Journal of General Virology (2015), 96, 2543–2556

DOI 10.1099/vir.0.000218

CX3CR1 is an important surface molecule for respiratory syncytial virus infection in human airway epithelial cells Tatiana Chirkova,1 Songbai Lin,13 Antonius G. P. Oomens,2 Kelsey A. Gaston,1 Seyhan Boyoglu-Barnum,14 Jia Meng,1 Christopher C. Stobart,1 Calvin U. Cotton,3 Tina V. Hartert,4 Martin L. Moore,1 Assem G. Ziady13 and Larry J. Anderson1 Correspondence

1

Department of Pediatrics and Children’s Healthcare of Atlanta, Emory University, Atlanta, Georgia, USA

2

Center for Veterinary Health Sciences, Oklahoma State University, Stillwater, Oklahoma, USA

3

Division of Pediatric Pulmonology, Case Western University, Cleveland, Ohio, USA

4

Division of Allergy, Pulmonary, and Critical Care Medicine, Department of Medicine and Vanderbilt Center for Asthma and Environmental Health Sciences Research, Vanderbilt University, Nashville, Tennessee, USA

Larry J. Anderson [email protected]

Received 17 February 2015 Accepted 11 June 2015

Respiratory syncytial virus (RSV) is a major cause of severe pneumonia and bronchiolitis in infants and young children, and causes disease throughout life. Understanding the biology of infection, including virus binding to the cell surface, should help develop antiviral drugs or vaccines. The RSV F and G glycoproteins bind cell surface heparin sulfate proteoglycans (HSPGs) through heparin-binding domains. The G protein also has a CX3C chemokine motif which binds to the fractalkine receptor CX3CR1. G protein binding to CX3CR1 is not important for infection of immortalized cell lines, but reportedly is so for primary human airway epithelial cells (HAECs), the primary site for human infection. We studied the role of CX3CR1 in RSV infection with CX3CR1-transfected cell lines and HAECs with variable percentages of CX3CR1-expressing cells, and the effect of anti-CX3CR1 antibodies or a mutation in the RSV CX3C motif. Immortalized cells lacking HSPGs had low RSV binding and infection, which was increased markedly by CX3CR1 transfection. CX3CR1 was expressed primarily on ciliated cells, and ,50 % of RSV-infected cells in HAECs were CX3CR1+. HAECs with more CX3CR1-expressing cells had a proportional increase in RSV infection. Blocking G binding to CX3CR1 with anti-CX3CR1 antibody or a mutation in the CX3C motif significantly decreased RSV infection in HAECs. The kinetics of cytokine production suggested that the RSV/CX3CR1 interaction induced RANTES (regulated on activation normal T-cell expressed and secreted protein), IL-8 and fractalkine production, whilst it downregulated IL-15, IL1-RA and monocyte chemotactic protein-1. Thus, the RSV G protein/CX3CR1 interaction is likely important in infection and infection-induced responses of the airway epithelium, the primary site of human infection.

INTRODUCTION Respiratory syncytial virus (RSV) is a major cause of severe pneumonia and bronchiolitis in infants and young children, and causes repeated respiratory infections 3Present address: Department of Paediatrics, Cincinnati Children’s Hospital, Cincinnati, Ohio, USA. 4Present address: Division of Viral Diseases, National Center for Immunization and Respiratory Diseases, CDC, Atlanta, Georgia, USA.

000218 G 2015 The Authors

Printed in Great Britain

throughout life (Falsey et al., 2005; Hall et al., 2009; Nair et al., 2010; Shay et al., 1999). The two RSV surface proteins, F and G, are key to the biology of infection, and are involved in RSV attachment and entry into cells, viral assembly, and induction of protective host immune responses. Both proteins bind to the cell surface through heparin-binding domains interacting with glycosaminoglycans (GAGs), particularly heparin sulfate proteoglycans (HSPGs) (Feldman et al., 1999, 2000). The F protein 2543

T. Chirkova and others

initiates virus–cell fusion and reportedly also binds to cell surface nucleolin (Tayyari et al., 2011). The G protein is not required for replication (Karron et al., 1997), but facilitates infection, and its presence increases virus infectivity in vitro and in vivo (Heminway et al., 1994). RSV binding to the cell surface appears to be most efficient through the G protein (Techaarpornkul et al., 2002), possibly because the G protein has higher affinity to GAGs, i.e. HSPG, than the F protein (Feldman et al., 2001). The G protein also has a fractalkine-like CX3C chemokine motif located at aa 182–186 and can bind to the fractalkine receptor CX3CR1 (Tripp et al., 2001). CX3CR1, a G-coupled transmembrane chemokine receptor, previously identified as GPR13 or V28 (Combadiere et al., 1998b; Imai et al., 1997), is expressed in various cell types, including epithelial (Jamieson et al., 2008), microglia (Harrison et al., 1998), smooth muscle (Perros et al., 2007) and different immune cells, such as monocytic, dendritic, NK- and T-cells (Bar-On et al., 2010; Imai et al., 1997; Nishimura et al., 2002). CX3CR1 expression was also documented in human and mouse lung tissues (Combadiere et al., 1998a; Song et al., 2008; Suzuki et al., 2012), although in mouse lungs the majority of CX3CR1+ cells were found to be of leukocyte origin (Kim et al., 2011). In addition, it has been shown recently that G protein binding to CX3CR1 is an important contributor to RSV infection of human primary airway epithelial cells, unlike conventional cell lines such as HeLa or Vero cells, which have abundant GAGs on their surface, but very low or no expression of CX3CR1 (Johnson et al., 2013). As airway epithelial cells are the primary site of RSV infection in humans, understanding the role of the G protein/CX3CR1 interaction in infection and host response in these cells is important to understanding human RSV disease pathogenesis. It has been shown that RSV infection of human airway epithelial cells (HAECs) induces strong cytokine/ chemokine responses, including production of IL-1b, IL-6, IL-8, IP-10, macrophage inflammatory protein (MIP)-1a, monocyte chemotactic protein (MCP)-1 and RANTES (regulated on activation normal T-cell expressed and secreted) (Fonceca et al., 2012; Ioannidis et al., 2012; Olszewska-Pazdrak et al., 1998; Oshansky et al., 2010; Tristram et al., 1998; Villenave et al., 2011), and these cytokines and released viral proteins influence the host immune response to RSV (Chirkova et al., 2013; Ioannidis et al., 2012; McNamara et al., 2013; Qin et al., 2011). RSV G protein has been shown to impair innate and memory immune responses, such as suppressing production of IFN-b (Shingai et al., 2008), lymphoproliferation of T-cells (Ray et al., 2001), and responsiveness of monocytes/macrophages and dendritic cells (Johnson et al., 2012; Polack et al., 2005). Some of these effects on the immune response were found to be a consequence of RSV G protein interaction with CX3CR1, including suppressed T-cell trafficking and cytotoxicity in the lungs of RSV-infected mice (Harcourt et al., 2006), and enhanced pulmonary inflammation in RSV-challenged, formalin-inactivated RSV-vaccinated mice (Haynes et al., 2544

2003). A decrease in respiratory rates in mice has also been linked to G binding to CX3CR1 (Tripp et al., 2003). In our previous work using an in vitro model of RSV infection, we showed RSV G protein/CX3CR1 interaction suppressed some human dendritic and memory T-cell responses (Chirkova et al., 2013). In our present study, we explored the role of the G/CX3CR1 interaction in RSV infection of primary HAECs and demonstrated that unlike in immortalized cell lines, in HAECs this interaction is important for infection and cellular response to RSV.

