Download Efficient infection of buffalo rat liver

Journal of General Virology (2009), 90, 187–196
DOI 10.1099/vir.0.004655-0
Efficient infection of buffalo rat liver-resistant cells
by encephalomyocarditis virus requires binding to
cell surface sialic acids
Monique Guy,1 Stefan Chilmonczyk,2 Catherine Crucière,1 Marc Eloit1
and Labib Bakkali-Kassimi1
1
Correspondence
UMR 1161 INRA, AFSSA, ENVA, Ecole Nationale Vétérinaire, 7 Avenue Général de Gaulle, 94704
Maisons-Alfort Cedex, France
Labib Bakkali-Kassimi
[email protected]
2
Unité de Virologie et Immunologie Moléculaires, INRA, 78350 Jouy-en-Josas Cedex, France
Received 5 June 2008
Accepted 9 September 2008
In contrast to the production of virus and cell lysis seen in baby hamster kidney cells (BHK-21)
infected with the strain 1086C of encephalomyocarditis virus (EMCV), in buffalo rat liver cells
(BRL) neither virus replication nor cytopathic effects were observed. After 29 passages in BRL
cells, each alternating with boosts of the recovered virus in BHK-21 cells, the virus acquired the
ability to replicate effectively in BRL cells, attaining virus titres comparable to those in BHK-21
cells and producing complete cell destruction. The binding of virus on BRL cells was increased
after adaptation and was similar to that observed on BHK-21 cells. Treatment of BRL cells with
sialidase resulted in an 87 % reduction in virus binding and inhibition of infection. Sequence
analyses revealed three mutations in the VP1 amino acid sequence of the adapted virus at
positions 49 (LysAGlu), 142 (LeuAPhe) and 180 (IleAAla). The residue 49 is exposed at the
surface of the capsid and is known to be part of a neutralization epitope. These results suggest
that the adaptation of EMCV to BRL cells may have occurred through a mutation in a neutralizing
site that confers to the virus a capacity to interact with cell surface sialic acid residues. Taken
together, these data suggest a link between virus neutralization site, receptor binding and cell
permissivity to infection.
INTRODUCTION
Encephalomyocarditis virus (EMCV) belongs to the genus
Cardiovirus within the family Picornaviridae that has a
worldwide distribution (Stanway et al. 2005). The virus was
first isolated from non-human primates and it is now
recognized as a cause of mortality in many host species.
Rodents are considered to be natural hosts of EMCV and
thought to be the primary reservoir and disseminators of
the virus (Acland & Littlejohns, 1975; Koenen et al. 1999;
Seaman et al. 1986).
The primary route of transmission for EMCV is believed to
be faecal–oral, through ingestion of food or water
contaminated with virus (Zimmerman et al. 1993).
Transplacental vertical transmission has been reported to
occur naturally in swine and baboons (Hubbard et al. 1992;
Kim et al. 1989). Whether an infection remains asymptomatic or induces clinical signs depends on virus factors,
including viral dose, virus strain and host factors including
species, age and immune status of the infected animal.
EMCV infection causes acute myocarditis and sudden
The GenBank/EMBL/DDBJ accession number for the sequence
reported in this study is DQ835185.
004655 G 2009 SGM
death in piglets, and reproductive failure in sows. Infection
of older pigs is asymptomatic. Although generally asymptomatic in rodents, EMCV infection can cause insulindependent diabetes mellitus or nervous disorders in mice
(Cerutis et al. 1989). The virus has been isolated from
various organs including heart, tonsils, kidney, lungs, liver,
spleen, small intestine and mesenteric lymph nodes, from
naturally or experimentally infected animals (Billinis et al.
2004; Dea et al. 1991; Maurice et al. 2002; Papaioannou
et al. 2003; Spyrou et al. 2004). Many isolates have been
reported to be antigenically indistinguishable (Kim et al.
1991). However, the pathogenicity of EMCV in various
animal species was found to be strain-dependent (Kim et al.
1989). The mechanisms involved in EMCV virulence and
pathogenicity are unknown.
The first step in any infection process is the attachment of
the virus to a cell membrane molecule representing the
cellular receptor or coreceptor for the virus. This virus–cell
interaction and the distribution of a virus receptor are
considered to be major determinants of viral host range,
tissue tropism and pathogenesis. Vascular cell adhesion
molecule 1 has been identified as a receptor for the D
variant of EMCV on murine vascular endothelial cells
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Printed in Great Britain
187
M. Guy and others
(Huber, 1994). Jin et al. (1994) identified a cell surface
70 kDa sialoglycoprotein as an attachment molecule for
EMCV on two permissive human cell lines, HeLa and
K562. EMCV is believed to bind to non-permissive human
erythrocytes by interaction with sialic acid residues of
glycophorin A, the major sialoglycoprotein of the erythrocyte surface membrane (Burness & Pardoe, 1983;
Tavakkol & Burness, 1990). Together, these data point to
a crucial role for sialic acid in binding of EMCV to cells.
Previous experiments have demonstrated that while EMCV
replicates in numerous animal and human cells, a panel of
rat cells were relatively resistant to EMCV infection (Kelly
et al., 1983; Su et al., 2003). In an effort to gain
understanding of the molecular mechanisms underlying
resistance to EMCV infection and expansion of the host
range, we examined the step(s) in the EMCV replication
cycle that were blocked in buffalo rat liver-resistant cells.
We analysed the virus changes that led to the adaptation of
these cells by comparing the genomic sequences between
the parental and the adapted viruses. We demonstrated
that the block in EMCV replication in these cells occurs at
the level of virus attachment. Adaptation of EMCV to
produce a rapid cytopathic effect (CPE) in BRL (buffalo rat
liver) cells correlated with an increase in virus binding to
cell surface sialic acid residues and introduction of amino
acid mutations in the VP1 capsid protein. One of these
mutations occurred at an amino acid residue exposed at
the surface of the viral capsid and known to be part of
neutralizing epitope, suggesting a close relationship
between a receptor-binding site and a virus neutralization
epitope.
