Abstract
JC virus (JCV) belongs to the polyomavirus family of double-stranded DNA viruses and in humans causes a demyelinating disease of the central nervous system, progressive multifocal leukoencephalopathy. Its hemagglutination activity and entry into host cells have been reported to depend on an N-linked glycoprotein containing sialic acid. In order to identify the receptors of JCV, we generated virus-like particles (VLP) consisting of major viral capsid protein VP1. We then developed an indirect VLP overlay assay to detect VLP binding to glycoproteins and a panel of glycolipids. We found that VLP bound to sialoglycoproteins, including α1-acid glycoprotein, fetuin, and transferrin receptor, and that this binding depended on α2-3-linked sialic acids and N-linked sugar chains. Neoglycoproteins were synthesized by using ovalbumin and conjugation with oligosaccharides containing the terminal α2-3- or α2-6-linked sialic acid or the branched α2-6-linked sialic acid. We show that the neoglycoprotein containing the terminal α2-6-linked sialic acid had the highest affinity for VLP, inhibited the hemagglutination activity of VLP and JCV, and inhibited the attachment of VLP to cells. We also demonstrate that VLP bound to specific glycolipids, such as lactosylceramide, and gangliosides, including GM3, GD2, GD3, GD1b, GT1b, and GQ1b, and that VLP bound weakly to GD1a but did not bind to GM1a, GM2, or galactocerebroside. Furthermore, the neoglycoprotein containing the terminal α2-6-linked sialic acid and the ganglioside GT1b inhibited JCV infection in the susceptible cell line IMR-32. These results suggest that the oligosaccharides of glycoproteins and glycolipids work as JCV receptors and may be feasible as anti-JCV agents.
The human neurotopic polyomavirus JC virus (JCV) is the causative agent of a fatal demyelinating disease known as progressive multifocal leukoencephalopathy (PML). To date, the mortality rate from PML is high, and there is no appropriate therapy for treating PML. The viral route of infection and the mechanisms of viral spread have not been conclusively identified; however, as JCV is frequently detected in untreated urban sewage and in sewage-exposed shellfish, a fecal-oral route of transmission has been suggested (4). Recently, JCV DNA was detected in several organs, including the tonsils and upper and lower gastrointestinal tracts (13, 22). Given that JCV ingested by the oral route may enter the intestinal wall and Peyer's patches and thus peripheral blood lymphocytes and various organs (4), the cell surface receptor for JCV is one of the potential therapeutic targets for controlling JCV infection.
The JCV receptor has been described as a glycoprotein containing terminal α2-6-linked sialic acid, on the basis of the finding that sialidase but not α2-3-specific sialidase inhibited infection of glial cells by JCV. In addition, the N glycosylation inhibitor tunicamycin inhibited JCV infection (12). Certain viruses utilize the sialooligosaccharides of glycoproteins and glycolipids for their attachment to host cells and strictly recognize cell surface sialic acid linkages and sugar sequences as their receptors. For example, human influenza virus A/PR/8/34 bound most effectively to lacto-series and neolacto-series gangliosides carrying sugar chains containing the terminal Neu5Ac α2-3Gal sequence, followed by ganglio-series and hematoside-series gangliosides. Human influenza virus B/Lee/40 bound lacto-series and neolacto-series gangliosides containing the Neu5Ac α2-6Gal sequence (29). Human parainfluenza types 1 and 3 preferentially bound gangliosides containing branched N-acetyllactosaminoglycans in their core structure with terminal Neu5Ac α2-3Gal (27). The structure of sialooligosaccharides recognized by JCV, however, has not been identified, partly because JCV is a slowly growing virus and it is difficult to obtain adequate amounts of the virus for the identification of its receptor.
JCV consists of three capsid proteins, VP1, VP2, and VP3. VP1, encoded by the late region of JCV, is the major capsid protein forming the outer surface of the virion. It has been reported that VP1 plays a major function in the attachment of JCV to cells because anti-VP1 antibody suppresses viral entry and infection of cells (9, 26) and recombinant VP1 derived from Escherichia coli or insect cells can form virus-like particles (VLP) (9, 19). In addition, JC VLP bind to cells via sialic acid in a manner similar to that of native JCV (26). Thus, we hypothesized that receptors of JCV would be VP1-binding molecules, and we developed a new method, an indirect VLP overlay assay, to determine VP1-binding molecules, such as glycoproteins or glycolipids. In this study, we determined the structures of the sialooligosaccharides recognized by VP1. Furthermore, we investigated whether neoglycoproteins and glycolipids containing sialic acids would suppress JCV infection.
MATERIALS AND METHODS
Reagents and cell culture conditions.
GT1b and GD3 were purchased from Wako Chemicals (Tokyo, Japan). Ovalbumin (OVA), disialyllacto-N-tetraose (DSLNT), and biotinylated Sambucus sieboldiana agglutinin (SSA) were obtained from Seikagaku Corp. (Tokyo, Japan). Horseradish peroxidase (HRP)-conjugated streptavidin was obtained from Nichirei (Tokyo, Japan). Sialylneolacto-N-tetraose c (LSTc), GQ1b, and transferrin receptor (human placenta derived) were obtained from Calbiochem (San Diego, Calif.). Sialyllacto-N-tetraose a (LSTa) was obtained from Oxford Glycosciences (Abingdon, United Kingdom). Sialyllacto-N-tetraose b (LSTb) was obtained from Dextra Laboratories (Reading, United Kingdom). An enhanced chemiluminescence (ECL) kit was obtained from Amersham Pharmacia Biotech (Piscataway, N.J.). Gangliosides, including GM1a, GM2, GM3, GD1a, and GD1b, galactocerebroside (GalCer), lactosyl ceramide (LacCer), fetuin, bovine serum albumin (BSA), and α1-acid glycoprotein (human serum derived) were obtained from Sigma (St. Louis, Mo.). Human neuroblastoma cell line IMR-32 (JCRB 9050) was provided by the Health Science Research Resources Bank (Tokyo, Japan) (30), and human glial SVG-A cells were kindly provided by W. J. Atwood (12). They were grown in Dulbecco's modified essential medium (DMEM) supplemented with 10% fetal calf serum (FCS) and antibiotics (penicillin and streptomycin) (Sigma) in 5% CO2. Native JC virus and rabbit anti-VP1 and antiagnoprotein antibodies were prepared as previously described (16, 26).
