JOURNAL OF VIROLOGY, Oct. 2010, p. 9677–9684
Copyright © 2010, American Society for Microbiology. All Rights Reserved.
Vol. 84, No. 19
Role of N-Linked Glycosylation of the 5-HT2AReceptor
in JC Virus Infection?
Melissa S. Maginnis, Sheila A. Haley, Gretchen V. Gee, and Walter J. Atwood*
Department of Molecular Biology, Cell Biology and Biochemistry, Brown University, Providence, Rhode Island 02912
Received 5 May 2010/Accepted 12 July 2010
JC virus (JCV) is a human polyomavirus and the causative agent of the fatal demyelinating disease
progressive multifocal leukoencephalopathy (PML). JCV infection of host cells is dependent on interactions
with cell surface asparagine (N)-linked sialic acids and the serotonin 5-hydroxytryptamine2Areceptor (5-
HT2AR). The 5-HT2AR contains five potential N-linked glycosylation sites on the extracellular N terminus.
Glycosylation of other serotonin receptors is essential for expression, ligand binding, and receptor function.
Also, glycosylation of cellular receptors has been reported to be important for JCV infection. Therefore, we
hypothesized that the 5-HT2AR N-linked glycosylation sites are required for JCV infection. Treatment of
5-HT2AR-expressing cells with tunicamycin, an inhibitor of N-linked glycosylation, reduced JCV infection.
Individual mutation of each of the five N-linked glycosylation sites did not affect the capacity of 5-HT2AR to
support JCV infection and did not alter the cell surface expression of the receptor. However, mutation of all
five N-linked glycosylation sites simultaneously reduced the capacity of 5-HT2AR to support infection and
altered the cell surface expression. Similarly, tunicamycin treatment reduced the cell surface expression of
5-HT2AR. Mutation of all five N-linked glycosylation sites or tunicamycin treatment of cells expressing
wild-type 5-HT2AR resulted in an altered electrophoretic mobility profile of the receptor. Treatment of cells
with PNGase F, to remove N-linked oligosaccharides from the cell surface, did not affect JCV infection in
5-HT2AR-expressing cells. These data affirm the importance of 5-HT2AR as a JCV receptor and demonstrate
that the sialic acid component of the receptor is not directly linked to 5-HT2AR.
The initial interaction between virus and host occurs via
molecular interactions of viral attachment proteins and recep-
tors on host cells. Therefore, receptor recognition is a critical
host cell determinant and may play a key regulatory role in
viral pathogenesis. The polyomavirus JC virus (JCV) is a ubiq-
uitous human pathogen (21, 25, 32) that is initially subclinical
yet establishes a persistent infection in the kidney (11). In
immunosuppressed individuals JCV can become reactivated,
leading to infection in the central nervous system (CNS) (13–
15, 20), where the virus specifically targets glial cells, including
astrocytes and the myelin-producing cells, oligodendrocytes
(40, 48). JCV infection and cytolytic destruction of oligoden-
droglia cause the fatal disease progressive multifocal leukoen-
cephalopathy (PML) (1, 22). The most common cause of PML
is associated with human immunodeficiency virus (HIV) and
AIDS (10, 23). However, in recent years PML has been re-
ported in patients receiving immunosuppressive therapies for
autoimmune diseases such as Crohn’s disease (44), multiple
sclerosis (MS) (24, 26, 28, 47), systemic lupus erythematosus
(5, 33), and rheumatoid arthritis (5, 19, 37). The prognosis of
PML is bleak, as the disease progresses rapidly and usually
proves fatal within 1 year of the onset of symptoms. While
current treatment options for PML are limited (23), recent
studies suggest that mirtazapine, a serotonin receptor antago-
nist, may be capable of slowing the progression of PML (6, 27,
JCV has a nonenveloped, icosahedral capsid that encap-
sidates a circular double-stranded DNA (dsDNA) genome
(39). JCV attachment to cells is mediated by an N-linked
glycoprotein with either ?(2,3)- or ?(2,6)-linked sialic acid (16,
31), suggesting that N-linked glycosylation of cellular receptors
is important for JCV infection. N-linked glycosylation is a
posttranslational process by which oligosaccharides are added
to asparagine residues, and this modification is important for
protein processing, folding, expression, and function (43). Pre-
vious studies from our laboratory revealed that the JCV also
requires the serotonin 5-hydroxytryptamine2Areceptor (5-
HT2AR) to mediate JCV infection (18, 35, 38), while others
report that JCV infection can occur in the absence of 5-HT2AR
(7, 8). 5-HT2AR is a seven-transmembrane-spanning G-pro-
tein-coupled receptor that belongs to a large family of 5-HT
serotonin receptors. 5-HT2AR is abundantly expressed on cells
in the brain (4), including glial cells (3), and in the kidney (4),
which parallels the sites of JCV infection. N-linked glycosyla-
tion plays a key regulatory role in the function of serotonin
receptors. Mutation of N-linked glycosylation sites in human
5-HT3AR and 5-HT5AR results in decreased expression at the
plasma membrane, which is critical for receptor function (17,
34). N-linked glycosylation of murine 5-HT3AR regulates
plasma membrane targeting, ligand binding, Ca2?flux, and
receptor trafficking (36), suggesting that glycosylation is essen-
tial for expression and function of serotonin receptors.
While previous studies have concluded that JCV utilizes an
N-linked glycoprotein with ?(2,3)-linked sialic acid (31) or
?(2,6)-linked sialic acid (16) and 5-HT2AR (18) to initiate
infection in host cells, the mechanism(s) by which JCV engages
its cellular receptors and the importance of receptor glycosyl-
ation remain unclear. 5-HT2AR contains potential asparagine
* Corresponding author. Mailing address: Department of Molecular
Biology, Cell Biology and Biochemistry, Brown University, 70 Ship
Street, Providence, RI 02903. Phone: (401) 863-3116. Fax: (401) 863-
9653. E-mail: Walter_Atwood@brown.edu.
?Published ahead of print on 21 July 2010.
(N)-linked glycosylation sites, five of which are predicted to be
expressed in the extracellular amino-terminal region, where
they could be accessible to the virus (2). The goal of this study
was to determine whether potential N-linked glycosylation
sites expressed in 5-HT2AR are required for JCV infection. We
found that N-linked glycosylation of 5-HT2AR is important for
receptor expression but not necessary for JCV infection.