RESULTS RSV binding to CX3CR1 is HSPG-independent, and increases virus internalization and infectivity To study the role of CX3CR1 in HSPG-independent RSV binding and internalization, HSPG-expressing (CHO-K1 and BEAS-2b) and HSPG-deficient CHO-K1DHSPG (pgsD-677) cell lines were transfected with human CX3CR1 (Fig. 1a). About 25–35 % of cells transfected and expressed moderate- to high-density surface CX3CR1. Endogenous CX3CR1 expression was detected in 1.2– 4.8 % of mock-transfected cells. A RSV binding assay showed that, as had been reported previously (Martı´nez & Melero, 2000), lack of HSPG on the cell surface significantly decreased RSV binding (Fig. 1b). Transfection with human CX3CR1 substantially increased the percentage of CHO-K1DHSPG cells with bound virus, from 1.7 % in mock-transfected to 11 % in CX3CR1transfected cells. To confirm the HSPG-independent RSV binding to CX3CR1 in HAECs, CX3CR1-transfected BEAS-2b cells were treated with heparinase I before the assay (Fig. 1c). Treatment with heparinase I has been shown to be the most efficient way to cleave heparin sulfate chains from cell surface HSPG (Martı´nez & Melero, 2000). In these experiments, in addition to RSV A2, we included two relatively low-passage group A strains and two group B strains. For all five strains, heparinase I dramatically decreased RSV binding to mock-transfected BEAS-2b. CX3CR1 transfection significantly increased RSV binding to heparinase-treated cells. The strain-specific increase of RSV binding with CX3CR1-transfected heparinase-I-treated cells correlated with binding in untreated cells, with 37–58 % of binding activity being restored. A virus internalization assay was performed with carboxyfluorescein diacetate succinimidyl ester (CFDASE)-labelled RSV. Upon internalization, the intracellular esterases cleave diacetate groups in CFDASE, thus converting it to the fluorescent carboxyfluorescein succinimidyl ester (CFSE) form (Drobni et al., 2003). Cells exposed to RSVCFDASE at 4 uC (Fig. 1d, dotted lines) exhibited low fluorescence (similar to levels at 0 h), whilst cells exposed to the virus at 37 uC had a progressive increase in percentage of CFSE-positive cells from 0 to 4 h post-infection (p.i.) Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs CX3CR1 transfection

(a) 50

RSV binding to CHO-K1

(b) 30

CHO-K1

Mock

*

CHO-K1ΔHSPG 40

25 RSV 20

30 0

1

2

10

3

10

10

4

10

10

5

BEAS-2b

20

RSV+ cells (%)

CX3CR1+ cells (%)

Mock

hCX3CR1-transfected

10

15

* 10

5

0

0

CHO-K1 ΔHSPG

CHO-K1

BEAS-2b

1

2

10

3

10

4

10

0

5

10

10

CHO-K1

CX3CR1

Untreated

hCX3CR1

CHO-K1ΔHSPG RSV binding to BEAS-2b cells

(c)

RSV strains:

60

RSV internalization assay

(d) 14

A/A2

*

*

12

A/1997/12-35

B/WV/14617/85

40

B/8/60

*

* * *

20

*

* *

*

10 CFSEhigh cells (%)

RSV+ cells (%)

A/TN/13/2-26

* 8

*

6

* 4

* *

2 hCX3CR1-tr Heparinase Ι RSV

– – –

– – +

– + +

+ + +

*Δ

0

1

2

4

Time (h)

Fig. 1. Association of RSV binding and internalization with presence of HSPG and CX3CR1 on the cell surface. HSPGexpressing cells (CHO-K1 and BEAS-2b) and HSPG-deficient cells (CHO-K1DHSPG or BEAS-2b cells pre-treated with heparinase I) transfected or mock-transfected with human CX3CR1 were exposed to RSV. (a) Percentage of CX3CR1expressing cells after CX3CR1 or mock transfection. CHO-K1 and CHO-K1DHSPG FACS plot: grey area, CHO-K1; dotted line, untreated CHO-K1DHSPG; dashed line, vector-transfected CHO-K1DHSPG; solid line, CX3CR1-transfected CHOK1DHSPG. BEAS-2b FACS plot: grey area, untreated; dotted line, vector-transfected; solid line, CX3CR1-transfected cells. (b) Percentage of cells with bound RSV after exposure to virus at m.o.i. 5. *P,0.05 versus pgsD-677; unpaired t-test. (c) Percentage of cells with bound RSV in BEAS-2b cells transfected with human CX3CR1 (hCX3CR1-tr), treated with heparinase I and exposed to different RSV strains (exposure dose equal to 15 mg RSV G protein, m.o.i. 0.05–4.5). *P,0.05 versus heparinase-I-treated cells; unpaired t-test. (d) Cells were exposed to CFDASE-labelled RSV for different times at 37 (solid lines) or 4 8C (dotted line) as a negative control. The percentage of CFSE+ cells (cells with internalized viral particles) was measured by flow cytometry. CHO-K1, X; CX3CR1-transfected CHO-K1, #; CHO-K1DHSPG, m; CX3CR1-transfected CHO-K1DHSPG, D. *P,0.05 versus 4 8C exposure, DP,0.05 versus untransfected cells.

http://vir.sgmjournals.org

2545

T. Chirkova and others

(Fig. 1d, solid lines). Transfection with human CX3CR1 significantly increased internalization of RSV in CHOK1DHSPG cells. Characterization of primary human airway epithelial cultures with differential percentage of CX3CR11 cells Seven different donor human airway epithelial cultures, two cystic fibrosis (CF) and five non-CF, were tested in the present study. To account for possible variations in the distribution of major cells types (i.e. ciliated, goblet and basal cells) and surface molecules that might affect RSV infection of HAECs, we determined the percentage of ciliated cells (b-tubulin IV+ or FoxJ1+), secretory cells (mucin-5AC+ or positive by Jacalin lectin binding) and basal cells (CD151+), and expression of HSPG (syndecan-1) and CX3CR1. We found HAECs from different donors had similar distributions of the major airway epithelial cell subtypes. The majority of both CF and non-CF HAECs (*74–85 %) had characteristics of ciliated cells, i.e. expressed both b-tubulin IV, a constituent of motile cilia (Jensen-Smith et al., 2003; Million et al., 1999), and FoxJ1, a forkhead box (f-box)

transcription factor required for ciliogenesis and cilia assembly (Blatt et al., 1999; You et al., 2004). It has been noted previously that ciliated cells are the major target for RSV infection (Zhang et al., 2002). About 13–27 % of HAECs had a secretory phenotype based on binding of Jacalin, a plant lectin which binds to O-glycosylated proteins such as mucins (Tachibana et al., 2006). Some of these secretory cells (9–18 %) also stained positive for mucin5AC. A very small percentage of cells (0.15–2.5 %) were positive for CD151, a tetraspanin specifically expressed on keratinocytes, including airway basal cells (Sincock et al., 1997). Thus, CF and non-CF HAEC cultures were composed mainly of fully differentiated airway epithelial cells of a ciliated or secretory phenotype. HAECs from different donors also had similar levels of surface HSPG expression as indicated by levels of syndecan-1 (Fig. 2a), a transmembrane HSPG family member and a major source of heparin sulfates on the cell surface (Bernfield et al., 1999; Park et al., 2000). The percentage of CX3CR1+ cells was, however, different amongst donors. CX3CR1 was predominantly detected on the surface of ciliated cells (FoxJ1+ subset; Fig. 2c), and the percentage of CX3CR1+ cells varied significantly between donors (Fig. 2b; two representative donors). We divided HAEC donors into two

(a)

HAECs CX3CR1high

(b)

HAECs CX3CR1low

15.2%

200K SSC

150K

105 2.7% 104 FoxJ1

200K

150K

100K

100K

50K

50K

CX3CR1 in ciliated cells

(c)

250K

250K

SSC

HAECs isotype control

HAECs CX3CR1low

HAECs CX3CR1high

103 102 101

101 102

103 104 CX3CR1

105

101

102

103 104 CX3CR1

105

101 102

103 104 CX3CR1

105

Fig. 2. Expression of syndecan-1 and CX3CR1 on HAECs. Data presented from one CX3CR1high donor (CF) and one CX3CR1low donor. (a) HAECs grown at the air/liquid interface were fixed on the membrane inserts, and stained for syndecan-1 (red) and nuclei (DAPI, blue). Bar, 100 mm. (b) Percentages of CX3CR1+ cells from HAEC donors. (c) CX3CR1high HAECs stained for surface CX3CR1 and intracellular FoxJ1. 2546

Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs

groups, i.e. CX3CR1high (two CF and three non-CF donors with the percentage of CX3CR1+ cells ranging from 10 to 26 %) and CX3CR1low (two non-CF donors with 2–7 % CX3CR1-expressing cells). With similar levels of surface HSPG and different percentages of CX3CR1+ cells, CX3CR1high and CX3CR1low HAECs provide an opportunity to study the effect of CX3CR1 on RSV infection in HAECs. Percentage of CX3CR1 cells correlates with RSV infection in HAECs The level of RSV infection in HAECs increased in a dosedependent manner (Fig. 3a) without obvious signs of cytopathology (data not shown). RSV was only detected in ciliated cells and 62–76 % of RSV+ cells were also CX3CR1+ (Fig. 3c). Imaging flow cytometry showed localization of CX3CR1 and RSV antigens on the apical surface of the ciliated cells, i.e. the cell surface with cilia. RSV antigens have previously been noted to be located close to the apical surface in HAECs (Zhang et al., 2002). These data correlated with the finding that infectious progeny virus (Fig. 3b) and viral RNA (not shown) were detected only in the apical washes of infected HAECs. CX3CR1high HAECs had a higher percentage of RSV-infected cells (Fig. 4a), a higher relative amount of viral RNA (data not shown) and more cell death than CX3CR1low HAECs (Fig. 4b). However, the percentages of RSV+ cells within the subset of CX3CR1+ cells were similar between both HAEC cultures (25–30 %; Fig. 4a). Overall, in all seven tested HAEC donors, the percentages of RSV-infected cells correlated with percentages of CX3CR1-expressing cells with a correlation coefficient of 0.76 (Fig. 4c). Alteration of the RSV G protein CX3C motif decreases virus binding and replication To confirm the role of the RSV G protein CX3C motif in binding and infection of HAECs, we assessed the effects of the virus with a mutation in the CX3C motif (RSV CX4C) and anti-CX3CR1 antibody on binding and infection. In the RSV attachment assay, anti-syndecan-1 antibody treatment showed the greatest decrease in RSV binding to HAECs (Fig. 5). Anti-CX3CR1 antibody treatment also decreased RSV binding, although less than anti-syndecan-1 treatment. CX4C showed impaired attachment to HAECs and treatment with anti-CX3CR1 antibody did not significantly change the binding of CX4C. Note that neither blocking with anti-CX3CR1 nor anti-syndecan-1 antibody eliminated all binding. These findings suggested that both HSPG and CX3CR1 contributed to RSV attachment to HAECs, but may not have been the only binding molecules for the virus. Analysis of the kinetics of RSV infection in CX3CR1high HAECs at day 2–8 p.i. showed significantly less replication with the CX4C compared with the CX3C virus (Fig. 6a). http://vir.sgmjournals.org

We found that most of the difference in CX3C and CX4C replication occurred in the CX3CR1+ but not the CX3CR1– population (Fig. 6a, compare panels 2 and 3). The percentage of CX3CR1+ cells remained stable through days 2– 6 and decreased at day 8 p.i, but the difference was not statistically significant. Compared with mock and CX4C samples, cells infected with CX3C virus had a slight increase in the percentage of CX3CR1+ cells on day 4 and 6 p.i., but the difference was not statistically significant (data not shown). In addition, the percentage of RSV+ cells in CX3CR1– cells was similar for the two viruses at days 2, 6 or 8 p.i. The correlation coefficient between RSV and CX3CR1+ cells was 0.20 for the CX4C virus and 0.76 for the CX3C virus. The percentage of RSV+ cells for the CX4C virus did not differ significantly for CX3CR1high and CX3CR1low HAEC cultures (Fig. 6b). In addition, the percentage of RSV+ cells in CX3C-infected but not CX4C-infected cultures was significantly decreased at day 4 p.i., by blocking CX3CR1 with a polyclonal antiCX3CR1 antibody (Fig. 6c). RSV G protein CX3C motif affects cytokine responses in HAECs CX3CR1 interaction with its ligand fractalkine induces downstream signalling leading to NFkB activation followed by production of fractalkine itself and some other proinflammatory cytokines (Chandrasekar et al., 2003). As the RSV G protein CX3C motif has structural similarity to the receptor-binding site of fractalkine and has been shown to mimic other fractalkine functions (Tripp et al., 2001), we also looked at the effect of G protein binding to CX3CR1 on the production of certain cytokines in airway epithelial cells. RSV infection induced multiple cytokines and chemokines in HAECs (Table 1). Consistent with their inhibition by the viral NS1 and NS2 proteins, neither CX3C or CX4C RSV induced IFN-a production in the HAECs. In contrast, both viruses induced significant production of IFN-l1 (IL-29) and IFN-l2 (IL-28a). Overall, production of cytokines was higher for CX3C than CX4C RSV. Some cytokines [e.g. IL-8, MIG (monokine induced by IFN-c), fractalkine] were released only during CX3C virus infection, suggesting that RSV/CX3CR1 interaction induced their production. This conclusion was supported by a significant decrease of the production of these cytokines after anti-CX3CR1 antibody treatment of HAECs before CX3C RSV infection (Table 1). For other cytokines and chemokines, the higher levels in CX3C-infected cultures appeared to be related to greater replication of CX3C than CX4C RSV in the HAECs (Fig. 6a). To account for the potential effect of virus replication on differences in cytokine production, we calculated the ratio (RCV) between the rate of increase in cytokine production and rate of increase in virus replication over time (see Methods). The RCV values for CX3C and CX4C viruses (Fig. 7) suggested that RANTES was likely induced more effectively by 2547

T. Chirkova and others

(a)

(b)

RSV replication, day 4 p.i. 9

5.0

Linear trend P=0.0014

Linear trend P=0.0018 *

Viral titer [log10(TCID50 ml–1)]

8 * 7 RSV+ cells (%)

RSV inapical washes, day 3 p.i.

6

*

5 4 3

*

4.5

4.0

3.5

3.0

2 2.5 1 0

Mock 0.6 m.o.i. 4 m.o.i. 6 m.o.i. UV-RSV

2.0

Mock

0.6 m.o.i. 4 m.o.i. 6 m.o.i. UV-RSV

(c) R1

R2

R3

R4

R2

FoxJ1

104 103 102 101 R4 102

R3 103

104

105

106

Merged

CX3CR1

105 R1

CX3CR1

RSV

FoxJ1

RSV

Fig. 3. Characterization of RSV infection in HAECs. CX3CR1high HAECs (one CF and one non-CF) were infected with RSV at the indicated m.o.i. Cell media (Mock) and UV-inactivated RSV (UV-RSV) were used as negative control. (a) Percentage of RSV-infected cells. *P,0.05; one-way ANOVA. (b) Viral titre in the apical washes of HAECs. *P,0.05; one-way ANOVA. (c) Regular and imaging flow cytometry of RSVs infected HAECs at day 8 p.i., stained for ciliated cell marker (FoxJ1), CX3CR1 and RSV F protein. Dot-plot shows cell subsets gated in ciliated (FoxJ1+) population: un-infected cells expressing CX3CR1 (CX3CR1+RSV2, R1), infected cells expressing CX3CR1 (CXCR1+RSV+, R2), infected cells without CX3CR1 expression (CX3CR12RSV+, R3) and un-infected cells (CX3CR12RSV2, R4). Imaging flow cytometry panels show individual cell images of the same subsets: green, RSV F protein; red, CX3CR1, purple, FoxJ1.

the CX3C virus (i.e. higher RCV value for CX3C than CX4C virus), whilst IL-1RA, IL-15 and MCP-1 were likely more effectively induced by CX4C RSV (higher RCV values for CX4C). 2548

DISCUSSION Our data show that in vitro the RSV G protein interaction with CX3CR1 on the cell surface of HAECs is more Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs

RSV+ within CX3CR1+ subset

RSV+ within total HAECs

(a) 8

30 ∗ SSC

130k 100k

25

70k

6

40k

50k 100k 150k 200k

FSC 4

250k

SSC-H

2

CX3CR1+

RSV+ cells (%)

RSV+ cells (%)

10k

20

15

10

200k 150k

5

100k 50k

0

101 102

HAECs CX3CR1high 16±2% CX3CR1+

105

0 HAECs CX3CR1high 16±2% CX3CR1+

CX3CR1

HAECs CX3CR1low 4±1% CX3CR1+

(b)

103 104

HAECs CX3CR1low 4±1% CX3CR1+

(c) 500

Cell death

Correlation between percentage of CX3CR1+ cells and RSV infection in HAECs 10

400

300 RSV+ cells (%)

Live/dead

8

200

6

4

100

CF HAECs 2

Non-CF HAECs

0 HAECs HAECs CX3CR1high CX3CR1low 13±1% CX3CR1+ 3±0.5% CX3CR1+

0 0

10

20

30

CX3CR1+ cells (%)

Fig. 4. Levels of CX3CR1+ cells correlate with RSV infection in HAECs. CX3CR1high (two CF and three non-CF) and CX3CR1low (two non-CF) HAECs were infected with RSV at m.o.i. 2 (grey bars) or mock-infected (white bars) and harvested from inserts on day 8 p.i. (a) Percentage of RSV-infected cells within total cell population (left panel) and CX3CR1+ subset (right panel). (b) Mean fluorescent intensity (MFI) of Fixable Viability Dye for three HAEC donors: one CF, one non-CF (CX3CR1high) and one non-CF (CX3CR1low). (c) Correlation between percentages of CX3CR1+ cells and RSV infection. Data represent individual insert measurements of HAECs (data from two to four inserts per donor).

important to the pathogenesis of human RSV than anticipated from its role in immortalized cell lines used to study RSV infection. The RSV G protein has been shown to bind to cells through HSPGs in epithelial cell lines as well as to http://vir.sgmjournals.org

bind to CX3CR1 in CX3CR1-transfected cell lines (Feldman et al., 1999; Tripp et al., 2001). In in vivo studies in animal models, the G protein/CX3CR1 interaction has been shown to affect a number of host immune responses 2549

