Increased cytokine response of rhinovirus-infected airway epithelial cells in chronic obstructive pulmonary disease

Abstract

Rationale: Airway inflammation is a central feature of chronic obstructive pulmonary disease (COPD). COPD exacerbations are often triggered by rhinovirus (RV) infection.

Objectives: We hypothesized that airway epithelial cells from patients with COPD maintain a proinflammatory phenotype compared with control subjects, leading to greater RV responses.

Methods: Cells were isolated from tracheobronchial tissues of 12 patients with COPD and 10 transplant donors. Eight patients with COPD had severe emphysema, three had mild to moderate emphysema, and one had no emphysema. All had moderate to severe airflow obstruction, and six met criteria for chronic bronchitis or had at least one exacerbation the previous year. Cells were grown at air-liquid interface and infected with RV serotype 39. Cytokine and IFN expression was measured by ELISA. Selected genes involved in inflammation, oxidative stress, and proteolysis were assessed by focused gene array and real-time polymerase chain reaction.

Measurements and main results: Compared with control subjects, cells from patients with COPD demonstrated increased mRNA expression of genes involved in oxidative stress and the response to viral infection, including NOX1, DUOXA2, MMP12, ICAM1, DDX58/RIG-I, STAT1, and STAT2. COPD cells showed elevated baseline and RV-stimulated protein levels of IL-6, IL-8/CXCL8, and growth-related oncogene-alpha/CXCL1. COPD cells demonstrated increased viral titer and copy number after RV infection, despite increased IL-29/IFN-lambda1, IL-28A/IFN-lambda2, and IFN-inducible protein-10/CXCL10 protein levels. Finally, RV-infected COPD cultures showed increased mRNA expression of IL28A/IFNlambda2, IL29/IFNlambda1, IFIH1/MDA5, DDX58/RIG-I, DUOX1, DUOX2, IRF7, STAT1, and STAT2.

Conclusions: Airway epithelial cells from patients with COPD show higher baseline levels of cytokine expression and increased susceptibility to RV infection, despite an increased IFN response.

Figure 1.

Histology of normal and chronic obstructive pulmonary disease (COPD) airway epithelial cell cultures. Passage one cells from normal lungs (A and B) or lungs of patients with COPD (C and D) were grown at air–liquid interface, and cells along with the membrane were fixed and embedded in paraffin. Cross-sections were stained with hematoxylin and eosin (A and C) or periodic acid Schiff (PAS) reagent (B and D). Number of PAS-positive cells per 100 μM was counted in five random fields in each culture and expressed as mean (±SEM) (E). Both normal cells and COPD cells differentiated into mucociliary phenotype. However, COPD cells show more PAS-positive cells. Images are representative of cultures from 10 normal donors and 12 patients with COPD. Arrows point to PAS positive cells. Data in E represent average and SEM (n = 10–12, *different from normal P < 0.05, t test).

Figure 2.

Proinflammatory cytokine production by normal and chronic obstructive pulmonary disease (COPD) airway epithelial cells under basal and stimulated conditions. Normal and COPD cells were grown at air–liquid interface for 3 weeks. Cultures were shifted to fresh hydrocortisone-free media and incubated for 24 hours. Cell cultures were infected apically with rhinovirus (RV) and incubated for another 24 hours at 33°C. Levels of IL-6 (A), IL-8/CXCL-8 (B), and growth-related oncogene-α/CXCL1 (C) present in basolateral media was determined by ELISA. COPD cells showed significantly higher basal levels of all three cytokines, which further increased with RV infection. Data represent the range and geometric mean (n = 10–12; *different from similarly treated normal cells, P ≤ 0.05, nonparametric analysis of variance with Dunn's post hoc test; ^†different from respective phosphate-buffered saline (PBS)–treated control group, P ≤ 0.05, Mann-Whitney test). UV-RV = ultraviolet-irradiated RV.

