodiated ion at m/z 2698 were identical between Fut8+/+ and Fut8−/− mice. At the m/z 2698, the N-glycans were sialylated, as indicated by the fragments ions at m/z 687.06, 2011.55, 2214.86, and 2376.47. At the m/z 2844.38, the N-glycans were core fucosylated, as suggested by the fragment ions at m/z 782.35, 985.46, 1147.39, 1309.48, 1471.43, 1674.52, 2157.71, and 2523.57. At the m/z 3530.63, the N-glycans were core fucosylated, as proposed by the fragment ions at m/z 782.26, 985.16, 1147.47, 1310.27, 1471.89, 1674.24, 2845.05, and 3120.58. These data indicated that a fucose is attached to the reducing end, GlcNAc, in these glycans and the sialylated N-glycan structure. In conclusion, MS profiling in combination with tandem mass spectrometry revealed that glycans with a core fucose disappeared and that α2,6-sialylation was increased in mouse sera glycoproteins by loss of Fut8.

MALDI-TOF/TOF tandem mass spectra of the selected N-glycans provide additional structural data.

Figure 2.
MALDI-TOF/TOF tandem mass spectra of the selected N-glycans provide additional structural data.

Signals presented as [M+H]+ molecular ion at m/z 2698, 2844, and 3530 from Fut8+/+ and m/z 2698 and, 2056 from Fut8−/− mice in Figure 1 were subjected to tandem MS, and the resulting MS/MS data are shown in (A)–(D), respectively. The horizontal arrows on the spectra indicate the losses from the molecular ion of the designated glycan moieties.

Figure 2.
MALDI-TOF/TOF tandem mass spectra of the selected N-glycans provide additional structural data.

Signals presented as [M+H]+ molecular ion at m/z 2698, 2844, and 3530 from Fut8+/+ and m/z 2698 and, 2056 from Fut8−/− mice in Figure 1 were subjected to tandem MS, and the resulting MS/MS data are shown in (A)–(D), respectively. The horizontal arrows on the spectra indicate the losses from the molecular ion of the designated glycan moieties.

Glycan patterns of serum glycoproteins are altered in Fut8−/− mice

We examined the glycan patterns of glycoproteins by lectin blotting of sera from Fut8+/+ and Fut8−/− mice (Figure 3 and Supplementary Table S1). AAL and LCA, which preferentially recognize fucose and α1,6-fucose (i.e. core fucose), respectively, were strongly detected in Fut8+/+ samples but were absent from the sera of Fut8−/− mice. A marked increase in α2,6-sialylation and bisecting GlcNAc was observed in Fut8−/− mouse sera by SNA and PHA-E blotting; this was accompanied by decreases in the signals corresponding to MAL II and PHA-L, which preferentially recognize α2,3-sialylation and multiantennary N-glycans (Figure 3A), also band intensities were analyzed in Figure 3B. In accordance with the MALDI-TOF MS data, α2,6-sialylated N-glycans (SNA binding) were enhanced in Fut8−/− samples.

Increased α2,6-sialylation level in Fut8−/− mice.

Figure 3.
Increased α2,6-sialylation level in Fut8−/− mice.

(A) Lectin blotting of pooled sera (1.5 μg of proteins) from Fut8+/+ (n = 4) and Fut8−/− (n = 4) mice were subjected to 10% SDS–PAGE and stained with Coomassie Brilliant Blue (CBB), and other electroblotted membranes were stained with AAL, LCA, SNA, PHA-E, MAL, and PHA-L lectins. AAL binds to 1-2/3/4/6-linked fucose residues. LCA binds to 1,6-linked fucose residues. SNA binds to 2,6-linked sialic acid residues. PHA-E binds with bisecting GlcNAc N-glycans. MAL II binds to 2,3-linked sialic acid residues. PHA-L binds to β1,6-GlcNAc-branched N-glycans. (B) The relative intensities of lectin staining of pooled sera from Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively (n = 3 individual experiments). Data represent mean ± S.D. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3.
Increased α2,6-sialylation level in Fut8−/− mice.

(A) Lectin blotting of pooled sera (1.5 μg of proteins) from Fut8+/+ (n = 4) and Fut8−/− (n = 4) mice were subjected to 10% SDS–PAGE and stained with Coomassie Brilliant Blue (CBB), and other electroblotted membranes were stained with AAL, LCA, SNA, PHA-E, MAL, and PHA-L lectins. AAL binds to 1-2/3/4/6-linked fucose residues. LCA binds to 1,6-linked fucose residues. SNA binds to 2,6-linked sialic acid residues. PHA-E binds with bisecting GlcNAc N-glycans. MAL II binds to 2,3-linked sialic acid residues. PHA-L binds to β1,6-GlcNAc-branched N-glycans. (B) The relative intensities of lectin staining of pooled sera from Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively (n = 3 individual experiments). Data represent mean ± S.D. *p < 0.05, **p < 0.01, ***p < 0.001.

Expressions levels of ST6GAL1 are not altered in Fut8−/− mice

The above results suggested that core fucose deficiency-induced α2,6-sialylation. In humans, glycomic changes in serum proteins can be subdivided into changes caused by the liver and those caused by B cells. We, therefore, quantified the mRNA levels of fucosyltransferases and sialyltransferases in liver tissue by real-time PCR to determine how FUT8 affects sialylation (Figure 4A). Fut8 transcript was absent from Fut8−/− mice, with concomitant down-regulation of FUT1, FUT7, FUT10, and FUT11 mRNA as compared with Fut8+/+ mice. We also observed that loss of Fut8 reduced the expression of ST3GAL3, ST3GAL4, ST3GAL5, and mannosyl (α-1,6-)-glycoprotein β-1,6-N-acetyl-glucosaminyltransferase (MGAT5). However, there were no differences in the expressions levels of ST6GAL1, neuraminidase 1, and sialic acid transporter, which were involved in the synthesis of α2,6-sialylation on N-glycan, between Fut8+/+ and Fut8−/− liver tissues. ST6GAL1 plays a critical role in the formation of α2,6-sialylation on N-glycan [38]. The absence of Fut8 and the lack of difference in ST6GAL1 protein level were confirmed by western blot analysis of liver homogenate and sera (Figure 4B), and lectin histochemical and immunohistochemical analyses (Figure 4C). These results indicate that the expression level of ST6GAL1 was not altered, although α2,6-sialylated N-glycans were up-regulated in Fut8−/− mice.

The expression level of ST6GAL1 was not altered in Fut8−/− mice.

Figure 4.
The expression level of ST6GAL1 was not altered in Fut8−/− mice.

(A) Expression levels of genes involved in protein fucosylation and sialylation and N-acetylglucosamine (GlcNAc) moieties were examined by real-time PCR using total RNA extracted from livers of Fut8+/+ and Fut8−/− mice. GAPDH served as a loading control. FUT, α1,2/3/6-fucosyltransferase; MGAT3, mannosyl (β-1,4-)-glycoprotein β-1,4-N-acetylglucosaminyltransferase; MGAT5, mannosyl (α-1,6-)-glycoprotein β-1,6-N-acetyl-glucosaminyltransferase; NEU, neuraminidase; SCT, sialic acid transporter; ST3GAL/ST6GAL, β-galactoside α2,3/6-sialyltransferase. Data represent mean ± S.D. (n = 3), *p < 0.05, **p < 0.01, ***p < 0.001. (B) Expression levels of protein of FUT8 and ST6GAL1 in liver homogenate (left) and serum (right) of Fut8+/+ and Fut8−/− mice by western blotting. GAPDH was detected as a control and Coomassie Brilliant Blue (CBB) staining was performed. (C) Immunohistochemical and lectin histochemical analyses of liver tissue from Fut8+/+ and Fut8−/− mice stained with anti-FUT8, anti-ST6GAL1 antibodies, LCA, and SNA lectin (left). Percentage of cells positive for FUT8, ST6GAL1, LCA, and SNA (right). Data represent mean ± S.D. (n = 3). *p < 0.05, **p < 0.01.