T. Chirkova and others

(a)

(b) 0.6 CX3C

Mock

CX4C

200k +

RSV 0.9%

RSV 2.5%

100k

∗ ∗

0.4

+

+

RSV 0.03%

150k

50k

∗ A490/640

10

2

103

104

105 106

CX3C+ anti-Syndecan

102 103

104

105 106

102 103

CX3C+ anti-CX3CR1

104

105 106

CX4C+ anti-CX3CR1

200k +

0.2

+

+

RSV 0.7%

150k

RSV 1.1%

RSV 1.3%

100k 50k

102

103

104

105 106

102 103

4

10

6 105 10

102

103

104

105 106

CX4C+antiCX3CR1

CX4C

CX3C+antiCX3CR1

CX3C+antiSyndecan

CX3C

HEp-2

Mock

0

Fig. 5. Assessment of RSV binding to the HAECs. CX3CR1high HAECs were exposed to RSV at m.o.i. 10 for 1 h at 4 8C. Some HAECs were pre-treated with anti-CX3CR1 or anti-syndecan-1 antibody before incubation with RSV. (a) RSV-specific ELISA. HEp-2 cells grown on membrane inserts and exposed to CX3C virus (black bar, HEp-2) were used as a positive control for RSV binding. *P,0.05; Mann–Whitney test. (b) Percentage of RSV+ cells measured by flow cytometry.

to infection (Harcourt et al., 2006; Haynes et al., 2003), likely through immune cells expressing high levels of CX3CR1 (Imai et al., 1997). The importance of G protein/CX3CR1 interactions has, however, not yet been determined for human RSV infection. Recently, it was noted that human primary airway epithelial cells express CX3CR1 and that G protein binding to CX3CR1 increases RSV replication in these cells (Johnson et al., 2013). Our data support and expand these observations. We show that G protein/ CX3CR1 interaction contributes to *50 % of RSV infectivity in HAECs, whilst other molecules (including HSPGs) contribute to the remaining RSV infectivity in these cells. This conclusion is based on the correlation between levels of CX3CR1+ cells and RSV infectivity, and the marked decrease in RSV infection of HAECs when RSV/CX3CR1 interaction is blocked by a mutated CX3C motif or by anti-CX3CR1 antibody. These findings are consistent with CX3CR1 being one of several RSV receptors on HAECs. Cell surface HSPGs are defined as syndecans or glypicans based on their protein core (Bernfield et al., 1999; Park et al., 2000). HSPGs contain heparin sulfate chains to which RSV binds (Hallak et al., 2000). Amongst the several HSPGs, syndecan-1 is the one primarily expressed on epithelial cells, including immortalized HAEC lines such as 2550

A549 (Hayashida et al., 2009; Pruessmeyer et al., 2010) and BEAS-2b (Chen et al., 2009), and has been shown to support RSV infection (Park et al., 2000). This suggested to us that syndecan-1 is likely involved in the binding and infection of RSV in airway epithelial cells. Although some authors have reported the lack of HSPGs on the surface of respiratory epithelium (Duan et al., 1998; Keswani et al., 2012), others have noted the presence of syndecan-1 on human normal bronchial epithelial cells (Zhang et al., 2011). Zhang et al. (2011) noted that only well-differentiated HAECs exhibited abundant expression of syndecan-1. We detected expression of syndecan-1 on HAECs, which was associated with RSV binding (Figs 2 and 5). It has been shown also that airway epithelial cells shed syndecan-1 upon injury and bacterial infection (Chen et al., 2009; Li et al., 2002; Park et al., 2001; Popova et al., 2006). It is possible that the contribution that CX3CR1 plays in RSV infection of HAECs is even greater with bacterial co-infection. We noticed that the percentage of CX3CR1+ cells varied amongst HAEC donors. This difference could be genetically determined, or result from a chronic pulmonary inflammation and lower levels of oxygenation known to affect expression of CX3CR1 in lung tissues (Hasegawa et al., 2005; McComb et al., 2008; Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs

8

RSV+, percentage of CX3CR1– 8

45 *Δ

Mock CX3C

*Δ

Mock

Mock

40

CX3C

CX4C

6 RSV+ cells (%)

RSV+, percentage of CX3CR1+

RSV+, percentage of total HAECs

(a)

CX4C

35

CX3C

*Δ

30

*Δ

25

*Δ

4 20 *Δ

4

*Δ

*Δ

*

*

15

2

10

2

*Δ

*

5 0

0 4

2

8

6

0 2

Time (day)

4 6 Time (day)

RSV replication, day 8 p.i

(b)

*Δ

2

8

0.060

4 6 Time (day)

6 *Δ

Relative amount of RNA

Mock CX3C

5

0.055

*Δ

8

CX3CR1 blocking, day 4 p.i

(c)

Percentage of RSV+ cells

8

6

CX4C

6

CX4C

4 * 0.045

RSV+ cells (%)

*

1/Ct

RSV+ cells (%)

4 0.050

3

2 2 0.040

0

CX3C

CX4C

HAECs CX3CR1high 18±2% CX3CR1+

CX3C

CX4C

HAECs CX3CR1low 4±1% CX3CR1+

0.035

1

0

Untreated

Anti-CX3CR1

HAECs CX3CR1high 18±2% CX3CR1+

Fig. 6. Altering the CX3C motif in the RSV G protein decreases RSV replication in HAECs. CX3CR1high and CX3CR1low HAECs were infected with RSV with an intact CX3C motif (CX3C) or mutated motif (CX4C) and maintained up to day 8 p.i. (a) Percentage of RSV-infected cells in CX3CR1high HAECs on days 2–8 p.i., measured within total cells, and CX3CR1+ and CX3CR1– subsets. *P,0.05 versus Mock, DP,0.05 versus CX4C, unpaired t-test, two-way ANOVA. (b) Percentage of RSV-infected cells and relative amount of viral RNA [real-time PCR, inverse cycle threshold (Ct) values]. *P,0.05 versus CX4C, DP,0.05 versus CX3CR1low HAECs, unpaired t-test. Data from six HAEC donors: one CF, three non-CF CX3CR1high and two non-CF CX3CR1low. (c) Percentage of RSV-infected cells in CX3CR1high HAECs pre-treated with anti-CX3CR1 antibody. *P,0.05 versus CX4C, DP,0.05 versus CX3C pre-treated with anti-CX3CR1 antibody; Mann–Whitney test. Data from one CF and three non-CF HAECs.

http://vir.sgmjournals.org

2551

T. Chirkova and others

Table 1. Cytokine and chemokine responses to RSV infection in HAECs Concentration at day 4 p.i. (pg ml21)

Cytokine

IL-1RA IL-6 IL-8 IL-15 Fractalkine IP-10 MIG MIP-1b MCP-1 RANTES IFN-a IFN-l1 IFN-l2

Mock

CX3C RSV

CX4C RSV

CX3C RSV+anti-CX3CR1 antibody

44¡5 46¡3 4859¡293 12¡2 66¡5 11¡2 7¡2 14¡2 120¡20 13¡1 27¡5 5¡1 59¡3

1600¡453* 105¡7* 18370¡1929* 739¡64* 243¡22* 3925¡453* 76¡11* 42¡2* 305¡31* 42¡3* 27¡5 683¡82* 458¡118*

375¡32 57¡8D 5210¡383D 182¡8D 66¡5D 180¡31D 7¡2D 23¡2D 178¡12D 16¡1D 27¡5 82¡3D 134¡12

794¡224 66¡2D 7653¡865D 373¡89*D 93¡13D 697¡204*D 10¡2D 19¡5D 218¡20 15¡2D 24¡3 160¡25D 191¡40

*P,0.05 versus mock. DP,0.05 versus CX3C; one-way ANOVA, Bonferroni’s multiple comparison test.