Preparation of JC VLP.
JC VLP were prepared as previously described (26). Briefly, the VP1 gene of JCV from pBR-Mad1 (17) was subcloned into the prokaryotic expression vector pET15-b (Novagen, Madison, Wis.). After confirmation of the inserted sequence, the plasmid was transformed in competent cells, BL21(DE3)/pLysS (Stratagene, La Jolla, Calif.). The expression of authentic VP1 was induced with 1.0 mM isopropyl-β-d-thiogalactopyranoside (IPTG) for 4 h at 30°C, and the mixture was centrifuged at 4,000 × g for 15 min. The pellet was resuspended in Tris-buffered saline (TBS) (20 mM Tris-HCl [pH 7.5], 150 mM NaCl) containing 1 mg of lysozyme/ml and kept on ice for 30 min, and then sodium deoxycholate was added to a final concentration of 2 mg/ml. After 10 min of incubation on ice, the sample was lysed by five cycles of sonication in 30-s bursts. The lysate was treated with DNase I (100 U/ml) for 30 min at 30°C and centrifuged at 10,000 × g for 10 min. The supernatant was centrifuged at 100,000 × g for 1 h at 4°C, and the pellet was resuspended in phosphate-buffered saline (PBS). The morphology of JC VLP with a diameter of 45 nm was confirmed by electron microscopy and immunoelectron microscopy. For conjugation of VLP with fluorescein isothiocyanate (FITC), 2.5 mg of VLP was dissolved in 0.1 M carbonate-bicarbonate buffer (pH 9.0), mixed with 250 μg of FITC (Sigma), and incubated at room temperature for 2 h. After centrifugation at 100,000 × g for 1 h at 4°C, the pellet was resuspended in TBS and centrifuged at 10,000 × g overnight, and the final pellet was resuspended in TBS.
Indirect VLP overlay assay. (i) Western blot VLP overlay assay.
Fetuin, α1-acid glycoprotein, and transferrin receptor (1 μg of protein) were separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) in a 10% gel. After the separated proteins were blotted onto a polyvinylidene fluoride (PVDF) membrane, the remaining binding site on the membrane was blocked with PBS containing 0.1% Tween 20 and 1% BSA (blocking solution 2) for 30 min. The membrane was incubated at 4°C for 2 h with purified VLP (5 μg/ml) resuspended in blocking solution 2. After four washes with ice-cold PBS containing 0.1% Tween 20 (PBST), the membrane was incubated at 4°C for 1 h with anti-VP1 antibody diluted 1:1,000 in PBST. Following ice-cold PBST washing, the membrane was incubated at 4°C for 1 h with HRP-conjugated F(ab′)2 goat anti-rabbit immunoglobulins (BioSource International, Camarillo, Calif.) diluted 1:5,000 in PBST. After four washes with ice-cold PBST, bound VLP were detected with the ECL kit.
(ii) TLC VLP overlay assay for glycolipids.
Gangliosides and other glycolipids (25, 50, and 100 pmol) were subjected to thin-layer chromatography (TLC) on silica gel plastic plates (Polygram Sil G; Macherey-Nagel, Düren, Germany) by using a solvent system of chloroform-methanol-12 mM aqueous magnesium chloride (5:4:1). The VLP overlay and the immunochemical detection of VLP on the plates were performed by using a modification of a method described previously (29). Briefly, the chromatograms were blocked with PBS containing 1% OVA and 1% polyvinylpyrrolidone (PVP) (blocking solution 1) at room temperature for 1 h. The plates were washed four times with PBS and incubated at 4°C for 2 h with purified VLP (5 μg/ml) resuspended in blocking solution 1. After removal of the VLP suspension by suction, each plate was washed four times with ice-cold PBS to remove unbound VLP and incubated at 4°C for 1 h with anti-VP1 antibody diluted 1:1,000 in PBS containing 1% PVP. After removal of the antibody solution by suction, the plates were washed four times with ice-cold PBS and incubated at 4°C for 1 h with HRP-conjugated F(ab′)2 goat anti-rabbit immunoglobulins diluted 1:5,000 in PBS containing 1% PVP. The plates were washed four times with ice-cold PBS, and VLP bound to the plates were visualized by incubation with an immunostaining reagent containing N,N-diethyl-p-phenylenediamine with monohydrochloride and 4-chloro-1-naphthol (6). Glycolipids (0.5 nmol) were visualized with orcinol reagent.
Synthesis of neoglycoproteins by coupling sialooligosaccharides with OVA.