Figure 3.

Viral load in rhinovirus (RV)-infected airway epithelial cells. Mucociliary-differentiated normal and chronic obstructive pulmonary disease (COPD) cultures were shifted to hydrocortisone-free media and incubated for 24 hours. Cells were then infected apically with RV or ultraviolet-irradiated RV (UV-RV) and incubated for 24 or 48 hours. The apical surface was rinsed with sterile phosphate-buffered saline and RNA was isolated from the cells. (A) Viral RNA copy number was determined by quantitative real-time polymerase chain reaction and normalized to 18S RNA. (B) Apical rinses from the cultures were used to determine the load of infectious virus, which was expressed as 50% tissue culture infective dose per milliliter. COPD cells showed a higher viral load than normal cultures at both 24 and 48 hours. Open circles = normal 24 h; open squares = normal 48 h; solid circles = COPD 24 h; solid squares = COPD 48 h. Data represent the range and geometric mean (n = 9–12, *P ≤ 0.05, Mann-Whitney test).

Figure 4.

IFN and IP-10 expression in normal and chronic obstructive pulmonary disease (COPD) airway epithelial cells in response to rhinovirus (RV) infection. Cell cultures were infected with RV, as described in Figure 3 legend, and incubated for 24 or 48 hours. Levels of IFN-λ1 (A), IFN-λ2 (B), and IP-10 (C) in the basolateral media were determined by ELISA. Both normal and COPD cells showed increased levels of IFN-λ1, IFN-λ2, and IFN-inducible protein-10 in response to RV infection, although COPD cells showed significantly higher absolute concentrations of both cytokines than normal cells. Data represent range and geometric mean (n = 9–12; *different from normal cells, P ≤ 0.05, nonparametric analysis of variance with Dunn's post hoc test; ^†different from respective phosphate-buffered saline [PBS]–treated control group, P ≤ 0.05, Mann-Whitney test).

Figure 5.

Airway epithelial cell mRNA levels of oxidative stress–related genes, MMP and TIMP, in normal and chronic obstructive pulmonary disease (COPD) cells under unstimulated conditions. Total RNA was isolated from mucociliary-differentiated cultures and subjected to focused gene arrays. Results are expressed as fold change compared with housekeeping genes. Compared with normal cells, COPD cells showed significant increases in the expression of mRNAs encoding Nox1 (A), DuoxA2 (B), and matrix metalloproteinase (MMP) 12. (C) In contrast, nitric oxide synthase (NOS) 2 mRNA levels were decreased in COPD cells. Data represent the range and geometric mean (n = 9–11; P values calculated by Mann-Whitney test).

Figure 6.

Airway epithelial cell mRNA expression of ICAM-1, TREM1, IFIH1/MDA-5, DDX58/RIG-I, and antiviral proteins under basal conditions. RNA isolated from normal and chronic obstructive pulmonary disease (COPD) mucociliary-differentiated cultures was subjected to focused gene arrays, and the results expressed as fold change over housekeeping genes. Compared with normal, COPD cells show significant increases in ICAM-1 (A), TREM-1 (B), DDX58/RIG-I (D), IFN-α (E), OAS (F), OASL (G), STAT1 (H), and STAT2 (I). Data represent range and geometric mean (n = 9; P values calculated by Mann-Whitney test).

Figure 7.

Airway epithelial cell mRNA expression of HDAC and gene repressors, SIRT1 and NCOR. Total RNA from normal and chronic obstructive pulmonary disease (COPD) cells was subjected to real-time polymerase chain reaction and the data expressed as fold change over the housekeeping gene, G3PDH. mRNA levels of HDAC2 (A) and HDAC4 (B) levels were similar in COPD and normal cells. In contrast, SIRT1 (C) and NCOR (D) gene expression showed a trend toward a reduction in COPD cells. Data represent range and geometric mean (n = 9, P values were calculated by Mann-Whitney test).