Upload
ralf-arnold
View
213
Download
0
Embed Size (px)
Citation preview
www.elsevier.com/locate/yviro
Virology 330 (20
Respiratory syncytial virus deficient in soluble G protein induced an
increased proinflammatory response in human lung epithelial cells
Ralf Arnold*, Brigitte Kfnig, Hermann Werchau, Wolfgang Kfnig
Institute of Medical Microbiology, Otto-von-Guericke-University, Magdeburg, Germany
Received 27 May 2004; returned to author for revision 4 August 2004; accepted 5 October 2004
Abstract
Respiratory syncytial virus (RSV) is worldwide the single most important respiratory pathogen in infancy and early childhood. The G
glycoprotein of RSV, named attachment protein, is produced by RSV-infected lung epithelial cells in both a membrane-anchored (mG
protein) and a soluble form (sG protein) that is secreted by the epithelial cell. Currently, the biological role of the sG protein in primary
RSV infection is still elusive. Therefore, we analyzed the inflammatory response of human lung epithelial cells (A549) infected either with
wild-type RSV (RSV-WT) or a spontaneous mutant thereof deficient in the production of secreted G protein (RSV-DsG). Our data reveal
that RSV-DsG, in comparison to RSV-WT, induced an increased cell surface expression of ICAM-1 on A549 cells and an enhanced
release of the chemokines IL-8 and RANTES after 20 h postinfection. The increased protein expression pattern correlated with an
enhanced mRNA level encoding for ICAM-1, IL-8, and RANTES, respectively. Furthermore, epithelial cells infected with RSV-DsG
showed a more increased binding activity of the transcription factor NF-nB when compared to RSV-WT. In contrast, the mutant RSV-DsG
replicated less efficiently in A549 cells than RSV-WT. Our data suggest that RSV, in the course of an ongoing infection, reduces by the
production of sG protein the detrimental inflammatory response evolved by the infected resident lung epithelial cell and thereby supports
its own replication.
D 2004 Elsevier Inc. All rights reserved.
Keywords: ICAM-1; RANTES; IL-8; Inflammation; NF-nB; RSV; A549 cell; Soluble G protein
Introduction
Respiratory syncytial virus (RSV) is worldwide the
major cause of serious lower respiratory disease in
infancy and early childhood (McIntosh, 1997). By the
age of 2 years, nearly every child becomes infected with
RSV, and due to incomplete immunity adults get
reinfected throughout their life (Hall and McCarthy,
2000; Falsey and Walsh, 2000). The epithelial cell of
the respiratory mucosa is the main target for RSV.
Evidence accumulated that epithelial cells infected with
0042-6822/$ - see front matter D 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.virol.2004.10.004
* Corresponding author. Current address: Institute of Medical Micro-
biology, Otto-von-Guericke-University, Leipzigerstr. 44, 39120 Magde-
burg, Germany. Fax: +49 391 6713384.
E-mail address: [email protected] (R. Arnold).
RSV play an active role in orchestrating the early
inflammatory and immune responses of the host (Doma-
chowske et al., 2001). The release of chemokines, i.e.,
interleukin-8 (IL-8), regulated upon activation, normal T-
cell-expressed and -secreted (RANTES), macrophage
inflammatory protein-1a (MIP-1a), monocyte chemoattrac-
tant protein-1 (MCP-1) (Arnold et al., 1994; Olszewska-
Pazdrak et al., 1998), and the expression of the intercellular
adhesion molecule-1 (ICAM-1) (Arnold and Konig, 1996)
are important and early epithelial responses following RSV
infection. Inflammatory cells are then chemotactically
recruited along the chemotaxin gradient from the local
blood vessels via the mucosal tissue into the lumen of the
alveolar space (Everard et al., 1994). Especially polymor-
phonuclear neutrophil granulocytes (PMN) are subsequently
responsible for the acute inflammation of the lung (Wang
and Forsyth, 2000).
04) 384–397
R. Arnold et al. / Virology 330 (2004) 384–397 385
The envelope of RSV contains three transmembrane
surface proteins, i.e., the fusion glycoprotein (F protein),
responsible for fusion of the viral envelope with the plasma
membrane of the host target cell (Collins and Mottet,
1991), the G glycoprotein (G protein), mediating attach-
ment of the virus particle to the target cell (Levine et al.,
1987), and the small SH protein with not well-understood
functions (Bukreyev et al., 1997). The G and F glyco-
proteins are the major neutralization and protective antigens
(Routledge et al., 1988). The G protein rich in serine,
threonine, and proline residues is a highly glycosylated
type II transmembrane protein with an apparent Mr of 90
kDa (Collins et al., 2001). The specific immune response
elicited by the G protein appears to be linked to the
pathogenesis of RSV: Immunization of mice with a
vaccinia virus recombinant expressing the G protein
induces a CD4+ T-cell response with an increased TH2
component and development of an eosinophilic pulmonary
infiltrate (Openshaw, 1995). Evidence accumulated that an
epitope in the region of amino acid 184 to 198 is essential
for the induction of detrimental lung eosinophilia and
increased lung pathology (Sparer et al., 1998; Tebbey et al.,
1998; Varga et al., 2000). Recently, Tripp et al. (2001)
found that this central conserved region contains a CX3C
motif at amino acid positions 182 to 186 that is able to
compete with fractalkine for binding to CX3CR1, the
fractalkine receptor, thereby interfering with the fractalkine-
mediated leukocyte chemotaxis. Moreover, in a mouse
vaccination model, this CX3C motif was shown to be
responsible for some aspects of enhanced pulmonary
disease after live RSV infection (Haynes et al., 2003).
Konig et al. (1996) have shown that purified membrane-
anchored G protein (mG protein) modulates the cytokine
release of IL-10, IL-12, and TNF-a from PBMC.
Beside the mG protein, a smaller soluble form of the G
protein (sG protein) with an apparent Mr of 82 kDa is also
produced by infected cells (Hendricks et al., 1987). It is
generated by translation initiation at a second AUG in the
mRNA and subsequent proteolytic trimming (Hendricks et
al., 1988; Roberts et al., 1994). The sG protein is already
shed during the first 12 h postinfection, when mG protein is
still not detectable (Hendricks et al., 1988), suggesting that
sG protein may play some role during the early innate
immune response. With regard to its biological role in
primary RSV infection, only few data are available
demonstrating that the sG protein can bind via its conserved
CX3C motif, like the mG protein, to the fractalkine receptor
(CX3CR1) thereby inducing migration of leukocytes (Tripp
et al., 2001). Recently, it has been shown by barometric
whole-body plethysmography in a primary mouse infection
model that a spontaneous variant of human RSV only
deficient in the production of the sG protein (RSV-DsG)
induced a more increased airway inflammation and impaired
lung function when compared with the wild-type strain
(RSV-WT) expressing both mG and sG protein, respectively
(Schwarze and Schauer, 2004). Therefore, we hypothesized
whether the sG protein might have some influence on the
inflammatory epithelial cell response during the first 20 h
postinfection. Since A549 lung epithelial cells serve as a
well-established in vitro infection model for analyzing
proinflammatory epithelial cell responses in the course of
an ongoing RSV infection, we used A549 cells in our
infection studies. We analyzed the cell surface expression of
ICAM-1 and the release of the chemokines IL-8 and
RANTES from A549 cells infected either with RSV-WT
or RSV-DsG, respectively.
Our data show that the expression of proinflammatory
molecules (IL-8, RANTES, ICAM-1) and the binding
activity of the transcription factor NF-nB was more
increased 20 h postinfection when A549 cells were infected
with RSV-DsG instead of RSV-WT. In contrast, the mutant
RSV-DsG replicated less efficiently in A549 cells, suggest-
ing that sG protein might act as a viral evasion factor during
an ongoing innate immune response.
