The innate response to rhinovirus in bronchial epithelial cells is controlled by both TLR3 and RNA helicase mediated signalling pathways

Session IV – Edwards, Michael

Title of Contribution: The innate response to rhinovirus in bronchial epithelial cells is controlled by both TLR3 and RNA helicase mediated signalling pathways.

Author(s): Michael R. Edwards, Louise Slater, Nathan W. Bartlett, Jennifer J. Haas, Jie Zhu, Ross P. Walton, Annemarie Sykes, and Sebastian L. Johnston.
Affiliation(s): .Department of Respiratory Medicine, National Heart & Lung Institute, Wright Fleming Institute of Infection and Immunity, Imperial College London, MRC & Asthma UK Centre for Allergic Mechanisms of Asthma, London UK
and the Centre for Respiratory Infections, London UK
Abstract:

Asthma exacerbations are caused by rhinovirus (RV) infection. RV infects bronchial epithelial cells and induces both pro-inflammatory cytokines and type I and type III interferons (IFN). Recently, asthmatic bronchial epithelial cells have been shown to have reduced IFN levels compared to non-asthmatics although the pathways required for IFN production are not well studied. The endosomal receptor for dsRNA TLR3, and the intracellular RNA helicases RIG-I and MDA5 have all been implicated in the innate response to viral infections, however few studies have explored the relative roles of both TLR3 and the RNA helicases together. We investigated the expression patterns and role of TLR3, RIG-I and MDA5 in RV induced IFN in primary human bronchial epithelial cells (HBECs), using siRNA knockdown experiments, western blotting, RT-PCR, and RV infection in vivo. Using siRNA, knockdown of MDA5 and TLR3 reduced IFN-β (p<0.01), IFN-λ1 (p<0.001) and IFN-λ2/3 mRNA, (p<0.01) versus control siRNA, while knockdown of RIG-I reduced IFN-β mRNA only versus control siRNA (p<0.001). Western blotting showed that uninfected HBECs contained very little RIG-I and MDA5, but were strongly induced by 4-12h post RV1B infection or polyIC treatment at the mRNA and protein level. Furthermore, analysis of bronchial biopsys from normal human donors at baseline and at day 4 post experimental RV infection in vivo showed that the cilated and columnar epithelium had very little staining for RIG-I or MDA5, however both helicases were upregulated at day 4 post RV challenge. Knockdown of TLR3 or the adaptor TRIF by siRNA significantly reduced RV1B induced RIG-I (p<0.001) and MDA5 mRNA (p<0.01) versus control siRNA, and expression of constitutively active TRIF (∆TRIF) greatly induced RIG-I and MDA5 mRNA compared to empty vector control, indicating that RIG-I and MDA5 are directly induced by RV infection in a TLR3 dependent manner. Finally in IFNAR1 deficient mice, which can not signal via IFN-α/β and also do not produce RV induced IFN-λ in the airways, we demonstrated that RV1B infection produce RIG-I and MDA5 protein and mRNA at 8-16h post infection in a similar manner to wildtype controls. The data indicate that the innate response to RV infection requires co-ordinated responses by endosomal TLR3 and intracellular RNA helicases, and that RNA helicases are induced directly by RV infection early in the infection cycle in a TLR3 dependent, but IFN independent manner.