METHODS
Cells and virus. BHK-21 cells (baby hamster kidney) were
maintained in Eagle’s minimum essential medium with Earle’s salts
supplemented with 1 mM L-glutamine (Gibco-BRL), 1 % sodium
pyruvate (100 mM; Gibco-BRL), 1 % non-essential amino acids
(1006; Gibco-BRL), 100 U penicillin (Gibco-BRL) ml21, 100 mg
streptomycin (Gibco-BRL) ml21 and 10 % fetal calf serum (FCS;
Gibco-BRL).
BRL cells were maintained in Dulbecco’s modified Eagle’s medium
supplemented with 10 % FCS, 4500 mg glucose ml21, 1 mM
Glutamax (Gibco-BRL), 100 U penicillin ml21 and 100 mg streptomycin ml21.
The EMCV 1086C strain was kindly provided by Dr Frank Koenen
(Department of Virology, Section of Epizootic Diseases, CODACERVA, Groeselenberg 99, B-1180 Ukkel, Belgium). The strain was
initially isolated from a rat (Dr Frank Koenen, personal communication) and was propagated in BHK-21 cells. The virus stock was
produced by infection of BHK-21 cells as described previously
(Hammoumi et al., 2006).
Adaptation of 1086C strain to BRL cells. The EMCV 1086C strain
was propagated alternately in BHK-21 and BRL cell monolayers
grown in 25 cm2 tissue culture flasks until a variant virus population
promoting CPE on BRL cells had emerged. After each cell passage,
infected cells were subjected to three freeze–thaw cycles and clarified
by centrifugation for 15 min at 2500 g at 4 uC. The BRL cells were
188
infected at an m.o.i. of 100 TCID50 per cell and incubated for 48 h
and the BHK-21 cells were inoculated with 0.5 ml clarified suspension
of infected BRL cells and incubated for 24 h. At each passage, the
virus titre was determined on BHK-21 cells. The adapted virus was
then propagated on BRL cells and subjected to 15 cell passages. The
recovered virus was named 1086C Ad.
RNA transfection. Total RNA was extracted from the culture
supernatant of infected and uninfected BHK-21 cells using the
RNeasy Mini kit (Qiagen) according to the manufacturer’s
recommendation. Viral RNA was used to transfect BHK-21 or
BRL cell monolayers grown in 24-well plates by using TransFast
reagent (Promega). The transfection was performed according to
the manufacturer’s instructions, by using 2 vols TransFast for 1 mg
RNA.
Plaque formation assay. Cells were seeded in 35 mm dishes
(1.56105 cells per well) for 24 h at 37 uC. Medium was then
removed, 0.5 ml 10-fold serial dilutions of virus suspension was
added and the cells were incubated for 1 h at 37 uC. Subsequently,
supernatant was removed and 2 ml growth medium containing 1.5 %
low-melting point agarose was added. The dishes were incubated at
37 uC for 3 days and stained with crystal violet solution containing
4 % paraformaldehyde. After removal of the agarose plug, plates were
washed with water. Virus concentrations were expressed as a number
of p.f.u. ml21.
Indirect immunofluorescence assay. Cells were seeded at 1–
26104 cells ml21 in 96-well plates and infected with virus for l h at
37 uC. Monolayers were then incubated with growth medium for 5 h
at 37 uC. Growth medium was then removed and monolayers were
washed three times with PBS before fixation with 90 % cold acetone
for 20 min at 220 uC. Monolayers were then washed three times with
PBS and incubated with anti-EMCV monoclonal antibody (mAb) for
1 h at 37 uC. After three washes with PBS, fluorescein isothiocyanate
(FITC)-conjugated goat anti-mouse IgG (Biosys) was added and the
cells were incubated for 30 min at 37 uC. Monolayers were then
washed three times with PBS and once with distilled water. Cells were
observed by epifluorescence by using an Olympus IX-71 fluorescence
microscope equipped with an FITC filter.
Virus-binding assay using flow cytometry. Several anti-EMCV
mAbs were produced as described previously (Bakkali et al., 1994)
and tested by flow cytometry for their reactivity with the 1086C strain
adsorbed in BHK-21 cells. The neutralizing mAb (mAb 16E4) that
gave the best signal was selected.
Cells were maintained on ice during the assay in order to inhibit virus
uncoating and all solutions were chilled. Cells were seeded in 24-well
plates (56105 cells per well), washed three times with PBS and
incubated with virus at an m.o.i. of 104 p.f.u. per cell for 1 h under
agitation. Unbound virus was washed by rinsing three times with PBS
and cells were incubated with anti-EMCV 16E4 mAb for 30 min
under agitation. After washing twice with PBS and once with PBS
containing 0.05 % Tween 20, cells were incubated with goat antimouse IgG–FITC conjugate for 30 min under agitation. Cells were
rinsed twice with PBS containing 0.05 % Tween 20 and once with PBS
and then fixed with 1 % paraformaldehyde. Samples were analysed by
a FACScan flow cytometer using CellQuestPro software (Becton
Dickinson). Each experiment was done in duplicate. For each sample
16104 individual cells were recorded using a dot-plot combination of
low-angle forward scattered and right-angle scattered laser light.
Mean fluorescence intensity of virus binding was determined by
subtracting the mean fluorescence intensity of the uninfected cells
incubated with mAb 16E4 and secondary antibody from the mean
fluorescence intensity of infected cells.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Journal of General Virology 90
Adaptation of EMCV to BRL cells
Effect of sialidase treatment on virus infection and attachment.