To investigate whether α2-6-linked sialic acid contributes to the binding of VLP to glycoproteins, OVA was conjugated with various oligosaccharides containing α2-6- or α2-3-linked sialic acid. LSTa, LSTb, LSTc, and DSLNT were dissolved in H2O and added to the reaction mixture (0.25 M potassium phosphate [pH 8.0], 4 mg of OVA, 10 mg of NaBH3CN/ml) at a final concentration of 1 mg/ml each. These reaction mixtures were incubated and evaporated at 37°C overnight. On the next day, the products were dissolved in H2O and dialyzed against H2O twice. Insoluble materials were precipitated by centrifugation at 20,000 × g and 4°C for 10 min, and the supernatants were used as neoglycoproteins. The concentrations of sialooligosaccharides conjugated with OVA were determined by measurement of the free sialic acid concentrations of the neoglycoproteins after treatment with 0.1 M H2SO4 at 80°C for 1 h, in accordance with the method described by Aminoff (1). Neoglycoproteins (200 pmol of sialooligosaccharides for protein staining and 100 pmol for the Western blot VLP overlay assay) were simultaneously resolved on SDS-polyacrylamide gels (10%). One gel was stained with Coomassie brilliant blue dye, and the other was subjected to the Western blot VLP overlay assay. To confirm the structures of the neoglycoproteins, we used a lectin blot with SSA. Each neoglycoprotein containing 100 pmol of sialooligosaccharide was blotted onto a PVDF membrane, treated with PBST, and incubated with 5 μg of biotinylated SSA/ml at room temperature for 1 h. After being washed with PBST, the membrane was incubated with diluted HRP-conjugated streptavidin at room temperature for 1 h, and the signal was detected with the ECL kit and analyzed with a LAS-1000 system (Fuji Film, Tokyo, Japan).
Sialidase and peptide N-glycosidase treatments of glycoproteins and glycolipids.
α1-acid glycoprotein, fetuin, and transferrin receptor (1 μg of protein) or neoglycoproteins (100 pmol of sialooligosaccharide) were incubated with 100 mM acetate buffer (pH 5.5) containing 0.1 U of sialidase (Arthrobacter ureafaciens; Nacalai Tesque, Kyoto, Japan)/ml or 0.5 U of α2-3-specific sialidase (Salmonella enterica serovar Typhimurium; Takara, Tokyo, Japan)/ml at 37°C for 1 h. For cleavage of N-linked sugar chains in glycoproteins, 5 μl of each protein solution (2 mg/ml) was mixed with 1 μl of denaturing buffer (1 M Tris-HCl [pH 8.6], 1% SDS, 1.5% 2-mercaptoethanol) and boiled at 100°C for 3 min. The sample was mixed with 2 μl of 5% NP-40-1.2 μl of H2O-0.8 μl of peptide N-glycosidase (final concentration, 0.4 mU; Takara) and incubated at 37°C for 16 h. Then, the binding of VLP to the enzyme-treated proteins (1 μg of protein) was examined by the Western blot VLP overlay assay.
Gangliosides and other glycolipids (100 pmol) were subjected to TLC, developed, and blocked as described above. The TLC plates were incubated at 37°C for 16 h with 10 mM acetate buffer (pH 5.0) in the presence or absence of 0.1 U of sialidase/ml. The plates were washed four times with PBS, reblocked with blocking solution 1 for 30 min, and incubated with VLP. Detection of VLP on the plates was performed as described above.
HA inhibition assay.
Neoglycoproteins (100 pmol of sialooligosaccharide/25 μl) and OVA were twofold serially diluted with 0.2% BSA-PBS on microtiter plates (Falcon, Franklin Lakes, N.J.). An equal volume of VLP or JCV (8 hemagglutination [HA] units) was added to each well, and the plates were incubated at 4°C for 1 h. Fifty microliters of 0.5% erythrocyte suspension was added to each well, and incubation was conducted at 4°C. The HA inhibition titer was defined as the maximum dilution of each of the proteins that caused HA inhibition after 3 h.
Inhibition of FITC-VLP binding to SVG-A and IMR-32 cells by neoglycoproteins.
Human fetal glial cell-derived SVG-A and IMR-32 cells were plated on micro-glass plates and cultured in DMEM containing 10% FCS for 1 day. FITC-VLP diluted 1:100 with DMEM were incubated with each neoglycoprotein (10 μM sialooligosaccharide [final concentration]) at 4°C for 1 h, and the mixtures were added to microwells on the glass plates. After 1 h, unbound FITC-VLP were removed by washing with PBS, and the FITC-VLP remaining on the cells were examined by fluorescence microscopy (Olympus, Tokyo, Japan). arsid8336957
Inhibition of JCV infection by neoglycoproteins and gangliosides.
In this study, we used IMR-32 cell-adapted JCV, which makes viral particles with an immunopositive reaction against anti-VP1 antibody and which has a type II transcription control region that is devoid of both the second TATA box and part of the third NF-1 binding site of the Mad-1 sequence (8) and that is similar to the reported transcription control region detected in peripheral blood leukocytes of AIDS patients (5). JCV was prepared as previously described (26). Briefly, JCV carrier cells (JCI cells) kept in cultures (14) were frozen and thawed three times, treated with 0.05 mg of neuraminidase type V (Sigma)/ml at 37°C for 16 h, incubated at 56°C for 30 min, and centrifuged at 1,000 × g for 10 min. The virus-containing supernatant was quantitated by an HA assay and stored at −80°C until use. IMR-32 cells (2 × 105) were plated on a 12-well plate (Falcon) and cultured in DMEM containing 10% FCS for 1 day. Ganglioside GT1b (40 nmol), LSTc-OVA (20 nmol of sialooligosaccharide and 25 μg of protein), or OVA (25 μg of protein) dissolved in 25 μl of TBS was incubated with JCV solution (256 HA units) on ice for 1 h. IMR-32 cells were inoculated with 150 μl of DMEM containing 2% FCS and 50 μl of virus mixture at 37°C. After 1 h, the cells were washed with PBS three times and cultured in 1 ml of DMEM containing 2% FCS at 37°C for 3 days. The monolayer culture in each well was washed with PBS three times and lysed with 50 μl of 1% Triton X-100-TBS containing 2 μg of aprotinin/ml. The cell lysates were centrifuged at 4°C for 10 min at 20,000 × g, and the supernatants (5 μg of protein) were separated on an SDS-polyacrylamide gel (12%). After blotting of the separated proteins onto a PVDF membrane, the membrane was immersed in 5% skim milk-PBST for 30 min and subsequently in antiagnoprotein antibody diluted 1:500 with PBST for 1 h. After two washes with PBST, the membrane was incubated for 1 h with HRP-conjugated F(ab′)2 goat anti-rabbit immunoglobulins diluted 1:2,000 in PBST. After four washes with PBST, the membrane was treated with the ECL kit. Quantitation of agnoprotein expression was performed with the LAS-1000 system. Agnoprotein levels in the cells were represented as intensities relative to that of the control sample without treatment of neoglycoproteins. The means and standard deviations from three independent experiments are reported.
