Inhibition of the TLR4 signalling pathway with TAK-242 reduces RSV infection and cytokine release in primary airway epithelial cells

Respiratory syncytial virus (RSV) infection is the leading cause of hospitalisation in children worldwide, but there is still no vaccine or anti-viral treatment available. RSV has been implicated in the development of respiratory diseases such as asthma. Toll like receptor 4 (TLR4) has been well characterised in the immune responses to RSV. However, the role of TLR4 in RSV infection remains unclear. To study RSV in the lung epithelium, where RSV preferentially infects ciliated cells, we used a well-differentiated primary airway epithelial cell (WD-PAEC) model: a pseudostratified epithelium that produces mucus and beating cilia. We demonstrate in this physiologically relevant model that TLR4 is a pro-viral factor. Inhibition of TLR4 using TAK-242 significantly reduces RSV titres in WD-PAECs in a dose-dependent manner but has no effect on RSV growth kinetics in a range of immortalised respiratory-derived cell lines. Specific inhibition of a range of downstream effectors of TLR4 signalling in the WD-PAEC model identified p38 MAPK as a pro-viral factor, whereas inhibition of MEK1/2 significantly increased RSV titres. Our data demonstrate a role for TLR4 in RSV infection and highlight the importance of biologically relevant models to study virus-host interactions. Author summary Respiratory Syncytial Virus (RSV) can cause severe respiratory infection in young children and is responsible for approximately 200,000 deaths worldwide every year. Despite decades of research since the identification of this virus in the 1950s there is still no vaccine or treatment available. Advances in research have led to the development of cell cultures that are very similar to the cells that line human airways. These cultures provide an opportunity to study how viruses interacts with airway cells in a representative model and may provide insights that traditional research models have not yet been able to answer. Using this experimental model we show that a drug, TAK-242 which targets a pathogen recognition receptor on the surface of cells, reduces growth of RSV and dampens the immune response to infection in these airway cells. Our data demonstrate potential targets for RSV treatments and also highlight the importance of using relevant experimental models.


Introduction 59
Respiratory syncytial virus (RSV) is responsible for over 3 million hospitalisations of infants 60 each year and is the primary cause of severe lower respiratory tract infections (LRTIs) in 61 children worldwide (1,2). There are currently no available vaccines or effective therapeutics 62 against RSV. As with many viruses, the innate immune system plays a critical role in defence against RSV 64 disease. Innate immune signalling pathways are triggered by the detection of pathogen associated molecular patterns (PAMPs) by surface and/or intracellular pathogen recognition 66 receptors (PRRs). For RNA viruses, toll-like receptor 3 (TLR3) has been well characterised as 67 a PRR that detects dsRNA and induces activation of IFN regulatory factor 3 (IRF3) and the 68 NF-κB pathway (3). TLR4 is also able to activate IRF3 and NF-κB pathways but is better 69 characterised as a receptor of lipopolysaccharide (LPS), a PAMP produced by many Gram-

89
In the clinic, TLR4 has been linked to severe RSV disease (18  Figure 2A). For all cell lines, TLR4 was predominantly 130 located in the cell membrane, whereas MD2 appeared more diffuse throughout the 131 cytoplasm. When infected with RSV-BT2a, both TLR4 and MD2 expression appeared to be 132 increased in A549 and BEAS-2B cells, but not in Calu-3 cells.

133
These cell lines were then treated with either TAK-242 (TLR4 inhibitor) or L48H37 (an MD2 134 inhibitor) at 10 μg/mL 6 h prior to infection with RSV BT2a or RSV A2-GFP. There was no 135 effect on virus titres throughout the 72 h time course for either RSV BT2a ( Figure 2B) or RSV 136 A2-GFP ( Figure 2C). These data are consistent with previous reports that found little evidence 137 that TLR4 played a significant role in RSV pathogenesis in immortalised cell lines (14).

138
Removal of heparan sulphate from A549 cells reduces RSV-A2 but not

147
To assess whether HS was masking any effect of TLR4 inhibition in cell lines, A549 cells were 148 treated with TAK-242 and/or treated with heparinase (to remove cell surface HS) prior to 149 infection with RSV (either BT2a or A2-GFP). There was no significant reduction in RSV-BT2a 150 titres following TAK-242, heparinase, or the combination treatment ( Figure 3A). In contrast,

151
RSV-A2-GFP growth kinetics, as measured by GFP expression, were significantly reduced 152 following heparinase treatment and the combination TAK-242 and heparinase treatment, but 153 not following TAK-242 alone ( Figures 3B and 3C). Viral titrations were consistent with a 154 reduction in RSV-A2-GFP titres in these conditions ( Figure 3A), but the results did not reach 155 significance.

inflammatory responses in WD-PAEC cultures.
158 The use of primary airway epithelial cells, isolated from human respiratory tracts, to form well- 181 WD-PAECs were treated with the MD2 inhibitor L48H37 for 6 h prior to infection with RSV.

182
There was a modest reduction in RSV titres from 24-96 hpi, but this was not statistically 183 significant ( Figure 5A). L48H37 binds to the binding pocket of MD2, which was shown to 184 prevent LPS binding and subsequent signal transduction (30). Structural modelling suggests 185 that L48H37 does not affect MD2 from interacting with TLR4.

186
Similarly, when WD-PAECs were treated for 1 h with heparinase prior to RSV infection, there 187 was no significant difference in viral production for either RSV-BT2a or RSV-A2-GFP ( Figure   188 5B).

317
Virus titration was performed via TCID 50 assay on HEp-2 (kindly provided by Prof. Ralph Tripp,

326
Appropriate secondary antibodies (AlexaFluor, Invitrogen) were applied for 1 h at 37°C prior 327 to mounting onto slides using mounting media containing DAPI (Vectashield, Vector Labs).

328
Images were obtained on a Leica SP5 confocal microscope or a Nexcelom Celigo as

341
Heparinase was resuspended in DMEM (with no supplements) at a working concentration of 342 5 mIU/mL prior to application to cells, which were incubated for 1 h at 37 °C.

Quantification of cytokines and chemokines 344
Concentrations of IFNλ1 from basolateral medium of WD-PAECs was determined using the 345 Platinum Sandwich ELISA kit (Invitrogen) following the manufacturer's protocol.