MATERIALS AND METHODS
Cells, viruses, and antibodies. HEK293A cells (ATCC, Manassas, VA) were
grown in Dulbecco’s minimal essential medium (DMEM) (Mediatech, Inc.,
Herndon, VA) supplemented with 5% fetal calf serum (FCS) (Atlanta Biologi-
cals, Lawrenceville, GA) and penicillin-streptomycin (Mediatech, Inc.). Gener-
ation of the virus strain Mad-1/SVE? was described previously (29, 30). JCV
infection in HEK293A cells was assessed using an antibody specific for JCV large
T antigen (T Ag) (PAB962). The PAB962 hybridoma produces a monoclonal
antibody (MAb) for JCV large T antigen that does not cross-react with simian
virus 40 (SV40) T antigen and was provided by the Tevethia laboratory (Penn
State University). The yellow fluorescent protein (YFP)-specific antibody
(ab290) used for immunoblot analysis and pan-cadherin antibody (ab16505) used
for confocal microscopy were purchased from Abcam (Cambridge, MA).
Analysis of 5-HT2AR potential glycosylation sites. 5-HT2AR protein was ana-
lyzed for topography and potential N-linked glycosylation sites using UniProt
Knowledgebase (UniProtKb) (http://www.uniprot.org).
Generation of 5-HT2AR–YFP fusion construct. The cDNA for human
5-HT2AR in pcDNA3.1 was purchased from the Missouri S&T cDNA Resource
Center (Rolla, MO) (www.cdna.org). To generate the 5-HT2AR–YFP fusion
construct, cDNA was PCR amplified with a 5? primer containing an XhoI site
followed by the Kozak sequence and the first 16 nucleotides of the open reading
frame (ORF) (HT2aR F, 5?-CTCGAGCACCATGGATATTCTTTGTG-3?) and
a 3? primer complementary to the last 19 nucleotides of the ORF with a BamHI
site in place of the stop codon (HT2aR R, 5?-GGATCCaaCACACAGCTCAC
CTTTTCA-3?; lowercase letters indicate linker nucleotides). This PCR product
was cloned into the pCRII vector using the TA cloning kit dual promoter
(Invitrogen, Carlsbad, CA). The product was digested with XhoI and BamHI and
ligated into the pEYFP-N1 vector (Clontech, Mountain View, CA) to fuse YFP
to the C terminus of the receptor. Directionality of cloning was confirmed by
sequencing using 640 ng of plasmid and 8 pmol of primers designed to recognize
YFP (5?33?): TGGCACCAAAATCAACGGG and CTTCAGGGTCAGCTT
GCC. Sequencing reactions were performed by Genewiz (New Brunswick, NJ)
and analyzed using MacVector.
Site-directed mutagenesis of 5-HT2AR N-linked glycosylation sites. Potential
N-linked glycosylation sites in 5-HT2AR were mutated using the QuikChange II
site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the man-
ufacturer’s instructions. PCR mutagenesis primers were designed using Strat-
agene’s web-based QuikChange primer design program and synthesized by In-
vitrogen. Primers used for mutagenesis (5?33?) were as follows: N8A, CACCA
GTG; N38A, GACTTTAACTCCGGAGAAGCTGCCACTTCTGATGCATTT
AACTG and CAGTTAAATGCATCAGAAGTGGCAGCTTCTCCGGAGTT
AAAGTC; N44A, GAGAAGCTAACACTTCTGATGCATTTGCCTGGACA
GTCGACT and AGTCGACTGTCCAGGCAAATGCATCAGAAGTGTTAG
CTTCTC; N51AF, AACTGGACAGTCGACTCTGAAGCTCGAACCAACCT
TTCCT and AGGAAAGGTTGGTTCGAGCTTCAGAGTCGACTGTCCA
GTT; and N54A, CGACTCTGAAAATCGAACCGCCCTTTCCTGTGAAGG
GTGC and GCACCCTTCACAGGAAAGGGCGGTTCGATTTTCAGAGTCG.
Plasmids were transformed in DH5? competent cells (Invitrogen) and S.O.C.
medium (Invitrogen). Mutations were confirmed by sequencing using 640 ng of
plasmid and 8 pmol of the YFP primers. Sequencing reactions were performed
by Genewiz and analyzed using MacVector.
Transfection of HEK293A cells. HEK293A cells were plated on 22- by 22-mm
glass coverslips (Fisher Scientific, Pittsburgh, PA) in six-well plates (Fisher Sci-
entific) in DMEM with 5% FCS overnight. Cells at ?90% confluence were
transfected with 1 ?g of DNA per well using Lipofectamine 2000 (Invitrogen).
Cells were incubated at 37°C for 4 h, and then the medium was replaced with
DMEM containing 5% FCS and penicillin-streptomycin and incubated at 37°C
for 20 h. Cells were monitored for transfection efficiency by YFP expression
using an Eclipse E800 epifluorescence microscope (Nikon Inc., Melville, NY) at
24 h posttransfection.
Treatment of cells with tunicamycin. At 4 h posttransfection, cells were
treated with 0.2 ?g/ml of tunicamycin (Sigma) (2) or an equivalent volume of
methanol (MeOH) (vehicle control) in complete DMEM. Cells were incubated
at 37°C for 24 h in the presence of tunicamycin. Cells were then infected as
described below, treated with MeOH to fix cells for confocal microscopy, or
harvested by mechanical scraping in phosphate-buffered saline (PBS) for immu-
Treatment of cells with neuraminidase. At 24 h posttransfection, cells were
treated with 0.8 U/ml of neuraminidase from Vibrio cholerae (Sigma Aldrich, St.
Louis, MO) in PBS (pH 6.5) with 1 mM MgCl2and 1 mM CaCl2at 37°C for 1 h.
Cells were then washed twice with minimal essential medium (MEM) containing
2% FCS prior to infection.
Treatment of cells with PNGase F. At 24 h posttransfection, cells were treated
with 2 or 10 U/?l of glycerol free PNGase F (N-glycosidase F) (New England
BioLabs, Ipswich, MA) in Hanks buffered salt solution (HBSS) (Invitrogen) at
37°C for 1 h (41). Cells were washed with MEM containing 2% FCS prior to
Indirect immunofluorescence assay of JCV infection. Cells were infected with
JCV at a multiplicity of infection (MOI) of 1 focus-forming unit (FFU)/cell in
MEM containing 2% FCS at 37°C for 1 h, DMEM containing 5% FCS and
penicillin-streptomycin was added, and cells were incubated at 37°C. At 48 h
postinfection, cells were washed with PBS and fixed either in 2% paraformalde-
hyde (PFA) at room temperature (RT) for 10 min or in MeOH at ?20°C for ?10
min. Cells were washed with PBS, permeabilized with 0.5% Triton X-100 at RT
for 15 min, and then blocked with 10% goat serum (MP Biomedicals, Solon, OH)
in PBS at RT for 1 h. Cells were stained with the JCV T-Ag-specific antibody
PAB962 (1:10) in PBS at 37°C for 1 h, washed with PBS, and then incubated with
a goat anti-mouse Alexa 594 antibody (1:500) in PBS (Invitrogen) and washed
with PBS. Coverslips were mounted on slides using Vectashield with DAPI
(4?,6?-diamidino-2-phenylindole) (Vector Labs, Burlingame, CA). Cells were
analyzed by epifluorescence microscopy (Nikon), and infected cells were quan-
titated based on nuclear T-Ag expression.