RESULTS
VLP bind to sialoglycoproteins.
JCV infection is thought to be mediated by sialoglycoproteins (12). To further study and confirm the oligosaccharide-binding specificities of JC VLP, we developed a Western blot VLP overlay assay. We first examined the binding of VLP to several known sialoglycoproteins, including α1-acid glycoprotein, fetuin, and transferrin receptor (2, 18, 31). We found that the binding of VLP to these glycoproteins was dependent on the presence of sialic acid, as treatment with sialidase abolished VLP binding (Fig. 1). Interestingly, the binding of VLP to these glycproteins was abolished by treatment with α2-3-specific sialidase, which prefers α2-3-linked sialic acid to α2-6-linked sialic acid, suggesting that VLP also bind to glycoproteins with α2-3-linked sialic acid in addition to those with α2-6-linked sialic acid (12). Next, to investigate whether N-linked oligosaccharides bind to VLP, these proteins were treated with peptide N-glycosidase, which removes N-linked oligosaccharides from glycoproteins. The binding of VLP to these glycoproteins was abolished by peptide N-glycosidase. These results, taken together, suggested that the N-linked sugar chains and the α2-3-linked sialic acids in glycoproteins are also involved in VLP binding to glycoproteins.
FIG. 1.
VLP overlay assay with sialoglycoproteins in the presence or absence of α2-3-specific sialidase, sialidase, and peptide N-glycosidase. For treatment with various enzymes, α1-acid glycoprotein, fetuin, and transferrin receptor (1 μg of protein) were incubated with 0.1 U of sialidase/ml or 0.5 U of α2-3-specific sialidase (S. enterica serovar Typhimurium)/ml at 37°C for 1 h. For cleavage of N-linked sugar chains in glycoproteins, 5 μl of each protein solution (2 mg/ml) was mixed with 2 μl of 5% NP-40-1.2 μl of H2O-0.8 μl of peptide N-glycosidase F and incubated at 37°C for 16 h. The binding of VLP to the enzyme-treated proteins was examined by the Western blot VLP overlay assay.
Neoglycoproteins consisting of OVA with sialic acids are bound to VLP, while OVA is not.
It has been reported that glycoproteins containing α2-6-linked sialic acid mediate JCV infection (12), but we could not confirm the involvement of α2-6-linked sialic acid in VLP binding to glycoproteins, as α2-6 sialyl linkage-specific sialidase was not available. To address this question, we synthesized neoglycoproteins by coupling OVA to sialooligosaccharides with various types of sialyl linkages. The structures of the sialooligosaccharides are shown in Table 1. LSTa has an α2-3-linked sialic acid at the terminal region of the oligosaccharide, LSTb has a branched α2-6-linked sialic acid at the internal region, LSTc has an α2-6-linked sialic acid at the terminal region, and DSLNT contains an α2-3-linked sialic acid at the terminal region and a branched α2-6-linked sialic acid at the internal region. After coupling of the sialooligosaccharides, conjugated OVA (LSTa-OVA, LSTb-OVA, LSTc-OVA, and DSLNT-OVA) showed a slower electrophoretic mobility than nontreated OVA in SDS-PAGE analysis, and reactivity to VLP had been acquired (Fig. 2a and b). As shown in Fig. 2b, LSTc-OVA with the terminal α2-6-linked sialic acid showed the most intense affinity for VLP.
TABLE 1.
Structures of sialooligosaccharides conjugated with OVA to synthesize neoglycoproteins
| Sialooligosaccharide | Structure |
|---|---|
| LSTa | Neu5Ac α2-3Gal β1-3GlcNAc β1-3Gal β1-4Glc |
| LSTb | Gal β1-3GlcNAc β1-3Gal β1-4Glc |
| 6 | |
| | | |
| Neu5Ac α2 | |
| LSTc | Neu5Ac α2-6Gal β1-4GlcNAc β1-3Gal β1-4Glc |
| DSLNT | Neu5Ac α2-3Gal β1-3GlcNAc β1-3Gal β1-4Glc |
| 6 | |
| | | |
| Neu5Ac α2 |
FIG. 2.
VLP overlay assay of neoglycoproteins containing different sialyl linkages. (a) Coomassie brilliant blue (CBB) staining of neoglycoproteins separated by SDS-PAGE. (b) Western blot VLP overlay assay of neoglycoproteins. (c) VLP overlay assay of neoglycoproteins after either sialidase or α2-3-specific sialidase treatment. (d) Lectin blotting of neoglycoproteins with SSA.
Next, to confirm that sialic acids in oligosaccharides are involved in VLP binding to neoglycoproteins, these neoglycoproteins were treated with sialidase and α2-3-specific sialidase and subjected to a VLP overlay assay. As expected, all neoglycoproteins lost their binding activities for VLP as a result of sialidase treatment, and LSTa-OVA lost binding activity as a result of α2-3-specific sialidase treatment (Fig. 2c). In order to confirm the structures of the neoglycoproteins, we carried out an experiment with SSA, because this lectin specifically recognizes Neu5Ac α2-6Gal β1-4GlcNAc β1 or Neu5Ac α2-6GalNAc structures. SSA exclusively recognized LSTc-OVA, indicating that the expected structures of the neoglycoproteins were present (Fig. 2d).