Figure 4.
The expression level of ST6GAL1 was not altered in Fut8−/− mice.

(A) Expression levels of genes involved in protein fucosylation and sialylation and N-acetylglucosamine (GlcNAc) moieties were examined by real-time PCR using total RNA extracted from livers of Fut8+/+ and Fut8−/− mice. GAPDH served as a loading control. FUT, α1,2/3/6-fucosyltransferase; MGAT3, mannosyl (β-1,4-)-glycoprotein β-1,4-N-acetylglucosaminyltransferase; MGAT5, mannosyl (α-1,6-)-glycoprotein β-1,6-N-acetyl-glucosaminyltransferase; NEU, neuraminidase; SCT, sialic acid transporter; ST3GAL/ST6GAL, β-galactoside α2,3/6-sialyltransferase. Data represent mean ± S.D. (n = 3), *p < 0.05, **p < 0.01, ***p < 0.001. (B) Expression levels of protein of FUT8 and ST6GAL1 in liver homogenate (left) and serum (right) of Fut8+/+ and Fut8−/− mice by western blotting. GAPDH was detected as a control and Coomassie Brilliant Blue (CBB) staining was performed. (C) Immunohistochemical and lectin histochemical analyses of liver tissue from Fut8+/+ and Fut8−/− mice stained with anti-FUT8, anti-ST6GAL1 antibodies, LCA, and SNA lectin (left). Percentage of cells positive for FUT8, ST6GAL1, LCA, and SNA (right). Data represent mean ± S.D. (n = 3). *p < 0.05, **p < 0.01.

De-core fucosylation enhances the catalytic activity of ST6GAL1

FUT8 regulates a variety of physiological functions by modifying target proteins [17]. Since ST6GAL1 is a core fucosylated protein [8], we speculated that its catalytic activity would be altered in the absence of Fut8. To address this question, we first confirmed whether ST6GAL1 protein was modified by core fucose by LCA staining of ST6GAL1 immunoprecipitated from Fut8+/+ and Fut8−/− mouse liver. We found that ST6GAL1 from Fut8−/− mice lacked core fucosylation (Figure 5A-i). To understand the detailed glycan structures on ST6GAL1, immunoprecipitated ST6GAL1 from Fut8+/+ and Fut8−/− liver were electrophoresed by SDS–PAGE and blotted onto PVDF membrane, and stained by DB-71 (Figure 5A-ii). The N-glycans released from the stained bands were permethylated and determined by MALDI-TOF MS. The MS spectrum and monosaccharide composition of the N-linked glycans of ST6GAL1 are summarized in Figure 5B and Table 4, respectively. All ST6GAL1 glycans in Fut8−/− mice that lacked a core fucose unit were confirmed (Figure 5B, lower), whereas ST6GAL1 glycans with and without core fucose were present in Fut8+/+ mice (Figure 5B, upper). However, only one sialylated glycans (signals 14) were identified in ST6GAL1 from Fut8−/− mice, while there were seven sialylated glycans in Fut8+/+ mice (Figure 5B, Table 4). The linkage types of these sialylated glycans should be α2,3-linked glycans [8]. These results were consistent with the fact that α2,3-sialylated N-glycans (Figure 3, MAL II lectin staining) were reduced in the serum of Fut8−/− mice.

ST6GAL1 activity was increased by the absence of core fucosylation in liver and serum of Fut8−/− mice.

Figure 5.
ST6GAL1 activity was increased by the absence of core fucosylation in liver and serum of Fut8−/− mice.

(A) Liver lysates from Fut8+/+ and Fut8−/− mice were immunoprecipitated with an anti-ST6GAL1 antibody. The immunoprecipitates were resolved by 10% SDS–PAGE and transferred to PVDF membrane, and probed with the LCA and anti-ST6GAL1 antibody (A-i), and stained with Direct Blue-71 (A-ii). (B) The N-glycans were released from bands corresponding to ST6GAl1 from Fut8+/+ and Fut8−/− mice (A-ii) and were analyzed by MALDI-TOF MS spectrum. Glycan peaks are numbered and summarized in Table 4. (C) Reverse-phase HPLC analyses the activity of FUT8 and ST6GAL1. FUT8 activity in liver membrane proteins and in sera from Fut8+/+ and Fut8−/− mice (C-i,iii) was examined using GDP-l-fucose and PABA labeled oligosaccharide as donor and acceptor substrate, respectively, as described in the ‘Materials and Methods.' The substrate (S1: PABA-G0) and FUT8 product (P1: PABA-G0F) were eluted at 6.5 and 13 min in liver and serum, respectively. Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P2 (PA-G2NA2) and P3 (PA-G2NA1) representing products of the ST6GAL1 reaction were eluted at 12 and 13 min in liver and serum, respectively. Glycan names indicate the presence of N-acetylneuraminic acid (NA1 and NA2), galactoses (G0 and G2), and fucose (F). The quantitative data are presented as means ± S.D. from three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Figure 5.
ST6GAL1 activity was increased by the absence of core fucosylation in liver and serum of Fut8−/− mice.

(A) Liver lysates from Fut8+/+ and Fut8−/− mice were immunoprecipitated with an anti-ST6GAL1 antibody. The immunoprecipitates were resolved by 10% SDS–PAGE and transferred to PVDF membrane, and probed with the LCA and anti-ST6GAL1 antibody (A-i), and stained with Direct Blue-71 (A-ii). (B) The N-glycans were released from bands corresponding to ST6GAl1 from Fut8+/+ and Fut8−/− mice (A-ii) and were analyzed by MALDI-TOF MS spectrum. Glycan peaks are numbered and summarized in Table 4. (C) Reverse-phase HPLC analyses the activity of FUT8 and ST6GAL1. FUT8 activity in liver membrane proteins and in sera from Fut8+/+ and Fut8−/− mice (C-i,iii) was examined using GDP-l-fucose and PABA labeled oligosaccharide as donor and acceptor substrate, respectively, as described in the ‘Materials and Methods.' The substrate (S1: PABA-G0) and FUT8 product (P1: PABA-G0F) were eluted at 6.5 and 13 min in liver and serum, respectively. Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P2 (PA-G2NA2) and P3 (PA-G2NA1) representing products of the ST6GAL1 reaction were eluted at 12 and 13 min in liver and serum, respectively. Glycan names indicate the presence of N-acetylneuraminic acid (NA1 and NA2), galactoses (G0 and G2), and fucose (F). The quantitative data are presented as means ± S.D. from three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