Results
Characterization of the RSV-DsG mutant
To demonstrate that RSV-DsG was only deficient in the
production of sG protein but still able to synthesize mG
protein, we performed immunoblot analysis for detection
of G protein. Since RSV was propagated in HEp-2 cells,
we used this cell line in our immunoblot studies. HEp-2
cells were infected with RSV-WT and RSV-DsG, respec-
tively, and cultured for 36 h. As shown in Fig. 1A, the
RSV-infected HEp-2 cells expressed the mG protein with
an apparent Mr of 90 kDa irrespective of whether they
were infected with RSV-WT (lane 2) or the mutant RSV-
DsG (lane 3). The immunoblots were probed with anti-G
protein mAb (12E2, 2 Ag/ml) (Roder et al., 2000). Next,
we analyzed the presence of sG in the cell supernatant 20
h postinfection. After this culture time, the sG protein
accounts for nearly 80% of the G protein released into the
medium (Hendricks et al., 1988). The sG protein was
enriched by means of ConA-sepharose extraction. As
presented in Fig. 1B, cell supernatants enriched for the
sG protein and analyzed by anti-G protein immunoblot
analysis revealed that only cell supernatant obtained from
RSV- WT-infected cells (lane 3) contained the sG protein
with an apparent Mr of 82 kDa. The supernatant from
RSV-DsG-infected cells, enriched for sG protein (lane 2),
did not contain any sG protein. Since supernatants were
precleared by high-speed centrifugation from infectious
virus particles and detached cells, the prepared cell
supernatants were not confounded by mG protein. In this
regard, we obtained no mG protein immunoblot signal in
the range of 90 kDa. For control, we coelectrophoresed
mG protein (lane 1), purified from RSV-WT-infected HEp-
2 cells 36 h postinfection. The different migration rates
of mG protein and sG protein due to their different
Fig. 1. Western blot analysis of membrane-anchored G protein (mG), soluble G protein (sG) and virus protein expression in HEp-2 cells infected with RSV-WT
and RSV-DsG. (A) Prepared cell lysates (10 Ag) from HEp-2 cells were separated by SDS-PAGE (8%) and analyzed by anti-G protein immunoblot. Mock-
infected cells (lane 1) and cells infected (m.o.i. = 3) with RSV-WT (lane 2) and RSV-DsG (lane 3), respectively, were cultured for 36 h. The molecular masses
(in kDa) of the coelectrophoresed molecular weight markers are depicted on the left (lane M). (B) mG protein (1 Ag) purified from HEp-2 cells infected with
RSV-WT for 36 h (lane 1) and sG protein (40 Al eluate) concentrated by ConA-sepharose 4B affinity chromatography from HEp-2 cell supernatants 20 h
following infection with RSV-DsG (lane 2) and RSV-WT (lane 3), respectively, were assessed by SDS-PAGE (6%). (C) HEp-2 cell lysates prepared from cells
infected with RSV-WT (lane 1) and RSV-DsG (lane 2) (m.o.i. = 3) 36 h postinfection were separated by SDS-PAGE (12%) and analyzed by anti-RSV
polyclonal antibody. Shown are representative blots out of three.
R. Arnold et al. / Virology 330 (2004) 384–397386
molecular weights are obvious on the performed 6% SDS-
PAGE. Thus, only mG protein was produced by RSV-
DsG-infected cells in contrast to cells infected with the
wild-type strain of RSV, which synthesized both the
membrane-anchored and the secreted form of G protein.
In order to demonstrate that both virus strains did not
differ with regard to the expression profile of other virus
proteins, we stained whole cell lysates prepared from RSV-
infected HEp-2 cells with a polyclonal anti-RSV Ab. As
shown in Fig. 1C, both RSV-WT and RSV-DsG showed an
identical virus protein expression pattern.
ICAM-1 expression on A549 cells infected with RSV-DsG
and RSV-WT
We and others have shown that RSV infection leads to
an increase of ICAM-1 cell surface expression on lung
epithelial cells (Arnold and Konig, 1996; Domachowske et
al., 2001). Thus, we thought to define the role of sG protein
in regulating RSV-induced ICAM-1 expression on infected
human lung epithelial cells. Confluent monolayers of A549
cells were infected with RSV at a m.o.i. of 1 and 3. We
analyzed ICAM-1 cell surface expression 20 h postinfection
since, as mentioned above, at that time the sG protein
accounts for nearly 80% of the G protein released into the
cell supernatant (Hendricks et al., 1988). As shown in Figs.
2A and B, the infection with RSV-DsG induced in
comparison to RSV-WT a more pronounced ICAM-1
expression pattern on A549 cells. The ICAM-1 expression
increased in a RSV-dose dependent manner (Fig. 2C). But
regardless of the infection dose used, the mutant RSV-DsG
induced always significantly more ICAM-1 expression on
A549 cells than RSV-WT. When infection time proceeded
up to 48 h, the RSV-induced expression of ICAM-1 was
similar when A549 cells were either infected (m.o.i. = 3)
with RSV-WT or RSV-DsG (RSV-WT: 1677 F 140 MFI;
RSV-DsG: 1317 F 184 MFI). Cells treated with UV-
inactivated RSV, i.e., RSV-WT and RSV-DsG, respectively,
did not show any ICAM-1 up-regulation, suggesting that
replication of both strains was a prerequisite for RSV-
induced ICAM-1 expression (data not shown). To study
whether the more increased ICAM-1 cell surface expression
on RSV-DsG-infected A549 cells might be a direct
consequence of a more increased cellular ICAM-1 mRNA
level, we determined the ICAM-1 mRNA level in a
quantitative manner. As presented in Fig. 2D, during the
first 24 h postinfection, the cells infected with RSV-DsG
possessed a higher amount of ICAM-1 mRNA when
directly compared with cells infected with RSV-WT.
Induction of RANTES release from A549 cells by RSV-DsG
and RSV-WT
It has been shown that human lung epithelial cells
infected with RSV in vitro or in vivo are able to secrete the
C–C chemokine RANTES into the cell supernatant
(Olszewska-Pazdrak et al., 1998). To compare the ability
of RSV-WT and RSV-DsG to induce the release of
RANTES from A549 cells, we analyzed the amount of
RANTES in the cell supernatants 20 h postinfection. The
cell supernatants harvested from RSV-DsG-infected cells
contained significantly more RANTES than supernatants
from A549 cells infected with RSV-WT (Fig. 3A). In
contrast, when the amount of RANTES was determined 48
h postinfection, both RSV-WT and RSV-DsG did not differ
with regard to their capability to induce the release of
RANTES from A549 cells (RSV-WT: 3.2 F 0.3 ng/ml;
RSV-DsG: 3.0 F 0.2 ng/ml). Similar to the experiments
concerning ICAM-1 gene expression, the expression of the
RANTES gene was quantitatively determined. Our data
show that the enhanced release of RANTES from A549
cells infected with RSV-DsG correlated with a more
pronounced RANTES mRNA level 20 h postinfection
(Fig. 3B).
Fig. 2. Effect of RSV-WT and RSV-DsG on ICAM-1 expression in A549 cells. The cell surface expression of ICAM-1 was measured by flow cytometry
analysis (A–C) and the cellular ICAM-1 mRNA amount was determined by quantitative real-time RT-PCR (D). (A, B) FACS histograms are shown as
representative results demonstrating different cell surface expression of ICAM-1 induced by RSV-WT (A) and RSV-DsG (B) 20 h postinfection (m.o.i. = 3).
(C) Cells infected at a different m.o.i. (1 and 3) showed a virus dose-dependent up-regulation of ICAM-1 20 h postinfection. Data are shown from mock-
infected cells (black bars), RSV-WT-infected cells (open bars), and RSV-DsG-infected cells (shaded bars), respectively. Results are means F SEM (n = 9).
Significant differences from mock-infected cells (black bars) are indicated by *P b 0.01. Significant differences from RSV-WT-infected cells (open bars) are
indicated by #P b 0.01. (D) ICAM-1 mRNA levels were analyzed 6–24 h postinfection (m.o.i. = 3) by TaqMan real-time RT-PCR. The housekeeping gene
GAPDH was used for normalization. The data are means of two independent experiments.
R. Arnold et al. / Virology 330 (2004) 384–397 387
Induction of IL-8 release from A549 cells by RSV-DsG and
RSV-WT
Evidence accumulated that A549 cells infected with RSV
secrete a substantial amount of IL-8 in a time- and RSV-
dose dependent manner (Arnold et al., 1994). Therefore, we
also analyzed the RSV-induced IL-8 synthesis in A549 cells.
Similar to the RANTES secretion pattern, RSV-DsG
induced a markedly increased release of IL-8 when
compared with RSV-WT-infected cells (Fig. 4A). These
results were obtained irrespective of the virus dose used.
As was already shown for the cell surface expression of
ICAM-1 and secretion of RANTES, when RSV-infected
cells (m.o.i. = 3) were incubated up to 48 h postinfection,
both virus strains induced a similar release of IL-8 (RSV-
WT: 45 F 4.9 ng/ml; RSV-DsG: 39 F 3.7 ng/ml). By using
real-time RT-PCR, we observed a significantly more
increased IL-8 mRNA level when A549 cells were infected
with RSV-DsG instead of RSV-WT (Fig. 4B).