As these neoglycoproteins had binding activity for VLP, we investigated whether they could inhibit the HA activity of either VLP or JCV. All neoglycoproteins had HA inhibition activities for JCV as well as VLP, and LSTc-OVA showed the highest inhibition activity (Table 2).
TABLE 2.
Inhibitory effects of neoglycoproteins and OVA against HA caused by VLP or JCVa
| Neoglycoprotein | HA inhibition titer of:
|
|
|---|---|---|
| VLP | JCV | |
| OVA | <8 | <8 |
| LSTa-OVA | 8 | 64 |
| LSTb-OVA | 8 | 32 |
| LSTc-OVA | 512 | 512 |
| DSLNT-OVA | 8 | 64 |
Either VLP or JCV was treated with neoglycoproteins at 4°C for 1 h, and HA activity was determined as described in Materials and Methods.
We next tested the ability of neoglycoproteins to inhibit VLP binding to cultured cells. FITC-labeled VLP were pretreated with neoglycoproteins and incubated with human fetal glial cell-derived SVG-A and IMR-32 cells. As shown in Fig. 3 and 4, OVA or the neoglycoproteins LSTa-OVA, LSTb-OVA, and DSLNT-OVA had no noticeable effects on VLP binding to both SVG-A and IMR-32 cells, but LSTc-OVA clearly inhibited the binding of VLP to the cellular surface. The results suggested that neoglycoprotein LSTc-OVA, which has a strong affinity for VLP, inhibits the attachment of both VLP and JCV to erythrocytes or cells and that the binding of LSTc-OVA to both VLP and JCV is biologically relevant.
FIG. 3.
Inhibitory effects of neoglycoproteins on FITC-VLP binding to SVG-A cells. FITC-VLP were incubated with each neoglycoprotein at 4°C for 1 h and then applied to SVG-A cells. After 1 h, FITC-VLP were observed by fluorescence microscopy. VLP were not observed exclusively in the cells treated with LSTc-OVA. Cont, control.
FIG. 4.
Inhibitory effects of neoglycoproteins on FITC-VLP binding to IMR-32 cells. As with SVG-A cells, VLP were not observed exclusively in the cells treated with LSTc-OVA. Cont, control.
VLP bind to glycolipids.
Major target tissues in the central nervous system for JCV infection are rich in gangliosides, which are glycolipids containing sialic acids. Gangliosides are also known to have an affinity for some viruses, including influenza virus, parainfluenza virus, rotavirus, and Sendai virus (3, 10, 24, 27, 28). In order to examine whether glycolipid-containing gangliosides are receptors for JCV, we developed a VLP overlay assay for glycolipids and examined the efficiency of VLP binding with various concentrations of glycolipids. VLP bound to 25 pmol of GM3, GD2, GD3, GD1b, GT1b, GQ1b, and LacCer and 100 pmol of GD1a (Fig. 5a), while GM1a, GM2, and GalCer did not bind to VLP. The amounts and positions of glycolipids on TLC plates were confirmed by orcinol staining (Fig. 5b), and incubation of glycolipids on TLC plates with anti-VP1 antibody but without VLP revealed no immunopositive signal (data not shown). The structures of the applied glycolipids are summarized in Table 3. To confirm the sensitivity of gangliosides on the silica plates for sialidase, they were treated with sialidase and subjected to a VLP overlay assay. After sialidase treatment, gangliosides containing sialic acids at the internal region had lost their binding activity for VLP, while GD3 and GM3 showed no difference in their binding capacity (Fig. 5c); the latter result was due to the fact that after sialidase treatment, GD3 and GM3 seemed to have been changed into LacCer, which bound to VLP (Fig. 5a and c and Table 3). Thus, we have shown that JC VLP bind to glycolipids as well as to the previously reported glycoproteins (12).
FIG. 5.
VLP overlay assay of glycolipids immobilized on TLC plates. (a) Various amounts (25, 50, and 100 pmol [lanes 1, 2, and 3, respectively]) of glycolipids were developed on TLC plates and used in a TLC VLP overlay assay as described in Materials and Methods. Applied glycolipids are indicated below the panels. (b and c) Orcinol staining and TLC VLP overlay assay of glycolipids (500 and 100 pmol, respectively) developed on TLC plates. In panel c, the TLC plates were treated with sialidase before the overlay of VLP. Lane 1, GM1a; lane 2, GM2; lane 3, GM3; lane 4, GD2; lane 5, GD3; lane 6, GD1a; lane 7, GD1b; lane 8, GT1b; lane 9, GQ1b; lane 10, GalCer; lane 11, LacCer.
TABLE 3.
Structures and binding affinities of glycolipids bound with VLP in the presence or absence of sialidasea
| Glycolipid | Structure | VLP binding affinity with the following treatment:
|
|
|---|---|---|---|
| Control | Sialidase | ||
| GalCer | Gal β1-1′Cer | − | − |
| LacCer | Gal β1-4Glc β1-1′Cer | + | + |
| GM3 | Neu5Ac α2-3Gal β1-4Glc β1-1′Cer | + | + |
| GD3 | Gal β1-4Glc β1-1′Cer | + | + |
| 3 | |||
| Neu5Ac α2-8Neu5Ac α2 | |||
| GM2 | GalNAc β1-4Gal β1-4Glc β1-1′Cer | − | − |
| 3 | |||
| 2α Neu5Ac | |||
| GD2 | GalNAc β1-4Gal β1-4Glc β1-1′Cer | + | − |
| 3 | |||
| 2α Neu5Ac8-2α Neu5Ac | |||
| GM1a | Gal β1-3 GalNAc β1-4Gal β1-4Glc β1-1′Cer | − | − |
| 3 | |||
| 2α Neu5Ac | |||
| GD1b | Gal β1-3 GalNAc β1-4Gal β1-4Glc β1-1′Cer | + | − |
| 3 | |||
| 2α Neu5Ac8-2α Neu5Ac | |||
| GD1a | Neu5Ac α2-3Gal β1-3GalNAc β1-4Gal β1-4Glc β1-1′Cer | ± | − |
| 3 | |||
| 2α Neu5Ac | |||
| GT1b | Neu5Ac α2-3Gal β1-3GalNAc β1-4Gal β1-4Glc β1-1′Cer | + | − |
| 3 | |||
| 2α Neu5Ac8-2α Neu5Ac | |||
| GQ1b | Gal β1-3GalNAc β1-4Gal β1-4Glc β1-1′Cer | + | − |
| 3 3 | |||
| Neu5Ac α2-8Neu5Ac α2 2α Neu5Ac8-2α Neu5Ac | |||
+, intense binding; ±, weak binding; −, no binding.