Table 4.
Summary of N-linked glycans released from ST6GAL1 of Fut8+/+ and Fut8−/− mice liver in MALDI-TOF MS
Peak no.Observed m/zCalculated m/z1Chemical composition2Stucture
Fut8+/+Fut8−/−
1579.74 1579.79 1579.78 (Man)2 + (Man)3(HexNAc)2  
1620.75 — 1620.77 (Hex)(HexNAc) + (Man)3(HexNAc)2  
1661.79 1661.82 1661.84 (HexNAc)2 + (Man)3(HexNAc)2  
1783.85 1783.90 1783.88 (Hex)3 + (Man)3(GlcNAc)2  
1835.87 — 1835.93 (HexNAc)2(Fuc) + (Man)3(HexNAc)2  
1865.89 1865.89 1865.94 (Hex)(HexNAc)2 + (Man)3(GlcNAc)2  
1987.91 1987.99 1987.98 (Hex)4 + (Man)3(GlcNAc)2  
2039.97 — 2040.03 (Hex)(HcxNAc)2(Fuc) + (Man)3(HexNAc)2  
2069.99 2070.03 2070.04 (Hex)2(HexNAc)2 + (Man)3(HexNAc)2  
10 2192.01 2192.09 2192.08 (Hex)5 + (Man)3(GlcNAc)2  
11 2244.04 — 2244.13 (Hex)2(HexNAc)2(Fuc) + (Man)3(GlcNAc)2  
12 2396.12 2396.17 2396.18 (Hex)6 + (Man)3(GlcNAc)2  
13 2431.13 — 2431.21 (Hex)2(HexNAc)2(NeuAc) + (Man)3(GlcNAc)2  
14 2461.12 2461.20 2461.22 (Hex)2(HexNAc)2(NeuGc) + (Man)3(GlcNAc)2  
15 — 2519.26 2519.26 (Hex)3(HexNAc)3 + (Man)3(HexNAc)2  
16 2605.18 — 2605.30 (Hex)2(HexNAc)2(Fuc)(NeuAc) + (Man)3(GlcNAc)2  
17 2635.21 — 2635.31 (Hcx)2(HexNAc)2(Fuc)(NeuGc) + (Man)3(GlcNAc)2  
18 2792.26 — 2792.38 (Hex)2(HexNAc)2(NeuAc)2 + (Man)3(GlcNAc)2  
19 2852.27 — 2852.41 (Mex)2(HexNAc)2(NeuGc)2 + (Man)3(MexNAc)2  
20 2966.32 — 2966.47 (Hex)2(HexNAc)2(Fuc)(NeuAc)2 + (Man)3(GlcNAc)2  
Peak no.Observed m/zCalculated m/z1Chemical composition2Stucture
Fut8+/+Fut8−/−
1579.74 1579.79 1579.78 (Man)2 + (Man)3(HexNAc)2  
1620.75 — 1620.77 (Hex)(HexNAc) + (Man)3(HexNAc)2  
1661.79 1661.82 1661.84 (HexNAc)2 + (Man)3(HexNAc)2  
1783.85 1783.90 1783.88 (Hex)3 + (Man)3(GlcNAc)2  
1835.87 — 1835.93 (HexNAc)2(Fuc) + (Man)3(HexNAc)2  
1865.89 1865.89 1865.94 (Hex)(HexNAc)2 + (Man)3(GlcNAc)2  
1987.91 1987.99 1987.98 (Hex)4 + (Man)3(GlcNAc)2  
2039.97 — 2040.03 (Hex)(HcxNAc)2(Fuc) + (Man)3(HexNAc)2  
2069.99 2070.03 2070.04 (Hex)2(HexNAc)2 + (Man)3(HexNAc)2  
10 2192.01 2192.09 2192.08 (Hex)5 + (Man)3(GlcNAc)2  
11 2244.04 — 2244.13 (Hex)2(HexNAc)2(Fuc) + (Man)3(GlcNAc)2  
12 2396.12 2396.17 2396.18 (Hex)6 + (Man)3(GlcNAc)2  
13 2431.13 — 2431.21 (Hex)2(HexNAc)2(NeuAc) + (Man)3(GlcNAc)2  
14 2461.12 2461.20 2461.22 (Hex)2(HexNAc)2(NeuGc) + (Man)3(GlcNAc)2  
15 — 2519.26 2519.26 (Hex)3(HexNAc)3 + (Man)3(HexNAc)2  
16 2605.18 — 2605.30 (Hex)2(HexNAc)2(Fuc)(NeuAc) + (Man)3(GlcNAc)2  
17 2635.21 — 2635.31 (Hcx)2(HexNAc)2(Fuc)(NeuGc) + (Man)3(GlcNAc)2  
18 2792.26 — 2792.38 (Hex)2(HexNAc)2(NeuAc)2 + (Man)3(GlcNAc)2  
19 2852.27 — 2852.41 (Mex)2(HexNAc)2(NeuGc)2 + (Man)3(MexNAc)2  
20 2966.32 — 2966.47 (Hex)2(HexNAc)2(Fuc)(NeuAc)2 + (Man)3(GlcNAc)2  

m/z values are monoisotopic. ‘—' indicates ‘not detected'.

1

The glycans were calculated as permethylated alditol and [M+Na]+;

2

Monosaccharide compositions were determined by database searching using GlycoMod (http://www.expasy.ch/tools/glycomod/). Monosaccharides are indicated as follows: Hex, hexose; HexNAc, N-acetylhexosamine; Fuc, fucose ; GlcNAc, N-acetylglucosamine ; NeuAc, N-acetylneuraminic acid ; NeuGc, N-glycolylneuraminic acid ; Man, mannose ; Gal, Galactose .

We then compared the enzymatic activity of FUT8 and ST6GAL1 from Fut8+/+ and Fut8−/− mouse liver and serum by HPLC. A typical elution pattern of the FUT8 and ST6GAL1 reaction product is shown in Figure 5C. The pattern was the same as that of authentic GnGn-bi-Asn-PABA, PA-monosialylated and -disialylated biantennary oligosaccharides. The amount of ST6GAL1 product gradually increased over time (0, 4, 6, and 8 h) (Supplementary Figure S1). FUT8 activity was abolished in Fut8−/− liver (Figure 5C-i), whereas ST6GAL1-mediated α2,6-hypersialylation was detected. Moreover, the level of α2,6-sialylation of products (P2) was higher in Fut8−/− than in Fut8+/+ liver (Figure 5C-ii). The enzymatic activities of both FUT8 and ST6GAL1 in serum from Fut8+/+ and Fut8−/− mice were similar to those in the liver (Figure 5C-iii,iv). These results demonstrate that the loss of Fut8 enhances the catalytic activity of ST6GAL1 in the serum and liver.

Some of the GTases have a catalytic preference for substrates [39]. There is no core fucosylation in the substrate of ST6GAL1 in Fut8−/− mice. Therefore, we tested whether ST6GAL1 has a preference for substrates that without core fucosylation by HPLC in vitro. As shown in Supplementary Figure S2, during the same reaction time, Only 3.36% of S1 (Fmoc-G2) remained, while 13.44% of S2 (Fmoc-G2-F) remained, their products are 28.25 + 68.39% and 48.92 + 37.64%, respectively. Both the decrease in substrates and the increase in the products can indicate that ST6GAL1 has a preference to non-core fucosylated substrates.

IgG glycosylation is altered by loss of Fut8

IgG is the most abundant glycoprotein in serum, and is a target for core fucosylation and α2,6-sialylation [40]. We, therefore, predicted that IgG in Fut8−/− mouse sera would show higher levels of sialylated N-glycan. For glycan analysis, the same quality of purified IgG derived from Fut8+/+ and Fut8−/− was subjected to SDS–PAGE and blotted onto PVDF membrane, and stained by DB-71 (Supplementary Figure S3). The N-glycans released from the stained bands were permethylated and determined by MALDI-TOF MS. We found that IgG glycans in Fut8−/− mice lacked a core fucose unit (Figure 6A, lower), whereas most of those in Fut8+/+ mice were modified with core fucose (Figure 6A, upper). Furthermore, signals 15 (m/z = 2257), 18 (m/z = 2461), and 20 (m/z = 2852) corresponding to sialylated glycans were observed only in the spectrum of IgG from Fut8−/− mice (Figure 6 and Supplementary Table S3), consistent with the overall variation in N-glycosylation of serum proteins (Figure 1). Analysis of IgG glycosylation was also performed by lectin blotting using AAL, LCA, SNA, and PHA-E lectins, as shown in Figure 6B, and band intensities analysis is presented in Figure 6C, the results consistent with the results of the lectin blot analysis of serum proteins (Figure 3).

IgG sialylation was increased in Fut8−/− mouse serum.

Figure 6.
IgG sialylation was increased in Fut8−/− mouse serum.