Confocal imaging of cell surface expression of 5-HT2AR–YFP. Cells were
washed with PBS at 24 h following transfection or tunicamycin treatment and
fixed in cold MeOH at ?20°C for ?10 min. Cells were then washed, permeab-
ilized using 1% Triton X-100 at RT for 5 min, and blocked in PBS with 5%
bovine serum albumin (BSA) and 0.1% Tween 20 (PBS-BT) at RT for 30 min.
Cells were stained with a plasma membrane marker (pan-cadherin-specific an-
tibody; Abcam) (4 ?g/ml) in PBS-BT at RT for 1 h, washed in PBS-BT, incubated
with a goat anti-rabbit Alexa 546 (Invitrogen) (1:1,000) in PBS-BT at RT for 1 h,
and washed in PBS-BT. Coverslips were rinsed with distilled water prior to being
mounted on slides using Vectashield with DAPI (Vector Labs). Cells were
analyzed using a Zeiss 510 confocal microscope equipped with a Meta detector
and Meta software (Carl Zeiss, New York, NY). Multiple images were captured
using a 63? objective. The cell surface expression of 5-HT2AR–YFP was quan-
titated using Metamorph imaging software (Universal Imaging, Downingtown,
PA). Images were separated by color and set to an inclusive threshold. A total of
35 individual cells per sample were outlined for measurement. Colocalization of
5-HT2AR–YFP was quantitated as a function of pan-cadherin (a plasma mem-
brane marker) to determine the percentage of 5-HT2AR–YFP expressed on the
Immunoblot analysis of 5-HT2AR glycosylation. HEK293A cells were trans-
fected as described above, washed with PBS, and harvested by mechanically
scraping cells in PBS. Cells were pelleted and resuspended in 50 ?l of cold 50
mM Tris (pH 7.6) with protease inhibitor cocktail (PIC) (Sigma) on ice. Samples
were sonicated using a 150 series Sonic Dismembrator (Fisher Scientific, Pitts-
burgh, PA) on a low-power setting equivalent to approximately 10 W on ice for
10 s and then pelleted at 19,060 ? g at 4°C for 10 min. Supernatants were
removed, and pellets (membrane fraction) were resuspended in cold Tris buffer
with PIC on ice and pelleted again. Pellets were resuspended in 50 ?l cold Tris
buffer without PIC and mixed with 50 ?l of sample buffer (Bio-Rad, Hercules,
CA). Samples were boiled at 95°C for 5 min, and 50 ?l of the sample was resolved
by SDS-PAGE on 10% Tris-HCl gels (Bio-Rad). Proteins were transferred to
polyvinylidene difluoride (PVDF) membranes (Bio-Rad) using a semidry Trans
Blot apparatus (Bio-Rad). Membranes were blocked in 5% milk–PBS-T over-
night and then incubated with ab290 (1:2,000) in 5% milk–PBS-T at RT for 1 h
and washed with PBS-T. Membranes were incubated with goat anti-rabbit Alexa
680 (Invitrogen) (1:2,000) in 5% milk–PBS-T at RT for 1 h and washed in PBS-T.
Membranes were rinsed in PBS and analyzed using the LiCor Odyssey (LiCor
Biosciences, Lincoln, NE). Images are shown in gray scale.
Statistical analysis. A two-tailed Student t test was performed using Microsoft
Excel. P values of ?0.05 were considered to be statistically significant.
9678 MAGINNIS ET AL.J. VIROL.
JCV infection of 5-HT2AR-expressing HEK293A cells is sen-
sitive to tunicamycin and neuraminidase. Previous studies
have indicated that JCV requires an N-linked glycoprotein for
infection (31). Thus, we wanted to determine the effect of
tunicamycin, an inhibitor of N-linked glycosylation, on JCV
infection in cells expressing 5-HT2AR. HEK293A cells, a
poorly permissive cell line, were transfected with a 5-HT2AR–
YFP fusion construct or YFP alone, treated with tunicamy-
cin, and then infected with JCV (Fig. 1A). Tunicamycin treat-
ment decreased JCV infection in 5-HT2AR–YFP-expressing
HEK293A cells, suggesting that N-linked glycosylation may be
important for JCV infection. HEK293A cells expressing either
5-HT2AR-pcDNA3.1 or 5-HT2AR–YFP support JCV infection
equivalently, suggesting that the appendage of a YFP fusion
does not affect the capacity of 5-HT2AR to support JCV infec-
tion (data not shown). To assess whether JCV infection of
5-HT2AR–YFP-expressing HEK293A cells is sialic acid depen-
dent, transfected cells were treated with neuraminidase to re-
move cell surface sialic acids and then infected with JCV (Fig.
1B). Neuraminidase treatment of 5-HT2AR–YFP-expressing
HEK293A cells reduced JCV infection, suggesting that infec-
tion is sialic acid dependent.
The 5-HT2Areceptor contains potential N-linked glycosyla-
tion sites that are surface exposed. Given that JC virus utilizes
an N-linked glycoprotein with ?(2,3)- or ?(2,6)-linked sialic acid
and 5-HT2AR to infect host cells and that JCV infection is sen-
sitive to tunicamycin (Fig. 1), we analyzed the nucleotide se-
quence of 5-HT2AR for the presence of potential N-linked gly-
cosylation sites. A potential N-linked glycosylation site is denoted
by the motif Asn-X-Ser/Thr, where X is any amino acid. Upon
sequence analysis of 5-HT2AR, we observed eight potential N-
linked glycosylation sites at residues N8, N38, N44, N51, N54,
N75, N107, and N384 (Fig. 2A). However, topographical analysis
of the seven-transmembrane receptor revealed that N75 is local-
ized at the plasma membrane and that sites N107 and N384 are
expressed on intracellular loops, suggesting that these residues
FIG. 1. JCV infection of 5-HT2AR–YFP-expressing HEK293A
cells is tunicamycin sensitive and sialic acid dependent. (A) HEK293A
cells were transfected with YFP or 5-HT2AR–YFP and then treated
with 0.2 ?g/ml of tunicamycin or MeOH (vehicle control) prior to
infection with JCV. (B) HEK293A cells were transfected with YFP or
5-HT2AR–YFP and then, at 24 h posttransfection, treated with 0.8
U/ml of neuraminidase at 37°C for 1 h prior to infection with JCV. At
48 h postinfection, cells were fixed and stained using JCV T-Ag-
specific MAb (PAB962) and goat anti-mouse Alexa 594. Cells were
analyzed by epifluorescence microscopy, and cells expressing nuclear T
Ag were scored as infected cells. Data represent the average number
of infected cells in a 20? field for nine fields of view for triplicate
samples. Error bars indicate standard deviations.*, P ? 0.05.