Activation of NF-jB by RSV-DsG and RSV-WT
Following infection of human lung epithelial cells with
RSV, especially the activation of the transcription factor
NF-nB leading to an increased gene expression of ICAM-1,
IL-8, and RANTES has been recently reported (Chini et al.,
1998; Garofalo et al., 1996; Thomas et al., 1998). To
determine whether the observed increased mRNA levels
encoding for ICAM-1, IL-8 and RANTES subsequent to
infection with RSV-DsG might be associated with a further
increased activation of NF-nB, we analyzed the binding
activity of Rel A (p65) and NF-nBI (p50) in A549 cells by
means of a quantitative NF-nB binding assay. The
activation of NF-nB was determined after 6 and 20 h
postinfection because RSV activates NF-nB late during
infection. The binding activity of p65 and p50 in RSV-
infected A549 cells is presented in Fig. 5. When compared
to uninfected cells (medium control) both transcription
factors showed a markedly increased binding activity in
A549 cells 24 h after infection with RSV-WT. Also, when
determined 6 h postinfection, a small but detectable
increased binding activity of p65 and p50 was observed.
However, with regard to activation of NF-nB (p65/p50)
binding, RSV-DsG was approximately twice as effective as
RSV-WT (Figs. 5A,B). A nuclear extract prepared from
Jurkat cells (supplied by the manufacturer) served as an
internal positive control (Figs. 5A,B: see control). Se-
quence specificity of the test was monitored by competition
with free wild-type NF-nB-consensus oligonucleotides (wt)as well as mutated NF-nB-consensus oligonucleotides (mt).
As shown, both protein DNA complexes, i.e., p65 and p50,
were specific because soluble oligonucleotides encoding for
the wild-type NF-nB-consensus binding site competed
completely with the binding activity of the titer-plate-
Fig. 3. Effect of RSV-WT and RSV-DsG on RANTES expression in A549
cells. The release of RANTES was determined by ELISA (A) and the
level of RANTES mRNA was quantified by Quantikine RANTES mRNA
ELISA (B). Cells were infected at a m.o.i. of 3 and cultured for 20 h. (A)
The release of RANTES from mock-infected cells (black bars) was
compared with the release from RSV-WT-infected cells (open bars) and
RSV-DsG-infected cells (shaded bars), respectively. Results are means FSEM (n = 6). (B) The amount of RANTES mRNA was quantified 20 h
postinfection. Results are means F SEM (n = 5). Significant differences
from mock infected cells (black bars) are indicated by *P b 0.01.
Significant differences from RSV-WT infected cells (open bars) are
indicated by #P b 0.01.
Fig. 4. Effect of RSV-WT and RSV-DsG on IL-8 expression in A549
cells. The release of IL-8 into the cell supernatant (A) and the amount of
cellular IL-8 mRNA (B) were quantified from mock-infected cells (black
bars), RSV-WT-infected cells (open bars), and RSV-DsG-infected cells
(shaded bars). (A) IL-8 was measured by ELISA 20 h postinfection in the
cell supernatants of mock-infected A549 cells and cells infected at a
m.o.i. of 1 and 3, respectively. Results are means F SEM (n = 5). (B)
The amount of IL-8 mRNA in mock-infected A549 cells and cells
infected with RSV (m.o.i. = 3) was determined by Taqman real-time RT-
PCR 20 h postinfection. Results are means F SEM (n = 7). *P b 0.01,
vs. mock infected cells (black bars); #P b 0.01, vs. RSV-WT-infected
cells (open bars).
R. Arnold et al. / Virology 330 (2004) 384–397388
bound DNA encoding for the NF-nB consensus binding
site (Figs. 5A,B: see control + wt). In contrast, soluble
oligonucleotides encoding for a mutated NF-nB-consensusbinding site did not compete with the immobilized DNA
for the binding of the activated transcription factors (Figs.
5A,B: see control + mt).
Replication of RSV-DsG and RSV-WT
It has been shown that replication of RSV is a
prerequisite for the RSV-induced activation of NF-nB(Bitko et al., 1997; Fiedler et al., 1996). We hypothesized
whether the observed increased binding activity of NF-nBin RSV-DsG-infected A549 cells might be a direct
consequence of a more increased replication rate of RSV-
DsG when compared to RSV-WT. We therefore analyzed the
growth of RSV-DsG and RSV-WT in A549 cells. First, we
determined the amount of infectious virus particles released
into the cell supernatants. As can be seen from Fig. 6A, the
virus strains RSV-DsG and RSV-WT showed a different
growth rate in A549 cells. During 96 h postinfection
(m.o.i. = 0.01), the mutant strain replicated less efficiently
in A549 cells. The growth was reduced nearly 10-fold when
compared to the growth rate of RSV-WT. To exclude that
RSV-DsG accumulated intracellularly instead of budding
from the cell, we analyzed the virus titer in the prepared cell
lysates and the corresponding supernatants 36 h post-
infection (m.o.i. = 1). As depicted in Fig. 6B, A549 cells
infected with the RSV-DsG strain possessed intracellularly
less infectious virus particles, suggesting that the mutant
RSV-DsG was not retained inside the cells to a higher
degree than the RSV-WT. Obviously, these data suggest that
the increased binding activity of NF-nB in RSV-DsG-
infected cells might not be explained by an increased
proliferation rate.
Fig. 5. NF-nB (p65, p50) binding activity in RSV-infected A549 cells. A549 cells were infected with RSV-WT and RSV-DsG, respectively at a m.o.i. of 3.
After 6 and 20 h, prepared cell lysates were assayed for p65 activity (A) and p50 activity (B). Jurkat nuclear extract provided by the manufacturer served as
positive control. To demonstrate specificity of the NF-nB binding assay, wild-type (wt) and mutant (mt) NF-nB oligonucleotides were used as competitors
during the binding reaction (shaded bars). Results are means F SEM (n = 8). Significant differences from mock-infected cells (black bars) are indicated by
*P b 0.01. Significant differences from RSV-WT infected cells (open bars) are indicated by #P b 0.01.
R. Arnold et al. / Virology 330 (2004) 384–397 389
Next, we asked whether the increased binding activity of
NF-nB might be due to an increased cellular virus protein
load. Therefore, we assayed the cell surface expression of
mG and mF protein on A549 cells infected with RSV-WT
and RSV-DsG, respectively. The reduced production of
infectious RSV-DsG particles correlated with a diminished
expression of mG- and mF protein on A549 cells 24 h
Fig. 6. Growth of RSV-WT and RSV-DsG in A549 cells. (A) Cells were infected
and analyzed by plaque assay on HEp-2 cells. Results are means from two exp
(m.o.i. = 1) and 36 h postinfection aliquots of cell supernatants and cell lysates wer
SEM (n = 5). *P b 0.01 vs. RSV-WT infected cells (open bars).
postinfection with RSV-DsG (Fig. 7A). Furthermore, a
reduced expression of mF and mG protein on A549 cells
infected with the mutant strain was still observed after 36 h
postinfection (data not shown). Additionally, like mG and
mF protein, the cell surface staining of the mSH protein
verified a reduced expression of this protein on RSV-DsG-
infected cells (data not shown). Again, to exclude that cells
at a m.o.i. of 0.01. Supernatants were harvested at the indicated time points
eriments performed in duplicate. (B) A549 cells were infected with RSV
e assayed for the presence of infectious virus particles. Results are meansF
Fig. 7. Comparison of the expression of mF and mG protein and fusion
activity of RSV-WT and RSV-DsG in A549 cells. (A) The amount of cell
surface expression of mG and mF protein was determined by FACS
analysis. Results are means F SEM (n = 6). *P b 0.01, vs. RSV-WT
infected cells (open bars). (B) The percentage of cells that stained positive
for mG and mF protein was determined by FACS analysis. Results are
means F SEM (n = 6). *P b 0.01, vs. RSV-WT infected cells (open
bars). (C) Relief of self-quenching of aminofluorescein-labeled virus
bound to A549 cells. A representative experiment performed in triplicate
is shown (n = 3).
R. Arnold et al. / Virology 330 (2004) 384–397390
infected with RSV-DsG do not retain synthesized G- and F
protein to a higher degree than RSV-WT-infected cells, we
stained permeabilized cells for these proteins. Our data
showed that these virus proteins were not stored within cells
either infected with RSV-DsG and RSV-WT, respectively
(data not shown). But when cells were stained for mG and
mF protein, we obtained less positive cells in case of
infection with RSV-DsG, suggesting that the mutant either
infects their target cell less efficiently or replicates more
slowly (Fig. 7B). In this regard, we analyzed whether
different fusion rates between the virus envelope and the
target cell membrane may occur. The fusion of RSV labeled
with aminofluorescein was measured by means of a
fluorescence dequenching assay. As depicted in Fig. 7C,
the fluorescence dequenching curves obtained with RSV-
DsG and RSV-WT were identical. Therefore, we concluded
that different fusion kinetics do not account for the different
replication rates of both virus strains.
Discussion
It has been shown that infection of human lung epithelial
cells with RSV induces expression of ICAM-1, RANTES,
and IL-8. In this study, we supply evidence that the
expression of these inflammatory mediators were even
more up-regulated in A549 cells when cells were infected
with a mutant strain RSV-DsG, lacking the production of sG
protein. This different expression pattern was observed 20 h
postinfection and correlated with an increased binding
activity of the transcription factor NF-nB and a diminished
growth rate in A549 cells. However, these different cell
responses were time constricted since with prolonged
culture time, i.e., 48 h postinfection, both RSV-DsG and
RSV-WT induced the same expression of ICAM-1,
RANTES, and IL-8, respectively.