BRL cell monolayers grown in 24-well plates at 56105 cells per well
were washed with serum-free medium and then incubated for 1 h at
37 uC with Vibrio cholerae sialidase (Sigma) at a final concentration of
25 mU ml21 in PBS pH 5.4. Untreated cells were incubated in the
same conditions with PBS pH 5.4 alone. After three washing steps,
cells were infected for 1 h at 37 uC at an m.o.i. of 2.5 p.f.u. per cell.
Cells were washed again three times and complete medium was
added. CPE was monitored by light microscopy after 24 h of
incubation at 37 uC. The effect of sialidase on virus infection was also
monitored in a plaque reduction assay. In this experiment, a plaque
formation assay was performed as described above on sialidasetreated cells and for comparison on untreated cells.
The effect of sialidase on virus attachment was analysed by
performing virus-binding assays on cells treated with sialidase, as
described above. The results are expressed as the percentage of
binding on sialidase-treated cells relative to binding on untreated
cells.
Sequencing. RNA was extracted from BHK-21 cells infected with the
1086C strain or from BRL cells infected with the adapted virus by
using RNeasy Mini kit (Qiagen) according to the manufacturer’s
instructions. RNA was used for amplification of EMCV genomic
fragments using one-step RT-PCR kit (Qiagen) and EMCV-specific
primers (data not shown). Amplified DNA fragments were purified
and both strands were sequenced directly by using the dye chain
termination method. Nucleotide sequences were deduced from two
independent PCR fragments and analysed on an IBM compatible
personal computer by using program Lasergene99 (DNASTAR). The
nucleotide sequence was submitted to the GenBank database and
assigned the accession number DQ835185.
RESULTS
BRL cells are resistant to EMCV infection
Infection of BHK-21 cells with the 1086C strain at an m.o.i.
as low as 1 p.f.u. per cell resulted in complete destruction
of cell monolayers within 24 h. In contrast, no CPE was
observed 72 h post-infection when the virus was inoculated
into BRL cells, even at a high m.o.i. In a plaque formation
assay, the virus produced large plaques on BHK-21 cells,
whereas no plaques were observed on BRL cells (Fig. 1a).
To determine whether BRL cells supported virus growth
without CPE, BHK-21 and BRL cells were infected with
EMCV at an m.o.i. of 10. The cells were harvested every
hour during a 9 h period and the virus titre was
determined by end-point dilution on permissive BHK-21
cells. Virus production in BHK-21 cells was observed as
early as 4 h after infection and maximum titre was reached
by 8 h post-infection (Fig. 1b). In the BRL cells, the virus
titre remained relatively stable after 9 h of incubation, most
likely corresponding to the input virus. Furthermore,
examination of cells by immunofluorescence following
virus inoculation (m.o.i. of 100) revealed that in infected
BHK-21 cells viral antigens were detected in .90 % of cells,
whereas in BRL cells only a few cells (,1 %) expressed viral
antigens (Fig. 1c). These results indicate that BRL cells are
resistant to infection by the EMCV 1086C strain.
Restriction in BRL cells infection occurred at an
early stage of viral cycle
The block in EMCV infection in BRL cells may occur at
one or several stages of viral cycle. The first step of EMCV
infection consists of the attachment of virus particles to a
cell surface receptor or coreceptor and internalization of
viral RNA. In order to determine whether this stage of
virus cycle is a limiting step in infection of BRL cells, cells
were transfected with viral RNA and then monitored for
the production of viral antigens, the production of
infectious particles and CPE. In both BHK-21- and BRLtransfected cells, viral antigens were detected by immuno-
Fig. 1. Resistance of BRL cells to EMCV
infection. (a) Plaque formation assay. Cells
were inoculated with 10-fold serial dilutions of
1086C virus and incubated for 72 h at 37 6C.
The picture illustrates the plaque produced on
BHK-21 cells with the virus suspension diluted
at 10”8 and on BRL cells with undiluted virus.
(b) Virus growth kinetics of 1086C strain in
BHK-21 and BRL cells. The mean±SD of three
experiments carried out in parallel is shown. (c)
Production of viral antigen in BHK-21 and BRL
cells at 5 h after infection with 1086C strain
(m.o.i. of 100). Antigen was detected by
indirect immunofluorescence assay using
anti-EMCV mAb 16E4 and FITC-conjugated
anti-mouse IgG (IF, UV light microscope;
B&W, visible light microscope).
http://vir.sgmjournals.org
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
189
M. Guy and others
fluorescence as early as 5 h post-transfection (Fig. 2a).
Analysis of infectious virus production in transfected BHK21 and BRL cells revealed comparable kinetics of virus
production between the cell types (Fig. 2b). The kinetic
curves represent a single replicative cycle since more than
90 % of the cells were infected at 5 h post-transfection (Fig.
2a). Infectious virus was produced as early as 2 h after
transfection and maximum titre was reached by 8 h after
transfection. At 24 h after transfection, similar quantities of
infectious virus were detected in cell culture supernatant of
both cells, indicating that virus particles are released from
BRL cells. However, no plaque formation was observed in
transfected BRL cells grown under agar, whereas in
transfected BHK-21 cells the progeny virus produced large
plaques (data not shown).
Taken together, these results indicate that EMCV genomic
RNA is able to replicate in BRL cells and to produce
infectious virus, but then the progeny virus is unable to
spread from cell to cell. Thus, EMCV infection of BRL cells
is blocked at an early stage of virus cycle, that is, virus
attachment and/or internalization.
In order to localize the block in BRL infection we
compared the binding efficiency of EMCV to BHK-21
and BRL cells. As shown in Fig. 2(c), EMCV bound 10
times more efficiently to BHK-21 cells than to BRL cells.