FIG. 2. Mutation of individual 5-HT2AR N-linked glycosylation
sites does not alter JCV infection. (A) Schematic diagram of the
seven-transmembrane-spanning human 5-HT2AR, with potential N-
linked glycosylation sites (Asn-X-Ser/Thr) indicated by position in the
protein sequence. The highlighted sites were chosen for site-directed
mutagenesis in which the Asn residues were mutated to Ala.
(B) HEK293A cells were transfected with YFP, 5-HT2AR–YFP, or
individual N-linked glycosylation site mutants and then infected with
JCV at 24 h posttransfection. At 48 h postinfection, cells were fixed
and stained using JCV T-Ag-specific MAb (PAB962) and goat anti-
mouse Alexa 594. Cells were analyzed by epifluorescence microscopy,
and cells expressing nuclear T Ag were scored as infected cells. Data
represent the average number of infected cells in a 20? field of view
for 12 fields of view for triplicate samples from three independent
experiments. Error bars represent standard errors of the means
VOL. 84, 2010 5-HT2ARECEPTOR GLYCOSYLATION IN JCV INFECTION9679
may not be accessible to ligands. The potential N-linked glycosyl-
ation sites N8, N38, N44, N51, and N54 are expressed on the
extreme N-terminal region of the receptor in an area presumably
accessible to JCV. These residues were selected for further char-
acterization of their role in JCV infection.
Mutation of individual N-linked glycosylation sites does not
affect the capacity of 5-HT2AR to support JCV infection. The
presence of five potential N-linked glycosylation sites in a surface-
exposed N-terminal region of 5-HT2AR suggested that JC virus
might utilize potential N-linked glycosylation sites in 5-HT2AR to
mediate infection of host cells. To test this possibility, asparagine
residues in potential N-linked glycosylation sites (Asn-X-Ser/Thr)
were mutated to alanine residues by site-directed mutagenesis in
the 5-HT2AR–YFP fusion construct. Wild-type and N-linked gly-
cosylation mutant 5-HT2AR–YFP constructs were transfected
into HEK293A cells and tested for the capacity to support JCV
infection (Fig. 2B). Mutation of individual N-linked glycosylation
sites in 5-HT2AR–YFP did not affect the capacity of 5-HT2AR–
YFP to support JCV infection. These data suggest that JC virus
does not utilize a specific N-linked glycosylation site to mediate
infection in a 5-HT2AR-dependent manner. However, it was nec-
essary to determine whether the cell surface expression of the
serotonin receptor was affected by mutation of the potential gly-
cosylation sites, as previous studies have found that point muta-
tion of N-linked glycosylation sites in serotonin receptors inter-
fered with cell surface expression and receptor functions (17, 34).
To test whether mutation of potential N-linked glycosylation sites
affected the cell surface expression of 5-HT2AR–YFP, HEK293A
cells transfected with wild-type and mutant constructs were fixed
at 24 h posttransfection, stained with the plasma membrane
marker pan-cadherin, and analyzed by confocal microscopy (Fig.
3A). 5-HT2AR–YFP was observed at the cell surface as well as in
sion was analyzed as a function of pan-cadherin expression using
Metamorph imaging analysis to determine the percentage of
5-HT2AR that was expressed at the cell surface (Fig. 3B). Cell
surface expression of the wild-type and mutant constructs was
determined to be equivalent, suggesting that mutation of individ-
ual N-linked glycosylation sites in 5-HT2AR–YFP does not affect
the normal localization of the receptor.
Mutation of all N-linked glycosylation sites reduces the ca-
pacity of 5-HT2AR to support JCV infection. The observation
that JCV infection of 5-HT2AR–YFP-expressing HEK293A
cells is tunicamycin sensitive indicates that glycosylation of
5-HT2AR may be important for JCV infection, yet mutation of
individual glycosylation sites had no effect on JCV infection
(Fig. 2B). Thus, a 5-HT2AR–YFP construct in which all five
N-linked glycosylation sites were mutated to alanine residues
(5-HT2AR–YFP–N8/38/44/51/54A) was generated and tested
for the capacity to support JCV infection (Fig. 4). Mutation of
FIG. 3. Mutation of individual 5-HT2AR N-linked glycosylation sites does not alter the cell surface expression of 5-HT2AR–YFP. HEK293A
cells were transfected with 5-HT2AR–YFP or individual N-linked glycosylation site mutants. At 24 h posttransfection, cells were fixed and stained
for pan-cadherin (plasma membrane marker), and coverslips were mounted to slides using Vectashield with DAPI. Cells were analyzed for
5-HT2AR–YFP (green) and pan-cadherin (red) under a 63? objective by confocal microscopy using a Zeiss 510 microscope equipped with a Meta
detector and LSM software. (A) Representative confocal micrograph images. (B) Cell surface quantitation of 35 individual cells expressing
5-HT2AR or individual N-linked glycosylation site mutants.
FIG. 4. Mutation of all 5-HT2AR N-linked glycosylation sites re-
duces JCV infection. HEK293A cells were transfected with YFP,
5-HT2AR–YFP or 5-HT2AR–YFP–N8/38/44/51/54A and infected with
JCV. At 48 h postinfection, cells were fixed and stained using JCV
T-Ag-specific MAb (PAB962) and goat anti-mouse Alexa 594. Cells
were analyzed by epifluorescence microscopy, and cells expressing
nuclear T Ag were scored as infected cells. Data represent the average
number of infected cells in a 20? field for nine fields of view for
triplicate samples. Error bars indicate standard deviations.*, P ? 0.05.
9680MAGINNIS ET AL. J. VIROL.
all potential N-linked glycosylation sites in 5-HT2AR resulted
in a significant reduction in JCV infection. These data indicate
that glycosylation of 5-HT2AR is required for the receptor to
support JCV infection.
Mutation of all N-linked glycosylation sites alters the cell
surface localization of 5-HT2AR–YFP. Previous studies have
found that the process of N-linked glycosylation is important
for targeting the serotonin receptor to the plasma membrane
(17, 34, 36). To test whether mutation of all potential N-linked
glycosylation sites affected the cell surface expression of
5-HT2AR–YFP, HEK293A cells were transfected with wild-
type 5-HT2AR–YFP or 5-HT2AR–YFP–N8/38/44/51/54A and
either treated with tunicamycin or left untreated. Cells were
fixed at 24 h posttransfection, stained for pan-cadherin as a
plasma membrane marker, and analyzed by confocal micros-
copy (Fig. 5A). 5-HT2AR–YFP was observed at the cell surface
as well as in other subcellular compartments, as seen in Fig. 3.