2552

4

CX3C CX4C

RCV

3

2 ∗

∗

∗

1

∗

∗

∗

∗

IL8 IL1 RA 5 N TE S M IG M IP -1 M β C P1 IP -1 IF 0 N -λ 2 IF N -λ 1

A

e

1R

in ta l

IL-

IL6

0 ac

Our data also suggest that the G protein/CX3CR1 interaction, similar to the fractalkine/CX3CR1 interaction, affects the cellular inflammatory response to infection. Chandrasekar et al. (2003) showed that CX3CR1, being a G-coupled transmembrane chemokine receptor, conducts the signal through a cascade of kinases, including phosphoinositide 3-kinase, phosphoinositide-dependent kinase 1, Akt and IkB kinase, that leads to NFkB activation and fractalkine production (Chandrasekar et al., 2003). Interestingly, the fractalkine/CX3CR1 interaction has been linked to both pro- and anti-inflammatory effects depending on the cell type. For example, stimulation of CX3CR1 induced production of IL-1b and TNF-a in microglia cells and neurons (Johnston et al., 2004; Zhang et al., 2012), whilst loss of CX3CR1 intensified production of TNF-a, MCP-1, MIP1a and RANTES in mouse liver macrophages (Aoyama et al., 2010). Our results showed that only WT RSV, but not CX4C virus, induced production of fractalkine, suggesting that RSV G protein works as a ligand-agonist for the epithelial CX3CR1 in HAECs (Table 1). The higher replication of CX3C in HAECs confounds the interpretation of differences in cytokine production compared with the CX4C virus. We addressed this concern by relating the rate of cytokine increase to the rate of virus replication (Fig. 7). This analysis showed qualitative differences likely related to the G protein/CX3CR1 interaction and not just virus replication; this difference merits further study. Our observations are consistent with Oshansky et al. (2010), who reported that purified RSV G protein elicited production of IL-1a and RANTES, and virus with a deleted G gene had lower IL-8 and higher IP-10 and MCP-1 levels in HAECs. Elevation levels of IL-1RA, IL-8, MCP-1, MIPs, IP-10 and RANTES have also been reported in nasopharyngeal aspirates of

infants with RSV bronchiolitis, and higher levels correlated with the severity of disease (Bermejo-Martin et al., 2007; Welliver et al., 2002). Unlike our previous studies with A549 cells in which the CX3C virus produced significantly less IFN type III than the CX4C RSV (Chirkova et al., 2013), production of IFN-l in these HAECs seemed to be independent of the G protein CX3C/CX3CR1 interaction.

Fr

Perros et al., 2007; Rimaniol et al., 2003; Song et al., 2008) or other factors.

Fig. 7. Cytokine responses in RSV-infected HAECs. CX3CR1high HAECs were infected with RSV with an intact CX3C motif (CX3C) or mutated motif (CX4C). Graph represents ratios (RCV) between cytokine production curve slope and viral replication (amount of viral RNA) curve slope. The slopes were calculated for the day 2–4 p.i. segment when both cytokine production and viral replication represented stable exponential growth. *P,0.05 versus CX4C; unpaired t-test and two-way ANOVA. Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs

As with infection of cells, the role of CX3C in type III IFN production may vary with cell type. In conclusion, RSV G protein interaction with CX3CR1, expressed on airway ciliated cells, facilitates virus attachment and infection of HAECs, and also modulates cell responses to infection. These findings, in conjunction with previous reports, suggest that CX3CR1 is one of several receptors for RSV, and along with GAGs and other molecules, such as nucleolin, provides RSV with multiple ways of attachment and infection of respiratory epithelium. As airway epithelial cells are an important site for human RSV replication, the RSV G protein/CX3CR1 interaction is likely an important contributor to human RSV infection and disease.

METHODS Cell cultures and transfection. Primary human bronchial epithelial cells from CF and normal patients were purchased from ChanTest and Lonza Group or kindly provided by Dr C. U. Cotton (Case Western Reserve University, Cleveland, OH, USA), and cultured as described previously (Chen et al., 2008; Ziady et al., 2012). Briefly, cells were grown on Costar Transwell or Snapwell inserts (Corning) until confluent, then transferred in the air/liquid interface where cells were maintained in primary airway medium containing 2 % Ultroser G (Pall Biosepra) or in Lonza formulated differentiation media until they differentiated and had resistance measurements w500 V.

HEp-2, CHO-K1, pgsD-677 and BEAS-2b (all ATCC) were maintained at 37 uC and 5 % CO2 in MEM (Gibco), F-12K (ATCC) or RPMI (Gibco) media, respectively, supplemented with 10 % FBS (Thermo Fisher Scientific), penicillin/streptomycin and L -glutamine. For the transfection studies, cells were grown in 24- or six-well plates up to 80 % confluency, and transfected with human CX3CR1 expression plasmid (kindly provided by Dr R. A. Tripp, University of Georgia, Athens, GA, USA) with Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. Transfection efficiency (percentage of CX3CR1+ cells) was measured by flow cytometry. HAEC characterization. HAECs were harvested from the inserts

by replacing the lower chamber media with trypsin/0.05 % EDTA (Gibco) and washing cells off the insert membrane with PBS. For surface markers, cells were stained with fluorochrome-labelled anti-CX3CR1 (BioLegend), anti-CD151 (eBioscience) and Jacalin (Vector). For intracellular markers, cells were fixed with BD FACS Lysing Solution, permeabilized with BD FACS Permeabilization Solution 2, blocked with PBS containing 5 % BSA and stained with primary antibody [mouse anti-human FoxJ1 (eBioscience), mouse anti-mucin-5AC (Santa Cruz Biotechnology) or goat anti-b-tubulin IV (Abcam)] and fluorochrome-labelled secondary antibody [donkey anti-mouse IgG (H+L) antibody (Invitrogen) or donkey anti-goat IgG (H+L) antibody (Abcam)]. All washes and incubations were performed on ice. Flow cytometry analysis was performed using a four-laser BD LSR-II (BD Biosciences) and FlowJo software (Tree Star). Positive signal and percentage of cells expressing markers were assessed against cells stained with isotype control antibody. For the Jacalin-binding test, HEp-2 cells stained with FITC-labelled Jacalin were used as a negative control. Expression of syndecan-1 on HAECs was determined by fluorescence microscopy. Without disrupting the cell sheet, cells were thoroughly washed with PBS containing 0.5 % BSA, fixed with BD FACS Lysing Solution, washed with PBS/0.5 % BSA, blocked with PBS containing 5 % BSA and stained with anti-syndecan-1 antibody (phycoerythrinlabelled anti-CD138; BioLegend). Membranes were cut out of inserts, http://vir.sgmjournals.org

placed on microscope slides, mounted with DAPI-containing Fluoromount-G (eBioscience) and analysed using an EVOS FL Cell Imaging System (Life Technologies). Viruses, binding, internalization and infection. RSV strains group

A (A/) and group B (B/) included: A/A2 and B/WV/14617/85 (acquired through ATCC), strain B/8/60 (kindly provided by Dr U. Power, Queen’s University, Belfast, UK), and clinical strains A/1997/12-35 (passage 12) and A/TN/13/2-26 (passage 4) (isolated from children seen at Vanderbilt University, Nashville, TN, USA). The amount of G protein in these virus stocks was quantified by ELISA (see below). Two RSV A2-derived strains, WT and CX3C mutant, were generated with a reverse genetics system as described previously (Mitra et al., 2012). The WT strain (designated CX3C) has an intact CX3C motif (182CWAIC186) in the G protein and is parent to a mutant strain (designated CX4C) which has an alanine A186 insertion in the CX3C motif. Viruses were grown in HEp-2 cells, purified by pelleting through 20 % sucrose cushion at 16 000 g for 2 h and stored at 280 uC. For the RSV internalization studies, virus was labelled with CFDASE (eBioscience) as described previously (Drobni et al., 2003) by incubating for 20 min at room temperature with CFDASE at a final concentration of 5 mM; labelled virus was purified through a 20 % sucrose cushion and stored at 280 uC. For the RSV-binding assay, CHO-K1, pgsD-677 and BEAS-2b cells were harvested from plates and exposed to virus in suspension at 4 uC for 1 h. For heparinase I pre-treatment, cells were resuspended in heparinase I solution (1 U ml21) (Sigma-Aldrich) and incubated for 1 h at room temperature before addition of RSV. For the RSV-binding assay with HAECs, cells were either left on plate/inserts or treated with trypsin/EDTA and harvested in suspension, untreated or pre-treated with CX3CR1 or syndecan-1-blocking antibody and exposed to RSV at 4 uC for 1 h. The cells were then washed extensively with cold PBS and virus binding was assessed: for cells left on the plate/inserts, by RSV-specific ELISA after fixation with BD FACS Lysing Solution (BD Biosciences); for cells in suspension, by flow cytometry (see below). For the RSV entry assay, cells were inoculated with CFDASElabelled RSV at m.o.i. 0.5 and incubated for up to 4 h at 37 uC. After incubation, the cells were harvested from plates with trypsin/EDTA and the fluorescence of intracellular CFSE was measured by flow cytometry. Cells exposed to CFDASE-labelled RSV at 4 uC were used as negative control for viral internalization. For RSV infection, cells were inoculated with different RSV strains at the indicated m.o.i. determined by virus titration in HEp-2 cells; m.o.i. 2.0 was used to compare RSV strains and anti-CX3CR1 antibody treatment. HAECs were washed with PBS and incubated for 2 h at 37 uC with viral inoculum diluted in PBS, or plain PBS as a mock infection, added to the apical surface of the cells. After the incubation, the virus inoculum was aspirated, the apical surface was washed with PBS and the basolateral media was changed. RSV-infected HAECs were incubated for up to 8 days at 37 uC and 5 % CO2. For the experiments with anti-CX3CR1 or anti-syndecan-1 antibody treatment, cells were incubated for 2 h with polyclonal anti-human CX3CR1 antibody (100 mg ml21) (eBioscience) or anti-syndecan-1 (20 mg ml21) (BioLegend) antibody. Before using the antibody, sodium azide was removed from the solution by buffer exchange through Zeba Spin Desalting Columns (Fisher Scientific). RSV titres in HAEC apical washes and bottom supernatants.