(A) Mass spectra of permethylated glycans of IgG immunoprecipitated from pooled sera (n = 4) of Fut8+/+ (upper) and Fut8−/− (lower) mice. Glycan signals were shown as boldface numerals and were summarized in Supplementary Table S3. # indicates glycan internal standard. (B) Lectin blotting of the same quality of purified IgG from pooled sera (n = 4) of Fut8+/+ and Fut8−/− mice were subjected to 10% SDS–PAGE and stained with Coomassie Brilliant Blue (CBB), and other electroblotted membranes were stained with AAL, LCA, SNA, and PHA-E lectins. AAL bind to 1-2/3/4/6-linked fucose residues. LCA bind to 1,6-linked fucose residue. SNA bind to 2-6-linked sialic acid residues. PHA-E bind with bisecting GlcNAc N-glycans. (C) The relative intensities of lectin staining of purified IgG from Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively (n = 3 individual experiments). Data represent mean ± S.D. **p < 0.01, ***p < 0.001.

Figure 6.
IgG sialylation was increased in Fut8−/− mouse serum.

(A) Mass spectra of permethylated glycans of IgG immunoprecipitated from pooled sera (n = 4) of Fut8+/+ (upper) and Fut8−/− (lower) mice. Glycan signals were shown as boldface numerals and were summarized in Supplementary Table S3. # indicates glycan internal standard. (B) Lectin blotting of the same quality of purified IgG from pooled sera (n = 4) of Fut8+/+ and Fut8−/− mice were subjected to 10% SDS–PAGE and stained with Coomassie Brilliant Blue (CBB), and other electroblotted membranes were stained with AAL, LCA, SNA, and PHA-E lectins. AAL bind to 1-2/3/4/6-linked fucose residues. LCA bind to 1,6-linked fucose residue. SNA bind to 2-6-linked sialic acid residues. PHA-E bind with bisecting GlcNAc N-glycans. (C) The relative intensities of lectin staining of purified IgG from Fut8+/+ and Fut8−/− mice were indicated by black and gray columns, respectively (n = 3 individual experiments). Data represent mean ± S.D. **p < 0.01, ***p < 0.001.

IgG is produced by B cells; ST6GalI is particularly highly expressed in the liver, there are also high expression in the lactating mammary gland, intestinal epithelia of newborn animals, and B cells [41]. We, therefore, investigated whether the change in IgG sialylation is the result of regulation by ST6GAL1 from B cells. We first confirmed that there was no difference in sialylation in splenic B cells from Fut8−/− and Fut8+/+ mice by SNA staining and FACS analysis (Figure 7A). Furthermore, there was little difference in ST6GAL1 expression (Figure 7B) and activity (Figure 7C) in splenic B cell lysates from Fut8+/+ and Fut8−/− mice, as determined by western blotting and HPLC analysis, respectively. These results indicate that the increase in sialylated IgG level is unrelated to ST6GAL1 secreted by B cells, this is in agreement with the previous observation that IgG sialylation occurs independently of the secretory pathway in B cells [42].

ST6GAL1 activity was not altered in B cells of Fut8−/− mice.

Figure 7.
ST6GAL1 activity was not altered in B cells of Fut8−/− mice.

(A) Flow cytometry analysis of cell-surface core fucosylation and α2,6-sialylation levels in CD45R+ B cells isolated from the spleen of Fut8+/+ and Fut8−/− mice. Gray shading represents the control. (B) Western blot analysis of ST6GAL1 expression in splenic B cell lysates from Fut8+/+ and Fut8−/− mice. GAPDH served as a loading control. (C) Reverse-phase HPLC analysis of ST6GAL1 activity in B cell lysates of Fut8+/+ and Fut8−/− mice. Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P2 (PA-G2NA2) representing the product of the ST6GAL1 reaction was eluted at 12.8 min.

Figure 7.
ST6GAL1 activity was not altered in B cells of Fut8−/− mice.

(A) Flow cytometry analysis of cell-surface core fucosylation and α2,6-sialylation levels in CD45R+ B cells isolated from the spleen of Fut8+/+ and Fut8−/− mice. Gray shading represents the control. (B) Western blot analysis of ST6GAL1 expression in splenic B cell lysates from Fut8+/+ and Fut8−/− mice. GAPDH served as a loading control. (C) Reverse-phase HPLC analysis of ST6GAL1 activity in B cell lysates of Fut8+/+ and Fut8−/− mice. Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P2 (PA-G2NA2) representing the product of the ST6GAL1 reaction was eluted at 12.8 min.

Hyper core fucosylation and reduced activation of ST6GAL1 in RA patients

The above results show that the loss of core fucose increases α2,6-sialylation of serum glycoproteins in Fut8−/− mice. It was recently reported that core fucosylation was enhanced in native IgG complexes from RA patients [14], with a corresponding decrease in sialylated IgG [16]. To clarify whether FUT8 modulates ST6GAL1 activity, we compared FUT8 expression in the sera of RA patients and healthy subjects. Because we were unable to obtain synovial fluid from healthy individuals for comparison with RA patients, OA patients were used as a comparison reference. Core fucosylation was increased, whereas α2,6-sialylation was decreased in RA patient's sera relative to healthy individuals, as determined by LCA and SNA lectin blot analysis (Figure 8A). Furthermore, compared with healthy controls, reduced sialylation and increased core fucosylation trends were observed in the sera of RA and OA patients (Figure 8A). Moreover, the expression levels of both core fucosylation and sialylation in synovial fluid were significantly higher in RA patients than in OA subjects (Figure 8B). In total, alterations of sialylation and core fucosylation in the synovial fluid between RA and OA patients were consistent with those in serum. Thus, it is reasonable to conclude that alterations of glycosylation in synovial fluid from RA patients were consistent with those in serum, with down-regulated sialylation and up-regulated core fucosylation glycans. Additionally, the expression of FUT8 and ST6GAL1 in RA patients were also tested. Results showed that FUT8 was increased in the pooled sera of RA patients, while ST6GAL1 level was similar between the two groups (Figure 8C, n = 6 per group). However, the activity of ST6GAL1 was decreased in the pooled sera of RA patients (Figure 8D). Thus, the expression of FUT8 is associated with the activity of ST6GAL1.

The expression level of α2,6-sialylation and the activation of ST6GAL1 were decreased in RA patients.

Figure 8.
The expression level of α2,6-sialylation and the activation of ST6GAL1 were decreased in RA patients.

(A) Lectin blotting analyses of glycoproteins in individual sera of healthy subjects, osteoarthritis (OA), and rheumatoid arthritis (RA) patients; 6 μg of proteins from the serum of healthy subjects (lane 1–3), OA patients (lane 4–6), and RA patients (lane 7–9) was subjected to 10% SDS–PAGE and stained with LCA and SNA lectins (left). (B) Detection of core fucosylation (LCA) and α2,6-sialylation (SNA) expression in synovial fluid from OA and RA patients; 10 μg of synovial fluid proteins from OA (lane 1–3) and RA patients (lane 4–6) was subjected to 10% SDS–PAGE and stained with LCA and SNA lectins (left). Coomassie Brilliant Blue (CBB) staining was performed as a control. (C) Western blot analysis of FUT8 (upper) and ST6GAL1 (lower) expression in pooled sera from healthy subjects and RA patients (n = 6 per group, right). (D) Detection of ST6GAL1 activities by reverse-phase HPLC analysis of reaction mixtures catalyzed by pooled sera as enzyme source from healthy subjects and RA patients (n = 6 per group). Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P3 (PA-G2NA1) representing the product of the ST6GAL1 reaction was eluted at 14.5 min. The quantitative data are presented as means ± S.D. from three independent experiments (*p < 0.05, **, p < 0.01), healthy subjects and RA patients were indicated by black and gray columns, respectively.

Figure 8.
The expression level of α2,6-sialylation and the activation of ST6GAL1 were decreased in RA patients.