The RSV-induced bronchiolitis is characterized by
peribronchial infiltration of leukocytes and epithelial
necrosis (Faden et al., 1984) as well as prominent
accumulation of PMN in the airway lumen (Everard et al.,
1994). It has been shown that the enhanced expression of
ICAM-1 on RSV-infected human epithelial cells is respon-
sible for an increased adhesion of neutrophils (Stark et al.,
1996) leading consequently to epithelial cell damage and
death (Wang et al., 1998). Therefore, in primary RSV
infection, lung epithelial cells infected with RSV-DsG,
expressing a further increased level of cell surface ICAM-
1, seem to be more susceptible to cell damage than RSV-
WT-infected cells. Whether the increased ICAM-1 cell
surface expression on RSV-DsG-infected cells might be
partly a consequence of an altered cell surface shedding is
still not known. However, the increase in the cellular mRNA
level encoding for ICAM-1 together with the enhanced
binding activity of NF-nB (as discussed below) suggest that
the enhanced cell surface expression is primarily mediated
by de novo protein synthesis.
The chemokine RANTES, a member of the C–C
chemokine family, is a chemotactic and activation factor
for monocytes, basophils, and eosinophils, with no activity
on neutrophils (Rot et al., 1992). With regard to our data,
showing an enhanced release of RANTES from RSV-DsG-
infected A549 cells during the first day postinfection, one
may hypothesize whether the above-mentioned cells may
accumulate earlier and/or to a more increased amount in the
RSV-DsG-infected lung.
The C–X–C chemokine IL-8 is mainly a chemotactic
factor and activator for neutrophils (Baggiolini et al., 1989).
The high amount of IL-8 released from lung epithelial cells
R. Arnold et al. / Virology 330 (2004) 384–397 391
may be primarily responsible for the prominent accumu-
lation of PMN in the alveolar space of the infected child
(Everard et al., 1994). Similar to the increased release of
RANTES, one is tempted to speculate whether the enhanced
release of IL-8 after infection with RSV-DsG may contribute
to a further increased recruitment of neutrophils into the
infected lung.
Several recently published studies have shown that
chemokines, such as IL-8, RANTES, and MIP-1a are
induced in respiratory secretions collected from RSV-
infected patients, and that their amount correlated with
disease severity (Noah et al., 2002; Smyth et al., 2002;
Teran et al., 1999). Furthermore, instead of T helper
cytokines, the chemokines MIP-1a and RANTES seem to
be better predictive markers for severe bronchiolitis and
recurrent wheezing (Chung and Kim, 2002; Garofalo et al.,
2001). Quite recently, Schwarze and Schauer (2004)
analyzed this mutant RSV-DsG strain in a murine model
of primary infection. Following infection with RSV-DsG, a
more impaired lung function, an enhanced airway inflam-
mation, and increased lung virus titers were observed when
compared with infection of RSV-WT. Also, the numbers of
pulmonary eosinophils, macrophages, and neutrophils were
more increased in mice infected with RSV-DsG. Therefore,
future studies in this in vivo infection model have to address
the question whether lung epithelial cells infected with
RSV-DsG are mainly responsible for this impaired infection
course.
The transcription factor NF-nB is one of the pivotal
regulators of proinflammatory gene expression, i.e., it
induces the transcription of proinflammatory cytokines,
chemokines, and adhesion molecules (Tak and Firestein,
2001). NF-nB is a collective name for a family of dimeric
transcription factors consisting of five members: RelA
(p65), RelB, c-Rel, NF-nB1 (p50), and NF-nB2 (p52).
They form various hetero- and homodimers. The RelA/NF-
nB1 (p65/p50) heterodimer is the most commonly found in
activated cells. It has been shown that the RSV-induced
expression of RANTES and IL-8 is highly dependent on the
activation of NF-nB (Casola et al., 2001; Garofalo et al.,
1996; Thomas et al., 1998). We observed that infection with
RSV-DsG as well as RSV-WT led to a more pronounced
binding activity of RelA and NF-nB1, respectively. Thissuggests that NF-nB heterodimers consisting of p65/p50
were formed regardless of the used RSV strain. That p65/
p50 complexes are primarily formed in RSV-infected cells is
in agreement with recently published results (Bitko et al.,
1997; Garofalo et al., 1996). The observed enhanced
binding activity of NF-nB after infection with RSV-DsG
correlated with a further increase of the cellular mRNA level
encoding for ICAM-1, IL-8, and RANTES, respectively.
Since all three genes under investigation have been shown
to be induced by RSV in a NF-nB-dependent manner, our
data suggest that RSV-DsG might lead to an increased
ICAM-1, IL-8, and RANTES gene expression via the
observed increased binding activity of NF-nB. Obviously,
the transcription rates of these genes are regulated by
additional transcription factors (Casola et al., 2000; Chini et
al., 1998, 2001), and the expression of the RANTES- and
IL-8 gene is further regulated by modulating mRNA
stability (Hoffmann et al., 2002; Koga et al., 1999).
Therefore, with regard to increased gene expression after
infection with RSV-DsG, additional mechanisms have to be
taken into account.
The mechanisms leading to the activation of NF-nB in
RSV-infected lung epithelial cells are still not known. In
A549 cells, infected with RSV-DsG, we observed no
intracellular accumulation of mG protein due to the lack
of secreted G protein (data not shown). Moreover, the
mutant strain replicated less efficiently when compared to
the wild-type strain. In this regard, one may exclude an
increased activation of NF-nB in RSV-DsG-infected A549
cells due to an enhanced cellular virus load and/or
replication rate (Fiedler et al., 1996). Similar to UV-
inactivated wild-type strain, UV-inactivated RSV-DsG did
not induce any expression of the proinflammatory mediators
under investigation (data not shown). Together with our data
showing identical fusion kinetics of both virus strains, we
conclude that adherence/fusion phenomena by themselves
cannot be responsible for the increased activation of NF-nBobserved in A549 cells infected with the mutant strain. In
this regard, we also exclude that the two point mutations in
the F gene of the mutant strain do not influence the infection
behavior of RSV-DsG.
Quite recently, Bose et al. (2003) demonstrated that the
activation state of NF-nB, at the beginning of a RSV
infection, controls the replication of RSV. When A549 cells
were preactivated with TNF-a and IL-1h, respectively, andsubsequently infected with RSV, the replication of the virus
was significantly inhibited due to the enhanced activation
state of NF-nB. One may hypothesize whether the reduced
replication rate of RSV-DsG in A549 cells might be at least
in part a direct consequence of the more increased activation
of NF-nB. The observed increased activation of NF-nB in
A549 cells 6 h postinfection with RSV-DsG supports this
thesis.
Recently, Teng et al. (2001) used reverse genetics to
construct a RSV-mutant deficient in the release of sG
protein. When compared with the wild-type strain, this
mutant did not differ markedly with regard to replication in
HEp-2 and Vero cells. However, when virus growth was
analyzed in the lower and upper respiratory tract of infected
BALB/c mice, a slightly reduced replication rate of the
mutant strain was observed. We obtained similar results
showing that the replication of RSV-DsG and RSV-WT did
not differ significantly in HEp-2 cells (data not shown). The
mechanisms for these cell-type-dependent replication rates
of RSV-DsG are still unknown. Both A549 cells and HEp-2
cells secrete a variety of cytokines following infection with
RSV (Arnold et al., 1994; Harrison et al., 1999). Whether
differences in the amount or cytokine profile secreted by the
infected epithelial cell type are responsible for the different
R. Arnold et al. / Virology 330 (2004) 384–397392
growth rates of RSV-DsG in A549 cells and HEp-2 cells
remains to be determined (Merolla et al., 1995).
The soluble G protein is capable of binding to epithelial
cell surfaces, and can substitute for the mG protein in the
course of infection (Escribano-Romero et al., 2004; Teng et
al., 2001). To analyze whether the sG protein might
influence the expression pattern of ICAM-1 in an autocrine
manner, we incubated RSV-DsG-infected A549 cells in the
presence of conditioned HEp-2 cell supernatant (1:2 diluted)
that was harvested from confluent cell monolayers (75 cm2,
10 ml supernatant) 20 h postinfection with RSV-WT
(m.o.i. = 3). The A549 cells were cultured for 20 h and
expression of ICAM-1 was determined thereafter. These
experiments clearly showed that ICAM-1 expression was
not diminished due to the added cell conditioned medium.