This result suggests that the resistance of BRL cells to
EMCV infection is likely due to a low attachment of virus
particles to cell surface molecules.
EMCV adaptation to BRL cells
To gain a better understanding of the mechanism(s)
involved in the block of EMCV infection, we attempted to
adapt the 1086C strain to BRL cells, in order to study the
viral modifications that led to this adaptation. First, assays
of adaptation by successive passages of EMCV in BRL cells
were unsuccessful. Virus titre decreased after each passage
and infectious virus was no longer recovered from
inoculated cells after the fourth passage. To circumvent
this complication, the 1086C strain was repeatedly
passaged in BRL cells with alternate passages in BHK-21
cells to boost viral yield. To monitor virus adaptation, the
virus titre was determined after each cell passage. As shown
in Fig. 3(a), the titre of infectious virus produced in BRL
cells increased with passages and approached that obtained
in BHK-21 cells. At the fourth alternate passage, CPE
became evident in infected BRL cells 48 h after inoculation.
Nevertheless, the produced virus was unable to produce
CPE in BRL cells when it was inoculated directly into these
cells. After 29 alternate passages, the virus was successfully
propagated 15 times in BRL cells and, at each cell passage,
complete destruction of cells was observed 24 h after
inoculation. Plaque formation assays showed that the
adapted virus produced plaques in infected BRL cells,
demonstrating its ability to spread from cell to cell (Fig.
3b). Furthermore, the growth kinetics of the adapted virus
in BRL cells and in BHK-21 cells, infected at an m.o.i. of
10, were similar (Fig. 3c). Virus production was detected
Fig. 2. (a) Production of viral antigen in BHK21 and BRL cells at 5 h after transfection with
1 mg RNA (specific infectivity 1.6¾104 p.f.u.
mg”1) extracted from the culture supernatant of
BHK-21 cells infected with the 1086C strain.
Antigen was detected by immunofluorescence
microscopy with anti-EMCV mAb 16E4 and
FITC-conjugated anti-mouse IgG. Original
magnification ¾20. (b) Virus growth kinetics
of 1086C strain in BHK-21- and BRL-transfected cells with viral RNA. The mean±SD of
three experiments carried out in parallel is
shown. (c) 1086C binding to BHK-21 and
BRL cells. Virus binding was analysed by flow
cytometry using a concentrated virus suspension (m.o.i. of 104 p.f.u. per cell). Data are
given as mean fluorescence intensity of two
experiments carried out in parallel.
190
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Journal of General Virology 90
Adaptation of EMCV to BRL cells
Fig. 3. Adaptation of 1086C to BRL cells. (a) Virus was propagated repeatedly in BRL cells with alternate passages in BHK-21
cells. The amount of infectious virus produced after each passage was determined. The plot indicates the titre of the virus
collected from BRL cells (m) at the indicated passage. The plot indicates the titre of virus harvested after inoculation of BHK-21
cells (&) with the virus collected from the BRL cells at the indicated passage. (b) Plaque production by the adapted virus
(1086C Ad) in BHK-21 and BRL cells. The picture illustrates the plaque produced on BRL cells inoculated with the virus
suspension diluted at 10”7 and in BHK-21 cells inoculated with the virus suspension diluted at 10”9. (c) Virus growth kinetics of
the adapted virus in BHK-21 and BRL cells. The mean±SD of three experiments carried out in parallel is shown. (d) Production
of viral antigen in BHK-21 and BRL cells at 5 h after infection with the adapted virus (m.o.i. of 100). Antigen was detected by
immunofluorescence microscopy with anti-EMCV mAb 16E4 and FITC-conjugated anti-mouse IgG. Original magnification ¾20
after 4 h of infection and virus titre showed a 3 log increase
to reach a maximum titre by 8 h. At 5 h after inoculation,
viral antigen was detected by immunofluorescence in
.90 % of inoculated BRL cells (Fig. 3d). These results
demonstrate the ability of the adapted virus to infect BRL
cells and to replicate in these cells.
Adaptation correlated with an increase in virus
binding to sialic acid residues
RNA transfection experiments and cell-binding assays
suggested that restriction in BRL cell infection is likely
due to a low attachment of the virus to the cell surface.
Thus, we speculated whether the adaptation of 1086C to
BRL cells resulted in an increase in virus binding. As shown
in Fig. 4, the binding of the adapted virus to BRL cells was
similar to that observed in BHK-21 cells. Furthermore,
http://vir.sgmjournals.org
compared with the initial binding of the parental 1086C
strain, the attachment of the adapted virus to BRL cells was
10-fold higher than the initial binding. This result indicates
that the adaptation of 1086C strain to BRL cells correlated
with enhanced virus binding to cell surface molecules.
Sialic acid residues have been reported to be involved in
EMCV binding to mammalian cells. To determine whether
the adapted virus uses sialic acid for the attachment to and
infection of BRL cells, cells were digested with V. cholerae
sialidase, which hydrolyses both a(2-3) and a(2-6) linkages
of sialic acid, before binding and infection assays. Sialidase
treatment resulted in 87 % reduction in virus binding (Fig.
5a and b) and inhibited virus plaque formation (Fig. 5c)
and CPE production (Fig. 5d). These results indicate that
BRL cell surface sialic acid residues are required for the
attachment and infection of the adapted virus. Binding
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
191
M. Guy and others
exposed and are located in the bE strand and the bF–bG
loop, respectively.