However, in either cells transfected with 5-HT2AR–YFP and
treated with tunicamycin or cells transfected with 5-HT2AR–
YFP–N8/38/44/51/54A, a significant alteration in the cellular
localization of 5-HT2AR–YFP was observed by confocal mi-
croscopy (Fig. 5A). 5-HT2AR–YFP cell surface expression was
analyzed as a function of pan-cadherin expression using Meta-
morph imaging analysis to determine the percentage of the
receptor that was expressed at the plasma membrane (Fig. 5B).
Through imaging analysis it was determined that the cell
surface expression of 5-HT2AR–YFP–N8/38/44/51/54A is
reduced, and the cellular expression pattern of 5-HT2AR–
YFP–N8/38/44/51/54A parallels that observed in tunicamycin-
treated 5-HT2AR–YFP-expressing HEK293A cells (Fig. 5A
and B). These data demonstrate that glycosylation of 5-HT2AR
plays an important role in the proper cell surface expression of
the receptor. Further, these data suggest that a reduction in
JCV infection in cells expressing 5-HT2AR–YFP–N8/38/44/51/
54A or in cells treated with tunicamycin is likely due to a
reduction of 5-HT2AR expression at the cell surface.
Inhibition of N-linked glycosylation alters the electro-
phoretic mobility of 5-HT2AR–YFP. Glycosylation of serotonin
receptors has previously been assessed by immunoblot analysis
of tunicamycin-treated cells expressing 5-HT receptors and
5-HT the receptor N-linked glycosylation mutants (17, 34, 36).
To determine the glycosylation status of the human 5-HT2AR,
we analyzed the electrophoretic mobility of 5-HT2AR–YFP in
the presence and absence of tunicamycin. HEK293A cells
transfected with either YFP or 5-HT2AR–YFP were treated
with tunicamycin or MeOH (vehicle control), harvested, and
analyzed by immunoblot analysis using a YFP-specific antibody
(Fig. 6A). Cells expressing 5-HT2AR–YFP produced a major
YFP-immunoreactive species at approximately 50 kDa and
minor species at approximately 60, 70, and 75 kDa. Tunicamy-
cin treatment of cells resulted in an altered electrophoretic
mobility pattern of 5-HT2AR–YFP. The immunoreactive spe-
cies at ?75 kDa was reduced under tunicamycin treatment,
suggesting that this represents a glycosylated form of the re-
ceptor. To further probe the importance of N-linked glycosyl-
ation, wild-type and N-linked glycosylation site mutant
5-HT2AR–YFP constructs were expressed in HEK293A cells,
harvested, and subjected to immunoblot analysis (Fig. 6B). The
apparent electrophoretic mobility of 5-HT2AR–YFP was re-
duced by ?3 kDa by point mutations of individual N-linked
glycosylation sites. However, mutation of all five sites resulted
in a reduction in the ?75-kDa band as in the tunicamycin-
treated sample (Fig. 6B), suggesting that inhibition of
5-HT2AR N-linked glycosylation by either mutation or tunica-
mycin treatment results in an altered electrophoretic mobility
profile. These data show that 5-HT2AR is normally glycosyl-
ated and that glycosylation is critical for cell surface localiza-
FIG. 5. Mutation of all 5-HT2AR N-linked glycosylation sites reduces cell surface expression of 5-HT2AR–YFP. HEK293A cells were trans-
fected with 5-HT2AR–YFP or 5-HT2AR–YFP–N8/38/44/51/54A. Cells were untreated or treated with either MeOH or tunicamycin for 24 h. Cells
were fixed and stained for pan-cadherin (a plasma membrane marker), and coverslips were mounted to slides using Vectashield with DAPI. Cells
were analyzed for 5-HT2AR–YFP (green) and pan-cadherin (red) under a 63? objective by confocal microscopy using a Zeiss 510 microscope
equipped with a Meta detector and LSM software. (A) Representative confocal micrograph images. (B) Cell surface quantitation of individual cells
expressing 5-HT2AR–YFP treated with MeOH or tunicamcyin and 5-HT2AR–YFP–N8/38/44/51/54A. Error bars indicate standard deviations.*,
P ? 0.05.
VOL. 84, 2010 5-HT2ARECEPTOR GLYCOSYLATION IN JCV INFECTION 9681
JCV infection of 5-HT2AR–YFP-expressing HEK293A cells
is resistant to PNGase F treatment. Treatment of 5-HT2AR–
YFP-expressing HEK293A cells with tunicamycin (Fig. 1A) or
mutation of all five N-linked glycosylation sites (Fig. 4) results
in a decreased capacity of 5-HT2AR to support JCV infection
yet also reduces the cell surface expression of the receptor
(Fig. 5). Thus, to further address whether N-linked glycosyla-
tion of 5-HT2AR is required for JCV infection, cells were
treated with PNGase F to specifically remove N-linked oligo-
saccharides from the surface of 5-HT2AR–YFP-expressing
HEK293A cells (41) and then infected with JCV (Fig. 7A).
Interestingly, PNGase F had no effect on JCV infection. Im-
munoblot analysis revealed that PNGase F treatment of
5-HT2AR–YFP-expressing cells resulted in an electrophoretic
mobility pattern of 5-HT2AR (Fig. 7B) similar to that observed
in tunicamycin-treated cells, suggesting that the PNGase F
used in the experiments effectively removed cell surface N-
linked oligosaccharides. Taken together, these data indicate
that the sialic acid component of the JCV receptor is not linked
This study was conducted to determine whether JCV utilizes
potential N-linked glycosylation sites on the 5-HT2Areceptor
to mediate JCV infection. Treatment of 5-HT2AR–YFP-ex-
pressing cells with tunicamycin, an inhibitor of N-linked glyco-
sylation, blocked JCV infection. We identified potential N-
linked glycosylation sites in the N terminus of 5-HT2AR and
subjected them to site-directed mutagenesis and expression in
a poorly permissive cell line. Cells expressing 5-HT2AR with
individual point mutations in the N-linked glycosylation sites
were equally capable of supporting JCV infection as cells ex-
pressing wild-type 5-HT2AR, yet mutation of all five sites re-
sulted a significant reduction in the capacity of 5-HT2AR to
support infection. Further, mutation of all five 5-HT2AR N-
linked glycosylation sites resulted in decreased cell surface
expression of the receptor, yet mutation of individual sites did
not. While JCV infection of 5-HT2AR–YFP-expressing cells
was tunicamycin sensitive, tunicamycin also reduced the cell
surface expression of the receptor, suggesting that infection is
FIG. 6. Inhibition of 5-HT2AR glycosylation by tunicamycin treat-
ment or mutation alters the electrophoretic mobility of 5-HT2AR.