(A) Lectin blotting analyses of glycoproteins in individual sera of healthy subjects, osteoarthritis (OA), and rheumatoid arthritis (RA) patients; 6 μg of proteins from the serum of healthy subjects (lane 1–3), OA patients (lane 4–6), and RA patients (lane 7–9) was subjected to 10% SDS–PAGE and stained with LCA and SNA lectins (left). (B) Detection of core fucosylation (LCA) and α2,6-sialylation (SNA) expression in synovial fluid from OA and RA patients; 10 μg of synovial fluid proteins from OA (lane 1–3) and RA patients (lane 4–6) was subjected to 10% SDS–PAGE and stained with LCA and SNA lectins (left). Coomassie Brilliant Blue (CBB) staining was performed as a control. (C) Western blot analysis of FUT8 (upper) and ST6GAL1 (lower) expression in pooled sera from healthy subjects and RA patients (n = 6 per group, right). (D) Detection of ST6GAL1 activities by reverse-phase HPLC analysis of reaction mixtures catalyzed by pooled sera as enzyme source from healthy subjects and RA patients (n = 6 per group). Evaluation of ST6GAL1 activity with CMP-NeuAc and S2 (PA-G2) as donor and acceptor, respectively. P3 (PA-G2NA1) representing the product of the ST6GAL1 reaction was eluted at 14.5 min. The quantitative data are presented as means ± S.D. from three independent experiments (*p < 0.05, **, p < 0.01), healthy subjects and RA patients were indicated by black and gray columns, respectively.

Discussion

Our glycomic analysis using glycans released from the whole serum of Fut8+/+ and Fut8−/− mice revealed a clear increase in the levels of α2,6-sialylated N-glycans in Fut8-deficient mice. It was reported that ST6GAL1 enzyme in the mouse serum is mainly produced in the liver [43]. Although IgG is the most abundant glycoprotein in serum, less than 10% of IgG antibodies are sialylated, so sialylated N-glycans in serum are mainly provided by other glycoproteins, which are mainly synthesized by the liver. Therefore, we investigated the mRNA and protein levels of ST6GAL1 in the liver and found no difference between wild-type and mutant mice. These data prompted us to investigate how the altered expression of one glycosyltransferase affects the catalytic products of others. We found that α2,6-sialylated N-glycan content was increased in the serum of Fut8−/− mice, which may have been due to ST6GAL1's higher catalytic activity in their serum and liver, and its preference to non-core fucosylated substrates.

Glycan biosynthesis involves multiple GTases in a multistep process that conjugates carbohydrates to proteins and lipids. The specificity of enzymes for their donor and acceptor substrates determines the structures of sugar chains produced by a cell [44]. The substrate-specificity of FUT8 is tightly regulated because the addition of Gal or bisecting GlcNAc to N-glycans destroys their ability to serve as a substrate for FUT8 [39]. Many studies have described the interaction between two GTases. For example, overexpression of N-acetylglucosaminyltransferase (GnT)-III inhibited α2,3-sialylation but not α2,6-sialylation [45]. Meanwhile, Fut9−/− mice showed increased sialylation, mainly with N-glycolylneuraminic acid [46], and GnT-IVa knockdown decreased the level of core fucosylated triantennary glycan [47]. However, there is limited information on the mechanistic basis for these interactions. It was reported that the absence of core fucose enhances the bisecting GlcNAc level via increased expression of GnT-III mediated by Wnt/β-catenin signaling [48]. It is possible that FUT8 directly regulates the expression/function of other GTases. Indeed, the results of our MS and lectin blotting analyses revealed significant changes in the fucosylation of serum proteins in Fut8−/− mice with concomitant down-regulation of FUT1, FUT7, FUT10, and FUT11, which are putative targets of FUT8 [17]. These enzymes have no activity in the serum and liver of mice, but they are regulated at the transcriptional level by Fut8, and the molecular mechanism has not been clarified. In addition, in this report, we found that these mice had little outer-arm fucosylation as diagrammed in Table 3 and Figure 3. Previously, Brinkman-Van der Linden et al. [49] reported that the product of FUT6, is responsible for the α1-3 fucosylation of glycoproteins produced in the liver, which is a major source of α1-3 fucosyltransferase activity in human plasma. In the case of mice, Fut6 is a pseudo-gene, and therefore there were few numbers of α1-3 fucosylated oligosaccharides on serum glycoproteins. Havenaar et al. [50] demonstrated that in mouse liver only α1-6 fucosyltransferase activity was present; α1-2, α1-3, α1-4 fucosyltransferase activities were below detectable levels. Therefore, the Fut8-deficient mouse is a powerful tool to show that direct evidence for most of the fucosylation of hepatic glycoproteins is α1-6 linkages. The results of AAL blotting obtained in this study using serum from Fut8+/+ and Fut8−/− mice are in accordance with our previous lectin blot analysis for AAL using lysates of Fut8−/− mouse embryonic fibroblasts and whole brain tissue from Fut8−/− mice [48,51].

In the present study, we demonstrated that loss of FUT8 enhanced α2,6-sialylation of glycoproteins in the serum of mouse. ST6GAL1 has two sites for N-glycosylation [9], and core fucosylated N-glycans [8]. The two N-glycosylation sites (N146 and N158) are within its catalytic domain (63–403 amino acids) [9,44]. We also demonstrated the core fucosylation of ST6GAL1 in Fut8+/+ liver and the loss of ST6GAL1 core fucose unit in Fut8−/− mouse liver by lectin staining and MS (Figure 5A,B). Based on our findings, the loss of Fut8 enhances the catalytic activity of ST6GAL1 in the serum and liver, and ST6GAL1 has a preference to non-core fucosylated substrates. We also showed that FUT8 regulates ST3GAL3, ST3GAL4, and ST3GAL5 mRNA expression (Figure 4A) and that ST3GAL3 and ST3GAL4 generate α2,3-linked sialic acids on N-glycans, suggesting that the change in α2,3-sialylation is controlled at the level of transcription. However, we cannot exclude the possibility that α2,3-sialylation is decreased as a result of substrate competition, given that α2,3- and α2,6-sialyltransferase share a common pool of sugar chains [45]. Although FUT8 can regulate GTases, it is itself not regulated by GTases because it is not a glycoprotein. Our finding that α2,6-sialylation products are increased in Fut8−/− mice, suggests a negative regulatory interaction between α2,6-sialylation expression and core fucose structures; it is thus reasonable to suppose that altered expression of one GTase can affect the enzymatic activity of others.

IgG is the most abundant glycoprotein in serum, the core carbohydrate structure of which can include core fucose, bisecting GlcNAc, galactose, and α2,6-sialic acid moieties [52]. We found that α2,6-sialylation of serum IgG was increased in Fut8−/− mice, although there was no difference in ST6GAL1 enzymatic activity in splenic B cells of Fut8+/+ and Fut8−/− mice. These findings suggest that the increase in the sialylated IgG level is unrelated to ST6GAL1 secreted by B cells [42]. ST6GAL1 in mouse serum is mainly produced in the liver [43]. The platelets are the major transporter of blood-borne GTases, which are dynamically controlled by platelet activation, to remodel cell-surface glycans and alter cell behavior [53]. Moreover, activated circulating platelets can supply activated sialic acid-donor substrate to functionally drive extracellular ST6Gal-1 for α2,6-sialylation in vivo and in vitro, respectively [54,55]. Hence, we speculated α2,6-sialylation of serum IgG can be regulated by the liver and platelets through enzymes and sugar donors released into circulation, and it is consistent with the findings of previous studies [42]. Indeed, we demonstrated that the loss of Fut8 enhances the catalytic activity of ST6GAL1 in the serum and liver (Figure 5C). Thus, in this study, α2,6-sialylation of serum IgG was increased in Fut8−/− mice can be an extracellular event, and mediated by platelets.