DISCUSSION
Fig. 4. Comparison of 1086C and 1086C Ad virus binding to
BHK-21 and BRL cells. Cells (5¾105) were incubated with the
indicated virus (m.o.i. of 104 p.f.u. per cell) at 4 6C. Bound virus
was detected by a FACScan flow cytometer after incubation with
anti-EMCV mAb 16E4 and goat anti-mouse IgG–FITC and plotted
as mean fluorescence intensity. Values represent means from two
experiments carried out in parallel.
inhibition assays using BHK-21 cells revealed that the
attachment of the parental 1086C strain to BHK-21 cells
was not inhibited by sialidase treatment, whereas binding
of the adapted virus was reduced by more than 75 % (Fig.
5a and b). This indicates that the parental virus did not
bind to sialic acids, suggesting that the capacity of the
adapted virus to bind sialic acid resulted from virus
adaptation.
Characterization of the genomic mutations
selected during the adaptation of EMCV to BRLresistant cells
In order to identify the genomic modifications that led to
the adaptation of 1086C strain to BRL cells, we determined
and compared the complete coding sequence of the virus
before and after adaptation. Nucleotide sequence comparison revealed nine mutations in the adapted virus sequence:
one in VP2 (T123C), two in VP3 (C39T and C168T), four
in VP1 (A145C, G426T, A538G and T539C) and two in the
3D protein (C531T and T552C). Only mutations in VP1
led to an amino acid substitution. Thus, lysine 49 was
replaced by glutamine, leucine 142 by phenylalanine and
isoleucine 180 by alanine. None of these mutations are
located in the ‘pit’: the presumed receptor-binding site.
However, amino acid Lys49 is located on the outer surface
of the capsid in the loop connecting bB and bC strands
(Fig. 6). In contrast, residues 142 and 180 are not surface
192
Previous studies have demonstrated that while EMCV
replicates in numerous animal and human cells, a panel of
rat cells were relatively resistant to EMCV infection and
replication. Indeed, limited replication of virus was observed
in rat myocardial cells, rat C6 glioma cells and rat embryo
cells, while rat HTC hepatoma cells were completely
refractory to EMCV replication. None of the rat cell lines
developed cytopathic changes following virus inoculation
and only a small percentage of cells demonstrated viral
antigen 24 h after infection (Donta et al., 1986; Kelly et al.,
1983). Plagemann & Swim (1966) have also reported
resistance in other rat cells, Novikoff hepatoma cells
(N1S1), to Mengo virus, which is now considered to be a
strain of EMCV. In the present study we show that, in
contrast to BHK-21 cells, no CPE was observed in buffalo rat
liver cells (BRL) 72 h after inoculation with the 1086C strain
of EMCV. Furthermore, viral antigen and viral production
were not detected in the inoculated cells. These results
demonstrate the inability of EMCV to infect BRL cells. These
observations are confusing if we consider that rats are the
natural host and reservoir of EMCV and virus has been
isolated from different organs of experimentally infected
animals. One explanation of this cell resistance may be an
alteration in cellular receptor that may occur when cells
adapt to growth in established tissue culture. The finding
that transfection of BRL cells with viral RNA resulted in
production of viral antigen and release of infectious progeny
virus indicated that the restriction of EMCV replication in
BRL cells depends only on its inability to penetrate into the
cell. Thus, the block in virus infection of these cells occurs at
an early stage of virus replication, namely, attachment to cell
surface and/or uncoating. Analysis of binding by flow
cytometry showed that the 1086C strain exhibited low
binding to BRL cells compared with the permissive BHK-21
cells. This finding suggests that the low binding likely
contributes to resistance of BRL cells. However, we cannot
rule out the fact that the uncoating step may also be a
limiting step in BRL cell infection. Indeed, after comparing
the susceptibility of murine and rat cells to the M strain of
EMCV, Donta et al. (1986) have reported that the block in
infection of rat glial cells (C6) probably occurs at the level of
uncoating. But in this case, the binding of the M strain was
similar in murine and rat cell lines.
To characterize the mechanism(s) responsible for restriction of the EMCV growth in BRL cells more fully, we tried
to obtain a host range mutant adapted to promote a rapid
and complete CPE in these cells. The first attempts to adapt
the 1086C strain by successive blind passages in BRL cells
were unsuccessful. We observed a diminution in the
quantity of virus recovered from infected BRL cells after
each cell passage and the virus was lost after the fourth
passage (data not shown). A similar observation has been
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Journal of General Virology 90
Adaptation of EMCV to BRL cells
Fig. 5. Inhibition of 1086C Ad binding (a, b) and infection (c, d) by pretreatment of BRL cells with sialidase. Cells (5¾105) were
pretreated with sialidase at 25 mU ml”1 before binding or infection assays. Binding was analysed by flow cytometry and the
results are shown as mean percentage of bound virus of two experiments carried out in parallel (a) or as a histogram of cellassociated fluorescence (b). (c) Plaque inhibition assays on BRL cells pretreated with sialidase. (d) Inhibition of 1086C Ad
infection by sialidase treatment. V. cholerae sialidase-treated (right panels) and untreated (left panels) BRL cell cultures were
infected at an m.o.i. of 2.5 p.f.u. per cell. The picture illustrates CPE on the culture observed at 24 h post-infection by light
microscopy. Original magnification ¾20.
reported by Kelly et al. (1983) concerning adaptation of the
M strain of EMCV to rat embryo cells. This failure in virus
adaptation is probably due to selection of a very low
quantity of viral particles during the first passage that is
diluted during the next passages. To overcome this
difficulty, we attempted the adaptation by successive
passages of the 1086C strain in BRL cells each alternating
with a passage in BHK-21 cells to amplify the recovered
virus. By using this method, an increase in the virus
quantity recovered from BRL cells was observed after each
passage, indicating progressive adaptation of EMCV to
these cells. After 29 alternate passages, the variant obtained
caused complete CPE in BRL cells 24 h after infection and
produced high levels of infectious virus. Furthermore, the
kinetics of virus production in BRL cells were similar to
those observed in BHK-21 cells, and the virus was able to
spread from cell to cell as shown by the plaque formation
assay. The adapted virus was propagated 15 times in BRL
cells without losing its infectivity, demonstrating its stability.
