Differential antiviral activities of respiratory syncytial virus (RSV) inhibitors in human airway epithelium

Abstract

Objectives: We report the use of reconstituted 3D human airway epithelium cells (HuAECs) of bronchial origin in an air-liquid interface to study respiratory syncytial virus (RSV) infection and to assess the efficacy of RSV inhibitors in (pre-)clinical development.

Methods: HuAECs were infected with RSV-A Long strain (0.01 CCID50/cell, where CCID50 represents 50% cell culture infectious dose in HEp2 cells) on the apical compartment of the culture. At the time of infection or at 1 or 3 days post-infection, selected inhibitors were added and refreshed daily on the basal compartment of the culture. Viral shedding was followed up by apical washes collected daily and quantifying viral RNA by RT-qPCR.

Results: RSV-A replicates efficiently in HuAECs and viral RNA is shed for weeks after infection. RSV infection reduces the ciliary beat frequency of the ciliated cells as of 4 days post-infection, with complete ciliary dyskinesia observed by day 10. Treatment with RSV fusion inhibitors resulted in an antiviral effect only when added at the time of infection. In contrast, the use of replication inhibitors (both nucleoside and non-nucleoside) elicited a marked antiviral effect even when the start of treatment was delayed until 1 day or even 3 days after infection. Levels of the inflammation marker RANTES (mRNA) increased ∼200-fold in infected, untreated cultures (at 3 weeks post-infection), but levels were comparable to those of uninfected cultures in the presence of PC786, an RSV replication inhibitor, suggesting that an efficient antiviral treatment might inhibit virus-induced inflammation in this model.

Conclusions: Overall, HuAECs offer a robust and physiologically relevant model to study RSV replication and to assess the efficacy of antiviral compounds.

Figure 1.

The effect of the RSV fusion inhibitors GS-5806 and TMC353121 was evaluated in HuAEC by using an early therapeutic treatment regimen (days 1–5 post-infection) (a, left panel) or by using a prophylactic regimen (−2 h to day 4 post-infection) (a, right panel). Compounds were added to the basal medium at a concentration of 100-fold the in vitro EC₅₀. Apical washes were collected at selected timepoints post-infection and viral RNA was quantified by means of RT-qPCR. The treatment period is depicted in grey. The sensitivity threshold was determined by the highest RSV RNA signal quantified on uninfected, untreated cells (uninfected). Error bars represent the standard deviation of duplicates. (b) Ciliary beat frequency (CBF) was measured by high-speed video microscopy with a custom-made Mathlab script. Three areas per insert, i.e. six measurements per condition, were acquired at three data points, i.e. days 4–10 and day 15 post-infection. Average measurements per condition are represented as horizontal bars.

Figure 2.

HuAEC were infected with RSV-A (Long strain) and were treated with: (a) the nucleoside viral polymerase inhibitor ALS-8112 or its prodrug ALS-8176 at 100-fold the in vitro EC₅₀ by using either a late (days 3–7 post-infection, right panel) or early (days 1–5 post-infection, left panel) therapeutic treatment regimen; or (b) the replication inhibitor AZ-27 at concentrations 10- and 100-fold the in vitro EC₅₀; or (c) ribavirin at 10-fold the in vitro EC₅₀ with an early therapeutic treatment regimen; or (d) (left panel) the replication inhibitor PC786 at 100-fold the in vitro EC₅₀ by using a late therapeutic treatment regimen. The treatment period is depicted in grey. The sensitivity threshold was determined by the highest RSV-RNA signal quantified on uninfected, untreated cells (uninfected). Error bars represent the standard deviation of duplicates. HuAEC were harvested at day 24 post-infection and total cellular mRNA was extracted. Levels of RANTES mRNA were quantified by means of RT-qPCR (d, right panel). β-actin was used as an internal control. Fold change of the uninfected, untreated control was calculated with the ΔΔCt method. Error bars represent the median of three independent quantifications of the condition and each condition was performed in duplicate (i.e. six measurements).