Glucocorticosteroids enhance replication of respiratory viruses: effect of adjuvant interferon

Abstract

Glucocorticosteroids (GCS) are used on a daily basis to reduce airway inflammation in asthma and chronic obstructive pulmonary disease (COPD). This treatment is usually escalated during acute disease exacerbations, events often associated with virus infections. We examined the impact of GCS on anti-viral defences and virus replication and assessed supplementary interferon (IFN) treatment. Here, we report that treatment of primary human airway cells in vitro with GCS prior to rhinovirus (RV) or influenza A virus (IAV) infection significantly reduces the expression of innate anti-viral genes and increases viral replication. Mice given intranasal treatment with GCS prior to IAV infection developed more severe disease associated with amplified virus replication and elevated inflammation in the airways. Adjuvant IFN treatment markedly reduced GCS-amplified infections in human airway cells and in mouse lung. This study demonstrates that GCS cause an extrinsic compromise in anti-viral defences, enhancing respiratory virus infections and provides a rationale for adjuvant IFN treatment.

Figure 1. Treatment of primary human airway cells with GCS prior to respiratory virus infection reduces innate anti-viral responses.

Monolayers of PBEC, PAF or PAM were treated with GCS for 24 hrs, washed and then infected with IAV (left panels) or RV (right panels). RNA was extracted from cell monolayers at 24 or 48 hrs post-infection and RT-PCR performed. mRNA expression of (A–C) IFNα, IFNβ and IFNλ₁ genes and (D–F) ISGs IFIT1, ISG15, MxA, 2′5′OAS, viperin and RIG-I was measured. Data is relative to the expression of 18S and represents the mean of a minimum of 3 replicate samples ± SD. Data is representative of cultures from a minimum of 3 independent donors. Expression significantly reduced in virus-infected samples treated with GCS, * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test.

Figure 2. GCS pre-treatment of primary human airway cells reduces the production of pro-inflammatory cytokines following virus infection.

Monolayers of (A) PBEC, (B) PAF or (C) PAM were treated with GCS for 24 hrs, washed and then infected with RV or IAV. Concentrations of cytokines in cell culture supernatants were determined by CBA at 24 or 48 hrs post-infection. The detection limit of each mediator is indicated as a dotted line. Data represents the mean of a minimum of 3 replicate samples ± SD and is representative of cultures from a minimum of 3 independent donors. Levels significantly reduced in virus-infected samples treated with GCS, * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test.

Figure 3. GCS pre-treatment of primary human airway cells enhances viral replication; an effect blunted by IFN.

PBEC or PAF were treated with GCS for 24 hrs, washed and then infected with IAV (left panels) or RV (right panels). (A) Levels of infectious virus in cell supernatants were determined by plaque assay (IAV) or titration (RV). Data represent the mean ± SD from triplicate samples and are representative of a minimum of cultures from 3 independent donors. Infectious virus in cell supernatants significantly increased with GCS treatment, * p < 0.05, *** p < 0.001, Student's t-test. (B) At 1 hr following IAV or RV infection, monolayers were stimulated with 250 IU/ml of human IFNα₂, IFNβ or IFNλ₁ and levels of infectious virus in cell supernatants were determined at the time points indicated. The detection limit of the assay is indicated as a dotted line. Data represent the mean ± SD from triplicate samples and are representative of a minimum of cultures from 3 independent donors. Infectious virus in cell supernatants significantly reduced by addition of IFN in both GCS treated and mock-treated infected cells, * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test.

Figure 4. Mice treated with GCS prior to IAV infection develop severe disease.