Binding experiments indicated that the attachment of the
1086C strain to BRL cells was increased after adaptation to
http://vir.sgmjournals.org
reach a level similar to that observed in BHK-21 cells. Since
the transfection experiments with viral RNA have shown
that the virus is able to replicate in BRL cells, this
observation strongly suggests that adaptation likely
occurred through mutation of the virus capsid, which
enhanced virus binding to a cell surface receptor. Indeed,
sequence analyses of the adapted virus revealed three
mutations in the amino acid sequence of the VP1 protein:
lysine 49 was replaced by glutamine, leucine 142 by
phenylalanine and isoleucine 180 by alanine. The threedimensional structure of the capsid of Mengo virus has
previously been determined by X-ray crystallography (Luo
et al., 1987). Since 95 % of the amino acid sequence of
EMCV and Mengo virus is identical, it seems reasonable to
predict that the three-dimensional structure of the capsid is
conserved between the two viruses. According to the threedimensional structure of the Mengo capsid, none of the
three mutations are located in the putative receptor site.
Mutation 142 is located in the bE strand and mutation 180
in the bF–bG loop. These 2 aa are located in the inner
surface of the capsid and are thus unlikely to be involved in
virus binding. However, mutation 49 is located in the loop
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
193
M. Guy and others
cells (Lys1049AGln) and the mutation observed after
attenuation of EMCV on BHK-21 cells (Asn1062AAsp)
(Denis & Koenen, 2003), which were obtained in the
absence of immune selection, are the same as those that
permit the virus to escape neutralization by neutralizing
mAbs (Kobasa et al., 1995). Taken together, these data
suggest that the region of EMCV that corresponds to
neutralization epitope 4 also contributes to the interaction
of virus with cell surface receptor and virus pathogenicity.
This overlap between receptor-binding site and antigenic
site has also been reported for other picornaviruses like
Theiler’s virus (Jnaoui & Michiels, 1998), foot-and-mouth
disease virus (Martinez et al., 1997) and poliovirus (Harber
et al., 1995). For example, mutations adapting the DA1
strain of Theiler’s virus to L929 cells have been reported to
modify two clusters of amino acids that are known to be
part of the neutralizing epitopes. In addition, this virus
adaptation was associated with attenuation of virulence for
the mouse, suggesting a close relationship between receptor
binding, virus neutralization and virus virulence (Jnaoui &
Michiels, 1998).
Fig. 6. Capsid mutations responsible for virus adaptation to BRL
cells. (a) Shown is a pentamer of the EMCV capsid (seen from
exterior) in a space-filling view generated with Swiss-PdBViewer
software. The capsid fivefold axis is indicated by an arrow. The
triangle shows the capsid unit (protomer) and the capsid viral
proteins VP1, VP2 and VP3. The residue 49 of VP1 is shown in
white on each protomer. The residues 142 and 180 of VP1 are
located in the interior of the protomer and are not visible in this
picture. (b) Position of Lys49, Leu142 and Ile180 on a ribbon
diagram of the VP1 protein.
connecting the bB and bC strands of VP1, which is surface
exposed and is thus accessible for binding to a cell receptor.
Interestingly, this amino acid was reported to be part of the
Mengo virus neutralization site number 4 (Kobasa et al.,
1995), suggesting that neutralizing antibodies recognize a
region in EMCV and Mengo virus that is involved in virus
attachment. Furthermore, it has been shown that a strain of
EMCV inducing myocarditis in piglets that has been
attenuated by 210 passages in BHK-21 cells displayed a
mutation in residue 62 of VP1 (Denis & Koenen, 2003).
This amino acid is also localized in the loop connecting the
bB and bC strands of VP1 and was reported to be part of
the Mengo virus neutralization site number 4 (Kobasa et
al., 1995). Nelsen-Salz et al. (1996) have shown that
diabetogenic or non-diabetogenic phenotype of EMCV
variants is likely due to a mutation in residue 63 of VP1,
which is also localized in the neutralization site number 4.