Glycosylation changes in IgG of patients with RA are a very common clinical phenomenon. Decreased sialylation and elevated core fucosylation of Fc fragments of IgG molecules have been reported for RA in many studies [13,14,16]. Clarifying the interaction between core fucosylation and sialylation in RA patients is important from the standpoint of basic and clinical research. In this present study, we showed that core fucosylation was enhanced in sera of RA patients relative to healthy subjects, accompanied by reduced α2,6-sialylation. Moreover, due to our inability to obtain synovial fluid from healthy controls, we used synovial fluid from OA patients as a reference to explore the glycosylation alterations in the synovial fluid of RA patients. Alterations of core fucosylation and α2,6-sialylation in synovial fluid protein between RA and OA were consistent with those in sera, indicating that proteins from OA patients could act as a reference for RA patients. The above results clearly showed that core fucosylation was significantly up-regulated and sialylation was reduced in sera and synovial fluid of RA patients. Meanwhile, an enhanced expression of FUT8 and decreased activity of ST6GAL1 in serum from RA patients were observed. On the other hand, α2,6-sialylation was increased in Fut8−/− mouse sera. One explanation for these observations is that FUT8 mostly regulates the activity of ST6GAL1 with or without core fucosylation, although additional studies are required to evaluate this possibility.

In summary, we investigated the possible mechanism by which FUT8 modulates α2,6-sialylation of glycoproteins in sera of Fut8−/− mice. FUT8 deficiency not only effects the core fucosylation of N-glycans, but also the α2,6-sialylation of glycoproteins. This may be explained by the facts that the loss of Fut8 enhances the catalytic activity of ST6GAL1 in the serum and liver, and ST6GAL1 has a preference to non-core fucosylated substrates. There are over 200 glycosyltransferase genes in mammals; thus, there must be a complex regulatory network among GTases, which are responsible for the composition and structure of N-glycans. The results of the present study provide a new insight for the diagnosis and treatment of RA.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

This work was supported by National Natural Science Foundation of China (grant no. 31770859, 31570797, and 31670807), the Natural Science Foundation of Liaoning Province (grant no. 2015020273), and the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry (SRF for ROCS, SEM). This work was also supported by the Liaoning Provincial Program for Top Discipline of Basic Medical Sciences.

Author Contributions

W.D. conceived and designed the experiments; G.H. and Y.L. performed the experiments; G.H., Z.L., Y.L., G.L., S.S., J.G., A.K., and W.L. analyzed the data; W.D. and G.H. wrote the manuscript.

Acknowledgements

The authors thank Dr. Tonghui Ma (Dalian Medical University), for his helpful suggestions and review of the manuscript.