(A) Monolayers of primary mouse airway epithelial cells (mAEC) were treated with GCS for 24 hrs, washed and infected with IAV (HKx31, MOI 1). Levels of infectious virus in cell supernatants were determined by plaque assay at 24 and 48 hrs post-infection. Bars represent the mean ± SD from triplicate samples. Viral titer significantly increased with GCS treatment, *** p < 0.001, Student's t-test. (B) Groups of 5 C57BL/6 mice were treated with 20 μg GCS via the intranasal route 48 hrs and 24 hrs prior to inoculation with 10² PFU of IAV (HKx31). Mice received additional doses of GCS 24 hr post-infection and every 48 hrs thereafter. Uninfected mice treated with PBS or GCS were included for comparison. Mice were observed daily for weight loss (left panel). Mice which had lost ≥15% of their original body weight were euthanized. Data represents the mean % weight change ± SEM. Survival plots are shown (right panel). (C) Titers of infectious virus in lung (left panel) and nasal tissue (right panel) homogenates were determined by plaque assay at days 1, 3 and 5 post-infection. Bars represent the mean viral titer from a group of 5 mice ± SD. Virus titers from GCS-treated mice were significantly higher than those from PBS-treated mice, *** p < 0.001, one-way ANOVA and Tukey's post-test. Data is representative of a minimum of 2 independent experiments.

Figure 5. Treatment of mice with GCS alters the inflammatory response in the airways.

Groups of 5 C57BL/6 mice were treated with 20 μg GCS via the intranasal route 48 hrs and 24 hrs prior to inoculation with 10² PFU of IAV (HKx31). Mice received additional doses of GCS 24 hrs post-infection and every 48 hrs thereafter. Uninfected mice treated with PBS or GCS were included for comparison. At days 3 and 5 post-infection, mice were sacrificed and BAL performed. (A) BAL cells were examined by flow cytometry for the presence of CD45⁺ leukocytes, neutrophils (Neut), airway macrophages (AM), dendritic cells (DC) and inflammatory macrophages (IM). Bars represent the mean cell number ± SD. Data is representative of 2 independent experiments. Cell numbers from GCS-treated IAV-infected mice were significantly different compared to PBS-treated IAV-infected mice, * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test. (B) Concentrations of cytokines in BAL supernatants were determined by CBA at days 3 and 5 post-infection. Bars represent the mean concentration ± SD. The detection limit of each assay is indicated as a dotted line. Data is representative of 2 independent experiments. Levels from GCS-treated mice that were significantly elevated compared to PBS-treated, infected with IAV, * p < 0.05, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test.

Figure 6. GCS treatment of mice inhibits the induction of the anti-viral response following IAV infection.

Groups of 5–7 mice were treated with 20 μg GCS via the intranasal route 48 hrs and 24 hrs prior to inoculation with 10² PFU of IAV (HKx31). Mice received an additional dose of GCS 24 hrs post-infection. Uninfected PBS and GCS controls were included for comparison. (A) Mouse IFNα, IFNβ and IFNλ2 protein levels in BAL fluids 24 hrs post-infection were determined by ELISA. Bars represent the mean concentration ± SD. The detection limit of each assay is indicated as a dotted line. mRNA expression of (B) IFN genes Ifnα, Ifnβ and Ifnλ₂ and (C) anti-viral genes Ifit1, Ifiti2 and Isg15, in lung tissue 48 hrs post-infection was detected by RT-PCR. Data is relative to the expression of Gapdh. Bars represent the mean ± SD. Data is representative of 2 independent experiments. Significantly reduced compared to PBS-treated IAV-infected mice, ** p < 0.01, *** p < 0.001, one-way ANOVA and Tukey's post-test.

Figure 7. Intranasal administration of recombinant IFN following IAV infection improves the severe disease observed in GCS-treated mice.

Groups of 5 C57BL/6 mice were treated with 20 μg of GCS via the intranasal route 48 hrs and 24 hrs prior to inoculation with 10² PFU of IAV (HKx31). Mice received additional doses of GCS at 24 hrs post-infection and every 48 hrs thereafter. At day 1 post-infection, GCS-treated mice were administered 3000 IU of mouse IFNα₁, IFNβ or IFNλ₂ via the intranasal route. (A) Mice were weighed daily and data represents the mean % weight change ± SEM. Animals that had lost ≥15% of their original body weight were euthanized. (B) Survival curves are shown. (C) Titers of infectious virus in lung homogenates were determined by plaque assay at day 2 post-infection (24 hrs following IFN treatment). Bars represent the mean viral titer from a group of 5 mice ± SD. Virus titers from GCS-treated mice treated with IFN were significantly lower than GCS-treated mice not treated with IFN, *** p < 0.001, one-way ANOVA and Tukey's post-test.