The mutation permitting adaptation of the EMCV to BRL
194
Previous studies have shown that cell surface sialic acid
residues are involved in the interaction of EMCV with
different mammalian cells (Burness & Pardoe, 1983; Jin et
al. 1994; Tavakkol & Burness, 1990). Our finding that the
binding to and the infection of BRL cells by the adapted
virus are inhibited after the treatment of cells with sialidase
also demonstrated that sialic acids play a role in EMCV
infection of these cells. This is in agreement with previously
reported results showing that EMCV binding to rat C6
glioma cells was markedly reduced following exposure of
cells to sialidase (Donta et al., 1986). The fact that the
binding of the 1086C parental virus to BHK-21 cells was
not reduced after sialidase treatment, whereas the binding
of the adapted virus was reduced by more than 75 %,
suggests that adaptation resulted in a selection of virus
particles with binding affinity to cell surface sialic acid
residues and that the parental virus binds to a different
molecule not affected by sialidase treatment. These
observations suggest that the mutation K49Q might lead
to the acquisition of a new receptor-binding region in the
capsid of the adapted virus. Changes in receptor specificity
during the course of passage in cell culture have been
documented for picornaviruses. Jnaoui et al. (2002) have
shown that adaptation of the Theiler’s GDVII virus to
CHO-K1 cells correlated with the appearance of sialic acid
binding. Similar to our study, this binding capacity was
found to result from the mutation of a single amino acid
(VP1, 51) located in VP1 CD loop. Although the parental
foot-and-mouth disease virus enters cells through an RGDdependent integrin, multiply passaged virus acquires the
capacity to bind heparin and to infect human K-562 cells,
which do not express integrin avb3 (Baranowski et al.,
2000). The virus binds to two separate receptors with two
separate regions of the capsid that both overlap with
neutralization epitopes. Enterovirus 70 variant derived by
passage in rhesus monkey kidney cells (EV70-Rmk)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Journal of General Virology 90
Adaptation of EMCV to BRL cells
replicates poorly in HeLa cells and utilizes a sialylated
molecule other than DAF as cell receptor. In contrast,
passage of this variant in HeLa cells leads to isolation of
virus (EV70-Dne) that is dependent upon DAF for cell
entry and does not replicate in rhesus monkey kidney cells
(Kim & Racaniello, 2007). Human rhinovirus type 89 is
adapted to use heparin sulfate for cell entry after passages
in non-permissive and ICAM-1 receptor-deficient cells
(Reischl et al., 2001, Vlasak et al., 2005). Similar to the
EMCV-adapted mutant, one consistent amino acid change
was found within the BC loop of VP1, which is involved in
the binding of minor-group viruses to the low-density
lipoprotein receptors. Other mutations were also found in
capsid regions that are not surface exposed. The authors
have hypothesized that the mutations in buried sites of the
capsid may contribute to a loss of virus stability at low pH
and at elevated temperature, which might be required for
efficient uncoating in the absence of ICAM. Given the
buried location of the mutations leucine 142 by phenylalanine and isoleucine 180 by alanine that we found in the
BRL cell-adapted virus, we can speculate by analogy to
HRV89 that these mutations may induce a reduction in
EMCV stability allowing for easier uncoating. Further
studies are needed to understand further the requirement
of these mutations to EMCV infection in BRL cells.
In conclusion, we adapted EMCV to replicate and to
produce CPE in buffalo rat liver-resistant cells. This
adaptation appears to result from an increase in virus
binding and the acquisition of the ability to bind cell
surface sialic acid residues. Adaptation is accompanied by
the accumulation of mutations in virus neutralizing site,
suggesting a connection between virus antigenic site, virus
receptor binding and cell susceptibility.
Cerutis, D. R., Bruner, R. H., Thomas, D. C. & Giron, D. J. (1989).
Tropism and histopathology of the D, B, K, and MM variants of
encephalomyocarditis virus. J Med Virol 29, 63–69.
Dea, S. A., Bilodeau, R. & Martineau, G. P. (1991). Isolation of
encephalomyocarditis virus among stillborn and post-weaning pigs in
Quebec. Arch Virol 117, 121–128.
Denis, P. & Koenen, F. (2003). Molecular analysis of the capsid coding
region of a virulent encephalomyocarditis virus isolate after serial cell
passages and assessment of its virulence. Arch Virol 148, 903–912.
Donta, S. T., Shanley, J. D. & Damiano-Burbach, P. (1986).
Encephalomyocarditis virus can bind to and transfect non-permissive
cells. Arch Virol 88, 301–307.
Hammoumi, S., Crucière, C., Guy, M., Boutrouille, A., Messiaen, S.,
Lecollinet, S. & Bakkali-Kassimi, L. (2006). Characterization of a
recombinant encephalomyocarditis virus expressing the enhanced
green fluorescence protein. Arch Virol 151, 1783–1796.
Harber, J., Bernhardt, G., Lu, H. H., Sgro, J. Y. & Wimmer, F. (1995).
Canyon rim residues, including antigenic determinants, modulate
serotype-specific binding of polioviruses to mutants of the poliovirus
receptor. Virology 214, 559–570.
Hubbard, G. B., Soike, K. F., Butler, T. M., Carey, K. D., Davis, H.,
Butcher, W. I. & Gauntt, C. J. (1992). An encephalomyocarditis virus
epizootic in a baboon colony. Lab Anim Sci 42, 233–239.
Huber, S. A. (1994). VCAM-1 is a receptor for encephalomyocarditis
virus on murine vascular endothelial cells. J Virol 68, 3453–3458.
Jin, Y. M., Pardoe, I. U., Burness, A. T. & Michalak, T. I. (1994).
Identification and characterization of the cell surface 70-kilodalton
sialoglycoprotein(s) as a candidate receptor for encephalomyocarditis
virus on human nucleated cells. J Virol 68, 7308–7319.
Jnaoui, K. & Michiels, T. (1998). Adaptation of Theiler’s virus to L929
cells: mutations in the putative receptor binding site on the capsid
map to neutralization sites and modulate viral persistence. Virology
244, 397–404.
Jnaoui, K., Minet, M. & Michiels, T. (2002). Mutations that affect the
tropism of DA and GDVII strains of Theiler’s virus in vitro influence
sialic acid binding and pathogenicity. J Virol 76, 8138–8147.
Kelly, P. M., Shanley, J. D. & Sears, J. (1983). Replication of
ACKNOWLEDGEMENTS
encephalomyocarditis virus in various mammalian cell types. J Med
Virol 11, 257–264.
We thank Jennifer Richardson for her valuable suggestion.
Kim, M. S. & Racaniello, V. R. (2007). Enterovirus 70 receptor
utilization is controlled by capsid residues that also regulate host
range and cytopathogenicity. J Virol 81, 8648–8655.
REFERENCES
Kim, H. S., Christianson, W. T. & Joo, H. S. (1989). Pathogenic
Acland, H. M. & Littlejohns, I. R. (1975). Encephalomyocarditis virus
properties of encephalomyocarditis virus isolates in swine fetuses.
Arch Virol 109, 51–57.
infection of pigs. 1. An outbreak in New South Wales. Aust Vet J 51,
409–415.