Also, purified sG protein (1 Ag/ml) or fractalkine (20 ng/ml)
added to the media of RSV-DsG-infected A549 cells did not
reduce the RSV-DsG-induced ICAM-1 expression pattern
(data not shown). Therefore, we concluded that sG protein
may not modulate the cell surface expression of ICAM-1 in
an autocrine manner as it was shown for IL-1a with regard
to IL-8 and ICAM-1 (Patel et al., 1995, 1998). Whether
RSV-DsG might be able to induce a substantial amount of
IL-1a earlier than RSV-WT, mediating ICAM-1 and IL-8
expression, remains to be determined.
Evidence accumulated that detrimental G-protein-
specific immune responses play a central role in RSV-
induced lung disease and eosinophilia (Openshaw, 1995).
The priming of mice with vaccinia viruses that expressed
mG protein led to a release of IL-5 into lung supernatants
and lung eosinophilia when subsequently challenged with
RSV. Moreover, mice primed with vaccinia viruses express-
ing sG protein and challenged with RSV revealed an even
more enhanced amount of IL-5 in the cell supernatant
combined with greatest lung eosinophilia (Johnson et al.,
1998). An established TH2-like response might be advanta-
geous for the survival of the virus. Therefore, the ability of
RSV to produce sG protein may represent a viral strategy
for immunomodulation in the course of adaptive immunity.
Cane (2001) proposed that sG protein might also act as a
decoy receptor in the course of a RSV-specific antibody
immune response. But to date, no data are available
supporting this concept. Also, Ray et al. (2001) did not
find any direct influence of purified sG protein on RSV-
specific T cell responses.
On the contrary, a lot of data accumulated that support
the thesis that sG protein seems to play an important
regulatory role in the course of innate immunity. Recently,
Tripp et al. (2000) published that a RSV mutant, deficient in
the G- and SH protein, induced an increased expression of
chemokines in the lung of infected BALB/c mice. However,
this mutant also contained amino acid changes in the F and
L protein. Therefore, further experiments are necessary to
determine exactly the role of the sG protein in this mouse
infection model. With respect to our observed two amino
acid changes in the F protein of the mutant RSV-DsG, we
hypothesize that they are not responsible for the observed
mutant phenotype since RSV-WT and RSV-DsG showed
identical fusion rates with their target cell membrane. The
data published by Tripp et al. (2000) are in line with recently
published results showing an increased lung infiltration of
immune effector cells in mice infected with RSV lacking the
secretion of sG protein (Schwarze and Schauer, 2004).
Furthermore, evidence accumulated that sG protein might
act as a chemokine by itself via binding through its con-
served CX3C motif to the fractalkine receptor (CX3CR1)
expressed on immune effector cells (Tripp et al., 2001). In
contrast, whether sG protein might be able to complex
chemokines and thereby blocking their chemotactic activity
as it was shown for a variety of viral soluble glycoproteins
remains to be determined (Seet and McFadden, 2002).
With respect to our data represented herein, one may
suggest that RSV is able to reduce the inflammatory
response by the synthesis of sG protein. This regulatory
activity seems to be time constricted. To control the
inflammatory response established by the infected lung
epithelial cell should be advantageous for the virus.
Materials and methods
Cell culture
Human A549 pulmonary type II epithelial cells and
HEp-2 cells [American Type Culture Collection (ATCC),
Rockville, USA] were cultured in DMEM supplemented
with 5% (v/v) FBS, streptomycin (100 Ag/ml), and
penicillin (100 IU/ml) (Gibco BRL, Karlsruhe, Germany).
All cell cultures and prepared virus stocks were free of
chlamydia and mycoplasmic contamination.
Virus preparation and titration
The human long strain of RSV (ATCC) and a mutant
thereof deficient in the production of sG protein both
designated as RSV-WT and RSV-DsG, respectively, were
propagated and titrated in HEp-2 cells. Cells were cultured
in DMEM containing 0.5% (v/v) FBS, streptomycin
(100 Ag/ml), and penicillin (100 U/ml). Confluent mono-
layers of HEp-2 cells were infected with RSV (WT or DsG)
for 3 h (m.o.i. = 0.1) in DMEM. The monolayers were
washed, overlayed with DMEM (0.5% FBS), and incubated
at 37 8C in 5% CO2 atmosphere until cytopathic effect
reached ~80%. Thereafter, the supernatants were harvested
and cellular debris was removed by centrifugation (5000 �g, 20 min). RSV was concentrated by polyethylene glycol
precipitation (final 10%, 60 min, 4 8C) and purified by
means of discontinuous sucrose gradient centrifugation
(Ueba, 1978). To stabilize the purified virus particles, they
were resolved in 20% sucrose/NT-buffer [150 mM NaCl,
50 mM Tris–HCl (pH 7.5)] and stored at �80 8C. The virustiter was determined using the microplaque immunoperox-
R. Arnold et al. / Virology 330 (2004) 384–397 393
idase method (Cannon, 1987). Briefly, RSV stocks as well
as cell supernatants and cell lysates (prepared from 24-well
plates in a volume of 1 ml) were serially 10-fold diluted
onto confluent HEp-2 monolayers in microtiter plates. After
24 h, the monolayers were fixed with paraformaldehyde
(2%) and stained with mouse anti-G protein RSV polyclonal
antibody (Ab) (Biotrend, Kfln, Germany) followed by
HRP-conjugated rabbit anti-mouse immunoglobulins
(Dako, Hamburg, Germany). The RSV plaques were
identified by colorimetric staining and counted microscopi-
cally. Stock titer of the used virus stocks were adjusted to
108 plaque forming units (pfu)/ml.
Characterization of the spontaneous RSV-DsG mutant
The mutant of RSV (long strain) only deficient in the
production of sG protein (RSV-DsG) was isolated in the
Department of Medical Microbiology and Virology at the
Ruhr-University Bochum. The G gene possesses the greatest
genetic variability (Johnson et al., 1987). This inherent
quality was exploited for isolating a spontaneous mutant
deficient in the production of sG protein. For this purpose,
the RSV-WT was serially passaged at high multiplicity of
infection (undiluted) in HEp-2 cells. Twenty hours after
each passage, the cell supernatants were screened for the
presence of sG protein and the cells refed with fresh
medium. At the 7th passage, no sG protein could be
detected. The amount of sG protein was assessed by ELISA
(Roder et al., 2000). The isolated mutant (7th passage)
possess a point mutation in the G gene at position 157 (A to
G), which led to an exchange at amino acid 48 from
methionine to valine and therefore in an inactivation of the
second start codon in the ORF of the G gene. To exclude
any other mutations detrimental for the replication of the
mutant RSV-DsG, we performed sequence analysis of the
genes, known to be important for viral replication, i.e.,
NS1-, NS2-, M2- N-, P-, L-, and F gene. Gene sequences
obtained from our wild-type RSV (RSV-WT) strain were
directly compared with sequences of the isolated mutant
RSV-DsG. Virus RNA was isolated from supernatants from
infected HEp-2 cells using the RNeasy isolating kit
according to the instructions of the manufacturer (Qiagen,
Hilden, Germany). Primers for the NS1-, NS2-, M2-, N-, P-,
L-, F-, and the G gene of RSV (Metabion, Planegg-
Martinsried, Germany) were designed based on sequences
published previously (accession number M74568) and are
available upon request from the authors. PCR fragments
were constructed with sufficient overlap with adjacent
fragments such that there were no primer-masked regions.
Virus RNA was reverse transcribed at 50 8C for 30 min
using Superscript reverse transcriptase, oligo(dT)12–18 as
primer, and reaction conditions according to the protocols
supplied by the manufacturer (Invitrogen, Karlsruhe, Ger-
many). The resulting cDNA was used as template in the
PCR procedure to generate a series of overlapping PCR
products, spanning the desired RSV genes. The PCR
mixture (50 Al) consisted of cDNA, 2 Al of the respective
primers (50 pmol) each, 10 Al of 10 mM deoxyribonucleo-
side triphosphate (dNTPs), and 2 units of Taq polymerase
(Invitrogen). The cDNA was amplified during 40 cycles of
95 8C for 1 min, 53 8C for 2 min, 72 8C for 3 min, with a
final extension at 72 8C for 10 min. After examination on a
2% agarose gel, the RT-PCR products were purified using a
QIAquick purification kit (Qiagen) and sequenced directly
using the ABI Prism BigDyek Terminator Cycle Sequenc-
ing Ready Reaction kit [PE Applied Biosystems (PE),
Weiterstadt, Germany] on an ABI PRISM 3100 automatic
DNA sequencer according to the instructions of the
manufacturer. Sequence alignments were made using the
Sequencerk software V.4.0.5 from Gene Codes Coopera-
tion (Ann Arbor, MI, USA). When compared to RSV-WT,
the genes encoding for L-, NS1-, NS2-, M2-, P, and N
protein were unchanged in RSV-DsG. We observed two
point mutations in the F gene at position 227 (T to A) and
635 (G to A) leading to amino acid changes Val to Glu and
Ser to Asn, respectively. In addition to the point mutation at
position 157, the G gene of the mutant strain accumulated
27 point mutations leading to 18 amino acid changes.