Abbreviations

     
  • AAL

    Aleuria aurantia lectin

  •  
  • BSA

    bovine serum albumin

  •  
  • CD45R

    cluster of differentiation 45 receptor

  •  
  • CMP

    cytidine monophosphate

  •  
  • FACS

    fluorescence-activated cell sorting

  •  
  • FUT8

    fucosyltransferase 8

  •  
  • GDP

    guanosine diphosphate

  •  
  • GlcNAc

    N-acetylglucosamine

  •  
  • GTase

    glycosyltransferase

  •  
  • HPLC

    high-performance liquid chromatography

  •  
  • HRP

    horseradish peroxidase

  •  
  • IgG

    immunoglobulin G

  •  
  • LCA

    lens culinaris agglutinin

  •  
  • MACS

    magnetic-activated cell sorting

  •  
  • MAL II

    Maackia amurensis lectin II

  •  
  • MALDI-TOF

    matrix-assisted laser desorption ionization-time of flight

  •  
  • MS

    mass spectrometry

  •  
  • MS/MS

    tandem mass spectrometry

  •  
  • PA

    2-aminopyridine

  •  
  • PBS

    phosphate-buffered saline

  •  
  • PHA-E

    Phaseolus vulgaris erythroagglutinin

  •  
  • PHA-L

    Phaseolus vulgaris leucoagglutinin

  •  
  • PNGase F

    peptide-N-glycosidase F

  •  
  • PVDF

    polvinylidene difluoride

  •  
  • RA

    rheumatoid arthritis

  •  
  • SDS–PAGE

    sodium dodecyl sulfate-polyacrylamide gel electrophoresis

  •  
  • SNA

    Sambucus nigra lectin

  •  
  • ST6GAL1

    β-galactoside α2,6-sialyltransferase 1

  •  
  • TBS-T

    tris-buffered saline with 0.05% Tween-20

References

References
1
Wilson
,
J.R.
,
Williams
,
D.
and
Schachter
,
H.
(
1976
)
The control of glycoprotein synthesis: N-acetylglucosamine linkage to a mannose residue as a signal for the attachment of L-fucose to the asparagine-linked N-acetylglucosamine residue of glycopeptide from α1-acid glycoprotein
.
Biochem. Biophys. Res. Commun.
72
,
909
916
2
Ferrara
,
C.
,
Grau
,
S.
,
Jager
,
C.
,
Sondermann
,
P.
,
Brunker
,
P.
,
Waldhauer
,
I.
, et al (
2011
)
Unique carbohydrate-carbohydrate interactions are required for high affinity binding between FcgammaRIII and antibodies lacking core fucose
.
Proc. Natl. Acad. Sci. U.S.A.
108
,
12669
12674
3
Flores
,
A.
and
Marrero
,
J.A.
(
2014
)
Emerging trends in hepatocellular carcinoma: focus on diagnosis and therapeutics
.
Clin. Med. Insights Oncol.
8
,
71
76
4
Okuyama
,
N.
,
Ide
,
Y.
,
Nakano
,
M.
,
Nakagawa
,
T.
,
Yamanaka
,
K.
,
Moriwaki
,
K.
, et al (
2006
)
Fucosylated haptoglobin is a novel marker for pancreatic cancer: a detailed analysis of the oligosaccharide structure and a possible mechanism for fucosylation
.
Int. J. Cancer
118
,
2803
2808
5
Wang
,
X.
,
Inoue
,
S.
,
Gu
,
J.
,
Miyoshi
,
E.
,
Noda
,
K.
,
Li
,
W.
, et al (
2005
)
Dysregulation of TGF-β1 receptor activation leads to abnormal lung development and emphysema-like phenotype in core fucose-deficient mice
.
Proc. Natl. Acad. Sci. U.S.A.
102
,
15791
15796
6
Wang
,
X.
,
Gu
,
J.
,
Ihara
,
H.
,
Miyoshi
,
E.
,
Honke
,
K.
and
Taniguchi
,
N.
(
2006
)
Core fucosylation regulates epidermal growth factor receptor-mediated intracellular signaling
.
J. Biol. Chem.
281
,
2572
2577
7
Harduin-Lepers
,
A.
,
Krzewinski-Recchi
,
M.A.
,
Colomb
,
F.
,
Foulquier
,
F.
,
Groux-Degroote
,
S.
and
Delannoy
,
P.
(
2012
)
Sialyltransferases functions in cancers
.
Front. Biosci. (Elite Ed.)
4
,
499
515
8
Macnaughtan
,
M.A.
,
Tian
,
F.
,
Liu
,
S.
,
Meng
,
L.
,
Park
,
S.
,
Azadi
,
P.
, et al (
2008
)
13C-sialic acid labeling of glycans on glycoproteins using ST6Gal-I
.
J. Am. Chem. Soc.
130
,
11864
11865
9
Weinstein
,
J.
,
Lee
,
E.U.
,
McEntee
,
K.
,
Lai
,
P.H.
and
Paulson
,
J.C.
(
1987
)
Primary structure of β-galactoside α2,6-sialyltransferase. conversion of membrane-bound enzyme to soluble forms by cleavage of the NH2-terminal signal anchor
.
J. Biol. Chem.
262
,
17735
17743
PMID:
[PubMed]
10
Kitazume
,
S.
,
Tachida
,
Y.
,
Oka
,
R.
,
Shirotani
,
K.
,
Saido
,
T.C.
and
Hashimoto
,
Y.
(
2001
)
Alzheimer's β-secretase, β-site amyloid precursor protein-cleaving enzyme, is responsible for cleavage secretion of a Golgi-resident sialyltransferase
.
Proc. Natl. Acad. Sci. U.S.A.
98
,
13554
13559
11
Fast
,
D.G.
,
Jamieson
,
J.C.
and
McCaffrey
,
G.
(
1993
)
The role of the carbohydrate chains of galβ-1,4-GlcNAcα 2,6-sialyltransferase for enzyme activity
.
Biochim. Biophys. Acta
1202
,
325
330
12
Choy
,
E.
(
2012
)
Understanding the dynamics: pathways involved in the pathogenesis of rheumatoid arthritis
.
Rheumatology (Oxford)
51
,
v3-11
13
Mahajan
,
V.S.
and
Pillai
,
S.
(
2016
)
Sialic acids and autoimmune disease
.
Immunol. Rev.
269
,
145
161
14
Stumer
,
J.
,
Biermann
,
M.H.C.
,
Knopf
,
J.
,
Magorivska
,
I.
,
Kastbom
,
A.
,
Svard
,
A.
, et al (
2017
)
Altered glycan accessibility on native immunoglobulin G complexes in early rheumatoid arthritis and its changes during therapy
.
Clin. Exp. Immunol.
189
,
372
382
15
Rook
,
G.A.
,
Steele
,
J.
,
Brealey
,
R.
,
Whyte
,
A.
,
Isenberg
,
D.
,
Sumar
,
N.
, et al (
1991
)
Changes in IgG glycoform levels are associated with remission of arthritis during pregnancy
.
J. Autoimmun.
4
,
779
794
16
van de Geijn
,
F.E.
,
Wuhrer
,
M.
,
Selman
,
M.H.
,
Willemsen
,
S.P.
,
de Man
,
Y.A.
,
Deelder
,
A.M.
, et al (
2009
)
Immunoglobulin G galactosylation and sialylation are associated with pregnancy-induced improvement of rheumatoid arthritis and the postpartum flare: results from a large prospective cohort study
.
Arthritis Res. Ther.
11
,
R193
17
Ihara
,
H.
,
Tsukamoto
,
H.
,
Gu
,
J.
,
Miyoshi
,
E.
,
Taniguchi
,
N.
and
Ikeda
,
Y.
(
2014
)
Fucosyltransferase 8. GDP-Fucose N-Glycan Core α6-Fucosyltransferase (FUT8)
. In:
Handbook of Glycosyltransferases and Related Genes
(
Taniguchi
,
N.
,
Honke
,
K.
,
Fukuda
,
M.
,
Narimatsu
,
H.
,
Yamaguchi
,
Y.
and
Angata
,
T.
, eds.), pp.
581
596
,
Springer Japan
,
Tokyo
.
18
Abbott
,
K.L.
,
Nairn
,
A.V.
,
Hall
,
E.M.
,
Horton
,
M.B.
,
McDonald
,
J.F.
,
Moremen
,
K.W.
, et al (
2008
)
Focused glycomic analysis of the N-linked glycan biosynthetic pathway in ovarian cancer
.
Proteomics
8
,
3210
3220
19
Chen
,
C.Y.
,
Jan
,
Y.H.
,
Juan
,
Y.H.
,
Yang
,
C.J.
,
Huang
,
M.S.
,
Yu
,
C.J.
, et al (
2013
)
Fucosyltransferase 8 as a functional regulator of nonsmall cell lung cancer
.
Proc. Natl. Acad. Sci. U.S.A.
110
,
630
635
20
Muinelo-Romay
,
L.
,
Vazquez-Martin
,
C.
,
Villar-Portela
,
S.
,
Cuevas
,
E.
,
Gil-Martin
,
E.
and
Fernandez-Briera
,
A.
(
2008
)
Expression and enzyme activity of α(1,6)fucosyltransferase in human colorectal cancer
.
Int. J. Cancer
123
,
641
646
21
Gretschel
,
S.
,
Haensch
,
W.
,
Schlag
,
P.M.
and
Kemmner
,
W.
(
2003
)
Clinical relevance of sialyltransferases ST6GAL-I and ST3GAL-III in gastric cancer
.
Oncology
65
,
139
145
22
Lin
,
S.
,
Kemmner
,
W.
,
Grigull
,
S.
and
Schlag
,
P.M.
(
2002
)
Cell surface α2,6-sialylation affects adhesion of breast carcinoma cells
.
Exp. Cell Res.
276
,
101
110
23
Swindall
,
A.F.
,
Londono-Joshi
,
A.I.
,
Schultz
,
M.J.
,
Fineberg
,
N.
,
Buchsbaum
,
D.J.
and
Bellis
,
S.L.
(
2013
)
ST6Gal-I protein expression is upregulated in human epithelial tumors and correlates with stem cell markers in normal tissues and colon cancer cell lines
.
Cancer Res.
73
,
2368
2378
24
Wang
,
P.
,
Lee
,
W.
,
Lee
,
Y.
,
Juang
,
C.
,
Chen
,
Y.
,
Chao
,
H.
, et al (
2003
)
Enhanced expression of α 2,6-sialyltransferase ST6Gal I in cervical squamous cell carcinoma
.
Gynecol. Oncol.
89
,
395
401
25
Schultz
,
M.J.
,
Swindall
,
A.F.
and
Bellis
,
S.L.