For apical wash collection, 100 ml PBS was added to the apical surface of cell cultures, incubated for 15 min at 37 uC, and then collected and stored at 280 uC. The TCID50 was determined in a microtitre infectivity assay using an ELISA to detect virus replication as described previously (Anderson et al., 1985). Briefly, HEp-2 cells were grown in 2553

T. Chirkova and others

a 96-well microtitre plate, infected with 10-fold serial dilutions of the HAEC apical washes, and incubated at 37 uC and 5 % CO2 for 5 days. After incubation, plates were washed with PBS, fixed with 70 % ethanol, dried and stored at 4 uC until testing by the RSV-specific ELISA. RSV antigen ELISA. For the ELISA, fixed cells were washed with PBS

containing 0.05 % Tween-20 (PBS/Tw), blocked with PBS/Tw containing 5 % BSA, incubated for 2 h at 37 uC with goat anti-RSV antibody (Millipore), and for 1 h at 37 uC with HRP-conjugated donkey anti-goat antibody (Jackson ImmunoResearch), and developed using o-phenylenediamine (OPD) substrate. A490/640 was determined. For RSV G-protein-specific ELISA, plates were coated with RSV G protein (BEI Resources) at serial dilutions and virus stocks at 1 : 100 dilution in carbonate/bicarbonate buffer at 4 uC overnight, and then plates were blocked with PBS/Tw containing 5 %BSA, incubated for 1 h at 37 uC with human anti-G antibody (Trellis Bioscience), and with HRP-conjugated goat anti-human antibody (Jackson ImmunoResearch), and developed with OPD. Flow cytometry of RSV-infected cells. Cells harvested from plate/

Core, and was supported by the National Institutes of Health [1U19AI095227 (M. L. L., L. J. A.), 5R01HL109362 (A. G. Z.), 1R01AI087798 (M. L. L)], funding from Children’s Healthcare of Atlanta, support from the Immunology Core of Emory Children’s Paediatric Research Center and the Emory Vaccinology Training Grant T32 [5T32AI074492-03 (S. B. B.)]. L. J. A. consulted on RSV vaccines for MedImmune, Novartis, Crucell Holland and AVC LLC; received funding from Trellis RSV Holdings; and is co-inventor on CDC patents on RSV vaccines and treatment.

REFERENCES Anderson, L. J., Hierholzer, J. C., Bingham, P. G. & Stone, Y. O. (1985). Microneutralization test for respiratory syncytial virus based

on an enzyme immunoassay. J Clin Microbiol 22, 1050–1052. Aoyama, T., Inokuchi, S., Brenner, D. A. & Seki, E. (2010). CX3CL1-

CX3CR1 interaction prevents carbon tetrachloride-induced liver inflammation and fibrosis in mice. Hepatology 52, 1390–1400.

inserts as described above were transferred in a 96-well plate, washed with PBS, stained on ice with Fixable Viability Dye (eBioscience), washed with cold staining buffer (BD Biosciences), incubated with anti-CX3CR1 antibody (BioLegend), fixed with BD FACS Lysing Solution, permeabilized with BD FACS Permeabilization Solution 2, blocked with PBS containing 5 % BSA, incubated with anti-RSV F protein human mAb (kindly provided by MedImmune) and Alexa Fluor 488-labelled goat anti-human IgG (H+L) antibody (Invitrogen). Flow cytometry analysis was performed using a four-laser BD LSR-II and FlowJo software. Imaging flow cytometry was performed using ImageStreamX Mark II and IDEAS data analysis software (AMNIS; Millipore).

Bar-On, L., Birnberg, T., Lewis, K. L., Edelson, B. T., Bruder, D., Hildner, K., Buer, J., Murphy, K. M., Reizis, B. & Jung, S. (2010). CX3CR1+

RSV RNA detection in HAECs. The amount of viral RNA was determined by a real-time (RT)-PCR assay. Total RNA was extracted from supernatants using a Qiagen Total RNA Extraction kit (Qiagen) according to the manufacturer’s instruction and stored at 280 uC. RT-PCR was performed by using an AgPath-ID OneStep RT-PCR kit (Applied Biosystems) and an ABI 7700 sequence detector system (PE Applied Biosystems) as described previously (Chirkova et al., 2013). The cycle threshold (Ct) values, i.e. the number of cycles required to exceed the background level, were calculated.

sulfate proteoglycans. Annu Rev Biochem 68, 729–777.

Cytokine analysis. The supernatants from the bottom chamber were

harvested at different time points post-infection, stored at 280 uC and analysed by Luminex magnetic bead kits (Invitrogen and Millipore).

Data analysis. Statistical analysis and graph production used Prism software (GraphPad). Data are expressed as mean+ SEM of three or

more independent experiments with three replicates for each experimental condition. The Mann–Whitney test, unpaired t-test and twoway ANOVA were used for the comparison of the different treatments of HAECs and multiple cytokine responses. Pv0.05 was considered to be statistically significant. Cytokine kinetics analysis was performed by calculating the ratios (RCV) of slopes of cytokine production curves (pg cytokine ml21) and replication curves (relative amount of viral RNA in HAECs found by RT-PCR) between days 2 and 4 p.i., the exponential phase of both cytokine production and virus replication. The slopes were calculated using logarithmic transformation of the cytokine and relative RNA levels.

CD8alpha+ dendritic cells are a steady-state population related to plasmacytoid dendritic cells. Proc Natl Acad Sci U S A 107, 14745–14750. Bermejo-Martin, J. F., Garcia-Arevalo, M. C., De Lejarazu, R. O., Ardura, J., Eiros, J. M., Alonso, A., Matı´as, V., Pino, M., Bernardo, D. & other authors (2007). Predominance of Th2 cytokines, CXC

chemokines and innate immunity mediators at the mucosal level during severe respiratory syncytial virus infection in children. Eur Cytokine Netw 18, 162–167. Bernfield, M., Go¨tte, M., Park, P. W., Reizes, O., Fitzgerald, M. L., Lincecum, J. & Zako, M. (1999). Functions of cell surface heparan Blatt, E. N., Yan, X. H., Wuerffel, M. K., Hamilos, D. L. & Brody, S. L. (1999). Forkhead transcription factor HFH-4 expression is temporally

related to ciliogenesis. Am J Respir Cell Mol Biol 21, 168–176. Chandrasekar, B., Mummidi, S., Perla, R. P., Bysani, S., Dulin, N. O., Liu, F. & Melby, P. C. (2003). Fractalkine (CX3CL1) stimulated by

nuclear factor kappaB (NF-kappaB)-dependent inflammatory signals induces aortic smooth muscle cell proliferation through an autocrine pathway. Biochem J 373, 547–558. Chen, J., Kinter, M., Shank, S., Cotton, C., Kelley, T. J. & Ziady, A. G. (2008). Dysfunction of Nrf-2 in CF epithelia leads to excess intracellular

H2O2 and inflammatory cytokine production. PLoS One 3, e3367. Chen, P., Abacherli, L. E., Nadler, S. T., Wang, Y., Li, Q. & Parks, W. C. (2009). MMP7 shedding of syndecan-1 facilitates re-epithelialization

by affecting alpha2beta1 integrin activation. PLoS One 4, e6565. Chirkova, T., Boyoglu-Barnum, S., Gaston, K. A., Malik, F. M., Trau, S. P., Oomens, A. G. & Anderson, L. J. (2013). Respiratory syncytial

virus G protein CX3C motif impairs human airway epithelial and immune cell responses. J Virol 87, 13466–13479. Combadiere, C., Gao, J., Tiffany, H. L. & Murphy, P. M. (1998a).

Gene cloning, RNA distribution, and functional expression of mCX3CR1, a mouse chemotactic receptor for the CX3C chemokine fractalkine. Biochem Biophys Res Commun 253, 728–732. Combadiere, C., Salzwedel, K., Smith, E. D., Tiffany, H. L., Berger, E. A. & Murphy, P. M. (1998b). Identification of CX3CR1.A chemotactic

receptor for the human CX3C chemokine fractalkine and a fusion coreceptor for HIV-1. J Biol Chem 273, 23799–23804.