(A) HEK293A cells were transfected with YFP or 5-HT2AR–YFP and
then treated with 0.2 ?g/ml of tunicamycin or MeOH (vehicle control).
(B) HEK293A cells were transfected with wild-type or mutant
5-HT2AR–YFP. Cells were harvested at 24 h posttransfection, and cell
membrane fractions were collected, resolved by SDS-PAGE, trans-
ferred to a PVDF membrane, and blotted using a YFP-specific anti-
body and an anti-rabbit Alexa 680 antibody. Membranes were analyzed
using a LiCor Odyssey. Images are shown in gray scale. Arrows indi-
cate altered electrophoretic mobility in a ?75-kDa band.
FIG. 7. Treatment of 5-HT2AR–YFP-expressing HEK293A cells
with PNGase F has no effect on JCV infection. (A) HEK293A cells
were transfected with 5-HT2AR–YFP and then treated with 2 or 10
U/?l of PNGase F or HBSS (vehicle control) prior to infection with
JCV. At 48 h postinfection, cells were fixed and stained using JCV
T-Ag-specific MAb (PAB962) and goat anti-mouse Alexa 594. Cells
were analyzed by epifluorescence microscopy, and cells expressing
nuclear T Ag were scored as infected cells. Data represent the average
number of infected cells in a 20? field for nine fields of view for
triplicate samples. Error bars indicate standard deviations. (B) Re-
moval of 5-HT2AR N-linked oligosaccharides by PNGase F treatment
alters the electrophoretic mobility of 5-HT2AR. HEK293A cells were
transfected with YFP or 5-HT2AR–YFP and then treated with 2 U/?l
of PNGase F or HBSS (vehicle control) at 37°C for 1 h at 24 h
posttransfection. Cells were harvested, and cell membrane fractions
were collected, resolved by SDS-PAGE, transferred to a PVDF mem-
brane, and blotted using a YFP-specific antibody and an anti-rabbit
Alexa 680 antibody. Membranes were analyzed using a LiCor Odyssey.
Images are shown in gray scale. The arrow indicates altered electro-
phoretic mobility in a ?75-kDa band.
9682 MAGINNIS ET AL.J. VIROL.
reduced due to the decreased availability of the receptor.
Moreover, the electrophoretic mobility of 5-HT2AR was al-
tered by tunicamycin treatment or mutation of all 5-HT2AR
N-linked glycosylation sites but was unaffected by mutation of
individual N-linked glycosylation sites, suggesting that the re-
ceptor is normally glycosylated. PNGase F, which enzymati-
cally removes N-linked oligosaccharides from the cell surface,
does not affect JCV infection. Taken together, these data sug-
gest that N-linked glycosylation of 5-HT2AR is required for
proper cell surface expression yet is not required for JCV
infection. These findings provide a better understanding of
how JC virus engages host cell receptors to mediate infection.
This is also the first characterization of the glycosylation profile
of the human 5-HT2AR. Previous studies have suggested that
glycosylation of serotonin receptors is essential for plasma
membrane localization of the receptor, signaling events, and
ligand binding (17, 34, 36). For example, treatment of cells
expressing human 5-HT3AR with tunicamycin results in a re-
duction in the molecular weight of the receptor as assessed by
electrophoretic mobility. Further, mutation of the individual
N-linked glycosylation sites also results in a reduction in elec-
trophoretic mobility of ?3 kDa, suggesting that human
5-HT3AR is normally glycosylated by N-linked oligosaccha-
rides. Mutation of these sites results in a decreased capacity for
the receptor to support ligand binding, indicating an essential
role for N-linked glycosylation in the function of human
5-HT3AR (34). Thus, our data suggest that like for other se-
rotonin receptor family members, N-linked glycosylation reg-
ulates the expression and function of the human 5-HT2Are-
These data also show that N-linked glycosylation of
5-HT2AR is important for receptor expression but not required
for JCV infection. Point mutation of individual N-linked gly-
cosylation sites in 5-HT2AR did not affect the function of the
receptor in JCV infection, indicating that JCV does not engage
a specific N-linked sialic acid expressed on 5-HT2AR. Treat-
ment of 5-HT2AR-expressing HEK293A cells with neuramin-
idase significantly reduced infection, indicating that the mech-
anism by which 5-HT2AR mediates JCV infection is via sialic
acid. These data further indicate that the sialic acid component
is likely expressed on an alternate coreceptor.
JCV has a very restricted host cell tropism, infecting cells in
the kidney (9, 12), B lymphocytes of the bone marrow (9), and
oligodendrocytes and astrocytes in the CNS (40, 48). Thus, it
seems plausible that JCV could utilize sialic acid as a low-
affinity binding step to adhere to 5-HT2AR, and then the virus
may interact with specific residues in 5-HT2AR to mediate
entry into host cells. It is also possible that JCV engages sialic
acid on another molecule through a low-affinity binding step
and then interacts with 5-HT2AR to mediate entry. A recent
study reports that JCV isolates from PML patients contain
polymorphisms in sialic acid binding domains in the viral cap-
sid protein VP1 (42), suggesting that the mechanism by which
JCV engages host cell receptors in the brain may differ from
that in the kidney or tissue culture systems. Given that JCV
reactivation occurs during immunosuppression due to HIV/
AIDS or in patients receiving immunosuppressive therapies, it
is possible that the virus engages host cell receptors in a dif-
ferent manner due to selective pressures in the host.
This study provides new information regarding the role of
5-HT2AR in JCV infection and the development of PML. As
treatment options for PML are very limited, the prognosis is
dismal (23). However, recent case studies have found that
treatment of PML patients with mirtazapine (a serotonin
5-HT2A, 5-HT2C, and 5-HT3receptor antagonist used for
treatment of depression) reduced the normally rapid progres-
sion of PML (6, 27, 45, 46). Treatment of PML patients with
mirtazapine was most effective when a patient was treated at
the early onset of symptoms; therefore, this may provide a
viable treatment option to extend the lives of PML patients (6).
These clinical studies suggest that treatment of PML patients
with mirtazapine, a serotonin receptor antagonist, may effec-
tively inhibit JCV binding to 5-HT2AR on oligodendrocytes
and thereby reduce JCV spread in the brain. This highlights
the importance of understanding the molecular interactions
between JCV and the 5-HT2Areceptor. Future studies will
provide a platform for continued pharmacological develop-
ment of 5-HT2Areceptor antagonists and inhibitors to treat
and prevent PML.
