Anti-H7N9 avian influenza A virus activity of interferon in pseudostratified human airway epithelium cell cultures

H7N9 A/Anhui/1/2013 was obtained from the Chinese National Influenza Center, and all experiments with this virus were performed in approved enhanced biosafety level 3 (BSL-3) laboratories. A549 cells were cultured in Dulbecco’s modified Eagle medium (DMEM; Gibco, NY, USA) with 10% fetal bovine serum (Gibco).

Primary HAE cells were isolated from patients who underwent surgical lung resection for pulmonary diseases in Nanjing Children’s Hospital, as described previously [19]. HAE cells were plated onto type I and III collagen-coated six-well tissue culture plates and cultured in BEGM media (Lonza, Germany) supplemented with the required additives (Lonza). When the cells reached 80–90% confluence, traditional monolayer two-dimensional (2D) HAE cells were dissociated with trypsin, and 3 × 10⁵ cells were seeded on type IV collagen-coated 12-well transwell inserts (Costar, ME, USA). Medium was renewed for both the apical and basolateral surfaces every other day. The medium was then changed to air–liquid interface (ALI) medium (BEGM+DMEM+additives) until the HAE cells reached full confluence. After 5 days, the HAE cells were exposed to air, and only the basolateral compartment was cultured with ALI medium. The ALI culture was continued for 4–6 weeks, during which time the cells differentiated into 3D pseudostratified HAE cells. Prior to the experiments, all cultures were maintained at 37 °C in a 5% CO₂ incubator.

3D HAE cells that had been infected with 100 tissue culture infective doses (TCID₅₀) of H7N9 for 24 h were fixed with 4% paraformaldehyde (PFA) for 30 min at room temperature, followed by washing of both the apical and basolateral sides three times each with PBS. Subsequently, the fixed cells were permeabilized with 0.2% Triton X for 2 h and blocked with 5% bovine serum albumin for 1 h. For the simultaneous detection of H7N9 protein M2 and ciliated cells or tight junctions, anti-influenza A M2 rabbit polyclonal antibody (PA5–32233, 1:500, Thermo Fisher), and mouse monoclonal anti-β tubulin antibody (T4026, 1:100, Sigma) or mouse monoclonal anti-ZO-1 antibody (33–9100, 1:500, Invitrogen, CA, USA), respectively, were applied as primary antibodies. Goat-derived Dylight 594-labeled anti-mouse IgG (H + L) (35,511, 1:500, Thermo Fisher) and Dylight 488-labeled anti-rabbit IgG (H + L) (35,553, 1:500, Thermo Fisher) were applied as secondary antibodies. Nuclei were counterstained with DAPI. Finally, the filters with cells were excised from the insert and mounted under coverslips on glass slides with mounting medium. Confocal images were taken with an UltraView VoX confocal microscope (PerkinElmer, Boston, MA, USA).

The apical surface of 3D HAE cells was rinsed three times with PBS and then incubated with 100 μl of 0.25 μg/ml rhIFN-λ1 (R&D, USA) or 60 IU/ml rhIFN-α2b (Yuance, China) for 20 h at 37 °C in a 5% CO₂ incubator. After discarding the cell culture supernatant, the 3D HAE cells were inoculated on the apical surface with 100 μl of 1 TCID₅₀ of H7N9 per well and incubated for 1 h. Virus and rhIFNs diluents were prepared with ALI medium. At the end of the incubation, the unbound H7N9 viral particles were removed by rising with Hank’s buffer, and the basolateral compartment was supplemented with 1.5 ml of ALI medium for 24 h at 37 °C in a 5% CO₂ incubator. Subsequently, 200 μl of ALI medium was added to the apical surface, and the supernatants were collected after 30 min. Virus RNA was extracted using the QIAamp MinElute Virus Spin Kit (Qiagen, Hilden, Germany), and quantification of H7N9 viral yields was performed using AgPath-ID™ One-Step RT-PCR Reagents (ABI, USA) to analyze the anti-H7N9 effect of different rhIFNs.