ACKNOWLEDGEMENTS

Drobni, P., Mistry, N., McMillan, N. & Evander, M. (2003). Carboxy-

Authors would like to thank Dr R. C. Kauffman for technical assistance. This work was performed using the Emory Paediatric Flow Cytometry

fluorescein diacetate, succinimidyl ester labeled papillomavirus virus-like particles fluoresce after internalization and interact with heparan sulfate for binding and entry. Virology 310, 163–172.

2554

Journal of General Virology 96

CX3CR1 facilitates RSV infection in HAECs

Duan, D., Yue, Y., Yan, Z., McCray, P. B. Jr & Engelhardt, J. F. (1998).

Polarity influences the efficiency of recombinant adenoassociated virus infection in differentiated airway epithelia. Hum Gene Ther 9, 2761–2776. Falsey, A. R., Hennessey, P. A., Formica, M. A., Cox, C. & Walsh, E. E. (2005). Respiratory syncytial virus infection in elderly and high-risk

adults. N Engl J Med 352, 1749–1759.

Jamieson, W. L., Shimizu, S., D’Ambrosio, J. A., Meucci, O. & Fatatis, A. (2008). CX3CR1 is expressed by prostate epithelial cells and

androgens regulate the levels of CX3CL1/fractalkine in the bone marrow: potential role in prostate cancer bone tropism. Cancer Res 68, 1715–1722. Jensen-Smith, H. C., Luduen˜a, R. F. & Hallworth, R. (2003).

Feldman, S. A., Hendry, R. M. & Beeler, J. A. (1999). Identification of

Requirement for the betaI and betaIV tubulin isotypes in mammalian cilia. Cell Motil Cytoskeleton 55, 213–220.

a linear heparin binding domain for human respiratory syncytial virus attachment glycoprotein G. J Virol 73, 6610–6617.

Johnson, S. M., McNally, B. A., Ioannidis, I., Flano, E., Oomens, A. G., Partida-Sanchez, S. & Peeples, M. E. (2013). Identification of

Feldman, S. A., Audet, S. & Beeler, J. A. (2000). The fusion glycoprotein

CX3CR1 as a cellular receptor for respiratory syncytial virus on primary well-differentiated human airway epithelial cultures. In American Society for Virology 32nd Annual Meeting. University Park, PA.

of human respiratory syncytial virus facilitates virus attachment and infectivity via an interaction with cellular heparan sulfate. J Virol 74, 6442–6447. Feldman, S. A., Crim, R. L., Audet, S. A. & Beeler, J. A. (2001). Human

respiratory syncytial virus surface glycoproteins F, G and SH form an oligomeric complex. Arch Virol 146, 2369–2383.

Johnson, T. R., McLellan, J. S. & Graham, B. S. (2012). Respiratory

syncytial virus glycoprotein G interacts with DC-SIGN and L-SIGN to activate ERK1 and ERK2. J Virol 86, 1339–1347.

Fonceca, A. M., Flanagan, B. F., Trinick, R., Smyth, R. L. & McNamara, P. S. (2012). Primary airway epithelial cultures from children are

Johnston, I. N., Milligan, E. D., Wieseler-Frank, J., Frank, M. G., Zapata, V., Campisi, J., Langer, S., Martin, D., Green, P. & other authors (2004). A role for proinflammatory cytokines and

highly permissive to respiratory syncytial virus infection. Thorax 67, 42–48.

fractalkine in analgesia, tolerance, and subsequent pain facilitation induced by chronic intrathecal morphine. J Neurosci 24, 7353–7365.

Hall, C. B., Weinberg, G. A., Iwane, M. K., Blumkin, A. K., Edwards, K. M., Staat, M. A., Auinger, P., Griffin, M. R., Poehling, K. A. & other authors (2009). The burden of respiratory syncytial virus

Karron, R. A., Buonagurio, D. A., Georgiu, A. F., Whitehead, S. S., Adamus, J. E., Clements-Mann, M. L., Harris, D. O., Randolph, V. B., Udem, S. A. & other authors (1997). Respiratory syncytial virus (RSV)

infection in young children. N Engl J Med 360, 588–598.

SH and G proteins are not essential for viral replication in vitro: clinical evaluation and molecular characterization of a cold-passaged, attenuated RSV subgroup B mutant. Proc Natl Acad Sci U S A 94, 13961–13966.

Hallak, L. K., Spillmann, D., Collins, P. L. & Peeples, M. E. (2000).

Glycosaminoglycan sulfation requirements for respiratory syncytial virus infection. J Virol 74, 10508–10513. Harcourt, J., Alvarez, R., Jones, L. P., Henderson, C., Anderson, L. J. & Tripp, R. A. (2006). Respiratory syncytial virus G protein and G

protein CX3C motif adversely affect CX3CR1+T cell responses. J Immunol 176, 1600–1608.

Harrison, J. K., Jiang, Y., Chen, S., Xia, Y., Maciejewski, D., McNamara, R. K., Streit, W. J., Salafranca, M. N., Adhikari, S. & other authors (1998). Role for neuronally derived fractalkine in

Keswani, S. G., Balaji, S., Le, L., Leung, A., Katz, A. B., Lim, F. Y., Habli, M., Jones, H. N., Wilson, J. M. & Crombleholme, T. M. (2012). Pseudotyped AAV vector-mediated gene transfer in a

human fetal trachea xenograft model: implications for in utero gene therapy for cystic fibrosis. PLoS One 7, e43633. Kim, K. W., Vallon-Eberhard, A., Zigmond, E., Farache, J., Shezen, E., Shakhar, G., Ludwig, A., Lira, S. A. & Jung, S. (2011). In vivo structure/

function and expression analysis of the CX3C chemokine fractalkine. Blood 118, e156–e167.

mediating interactions between neurons and CX3CR1-expressing microglia. Proc Natl Acad Sci U S A 95, 10896–10901.

Li, Q., Park, P. W., Wilson, C. L. & Parks, W. C. (2002). Matrilysin shed-

Hasegawa, M., Sato, S., Echigo, T., Hamaguchi, Y., Yasui, M. & Takehara, K. (2005). Up regulated expression of fractalkine/CX3CL1

ding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury. Cell 111, 635–646.

and CX3CR1 in patients with systemic sclerosis. Ann Rheum Dis 64, 21–28.

Martı´nez, I. & Melero, J. A. (2000). Binding of human respiratory syncytial

Hayashida, A., Bartlett, A. H., Foster, T. J. & Park, P. W. (2009).

virus to cells: implication of sulfated cell surface proteoglycans. J Gen Virol 81, 2715–2722.

Staphylococcus aureus beta-toxin induces lung injury through syndecan-1. Am J Pathol 174, 509–518.

McComb, J. G., Ranganathan, M., Liu, X. H., Pilewski, J. M., Ray, P., Watkins, S. C., Choi, A. M. & Lee, J. S. (2008). CX3CL1 up-regulation

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. Heminway, B. R., Yu, Y., Tanaka, Y., Perrine, K. G., Gustafson, E., Bernstein, J. M. & Galinski, M. S. (1994). Analysis of respiratory

is associated with recruitment of CX3CR1+ mononuclear phagocytes and T lymphocytes in the lungs during cigarette smoke-induced emphysema. Am J Pathol 173, 949–961.

McNamara, P. S., Fonceca, A. M., Howarth, D., Correia, J. B., Slupsky, J. R., Trinick, R. E., Al Turaiki, W., Smyth, R. L. & Flanagan, B. F. (2013). Respiratory syncytial virus infection of airway epithelial

syncytial virus F, G, and SH proteins in cell fusion. Virology 200, 801–805.

cells, in vivo and in vitro, supports pulmonary antibody responses by inducing expression of the B cell differentiation factor BAFF. Thorax 68, 76–81.

Imai, T., Hieshima, K., Haskell, C., Baba, M., Nagira, M., Nishimura, M., Kakizaki, M., Takagi, S., Nomiyama, H. & other authors (1997).