However, the four cysteine residues within the conserved
gene sequence of the G protein, which is known to be
responsible for enhanced lung eosinophilia, were unchanged
(Sparer et al., 1998; Tebbey et al., 1998). Since both RSV-
DsG and RSV-WT showed identical fusion rates (see
Fig. 7C), we concluded that no additional mutations, except
the loss of the sG protein production, might influence the
biological behavior of the RSV-DsG.
Cell infection and stimulation
Confluent monolayers of A549 cells were washed with
DMEM, exposed to RSV for 2 h, washed and incubated
with fresh medium DMEM (2% FBS) for 20 and 48 h,
respectively. Thereafter, cells were harvested and immedi-
ately stained for FACS analysis. Corresponding cell super-
natants were collected and stored at �70 8C until cytokine
determination.
Preparation of soluble and membrane-anchored G protein
HEp-2 monolayers were infected with RSV-WT or
RSV-DsG (m.o.i. = 3), washed and cultured for 20 and
36 h, respectively. After 20 h, cell supernatants were
harvested, cleared by centrifugation (100000 � g, 30
min) and sG protein was prepared by ConA-sepharose 4B
affinity chromatography (Amersham Biosciences, Frei-
burg, Germany) (Johnson et al., 1998). Cellular proteins
were isolated from cells by suspension in RIPA buffer
(1� PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate,
0,1% SDS) containing 100 Ag/ml PMSF, 60 Ag/ml
aprotinin, and 1 mM sodium orthovanadate. Cell lysates
were centrifuged and the supernatants collected. Mem-
brane-bound G protein from RSV-infected HEp-2 cells
R. Arnold et al. / Virology 330 (2004) 384–397394
was prepared as previously described (Roder et al.,
2000).
Fluorescence dequenching assay
The virus envelope was labeled with the fluorescent
aminofluorescein (Molecular Probes, Leiden, Netherlands).
An aliquot (10 Al) of the aminofluorescein solution (50
mg/ml DMSO) was added to a RSV solution containing
5 � 106 pfu/500 Al HBSS (Miller and Hutt-Fletcher,
1992). Following incubation for 60 min at RT, the
unincorporated fluorescent dye was removed by gel
filtration on Sephadex G-75 columns using HBSS as the
eluting solution. The labeled virus was used for experi-
ments at the same day. A549 cells (5 � 105) were
incubated with RSV (5 � 104) in a volume of 100 AlHBSS for 2 h. Fluorescence intensity was measured on a
SpectraFluorPlus Reader (TECAN, Crailsheim, Germany)
with an excitation wavelength of 485 nm and an emission
wavelength of 535 nm. Each experiment was run in
triplicate. At the end of each experiment, Triton X-100
was added to 1% (v/v) into the reaction mixture, and the
maximum of fluorescence was obtained, which was
considered to be equivalent to 100% fusion of RSV with
target cells. The percentage of fusion (F) was determined
as follows: %F = 100 � (Fd � F0)/(Ft � F0), where
F0 = Em 535 nm of aminofluorescein labeled virus, Fd =
Em 535 nm after incubation with target cells, and Ft = Em
535 nm after addition of Triton X-100.
Western immunoblot
SDS-PAGE, semidry electroblotting, and detection of
bound sheep anti-mouse HRP-conjugated secondary Ab
(Dako) by ECL Plusk Western blotting detection system
(Amersham Biosciences) was performed as already
described (Roder et al., 2000). The expression of several
virus proteins was analyzed by means of HRP-labeled
polyclonal goat anti-RSV Ab (Biotrend). Protein molecular
weight markers were obtained from Amersham Bioscien-
ces. The chemiluminescence signal was measured by
means of Lumi-Imagerk F1 workstation with Lumi-
Analystk 3.x software from Roche Diagnostics (Mann-
heim, Germany).
Flow cytometry
The expression of ICAM-1 on cell surfaces of A549 cells
was measured with a FACSCalibur (BD Biosciences,
Heidelberg, Germany). Detachment and staining of cells
was performed as previously described (Arnold and Konig,
1996). PE-labeled mouse anti-human ICAM-1 mAb and
PE-labeled IgG isotype control Abs were obtained from BD
Biosciences. Monoclonal mouse anti-G protein- and anti-F
protein Abs were obtained from Biotrend. The binding of
primary Abs was detected by secondary staining with Cy3-
labeled AffiniPure Goat anti-mouse IgG (H+L) (Dianova,
Hamburg, Germany). By using the Fix and Perm kit for
intracellular staining (BD Biosciences), the cells were fixed
and permeabilized according to the manufacturer’s instruc-
tions. Data were analyzed by means of CellQuest software
(BD). The mean fluorescence intensity (MFI) of 10000 cells
was determined and corrected by subtraction of background
fluorescence of isotype control.
Cytokine measurements
Cytokines were measured by ELISA. RANTES was
determined according to the manufacturer’s instructions
(R&D Systems, Wiesbaden-Nordenstadt, Germany), and
IL-8 was determined as previously described (Arnold et al.,
1999).
NF-jB binding activity
The binding activity of NF-nB subunits Rel A (p65) and
NF-nB1 (p50) was measured using the Trans-AMk tran-
scription factor assay kit (Active Motif Europe, Rixensart,
Europe) according to the manufacturer’s recommendations
(Renard et al., 2001). Briefly, DNA encoding for the NF-nBconsensus binding site was bound to microtiter plates. The
binding of active forms of p65 or p50 to this immobilized
DNA was revealed by incubation with NF-nB specific
antibodies using ELISA technology. Whole cell lysates were
assayed for p65 and p50 binding activity. Optical density
was measured at 450 nm.
Quantikine mRNA ELISA and quantitative real-time
RT-PCR
The amount of RANTES mRNA was determined by
means of the Quantikine mRNA detection kit in accordance
with the manufacturer’s instructions (R&D Systems). The
amount of ICAM-1 and IL-8 mRNA was quantified by
Taqman real-time RT-PCR. The total cellular RNA from
uninfected and RSV-infected epithelial cells (5 � 105) was
extracted using the QiAmp 96 viral RNA Biorobotk kit on
the Biorobot 3000 System fromQiagen. Using the Gene Amp
5700 Sequence Detector (PE Applied Biosystems) the RNA
(1 Ag) was reverse transcribed to 2.5 Al cDNA and amplified
in a volume of 25 Al by means of TaqMan Universal PCR
Master Mix (PE Applied Biosystems) containing specific
primers for IL-8 (10 pmol) and a fluorescently labeled probe
(7 pmol) that were selected by means of Primer Express
Software version 2.0 [IL-8 primers: forward, 5V-CTTA-GATGTCAGT GCATAAAGACATACTCC-3V/reverse,5V-CTCAGCCCTCTTCAAAAACTTCTCCACA-3V/probe,5V-6-carboxy-fluorescein-TTGAGAGTGGACCACACTGC-6-carboxytetramethyl-rhodamine-3V (Metabion)]. The reac-
tions were incubated at 50 8C for 2 min followed by 95 8C for
10 min. The amplification profile was 95 8C for 15 s and 60
8C for 1 min for 45 cycles. For quantitative determination, a
R. Arnold et al. / Virology 330 (2004) 384–397 395
control plasmid encoding for IL-8 was simultaneously
amplified. For this purpose, amplified IL-8 cDNA,
prepared from LPS-stimulated PBMC by use of the
above-mentioned IL-8 primers, was eluted from an agarose
gel (2%) with QIAquick gel extraction kit (Qiagen). By
using Zero Blunt TOPO PCR cloning kit for sequencing
(Invitrogen), the extracted IL-8 cDNA was cloned in a
pCR 4Blunt-TOPO vector and multiplied in transformed E.
coli. Purification of plasmids was performed by using
Qiaprep Miniprep kit (Qiagen). Plasmids used in the range
of 107 to 2 � 1010 copies of IL-8 cDNA/2.5 Al were in
parallel amplified with the samples. Data were analyzed
with PE Biosystems 5700 Sequence detection system
software and obtained fluorescence signals, in a linear
range, were used as standard curve for determination of the
amount of IL-8 cDNA/2.5 Al cDNA. The primer/probe sets
for human ICAM-1 and the housekeeping gene glycer-
aldehyde-3-phosphate dehydrogenase (GAPDH) were pur-
chased as predeveloped kits from Qiagen. The comparative
CT method was used to determine relative quantities of
ICAM-1 in cell samples. The relative RNA amount was
analyzed by using the following equation: 2�DDCT were
DDCT = DCT,q � CT,cb. CT is defined as the CT value for
ICAM-1 minus the CT value for GAPDH for a given
sample; q is the unknown sample, and cb is the calibrator
(noninfected sample).