Bakkali, L., Guillou, R., Gonzague, M. & Crucière, C. (1994). A rapid
and sensitive chemiluminescence dot-immunobinding assay for
screening hybridoma supernatants. J Immunol Methods 170, 177–184.
Baranowski, E., Ruiz-Jarabo, C. M., Sevilla, N., Andreu, D., Beck, E. &
Domingo, E. (2000). Cell recognition by foot-and-mouth disease
virus that lacks the RGD integrin-binding motif: flexibility in
aphthovirus receptor usage. J Virol 74, 1641–1647.
Billinis, C., Leontides, L., Psychas, V., Spyrou, V., Kostoulas, P.,
Koenen, F. & Papadopoulos, O. (2004). Effect of challenge dose and
age in experimental infection of pigs with encephalomyocarditis virus.
Vet Microbiol 99, 187–195.
Burness, A. T. & Pardoe, I. U. (1983). A sialoglycopeptide from
human erythrocytes with receptor-like properties for encephalomyocarditis and influenza viruses. J Gen Virol 64, 1137–1148.
http://vir.sgmjournals.org
Kim, H. S., Christianson, W. T. & Joo, H. S. (1991). Characterization of
encephalomyocarditis virus isolated from aborted swine fetuses. Am J
Vet Res 52, 1649–1652.
Kobasa, D., Mulvey, M., Lee, J. S. & Scraba, D. G. (1995).
Characterization of Mengo virus neutralization epitopes. II.
Infection of mice with an attenuated virus. Virology 214, 118–127.
Koenen, F., Vanderhallen, H., Castryck, F. & Miry, C. (1999).
Epidemiologic, pathogenic and molecular analysis of recent encephalomyocarditis outbreaks in Belgium. Zentralbl Veterinarmed B 46,
217–231.
Luo, M., Vriend, G., Kamer, G., Minor, I., Arnold, F., Rossmann, M. G.,
Boege, U., Scraba, D. G., Duke, G. M. & Palmenberg, A. C. (1987).
The atomic structure of Mengo virus at 3.0 Å resolution. Science 235,
182–191.
Martinez, M. A., Verdaguer, N., Mateu, M. G. & Domingo, F. (1997).
Evolution subverting essentiality: dispensability of the cell attachment
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
195
M. Guy and others
Arg-Gly-Asp motif in multiply passaged foot-and-mouth disease
virus. Proc Natl Acad Sci U S A 94, 6798–6802.
Spyrou, V., Maurice, H., Billinis, C., Papanastassopoulou, M., Psalla, D.,
Nielen, M., Koenen, F. & Papadopoulos, O. (2004). Transmission and
Maurice, H., Nielen, M., Stegeman, J. A., Vanderhallen, H. & Koenen, F.
(2002). Transmission of encephalomyocarditis virus (EMCV) among
pathogenicity of encephalomyocarditis virus (EMCV) among rats. Vet
Res 35, 113–122.
pigs experimentally quantified. Vet Microbiol 88, 301–314.
Stanway, G., Brown, F., Christian, P., Hovi, T., Hyypiä, T., King, A. M. Q.,
Knowles, N. J., Lemon, S. M., Minor, P. D. & other authors (2005).
Nelsen-Salz, B., Zimmermann, A., Wickert, S., Arnold, G., Botta, A.,
Eggers, H. J. & Kruppenbacher, J. P. (1996). Analysis of sequence and
pathogenic properties of two variants of encephalomyocarditis virus
differing in a single amino acid in VP1. Virus Res 41, 109–122.
Papaioannou, N., Billinis, C., Psychas, V., Papadopoulos, O. &
Vlemmas, I. (2003). Pathogenesis of encephalomyocarditis virus
(EMCV) infection in piglets during the viraemia phase: a histopathological, immunohistochemical and virological study. J Comp Pathol
129, 161–168.
Family Picornaviridae. In Virus Taxonomy. Eighth Report of the
International Committee on Taxonomy of Viruses, pp. 757–778. Edited
by C. M. Fauquet, M. A. Mayo, J. Maniloff, U. Desselberger & L. A. Ball.
London: Elsevier/Academic Press.
Su, W., Ikegami, H., Nakayama, Y., Suzuki, K., Katayama, K.,
Nakayama, H. & Doi, K. (2003). Susceptibility of primary culture
neurons from rats of different ages to encephalomyocarditis (EMC)
virus infection. Exp Mol Pathol 75, 160–164.
Plagemann, P. G. & Swim, H. E. (1966). Replication of mengovirus. I.
Effect on synthesis of macromolecules by host cell. J Bacteriol 91,
2317–2326.
Tavakkol, A. & Burness, A. T. (1990). Evidence for a direct role for
Reischl, A., Reithmayer, M., Winsauer, G., Moser, R., Gösler, I. &
Blaas, D. (2001). Viral evolution toward change in receptor usage:
Vlasak, M., Goesler, I. & Blaas, D. (2005). Human rhinovirus type 89
variants use heparan sulfate proteoglycan for cell attachment. J Virol
79, 5963–5970.
adaptation of a major group human rhinovirus to grow in ICAM-1negative cells. J Virol 75, 9312–9319.
sialic acid in the attachment of encephalomyocarditis virus to human
erythrocytes. Biochemistry 29, 10684–10690.
Seaman, J. T., Boulton, J. G. & Carrigan, M. J. (1986).
Zimmerman, J., Schwartz, K., Hill, H. T., Meetz, M. C., Simonson, R. &
Carlson, J. H. (1993). Influence of dose and route on transmission
Encephalomyocarditis virus disease of pigs associated with a plague
of rodents. Aust Vet J 63, 292–294.
of encephalomyocarditis virus to swine. J Vet Diagn Invest 5, 317–
321.
196
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 04:18:53
Journal of General Virology 90