We thank members of the Atwood lab for critical discussions and
review of the manuscript. We thank Bethany O’Hara for technical
Work in our laboratory was supported by grants R01CA71878
(W.J.A.) and R01NS43097 (W.J.A.) and by Ruth L. Kirschstein Na-
tional Research Service Award F32NS064870 from the National Insti-
tute of Neurological Disorders and Stroke (M.S.M.). Confocal micros-
copy analysis was completed in the Leduc Bioimaging Facility at
Brown University. Immunoblot analysis was performed in the Center
for Genomics and Proteomics at Brown University, which is supported
by grant P20RR015578 (W.J.A.).
1. Astrom, K. E., E. L. Mancall, and E. P. Richardson, Jr. 1958. Progressive
multifocal leuko-encephalopathy; a hitherto unrecognized complication of
chronic lymphatic leukaemia and Hodgkin’s disease. Brain 81:93–111.
2. Backstrom, J. R., R. S. Westphal, H. Canton, and E. Sanders-Bush. 1995.
Identification of rat serotonin 5-HT2C receptors as glycoproteins containing
N-linked oligosaccharides. Brain Res. Mol. Brain Res. 33:311–318.
3. Bockaert, J., S. Claeysen, C. Becamel, A. Dumuis, and P. Marin. 2006.
Neuronal 5-HT metabotropic receptors: fine-tuning of their structure, sig-
naling, and roles in synaptic modulation. Cell Tissue Res. 326:553–572.
4. Bonhaus, D. W., C. Bach, A. DeSouza, F. H. Salazar, B. D. Matsuoka, P.
Zuppan, H. W. Chan, and R. M. Eglen. 1995. The pharmacology and distri-
bution of human 5-hydroxytryptamine2B (5-HT2B) receptor gene products:
comparison with 5-HT2A and 5-HT2C receptors. Br. J. Pharmacol. 115:622–
5. Carson, K. R., A. M. Evens, E. A. Richey, T. M. Habermann, D. Focosi, J. F.
Seymour, J. Laubach, S. D. Bawn, L. I. Gordon, J. N. Winter, R. R. Furman,
J. M. Vose, A. D. Zelenetz, R. Mamtani, D. W. Raisch, G. W. Dorshimer,
S. T. Rosen, K. Muro, N. R. Gottardi-Littell, R. L. Talley, O. Sartor, D.
Green, E. O. Major, and C. L. Bennett. 2009. Progressive multifocal leuko-
encephalopathy after rituximab therapy in HIV-negative patients: a report of
57 cases from the Research on Adverse Drug Events and Reports project.
6. Cettomai, D., and J. C. McArthur. 2009. Mirtazapine use in human immu-
nodeficiency virus-infected patients with progressive multifocal leukoen-
cephalopathy. Arch. Neurol. 66:255–258.
7. Chapagain, M. L., L. Sumibcay, U. Gurjav, P. H. Kaufusi, R. E. Kast, and
V. R. Nerurkar. 2008. Serotonin receptor 2A blocker (risperidone) has no
effect on human polyomavirus JC infection of primary human fetal glial cells.
J. Neurovirol. 14:448–454.
8. Chapagain, M. L., S. Verma, F. Mercier, R. Yanagihara, and V. R. Nerurkar.
2007. Polyomavirus JC infects human brain microvascular endothelial cells
independent of serotonin receptor 2A. Virology 364:55–63.
9. Chesters, P. M., J. Heritage, and D. J. McCance. 1983. Persistence of DNA
sequences of BK virus and JC virus in normal human tissues and in diseased
tissues. J. Infect. Dis. 147:676–684.
10. Cinque, P., I. J. Koralnik, S. Gerevini, J. M. Miro, and R. W. Price. 2009.
Progressive multifocal leukoencephalopathy in HIV-1 infection. Lancet In-
fect. Dis. 9:625–636.
VOL. 84, 2010 5-HT2ARECEPTOR GLYCOSYLATION IN JCV INFECTION 9683
11. Dorries, K. 1998. Molecular biology and pathogenesis of human polyoma- Download full-text
virus infections. Dev. Biol. Stand. 94:71–79.
12. Dorries, K., and V. ter Meulen. 1983. Progressive multifocal leucoencepha-
lopathy: detection of papovavirus JC in kidney tissue. J. Med. Virol. 11:307–
13. Dorries, K., E. Vogel, S. Gunther, and S. Czub. 1994. Infection of human
polyomaviruses JC and BK in peripheral blood leukocytes from immuno-
competent individuals. Virology 198:59–70.
14. Dubois, V., H. Dutronc, M. E. Lafon, V. Poinsot, J. L. Pellegrin, J. M.
Ragnaud, A. M. Ferrer, and H. J. Fleury. 1997. Latency and reactivation of
JC virus in peripheral blood of human immunodeficiency virus type 1-in-
fected patients. J. Clin. Microbiol. 35:2288–2292.
15. Dubois, V., M. E. Lafon, J. M. Ragnaud, J. L. Pellegrin, F. Damasio, C.
Baudouin, V. Michaud, and H. J. Fleury. 1996. Detection of JC virus DNA
in the peripheral blood leukocytes of HIV-infected patients. AIDS 10:353–
16. Dugan, A. S., M. L. Gasparovic, and W. J. Atwood. 2008. Direct correlation
between sialic acid binding and infection of cells by two human polyomavi-
ruses (JC virus and BK virus). J. Virol. 82:2560–2564.
17. Dutton, A. C., A. N. Massoura, T. J. Dover, N. A. Andrews, and N. M. Barnes.
2008. Identification and functional significance of N-glycosylation of the
5-ht5A receptor. Neurochem Int. 52:419–425.
18. Elphick, G. F., W. Querbes, J. A. Jordan, G. V. Gee, S. Eash, K. Manley, A.
Dugan, M. Stanifer, A. Bhatnagar, W. K. Kroeze, B. L. Roth, and W. J.
Atwood. 2004. The human polyomavirus, JCV, uses serotonin receptors to
infect cells. Science 306:1380–1383.
19. Fleischmann, R. M. 2009. Progressive multifocal leukoencephalopathy fol-
lowing rituximab treatment in a patient with rheumatoid arthritis. Arthritis
20. Houff, S. A., E. O. Major, D. A. Katz, C. V. Kufta, J. L. Sever, S. Pittaluga,
J. R. Roberts, J. Gitt, N. Saini, and W. Lux. 1988. Involvement of JC
virus-infected mononuclear cells from the bone marrow and spleen in the
pathogenesis of progressive multifocal leukoencephalopathy. N. Engl.
J. Med. 318:301–305.
21. Kean, J. M., S. Rao, M. Wang, and R. L. Garcea. 2009. Seroepidemiology of
human polyomaviruses. PLoS Pathog. 5:e1000363.
22. Khalili, K., and M. K. White. 2006. Human demyelinating disease and the
polyomavirus JCV. Mult. Scler. 12:133–142.