(
2012
)
Regulation of the metastatic cell phenotype by sialylated glycans
.
Cancer Metastasis Rev.
31
,
501
518
26
Aletaha
,
D.
,
Neogi
,
T.
,
Silman
,
A.J.
,
Funovits
,
J.
,
Felson
,
D.T.
,
Bingham
,
C.O.
,
Birnbaum
,
N.S.
, et al (
2010
)
Rheumatoid arthritis classification criteria: an American college of rheumatology/European league against rheumatism collaborative initiative
.
Arthritis Rheum.
62
,
2569
2581
27
Dong
,
W.
,
Matsuno
,
Y.
and
Kameyama
,
A.
(
2013
)
Serum protein fractionation using supported molecular matrix electrophoresis
.
Electrophoresis
34
,
2432
2439
28
De Haan
,
N.
,
Reiding
,
K.R.
,
Kristic
,
J.
,
Ederveen
,
A.L.H.
,
Lauc
,
G.
and
Wuhrer
,
M.
(
2017
)
The N-glycosylation of mouse immunoglobulin G (IgG)-fragment crystallizable differs between IgG subclasses and strains
.
Front. Immunol.
8
,
608
29
Xu
,
J.
,
Sun
,
H.
,
Huang
,
G.
,
Liu
,
G.
,
Li
,
Z.
,
Yang
,
H.
, et al (
2019
)
A fixation method for the optimisation of western blotting
.
Sci. Rep.
9
,
6649
30
Peirson
,
S.N.
,
Butler
,
J.N.
and
Foster
,
R.G.
(
2003
)
Experimental validation of novel and conventional approaches to quantitative real-time PCR data analysis
.
Nucleic Acids Res.
31
.
31
Ciucanu
,
I.
and
Costello
,
C.E.
(
2003
)
Elimination of oxidative degradation during the per-O-methylation of carbohydrates
.
J. Am. Chem. Soc.
125
,
16213
16219
32
Hennet
,
T.
,
Chui
,
D.
,
Paulson
,
J.C.
and
Marth
,
J.D.
(
1998
)
Immune regulation by the ST6Gal sialyltransferase
.
Proc. Natl. Acad. Sci. U.S.A.
95
,
4504
4509
33
Pousset
,
D.
,
Piller
,
V.
,
Bureaud
,
N.
,
Monsigny
,
M.
and
Piller
,
F.
(
1997
)
Increased α2,6 sialylation of N-glycans in a transgenic mouse model of hepatocellular carcinoma
.
Cancer Res.
57
,
4249
4256
PMID:
[PubMed]
34
Uozumi
,
N.
,
Teshima
,
T.
,
Yamamoto
,
T.
,
Nishikawa
,
A.
,
Gao
,
Y.E.
,
Miyoshi
,
E.
, et al (
1996
)
A fluorescent assay method for GDP-L-Fuc:N-acetyl-β-D-glucosaminide α1-6fucosyltransferase activity, involving high performance liquid chromatography
.
J. Biochem.
120
,
385
392
35
Koyota
,
S.
,
Ikeda
,
Y.
,
Miyagawa
,
S.
,
Ihara
,
H.
,
Koma
,
M.
,
Honke
,
K.
, et al (
2001
)
Down-regulation of the α-Gal epitope expression in N-glycans of swine endothelial cells by transfection with the N-acetylglucosaminyltransferase III gene modulation of the biosynthesis of terminal structures by a bisecting GlcNAc
.
J. Biol. Chem.
276
,
32867
32874
36
Wang
,
F.
,
Shen
,
X.
,
Li
,
S.
,
Chen
,
L.
,
Wang
,
Y.
,
Qin
,
J.
, et al (
2015
)
Splenocytes derived from young WT mice prevent AD progression in APPswe/PSENldE9 transgenic mice
.
Oncotarget
6
,
20851
20862
37
Furukawa
,
J.
,
Shinohara
,
Y.
,
Kuramoto
,
H.
,
Miura
,
Y.
,
Shimaoka
,
H.
,
Kurogochi
,
M.
, et al (
2008
)
Comprehensive approach to structural and functional glycomics based on chemoselective glycoblotting and sequential tag conversion
.
Anal. Chem.
80
,
1094
1101
38
Korekane
,
H.
,
Matsumoto
,
A.
,
Ota
,
F.
,
Hasegawa
,
T.
,
Misonou
,
Y.
,
Shida
,
K.
, et al (
2010
)
Involvement of ST6Gal I in the biosynthesis of a unique human colon cancer biomarker candidate, α2,6-sialylated blood group type 2H (ST2H) antigen
.
J. Biochem
148
,
359
370
39
Calderon
,
A.D.
,
Liu
,
Y.
,
Li
,
X.
,
Wang
,
X.
,
Chen
,
X.
,
Li
,
L.
, et al (
2016
)
Substrate specificity of FUT8 and chemoenzymatic synthesis of core-fucosylated asymmetric N-glycans
.
Org. Biomol. Chem.
14
,
4027
4031
40
Stadlmann
,
J.
,
Weber
,
A.L.
,
Pabst
,
M.
,
Anderle
,
H.
,
Kunert
,
R.
,
Ehrlich
,
H.J.
, et al (
2009
)
A close look at human IgG sialylation and subclass distribution after lectin fractionation
.
Proteomics
9
,
4143
4153
41
Dalziel
,
M.
,
Huang
,
R.Y.
,
Dall'Olio
,
F.
,
Morris
,
J.R.
,
TaylorPapadimitriou
,
J.
and
Lau
,
J.T.Y.
(
2001
)
Mouse ST6Gal sialyltransferase gene expression during mammary gland lactation
.
Glycobiology
11
,
407
412
42
Jones
,
M.B.
,
Oswald
,
D.M.
,
Joshi
,
S.
,
Whiteheart
,
S.W.
,
Orlando
,
R.
and
Cobb
,
B.A.
(
2016
)
B-cell-independent sialylation of IgG
.
Proc. Natl. Acad. Sci. U.S.A.
113
,
7207
7212
43
Appenheimer
,
M.M.
,
Huang
,
R.Y.
,
Chandrasekaran
,
E.V.
,
Dalziel
,
M.
,
Hu
,
Y.P.
,
Soloway
,
P.D.
, et al (
2003
)
Biologic contribution of P1 promoter-mediated expression of ST6Gal I sialyltransferase
.
Glycobiology
13
,
591
600
44
Paulson
,
J.C.
and
Colley
,
K.J.
(
1989
)
Glycosyltransferases. structure, localization, and control of cell type-specific glycosylation
.
J. Biol. Chem.
264
,
17615
17618
PMID:
[PubMed]
45
Lu
,
J.
,
Isaji
,
T.
,
Im
,
S.
,
Fukuda
,
T.
,
Kameyama
,
A.
and
Gu
,
J.
(
2016
)
Expression of N-Acetylglucosaminyltransferase III suppresses α2,3-sialylation, and its distinctive functions in cell migration are attributed to α2,6-sialylation levels
.
J. Biol. Chem.
291
,
5708
5720
46
Noro
,
E.
,
Togayachi
,
A.
,
Sato
,
T.
,
Tomioka
,
A.
,
Fujita
,
M.
,
Sukegawa
,
M.
, et al (
2015
)
Large-scale identification of N-glycan glycoproteins carrying lewis x and site-specific N-glycan alterations in Fut9 knockout mice
.
J. Proteome Res.
14
,
3823
3834
47
Nie
,
H.
,
Liu
,
X.
,
Zhang
,
Y.
,
Li
,
T.
,
Zhan
,
C.
,
Huo
,
W.
, et al (
2015
)
Specific N-glycans of hepatocellular carcinoma cell surface and the abnormal increase of core-α-1, 6-fucosylated triantennary glycan via N-acetylglucosaminyltransferases-IVa regulation
.
Sci. Rep.
5
,
16007
48
Kurimoto
,
A.
,
Kitazume
,
S.
,
Kizuka
,
Y.
,
Nakajima
,
K.
,
Oka
,
R.
,
Fujinawa
,
R.
, et al (
2014
)
The absence of core fucose up-regulates GnT-III and Wnt target genes: a possible mechanism for an adaptive response in terms of glycan function
.
J. Biol. Chem.
289
,
11704
11714
49
Brinkman-Van der Linden
,
E.C.
,
Mollicone
,
R.
,
Oriol
,
R.
,
Larson
,
G.
,
Van den Eijnden
,
D.H.
and
Van Dijk
,
W.
(
1996
)
A missense mutation in the FUT6 gene results in total absence of α3-fucosylation of human α1-acid glycoprotein
.
J. Biol. Chem.
271
,
14492
14495
50
Havenaar
,
E.C.
,
Hoff
,
R.C.
,
van den Eijnden
,
D.H.
and
van Dijk
,
W.
(
1998
)
Sialyl lewis(x) epitopes do not occur on acute phase proteins in mice: relationship to the absence of α3-fucosyltransferase in the liver
.
Glycoconj. J.
15
,
389
395
51
Fukuda
,
T.
,
Hashimoto
,
H.
,
Okayasu
,
N.
,
Kameyama
,
A.
,
Onogi
,
H.
,
Nakagawasai
,
O.
, et al (
2011
)
α1,6-fucosyltransferase-deficient mice exhibit multiple behavioral abnormalities associated with a schizophrenia-like phenotype: importance of the balance between the dopamine and serotonin systems
.
J. Biol. Chem.
286
,
18434
18443
52
Kaneko
,
Y.
,
Nimmerjahn
,
F.
and
Ravetch
,
J.V.
(
2006
)
Anti-inflammatory activity of immunoglobulin G resulting from Fc sialylation
.
Science
313
,
670
673
53
Lee-Sundlov
,
M.M.
,
Ashline
,
D.J.
,
Hanneman
,
A.J.
,
Grozovsky
,
R.
,
Reinhold
,
V.N.
,
Hoffmeister
,
K.M.
, et al (
2017
)
Circulating blood and platelets supply glycosyltransferases that enable extrinsic extracellular glycosylation
.
Glycobiology
27
,
188
198
54
Manhardt
,
C.T.
,
Punch
,
P.R.
,
Dougher
,
C.W.L.
and
Lau
,
J.T.Y.
(
2017
)
Extrinsic sialylation is dynamically regulated by systemic triggers in vivo
.
J. Biol. Chem.
292
,
13514
13520
55
Lee
,
M.M.
,
Nasirikenari
,
M.
,
Manhardt
,
C.T.
,
Ashline
,
D.J.
,
Hanneman
,
A.J.
,
Reinhold
,
V.N.
et al (
2014
)
Platelets support extracellular sialylation by supplying the sugar donor substrate
.
J. Biol. Chem.
289
,
8742
8748

Author notes

*

These authors contributed equally to the work.