Inhibition of JCV infection by gangliosides and neoglycoproteins.
Finally, we investigated whether sialooligosaccharides can inhibit viral infection by measurement of the level of expression of JCV-derived protein after inoculation of JCV pretreated with either GT1b or LSTc-OVA, which was shown to be biologically relevant (Fig. 3, 4, and 5). In order to estimate the inhibitory effects of LSTc-OVA and GT1b, we examined the levels of expression of agnoprotein, which consists of 71 amino acid residues with a molecular mass of approximately 8 kDa at the late protein coding region of JCV, because antiagnoprotein antibody is more sensitive than antibodies for other JCV proteins, such as VP1 and large T antigen, in our detection system; in addition, agnoprotein may be involved in the viral transcription of VP1 and large T antigen (15). The latter notions are also supported by our data that signals for VP1 and large T antigen were not detected by immunoblotting. As shown in Fig. 6, the expression of agnoprotein in JCV-infected IMR-32 cells was inhibited by pretreatment with GT1b (approximately an 80% decrease) and LSTc-OVA (approximately a 40% decrease), while the level of expression of agnoprotein was not altered by OVA. Thus, glycoproteins and glycolipids which contain sialic acids seem to play an important role in the attachment of JCV to the cellular surface.
FIG. 6.
Gangliosides and neoglycoproteins inhibit JCV infection in IMR-32 cells. JCV pretreated with either GT1b or LSTc-OVA at 4°C for 1 h was inoculated into IMR-32 cells and incubated at 37°C for 1 h. After 3 days, JCV agnoprotein was quantitated by Western blotting. (a) Representative Western blot of agnoprotein in IMR-32 cells. Cont, control. (b) Quantitation of agnoprotein in IMR-32 cells infected with pretreated JCV. Relative ratios of agnoprotein expression levels against the level of expression in the control infection are represented as a bar graph with the means and standard deviations from three independent experiments.
DISCUSSION
In this study, we investigated the binding of JC VLP to various glycoproteins and glycolipids. We showed that the binding of VLP to either glycoproteins or glycolipids is dependent on the sialyl linkage and the sugar chain sequence. Pretreatment of JCV with glycoproteins or gangliosides that showed biologically relevant binding to VLP decreased the infectivity of JCV for cells, indicating that these oligosaccharides contribute to JCV receptors. We also showed that VLP bind to both glycoproteins and glycolipids, an activity which may contribute to the broad distribution of JCV receptors in various cell lines (26).
The binding of α1-acid glycoprotein, fetuin, and transferrin receptor to VLP was inhibited by pretreatment with sialidase, α2-3-specific sialidase, and peptide N-glycosidase. These results suggest that JCV also binds to N-linked oligosaccharides containing α2-3-linked sialic acids. Fetuin has triantennary oligosaccharides containing two α2-3-linked sialic acids and one α2-6-linked sialic acid; α1-acid glycoprotein derived from human serum has tri- and tetraantennary oligosaccharides containing α2-3-linked sialic acids and α2-6-linked sialic acids; and the transferrin receptor derived from human placenta has triantennary oligosaccharides containing α2-3-linked sialic acids as the major structure of N-linked oligosaccharides (2, 18, 31). Although glycoproteins show heterogeneity in their sugar chain structures, tri- or tetraantennary N-linked oligosaccharides containing sialic acids may be involved in JCV infection. Fetuin and α1-acid glycoprotein retained α2-6-linked sialic acids after α2-3-specific sialidase treatment (data not shown) but lost the affinities for VLP. Thus, more than two sialic acids contained in N-linked oligosaccharides may be necessary for recognition by VLP. Transferrin has been reported to enter SVG-A cells by clathrin-mediated endocytosis similar to JCV (21). It was demonstrated that the transferrin receptor derived from human placenta bound to VLP but that the transferrin receptor immunoprecipitated from IMR-32 or human glial SVG-A cells did not bind to VLP (data not shown). As the oligosaccharide structures of glycoproteins have been reported to be different in tissues and cell types (20), the transferrin receptor derived from human brain tissue may bind to VLP. α1-Acid glycoprotein may also function as an inhibitor of JCV infection in human serum, since it has HA inhibition activity against JCV (data not shown).
To investigate whether sialic acids of glycoproteins are involved in binding to VLP, sialooligosaccharides containing α2-6-linked sialic acids at either the terminal or the internal region were coupled to OVA and examined for their affinity for VLP. The sialooligosaccharide containing α2-3-linked sialic acid at its terminal region was also coupled to OVA and investigated. All of these neoglycoproteins bound to VLP in a manner dependent on their sialic acids. Because DSLNT-OVA, which has α2-3-linked sialic acid at its terminal region and α2-6-linked sialic acid at its internal region, bound to VLP after α2-3-specific sialidase treatment, branched α2-6-linked sialic acid also seems to contribute to VLP binding. These results suggest that both α2-3-linked and α2-6-linked sialic acids at the terminal region and α2-6-linked sialic acid at the internal region of oligosaccharides are involved in the binding of glycoproteins to VLP. As shown in Table 2, neoglycoproteins had a higher titer against JCV than against VLP in several cases. The native JCV virion consists of VP2 and VP3 in addition to VP1, while VP1 constitutes about 75% of the capsid shell protein of VLP. As for the difference in HA inhibition titers between VLP and the JCV virion, we suppose that VP2 and VP3 may modulate the binding of VP1 to neoglycoproteins, since bovine papillomavirus type 1 L2, which is not essential for the assembly of the virion, plays a role in the HA of erythrocytes (23).