Statistics
All data were expressed as the mean F SEM. Statistical
significance analysis of data was performed using Student’s
t test (two-sided). A value of P b 0.05 was considered
significant.
Acknowledgments
This work was partially supported by grants from
Bundesministerium fqr Arbeit und Forschung (BMBF-
NBL3). The authors wish to thank Ms. Bettina Polte for
her excellent technical assistance in all the performed
experiments.
References
Arnold, R., Kfnig, W., 1996. ICAM-1 expression and low-molecular-
weight G-protein activation of human bronchial epithelial cells (A549)
infected with RSV. J. Leukocyte Biol. 60, 766–771.
Arnold, R., Humbert, B., Werchau, H., Gallati, H., Kfnig, W., 1994.
Interleukin-8, interleukin-6, and soluble tumour necrosis factor receptor
type I release from a human pulmonary epithelial cell line (A549)
exposed to respiratory syncytial virus. Immunology 82, 126–133.
Arnold, R., Rihoux, J.P., Kfnig, W., 1999. Cetirizine counter-regulates
interleukin-8 release from human epithelial cells (A549). Clin. Exp.
Med. 29, 1681–1691.
Baggiolini, M., Walz, A., Kunkel, S.L., 1989. Neutrophil-activating
peptide/interleukin-8, a novel cytokine that activates neutrophils. J.
Clin. Invest. 84, 1045–1049.
Bitko, V., Velazquez, A., Yang, L., Yang, Y.-C., Barik, S., 1997.
Transcriptional induction of multiple cytokines by human respiratory
syncytial virus requires activation of NF-nB and is inhibited by sodium
salicylate and aspirin. Virology 232, 369–378.
Bose, S., Kar, N., Maitra, R., DiDonato, J.A., Banerjee, A.K., 2003.
Temporal activation of NF-nB regulates an interferon-independent
innate antiviral response against cytoplasmic RNA viruses. Pro. Natl.
Acad. Sci. U.S.A. 100, 10890–10895.
Bukreyev, A., Whitehead, S.S., Murphy, B.R., Collins, P.L., 1997.
Recombinant respiratory syncytial virus from which the entire SH
gene has been deleted grows efficiently in cell culture and exhibits site-
specific attenuation in the respiratory tract of the mouse. J. Virol. 71,
8973–8982.
Cane, P.A., 2001. Molecular epidemiology of respiratory syncytial virus.
Rev. Med. Virol. 11, 101–116.
Cannon, M.J., 1987. Microplaque immunoperoxidase detection of infec-
tious respiratory syncytial virus in the lungs of infected mice. J. Virol.
Methods 16, 293–301.
Casola, A., Garofalo, R.P., Jamaluddin, M., Vlahopoulos, S., Brasier,
A.R., 2000. Requirement of a novel upstream response element in
respiratory syncytial virus-induced IL-8 gene expression. J. Immunol.
164, 5944–5951.
Casola, A., Garofalo, R.P., Haeberle, H., Elliot, T.F., Lin, R., Jamaluddin,
M., Brasier, A.R., 2001. Multiple cis regulatory elements control
RANTES promoter activity in alveolar epithelial cells infected with
respiratory syncytial virus. J. Virol. 75, 6428–6439.
Chini, B.A., Fiedler, M.A., Milligan, L., Hopkins, T., Stark, J.M., 1998.
Essential roles of NF-nB and C/EBP in the regulation of intercellular
adhesion molecule-1 after respiratory syncytial virus infection of human
respiratory epithelial cell cultures. J. Virol. 72, 1623–1626.
Chung, H.L., Kim, S.G., 2002. RANTES may be predictive of later
recurrent wheezing after respiratory syncytial virus bronchiolitis in
infants. Ann. Allergy, Asthma, Immunol. 88, 463–467.
Collins, P.L., Mottet, G., 1991. Post-translational processing and oligome-
rization of the fusion glycoprotein of human respiratory syncytial virus.
J. Gen. Virol. 72, 3095–3101.
Collins, P.L., Chanock, R.M., Murphy, B.R., 2001. Respiratory
syncytial virus. In: Fields, B.N., Knipe, D.M., Howley, P.M.,
Griffin, D.E., Lamb, R.A, Martin, M.A., Roizman, B., Straus,
S.E. (Eds.), Fields Virology, 4th ed. Lippincott Williams and
Wilkins, Philadelphia, pp. 1443–1485.
Domachowske, J.B., Bonville, C.A., Rosenberg, H.F., 2001. Gene
expression in epithelial cells in response to pneumovirus infection.
Respir. Res. 2, 225–233.
Escribano-Romero, E., Rawling, J., Garcia-Barreno, B., Melero, J.A., 2004.
The soluble form of human respiratory syncytial virus attachment
protein differs from the membrane-bound form in its oligomeric state
but is still capable of binding to cell surface proteoglycans. J. Virol. 78,
3524–3532.
Everard, M.L., Awarbrick, A., Wrightham, M., McIntyre, J., Dunkley, C.,
James, P.D., Sewell, H.F., Milner, A.D., 1994. Analysis of cells
obtained by bronchial lavage of infants with respiratory syncytial virus
infection. Arch. Dis. Child. 71, 428–432.
Faden, H., Hong, J.J., Ogra, P.L., 1984. Interaction of polymorphonuclear
leukocytes and viruses in humans: adherence of polymorphonuclear
leukocytes to respiratory syncytial virus-infected cells. J. Virol. 52,
16–23.
Falsey, A.R., Walsh, E.E., 2000. Respiratory syncytial virus infection in
adults. Clin. Microbiol. Rev. 13, 371–384.
Fiedler, M.A., Wernke-Dollries, K., Stark, J.M., 1996. Inhibition of viral
replication reverses respiratory syncytial virus-induced NF-nB activa-
tion and interleukin-8 gene expression in A549 cells. J. Virol. 70,
9079–9082.
Garofalo, R., Sabry, M., Jammaluddin, M., Yu, R.K., Casola, A., Ogra, P.L.,
Brasier, R., 1996. Transcriptional activation of the interleukin-8 gene by
R. Arnold et al. / Virology 330 (2004) 384–397396
respiratory syncytial virus infection in alveolar epithelial cells: nuclear
translocation of the RelA transcription factor as a mechanism producing
airway mucosal inflammation. J. Virol. 70, 8773–8781.
Garofalo, R.P., Patti, J., Hintz, K.A., Hill, V., Ogra, P.L., Welliver, R.C.,
2001. Macrophage inflammatory protein-1a (not T helper type 2
cytokines) is associated with severe forms of respiratory syncytial virus
bronchiolitis. J. Infect. Dis. 184, 393–399.
Hall, C.B., McCarthy, C.A., 2000. Respiratory syncytial virus. In: Mandell,
G.L., Benett, J.E., Dolin, R. (Eds.), Principles and Practice of Infectious
Diseases. Churchill Livingstone, Philadelphia, pp. 1782–1801.
Harrison, A.M., Bonville, C.A., Rosenberg, H.F., Domachowske, J.B.,
1999. Respiratory syncytial virus-induced chemokine expression in the
lower airways. Am. J. Respir. Crit. Care Med. 159, 1918–1924.
Haynes, L.M., Jones, L.P., Barskey, A., Anderson, L.J., Tripp, R.A., 2003.
Enhanced disease and pulmonary eosinophilia associated with formalin-
inactivated respiratory syncytial virus vaccination are linked to G
glycoprotein CX3C-CX3CR1 interaction and expression of substance P.
J. Virol. 77, 9831–9844.
Hendricks, D.A., Baradaran, K., McIntosh, K., Patterson, J.L., 1987.
Appearance of a soluble form of the G protein of respiratory syncytial
virus in fluids of infected cells. J. Gen. Virol. 68, 1705–1714.
Hendricks, D.A., McIntosh, K., Patterson, J.L., 1988. Further character-
ization of the soluble form of the G glycoprotein of respiratory syncytial
virus. J. Virol. 62, 2228–2233.
Hoffmann, E., Dittrich-Breiholz, O., Holtmann, H., Kracht, M., 2002.
Multiple control of interleukin-8 gene expression. J. Leukocyte Biol.
72, 847–855.
Johnson, P.R., Spriggs, M.K., Olmsted, R.A., Collins, P.L., 1987. The G
glycoprotein of human respiratory syncytial viruses of subgroups A and
B: extensive sequence divergence between antigenically related
proteins. Proc. Natl. Acad. Sci. U.S.A. 84, 5625–5629.