23. Kishida, S. 2007. Progressive multifocal leukoencephalopathy–epidemiol-
ogy, clinical pictures, diagnosis and therapy. Brain Nerve 59:125–137.
24. Kleinschmidt-DeMasters, B. K., and K. L. Tyler. 2005. Progressive multifo-
cal leukoencephalopathy complicating treatment with natalizumab and in-
terferon beta-1a for multiple sclerosis. N. Engl. J. Med. 353:369–374.
25. Knowles, W. A., P. Pipkin, N. Andrews, A. Vyse, P. Minor, D. W. Brown, and
E. Miller. 2003. Population-based study of antibody to the human polyoma-
viruses BKV and JCV and the simian polyomavirus SV40. J. Med. Virol.
26. Langer-Gould, A., S. W. Atlas, A. J. Green, A. W. Bollen, and D. Pelletier.
2005. Progressive multifocal leukoencephalopathy in a patient treated with
natalizumab. N. Engl. J. Med. 353:375–381.
27. Lanzafame, M., S. Ferrari, E. Lattuada, F. Corsini, R. Deganello, S. Vento,
and E. Concia. 2009. Mirtazapine in an HIV-1 infected patient with pro-
gressive multifocal leukoencephalopathy. Infez. Med. 17:35–37.
28. Linda, H., A. von Heijne, E. O. Major, C. Ryschkewitsch, J. Berg, T. Olsson,
and C. Martin. 2009. Progressive multifocal leukoencephalopathy after na-
talizumab monotherapy. N. Engl. J. Med. 361:1081–1087.
29. Liu, C. K., and W. J. Atwood. 2001. Propagation and assay of the JC virus.
Methods Mol. Biol. 165:9–17.
30. Liu, C. K., A. P. Hope, and W. J. Atwood. 1998. The human polyomavirus,
JCV, does not share receptor specificity with SV40 on human glial cells.
J. Neurovirol. 4:49–58.
31. Liu, C. K., G. Wei, and W. J. Atwood. 1998. Infection of glial cells by the
human polyomavirus JC is mediated by an N-linked glycoprotein containing
terminal alpha(2–6)-linked sialic acids. J. Virol. 72:4643–4649.
32. Major, E. O., K. Amemiya, C. S. Tornatore, S. A. Houff, and J. R. Berger.
1992. Pathogenesis and molecular biology of progressive multifocal leukoen-
cephalopathy, the JC virus-induced demyelinating disease of the human
brain. Clin. Microbiol. Rev. 5:49–73.
33. Molloy, E. S., and L. H. Calabrese. 2008. Progressive multifocal leukoen-
cephalopathy in patients with rheumatic diseases: are patients with systemic
lupus erythematosus at particular risk? Autoimmun Rev. 8:144–146.
34. Monk, S. A., J. M. Williams, A. G. Hope, and N. M. Barnes. 2004.
5-hydroxytryptamine3A receptor subunit. Biochem. Pharmacol. 68:1787–
35. O’Hara, B. A., and W. J. Atwood. 2008. Interferon beta1-a and selective
anti-5HT(2a) receptor antagonists inhibit infection of human glial cells by JC
virus. Virus Res. 132:97–103.
36. Quirk, P. L., S. Rao, B. L. Roth, and R. E. Siegel. 2004. Three putative
N-glycosylation sites within the murine 5-HT3A receptor sequence affect
plasma membrane targeting, ligand binding, and calcium influx in heterolo-
gous mammalian cells. J. Neurosci. Res. 77:498–506.
37. Rahmlow, M., E. A. Shuster, J. Dominik, H. G. Deen, Jr., D. W. Dickson,
A. J. Aksamit, Jr., H. A. Robles, and W. D. Freeman. 2008. Leflunomide-
associated progressive multifocal leukoencephalopathy. Arch. Neurol. 65:
38. Schaumburg, C., B. A. O’Hara, T. E. Lane, and W. J. Atwood. 2008. Human
embryonic stem cell-derived oligodendrocyte progenitor cells express the
serotonin receptor and are susceptible to JC virus infection. J. Virol. 82:
39. Shah, K. V., B. N. Fields, D. M. Knipe, and P. M. Howley. 1996. Fields
virology, 3rd ed., vol. 2. Lippincott-Raven Publishers, Philadelphia, Pa.
40. Silverman, L., and L. J. Rubinstein. 1965. Electron microscopic observations
on a case of progressive multifocal leukoencephalopathy. Acta Neuropathol.
41. Straumann, N., A. Wind, T. Leuenberger, and T. Wallimann. 2006. Effects of
N-linked glycosylation on the creatine transporter. Biochem. J. 393:459–469.
42. Sunyaev, S. R., A. Lugovskoy, K. Simon, and L. Gorelik. 2009. Adaptive
mutations in the JC virus protein capsid are associated with progressive
multifocal leukoencephalopathy (PML). PLoS Genet. 5:e1000368.
43. Trombetta, E. S. 2003. The contribution of N-glycans and their processing in
the endoplasmic reticulum to glycoprotein biosynthesis. Glycobiology 13:
44. Van Assche, G., M. Van Ranst, R. Sciot, B. Dubois, S. Vermeire, M. Noman,
J. Verbeeck, K. Geboes, W. Robberecht, and P. Rutgeerts. 2005. Progressive
multifocal leukoencephalopathy after natalizumab therapy for Crohn’s dis-
ease. N. Engl. J. Med. 353:362–368.
45. Verma, S., K. Cikurel, I. J. Koralnik, S. Morgello, C. Cunningham-Rundles,
Z. R. Weinstein, C. Bergmann, and D. M. Simpson. 2007. Mirtazapine in
progressive multifocal leukoencephalopathy associated with polycythemia
vera. J. Infect. Dis. 196:709–711.
46. Vulliemoz, S., F. Lurati-Ruiz, F. X. Borruat, J. Delavelle, I. J. Koralnik, T.
Kuntzer, J. Bogousslavsky, F. Picard, T. Landis, and R. A. Du Pasquier.
2006. Favourable outcome of progressive multifocal leucoencephalopathy in
two patients with dermatomyositis. J. Neurol. Neurosurg. Psychiatry 77:
47. Wenning, W., A. Haghikia, J. Laubenberger, D. B. Clifford, P. F. Behrens, A.
Chan, and R. Gold. 2009. Treatment of progressive multifocal leukoenceph-
alopathy associated with natalizumab. N. Engl. J. Med. 361:1075–1080.
48. Zurhein, G., and S. M. Chou. 1965. Particles resembling papova viruses in
human cerebral demyelinating disease. Science 148:1477–1479.
of N-glycosylation ofthe human
9684MAGINNIS ET AL.J. VIROL.