Given that GT1b clearly inhibited JCV infection of IMR-32 cells, gangliosides may be a potential candidate JCV receptor, in addition to being the target of VLP. Glycoproteins containing α2-6-linked sialic acids apparently have been shown to be JCV receptors by in vitro experiments (12); gangliosides may also mediate JCV infection in the central nervous system in vivo, because oligodendrocytes, which JCV can infect, contain gangliosides, such as GD1b and GT1b, which have an affinity for VLP (25). It was reported that biotinylated α2-6-linked sialic acid-specific lectin was detected in tonsil and brain tissues in which JCV DNA was detected (7; unpublished data). The distribution of various glycolipids and glycoproteins that had biologically relevant binding activity for JCV may be associated with the broad distribution of JCV receptors in various cell lines. Further investigation is needed to examine whether glycolipids are involved in JCV infection of the brain in vivo.
In this study, it was shown that neoglycoproteins and gangliosides showing biologically relevant binding to JCV inhibited JCV infection of SVG-A and IMR-32 cells, indicating that the oligosaccharides of glycoproteins and glycolipids function as JCV receptors. Multivalent ligands with these oligosaccharides could be efficient antiviral agents that inhibit both JCV infection and progressive multifocal leukoencephalopathy, similar to an anti-influenza virus agent, a synthetic glycopolymer, which had a higher inhibitory activity for influenza virus than did a monomer (11).
Acknowledgments
We thank M. Satoh and M. Sasada for technical assistance.
This study was supported in part by grants from the Ministry of Education, Science, Sports and Culture of Japan.
REFERENCES
- 1.Aminoff, D. 1961. Methods for quantitative estimation of N-acetylneuraminic acid and their application to hydrolysates of sialomucoids. Biochem. J. 81:384.. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Baenziger, J. U., and D. Fiete. 1979. Structure of the complex oligosaccharides of fetuin. J. Biol. Chem. 254:789-795. [PubMed] [Google Scholar]
- 3.Bergelson, L. D., A. G. Bukrinskaya, N. V. Prokazova, G. I. Shaposhnikova, S. L. Kocharov, V. P. Shevchenko, G. V. Kornilaeva, and E. V. Fomina-Ageeva. 1982. Role of gangliosides in reception of influenza virus. Eur. J. Biochem. 128:467-474. [DOI] [PubMed] [Google Scholar]
- 4.Bofill-Mas, S., M. Formiga-Cruz, P. Clemente-Casares, F. Calafell, and R. Girones. 2001. Potential transmission of human polyomaviruses through the gastrointestinal tract after exposure to virions or viral DNA. J. Virol. 75:10290-10299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ciappi, S., A. Azzi, R. De Santis, F. Leoncini, G. Sterrantino, F. Mazzotta, and L. Mecocci. 1999. Archetypal and rearranged sequences of human polyomavirus JC transcription control region in peripheral blood leukocytes and in cerebrospinal fluid. J. Gen. Virol. 80:1017-1023. [DOI] [PubMed] [Google Scholar]
- 6.Conyers, S. M., and D. A. Kidwell. 1991. Chromogenic substrates for horseradish peroxidase. Anal. Biochem. 192:207-211. [DOI] [PubMed] [Google Scholar]
- 7.Dimitrova, S., and W. J. Atwood. 2002. Distribution of the alpha (2-6)-linked sialic acid component of JCV receptor in normal human tissues. J. Neurovirol. 8:110. [Google Scholar]
- 8.Frisque, R. J., G. L. Bream, and M. T. Cannella. 1984. Human polyomavirus JC virus genome. J. Virol. 51:458-469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Goldmann, C., H. Petry, S. Frye, O. Ast, S. Ebitsch, K. D. Jentsch, F. J. Kaup, F. Weber, C. Trebst, T. Nisslein, G. Hunsmann, T. Weber, and W. Luke. 1999. Molecular cloning and expression of major structural protein VP1 of the human polyomavirus JC virus: formation of virus-like particles useful for immunological and therapeutic studies. J. Virol. 73:4465-4469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Holmgren, J., L. Svennerholm, H. Elwing, P. Fredman, and O. Strannegard. 1980. Sendai virus receptor: proposed recognition structure based on binding to plastic-adsorbed gangliosides. Proc. Natl. Acad. Sci. USA 77:1947-1950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kamitakahara, H., T. Suzuki, N. Nishigori, Y. Suzuki, O. Kanie, and C.-H. Wong. 1998. A lysoganglioside/poly-l-glutamic acid conjugate as a picomolar inhibitor of influenza hemagglutinin. Angew. Chem. Int. Ed. 37:1524-1528. [DOI] [PubMed] [Google Scholar]
- 12.Liu, C. K., G. Wei, and W. J. Atwood. 1998. Infection of glial cells by the human polyomavirus JC virus is mediated by an N-linked glycoprotein containing terminal alpha(2-6)-linked sialic acids. J. Virol. 72:4643-4649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Monaco, M. C., P. N. Jensen, J. Hou, L. C. Durham, and E. O. Major. 1998. Detection of JC virus DNA in human tonsil tissue: evidence for site of initial viral infection. J. Virol. 72:9918-9923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Nukuzuma, S., Y. Yogo, J. Guo, C. Nukuzuma, S. Itoh, T. Shinohara, and K. Nagashima. 1995. Establishment and characterization of a carrier cell culture producing high titres of polyoma JC virus. J. Med. Virol. 47:370-377. [DOI] [PubMed] [Google Scholar]