Million, K., Larcher, J., Laoukili, J., Bourguignon, D., Marano, F. & Tournier, F. (1999). Polyglutamylation and polyglycylation of alpha-

Identification and molecular characterization of fractalkine receptor CX3CR1, which mediates both leukocyte migration and adhesion. Cell 91, 521–530. Ioannidis, I., McNally, B., Willette, M., Peeples, M. E., Chaussabel, D., Durbin, J. E., Ramilo, O., Mejias, A. & Flan˜o, E. (2012). Plasticity and

virus specificity of the airway epithelial cell immune response during respiratory virus infection. J Virol 86, 5422–5436. http://vir.sgmjournals.org

and beta-tubulins during in vitro ciliated cell differentiation of human respiratory epithelial cells. J Cell Sci 112, 4357–4366. Mitra, R., Baviskar, P., Duncan-Decocq, R. R., Patel, D. & Oomens, A. G. (2012). The human respiratory syncytial virus matrix protein

is required for maturation of viral filaments. J Virol 86, 4432–4443. Nair, H., Nokes, D. J., Gessner, B. D., Dherani, M., Madhi, S. A., Singleton, R. J., O’Brien, K. L., Roca, A., Wright, P. F. & other authors

2555

T. Chirkova and others

(2010). Global burden of acute lower respiratory infections due to

respiratory syncytial virus in young children: a systematic review and meta-analysis. Lancet 375, 1545–1555. Nishimura, M., Umehara, H., Nakayama, T., Yoneda, O., Hieshima, K., Kakizaki, M., Dohmae, N., Yoshie, O. & Imai, T. (2002). Dual

functions of fractalkine/CX3C ligand 1 in trafficking of perforin+/ granzyme B+ cytotoxic effector lymphocytes that are defined by CX3CR1 expression. J Immunol 168, 6173–6180.

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-1alpha by lower airway epithelial cells and eosinophils infected with respiratory syncytial virus. J Virol 72, 4756–4764. Oshansky, C. M., Barber, J. P., Crabtree, J. & Tripp, R. A. (2010).

Song, Y., Coleman, L., Shi, J., Beppu, H., Sato, K., Walsh, K., Loscalzo, J. & Zhang, Y. Y. (2008). Inflammation, endothelial

injury, and persistent pulmonary hypertension in heterozygous BMPR2-mutant mice. Am J Physiol Heart Circ Physiol 295, H677–H690. Suzuki, F., Kubota, T., Miyazaki, Y., Ishikawa, K., Ebisawa, M., Hirohata, S., Ogura, T., Mizusawa, H., Imai, T. & other authors (2012). Serum level of soluble CX3CL1/fractalkine is elevated in

patients with polymyositis and dermatomyositis, which is correlated with disease activity. Arthritis Res Ther 14, R48. Tachibana, K., Nakamura, S., Wang, H., Iwasaki, H., Tachibana, K., Maebara, K., Cheng, L., Hirabayashi, J. & Narimatsu, H. (2006).

Elucidation of binding specificity of Jacalin toward O-glycosylated peptides: quantitative analysis by frontal affinity chromatography. Glycobiology 16, 46–53.

Respiratory syncytial virus F and G proteins induce interleukin 1alpha, CC, and CXC chemokine responses by normal human bronchoepithelial cells. J Infect Dis 201, 1201–1207.

for human respiratory syncytial virus. Nat Med 17, 1132–1135.

Park, P. W., Reizes, O. & Bernfield, M. (2000). Cell surface heparan sulfate

Techaarpornkul, S., Collins, P. L. & Peeples, M. E. (2002). Respiratory

proteoglycans: selective regulators of ligand-receptor encounters. J Biol Chem 275, 29923–29926. Park, P. W., Pier, G. B., Hinkes, M. T. & Bernfield, M. (2001).

Exploitation of syndecan-1 shedding by Pseudomonas aeruginosa enhances virulence. Nature 411, 98–102. Perros, F., Dorfmu¨ller, P., Souza, R., Durand-Gasselin, I., Godot, V., Capel, F., Adnot, S., Eddahibi, S., Mazmanian, M. & other authors (2007). Fractalkine-induced smooth muscle cell proliferation in

pulmonary hypertension. Eur Respir J 29, 937–943. Polack, F. P., Irusta, P. M., Hoffman, S. J., Schiatti, M. P., Melendi, G. A., Delgado, M. F., Laham, F. R., Thumar, B., Hendry, R. M. & other authors (2005). The cysteine-rich region of respiratory

Tayyari, F., Marchant, D., Moraes, T. J., Duan, W., Mastrangelo, P. & Hegele, R. G. (2011). Identification of nucleolin as a cellular receptor

syncytial virus with the fusion protein as its only viral glycoprotein is less dependent on cellular glycosaminoglycans for attachment than complete virus. Virology 294, 296–304. 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. Tripp, R. A., Dakhama, A., Jones, L. P., Barskey, A., Gelfand, E. W. & Anderson, L. J. (2003). The G glycoprotein of respiratory syncytial

virus depresses respiratory rates through the CX3C motif and substance P. J Virol 77, 6580–6584. Tristram, D. A., Hicks, W. Jr & Hard, R. (1998). Respiratory syncytial

syncytial virus attachment protein inhibits innate immunity elicited by the virus and endotoxin. Proc Natl Acad Sci U S A 102, 8996–9001.

virus and human bronchial epithelium. Arch Otolaryngol Head Neck Surg 124, 777–783.

Popova, T. G., Millis, B., Bradburne, C., Nazarenko, S., Bailey, C., Chandhoke, V. & Popov, S. G. (2006). Acceleration of epithelial cell

Villenave, R., O’Donoghue, D., Thavagnanam, S., Touzelet, O., Skibinski, G., Heaney, L. G., McKaigue, J. P., Coyle, P. V., Shields, M. D. & Power, U. F. (2011). Differential cytopathogenesis of

syndecan-1 shedding by anthrax hemolytic virulence factors. BMC Microbiol 6, 8. Pruessmeyer, J., Martin, C., Hess, F. M., Schwarz, N., Schmidt, S., Kogel, T., Hoettecke, N., Schmidt, B., Sechi, A. & other authors (2010). A disintegrin and metalloproteinase 17 (ADAM17) mediates

inflammation-induced shedding of syndecan-1 and -4 by lung epithelial cells. J Biol Chem 285, 555–564. Qin, L., Hu, C. P., Feng, J. T. & Xia, Q. (2011). Activation of lymphocytes

induced by bronchial epithelial cells with prolonged RSV infection. PLoS One 6, e27113. 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. Rimaniol, A. C., Till, S. J., Garcia, G., Capel, F., Godot, V., Balabanian, K., Durand-Gasselin, I., Varga, E. M., Simonneau, G. & other authors (2003). The CX3C chemokine fractalkine in allergic asthma and

rhinitis. J Allergy Clin Immunol 112, 1139–1146. Shay, D. K., Holman, R. C., Newman, R. D., Liu, L. L., Stout, J. W. & Anderson, L. J. (1999). Bronchiolitis-associated hospitalizations

among US children, 1980-1996. JAMA 282, 1440–1446. Shingai, M., Azuma, M., Ebihara, T., Sasai, M., Funami, K., Ayata, M., Ogura, H., Tsutsumi, H., Matsumoto, M. & Seya, T. (2008). Soluble G

protein of respiratory syncytial virus inhibits Toll-like receptor 3/4mediated IFN-beta induction. Int Immunol 20, 1169–1180. Sincock, P. M., Mayrhofer, G. & Ashman, L. K. (1997). Localization of

the transmembrane 4 superfamily (TM4SF) member PETA-3 (CD151) in normal human tissues: comparison with CD9, CD63, and alpha5beta1 integrin. J Histochem Cytochem 45, 515–525. 2556

respiratory syncytial virus prototypic and clinical isolates in primary pediatric bronchial epithelial cells. Virol J 8, 43. Welliver, R. C., Garofalo, R. P. & Ogra, P. L. (2002). Beta-chemokines,

but neither T helper type 1 nor T helper type 2 cytokines, correlate with severity of illness during respiratory syncytial virus infection. Pediatr Infect Dis J 21, 457–461. You, Y., Huang, T., Richer, E. J., Schmidt, J. E., Zabner, J., Borok, Z. & Brody, S. L. (2004). Role of f-box factor foxj1 in differentiation of

ciliated airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 286, L650–L657. Zhang, L., Peeples, M. E., Boucher, R. C., Collins, P. L. & Pickles, R. J. (2002). Respiratory syncytial virus infection of human airway epithelial

cells is polarized, specific to ciliated cells, and without obvious cytopathology. J Virol 76, 5654–5666. Zhang, H., Xia, T., Meng, H., Xue, M., George, S., Ji, Z., Wang, X., Liu, R., Wang, M. & other authors (2011). Differential expression of

syndecan-1 mediates cationic nanoparticle toxicity in undifferentiated versus differentiated normal human bronchial epithelial cells. ACS Nano 5, 2756–2769. Zhang, M., Xu, G., Liu, W., Ni, Y. & Zhou, W. (2012). Role of fractalkine/

CX3CR1 interaction in light-induced photoreceptor degeneration through regulating retinal microglial activation and migration. PLoS One 7, e35446. Ziady, A. G., Sokolow, A., Shank, S., Corey, D., Myers, R., Plafker, S. & Kelley, T. J. (2012). Interaction with CREB binding protein modulates

the activities of Nrf2 and NF-kB in cystic fibrosis airway epithelial cells. Am J Physiol Lung Cell Mol Physiol 302, L1221–L1231. Journal of General Virology 96