Johnson, T.R., Johnson, J.E., Roberts, S.R., Wertz, G.W., Parker,
R.A., Graham, B.S., 1998. Priming with secreted glycoprotein G of
respiratory syncytial virus (RSV) augments interleukin-5 produc-
tion and tissue eosinophilia after RSV challenge. J. Virol. 72,
2871–2880.
Koga, T., Sardina, E., Tidwell, R.M., Pelletier, M., Look, D.C., Holtzman,
M.J., 1999. Virus-inducible expression of a host chemokine gene relies
on replication-linked mRNA stabilization. Proc. Natl. Acad. Sci. U.S.A.
96, 5680–5685.
Kfnig, B., Streckert, H.J., Krusat, T., Kfnig, W., 1996. Respiratory
syncytial virus G-protein modulates cytokine release from human
peripheral blood mononuclear cells. J. Leukocyte Biol. 59, 403–406.
Levine, S., Klaiber-Franco, R., Pradiso, P.R., 1987. Demonstration that
glycoprotein G is the attachment protein of respiratory syncytial virus.
J. Gen. Virol. 68, 2521–2524.
McIntosh, K., 1997. Respiratory syncytial virus. In: Evans, A.S., Kaslow,
R.A. (Eds.), Viral Infections of Humans: Epidemiology and Control.
Plenum Publishing, New York, pp. 91–711.
Merolla, R., Rebert, N., Tsiviste, P.T., Hoffmann, P., Panuska, J.R., 1995.
Respiratory syncytial virus replication in human lung epithelial cells:
inhibition by tumor necrosis factor a and interferon h. Am. J. Respir.
Crit. Care Med. 152, 1358–1366.
Miller, N., Hutt-Fletcher, L.M., 1992. Epstein–Barr Virus enters B cells and
epithelial cells by different routes. J. Virol. 66, 3409–3414.
Noah, T.L., Ivins, S.S., Murphy, P., Kazachkova, I., Moats-Staats, B.,
Henderson, F.W., 2002. Chemokines and inflammation in the nasal
passages of infants with respiratory syncytial virus bronchiolitis. Clin.
Immunol. 104, 86–95.
Olszewska-Pazdrak, B., Casola, A., Saito, T., Alam, R., Crowe, S.E., Mei,
F., Ogra, P.L., Garofalo, R.P., 1998. Cell-specific expression of
RANTES, MCP-1, and MIP-1a by lower airway epithelial cells and
eosinophils infected with respiratory syncytial virus. J. Virol. 72,
4756–4764.
Openshaw, P.J., 1995. Immunity and immunopathology to respiratory
syncytial virus. The mouse model. Am. J. Respir. Crit. Care Med. 152,
59–62.
Patel, J.A., Kunimoto, M., Sim, T.C., Garofalo, R., Eliott, T., Baron, S.,
Ruuskanen, O., Chonmaitree, T., Ogra, P.L., Schmalstieg, F., 1995.
Interleukin-1a mediates the enhanced expression of intercellular
adhesion molecule-1 in pulmonary epithelial cells infected with
respiratory syncytial virus. Am. J. Respir. Cell Mol. Biol. 13, 602–609.
Patel, J.A., Jiang, Z., Nakajima, N., Kunimoto, M., 1998. Autocrine
regulation of interleukin-8 by interleukin-1a in respiratory syncytial
virus-infected pulmonary epithelial cells in vitro. Immunology 95,
501–506.
Ray, R., Hoft, D.F., Meyer, K., Brown, R., Lagging, L.M., Belshe, R.B.,
2001. Immunoregulatory role of secreted glycoprotein G from
respiratory syncytial virus. Virus Res. 75, 147–154.
Renard, P., Ernest, I., Houbion, A., Art, M., Le Calvez, H., Raes, M.,
Remacle, J., 2001. Development of a sensitive multi-well colorimetric
assay for active NFnB. Nucleic Acids Res. 29, e21–e25.
Roberts, S.R., Lichtenstein, D., Ball, L.A., Wertz, G.W., 1994. The
membrane-associated and secreted forms of the respiratory syncytial
virus attachment glycoprotein G are synthesized from alternative
initiation codons. J. Virol. 68, 4538–4546.
Roder, C., Krusat, T., Reimers, K., Werchau, H., 2000. Purification of
respiratory syncytial virus F and G proteins. J. Chromatogr., B: Biomed.
Sci. Appl. 737, 97–106.
Rot, A., Klieger, M., Brunner, T., Bishoff, S.C., Scall, T.J., Dahinden, C.A.,
1992. RANTES and macrophage inflammatory protein-1 induce the
migration and activation of normal human eosinophil granulocytes.
J. Exp. Med. 176, 1489–1495.
Routledge, E.G., Willcocks, M.M., Samson, A.C., Morgan, L., Scott, R.,
Anderson, J.J., Toms, G.L., 1988. The purification of four respiratory
syncytial virus proteins and their evaluation as protective agents against
experimental infection in BALB/c mice. J. Gen. Virol. 69, 293–303.
Schwarze, J., Schauer, U., 2004. Enhanced virulence, airway inflammation
and impaired lung function induced by respiratory syncytial virus
deficient in secreted G protein. Thorax 59, 517–521.
Seet, B.T., McFadden, G., 2002. Viral chemokine-binding proteins.
J. Leukocyte Biol. 72, 24–34.
Smyth, R.L., Mobbs, K.J., O’Hea, U., Ashby, D., 2002. Respiratory
syncytial virus bronchiolitis: disease severity, interleukin-8, and virus
genotype. Pediatr. Pulmonol. 33, 339–346.
Sparer, T.E., Matthews, S., Hussell, T., Rae, A.J., Garcia-Barreno, B.,
Melero, J.A., Openshaw, P.J.M., 1998. Eliminating a region of
respiratory syncytial virus attachment protein allows induction of
protective immunity without vaccine-enhanced lung eosinophilia.
J. Exp. Med. 187, 1921–1926.
Stark, J.M., Godding, V., Sedgwick, J.B., Busse, W.W., 1996. Respiratory
syncytial virus infection enhances neutrophil and eosinophil adhesion to
cultured respiratory epithelial cells. J. Immunol. 156, 4774–4782.
Tak, P.P., Firestein, G.S., 2001. NF-nB: a key role in inflammatory diseases.
J. Clin. Invest. 107, 7–11.
Tebbey, P.W., Hagen, M., Hancock, G.E., 1998. Atypic pulmonary
eosinophilia is mediated by a specific amino acid sequence of the
attachment G protein of respiratory syncytial virus. J. Exp. Med. 188,
1967–1972.
Teng, M.N., Whitehead, S.S., Collins, P.L., 2001. Contribution of the
respiratory syncytial virus G glycoprotein and its secreted and
membrane-bound forms to virus replication in vitro and in vivo.
Virology 289, 283–296.
Teran, L.M., Seminario, M.C., Shute, J.K., Papi, A., Compton, S.J., Low,
J.L., Gleich, G.J., Johnston, S.L., 1999. RANTES, macrophage-
inhibitory protein 1a and the eosinophil product major basic protein
are released into upper respiratory secretions during virus-induced
asthma exacerbations in children. J. Infect. Dis. 179, 677–681.
Thomas, L.H., Friedland, J.S., Sharland, M., Becker, S., 1998. Respiratory
syncytial virus-induced RANTES production from human bronchial
epithelial cells is dependent on nuclear factor-nB nuclear binding and is
inhibited by adenovirus-mediated expression of inhibitor of nBa.J. Immunol. 161, 1007–1016.
Tripp, R.A., Jones, L., Anderson, L.J., 2000. Respiratory syncytial virus G
R. Arnold et al. / Virology 330 (2004) 384–397 397
and/or SH glycoproteins modify CC and CXC chemokine mRNA
expression in the BALB/c mouse. J. Virol. 74, 6227–6229.
Tripp, R.A., Jones, L.P., Haynes, L.M., Zheng, H., Murphy, P.M.,
Anderson, L.J., 2001. CX3C chemokine mimicry by respiratory
syncytial virus G glycoprotein. Nat. Immunol. 2, 732–738.
Ueba, O., 1978. Respiratory syncytial virus. I. Concentration and
purification of the infectious virus. Acta Med. Okayama 32, 265–272.
Varga, S.M., Wissinger, E.L., Braciale, T.J., 2000. The attachment (G)
glycoprotein of respiratory syncytial virus contains a single dominant
epitope that elicits both Th1 and Th2 CD4+ T cell responses.
J. Immunol. 165, 6487–6495.
Wang, S.Z., Forsyth, K.D., 2000. The interaction of neutrophils with
respiratory epithelial cells in viral infection. Respirology 5, 1–9.
Wang, S.Z., Xu, H., Wraith, A., Bowden, J.J., Alpers, J.H., Forsyth, K.D.,
1998. Neutrophils induce damage to respiratory epithelial cells infected
with respiratory syncytial virus. Eur. Respir. J. 12, 612–618.