- 15.Okada, Y., S. Endo, H. Takahashi, H. Sawa, T. Umemura, and K. Nagashima. 2001. Distribution and function of JCV agnoprotein. J. Neurovirol. 7:302-306. [DOI] [PubMed] [Google Scholar]
- 16.Okada, Y., H. Sawa, S. Endo, Y. Orba, T. Umemura, H. Nishihara, A. C. Stan, S. Tanaka, H. Takahashi, and K. Nagashima. 2002. Expression of JC virus agnoprotein in progressive multifocal leukoencephalopathy brain. Acta Neuropathol. (Berlin) 104:130-136. [DOI] [PubMed] [Google Scholar]
- 17.Okada, Y., H. Sawa, S. Tanaka, A. Takada, S. Suzuki, H. Hasegawa, T. Umemura, J. Fujisawa, Y. Tanaka, W. W. Hall, and K. Nagashima. 2000. Transcriptional activation of JC virus by human T-lymphotropic virus type I Tax protein in human neuronal cell lines. J. Biol. Chem. 275:17016-17023. [DOI] [PubMed] [Google Scholar]
- 18.Orberger, G., R. Geyer, S. Stirm, and R. Tauber. 1992. Structure of the N-linked oligosaccharides of the human transferrin receptor. Eur. J. Biochem. 205:257-267. [DOI] [PubMed] [Google Scholar]
- 19.Ou, W. C., M. Wang, C. Y. Fung, R. T. Tsai, P. C. Chao, T. H. Hseu, and D. Chang. 1999. The major capsid protein, VP1, of human JC virus expressed in Escherichia coli is able to self-assemble into a capsid-like particle and deliver exogenous DNA into human kidney cells. J. Gen. Virol. 80:39-46. [DOI] [PubMed] [Google Scholar]
- 20.Parekh, R. B., A. G. Tse, R. A. Dwek, A. F. Williams, and T. W. Rademacher. 1987. Tissue-specific N-glycosylation, site-specific oligosaccharide patterns and lentil lectin recognition of rat Thy-1. EMBO J. 6:1233-1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pho, M. T., A. Ashok, and W. J. Atwood. 2000. JC virus enters human glial cells by clathrin-dependent receptor-mediated endocytosis. J. Virol. 74:2288-2292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ricciardiello, L., L. Laghi, P. Ramamirtham, C. L. Chang, D. K. Chang, A. E. Randolph, and C. R. Boland. 2000. JC virus DNA sequences are frequently present in the human upper and lower gastrointestinal tract. Gastroenterology 119:1228-1235. [DOI] [PubMed] [Google Scholar]
- 23.Roden, R. B., N. L. Hubbert, R. Kirnbauer, F. Breitburd, D. R. Lowy, and J. T. Schiller. 1995. Papillomavirus L1 capsids agglutinate mouse erythrocytes through a proteinaceous receptor. J. Virol. 69:5147-5151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Rolsma, M. D., H. B. Gelberg, and M. S. Kuhlenschmidt. 1994. Assay for evaluation of rotavirus-cell interactions: identification of an enterocyte ganglioside fraction that mediates group A porcine rotavirus recognition. J. Virol. 68:258-268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Schnaar, R. L., P. Longo, L. J. Yang, and T. Tai. 1996. Distinctive ganglioside patterns revealed by anti-ganglioside antibody binding to differentiating CG-4 oligodendrocytes. Glycobiology 6:257-263. [DOI] [PubMed] [Google Scholar]
- 26.Suzuki, S., H. Sawa, R. Komagome, Y. Orba, M. Yamada, Y. Okada, Y. Ishida, H. Nishihara, S. Tanaka, and K. Nagashima. 2001. Broad distribution of the JC virus receptor contrasts with a marked cellular restriction of virus replication. Virology 286:100-112. [DOI] [PubMed] [Google Scholar]
- 27.Suzuki, T., A. Portner, R. A. Scroggs, M. Uchikawa, N. Koyama, K. Matsuo, Y. Suzuki, and T. Takimoto. 2001. Receptor specificities of human respiroviruses. J. Virol. 75:4604-4613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Suzuki, Y., M. Matsunaga, and M. Matsumoto. 1985. N-acetylneuraminyllactosylceramide, GM3-NeuAc, a new influenza A virus receptor which mediates the adsorption-fusion process of viral infection. Binding specificity of influenza virus A/Aichi/2/68 (H3N2) to membrane-associated GM3 with different molecular species of sialic acid. J. Biol. Chem. 260:1362-1365. [PubMed] [Google Scholar]
- 29.Suzuki, Y., T. Nakao, T. Ito, N. Watanabe, Y. Toda, G. Xu, T. Suzuki, T. Kobayashi, Y. Kimura, A. Yamada, K. Sugawara, H. Nishimura, F. Kitame, K. Nakamura, E. Deya, M. Kiso, and A. Hasegawa. 1992. Structural determination of gangliosides that bind to influenza A, B, and C viruses by an improved binding assay: strain-specific receptor epitopes in sialo-sugar chains. Virology 189:121-131. [DOI] [PubMed] [Google Scholar]
- 30.Tumilowicz, J. J., W. W. Nichols, J. J. Cholon, and A. E. Greene. 1970. Definition of a continuous human cell line derived from neuroblastoma. Cancer Res. 30:2110-2118. [PubMed] [Google Scholar]
- 31.Yoshima, H., A. Matsumoto, T. Mizuochi, T. Kawasaki, and A. Kobata. 1981. Comparative study of the carbohydrate moieties of rat and human plasma alpha 1-acid glycoproteins. J. Biol. Chem. 256:8476-8484. [PubMed] [Google Scholar]