For A549 cells, the experimental method was similar to that used for 3D HAE cells, except that the A549 cells were cultured on 24-well plates and the applied virus medium was DMEM with added TPCK-trypsin (0.5 μg/ml). Nucleic acids were extracted and quantified once per day for 10 days. Real-time PCR assays were used for the detection of the HA gene from H7N9 A/Anhui/1/2013. The primer sequences were H7N9-F 5′-AGAAATGAAATGGCTCCTGTCAA-3′ and H7N9-R 5′-GGTTTTTTCTTGTATTTTTATATGACTTAG-3′, and the probe sequence was H7N9-P FAM-AGATAATGCATTCCCGCAGAT-TAMRA.

2D HAE cells were rinsed three times with PBS and then incubated with 100 μl of successive 4⁰- to 4⁹-fold dilutions of 4 μg/ml rhIFN-λ1 (R&D) or 1000 U/ml rhIFN-α2b (Yuance) for 20 h at 37 °C in a 5% CO₂ incubator. Next, the cells were challenged with 100 μl of 100 TCID₅₀ of H7N9 per well. After incubation for 1 h, the unbound H7N9 viral particles were removed by rising with Hank’s buffer, and the cells were supplemented with 100 μl of BEGM medium. Harvests were collected at 72 h post-inoculation, and H7N9 RNA extraction and nucleic acid quantification were performed as described above to analyze the anti-H7N9 effect of different rhIFNs. For 3D HAE cells and A549 cells, the experimental protocols were the same as that described for 2D HAE cells, but the medium used for cell culture was ALI medium and DMEM, respectively.

The apical surface and the basolateral chamber of 3D HAE cells were cultured with 4 μg/ml rhIFN-λ1 (R&D, USA) or 1000 IU/ml rhIFN-α2b (Yuance, China) for 20 h at 37 °C in a 5% CO₂ incubator (500 μl and 1.5 ml for the apical surface and basolateral chamber, respectively). The apical surface was washed three times with PBS, followed by inoculation with 100 μl of 100 TCID₅₀ of H7N9, and the basolateral chamber was maintained with 1.5 ml of ALI medium. After incubation for 1 h, the apical chamber was washed twice with PBS and maintained continuously with an ALI for 24 h. The cells were fixed in 4% PFA along with the entire transwell insert for hematoxylin-eosin (HE) staining.

A549 cells, 2D HAE cells, and well-differentiated pseudostratified 3D HAE cells were each incubated with 60 IU/ml rhIFN-α2b (Yuance) or 0.25 μg/ml rhIFN-λ1 (R&D) for 18 h at 37 °C in a 5% CO₂ incubator. After rinsing with PBS, the cells were lysed with Trizol, and the total RNA was extracted. Reverse transcription was performed using the Superscript® III First-Strand Synthesis System (Invitrogen). The relative expression levels of ISGs, including ISG15, MX1, and OAS1, were analyzed by a Taqman real-time PCR assay using the 2^-ΔΔCT method, and GADPH was used as the internal reference [20]. The sequences of primers and probes are shown in Table 1.

β-tubulin is an important component of the cytoskeleton and is a marker protein of cilia, so it can be used to distinguish between ciliated cells and non-ciliated cells. The tight junction protein zonula occludens-1 (ZO-1) is located at the apex of epithelial cells; it is a major component of tight junctions and plays an important role in maintaining pseudostratified cell layer integrity and barrier function. As shown by immunofluorescence staining with anti-β-tubulin IV (Fig. 1a) and anti-ZO1 (Fig. 1b) antibodies, H7N9 could infect both non-ciliated and ciliated cells. In contrast with the state of the cilia in the mock control, 3D HAE cells that were infected with H7N9 showed an obvious reduction of cilia. Additionally, the H7N9-infected 3D HAE cells showed a disassociation of the ZO-1, suggesting airway epithelial damage.

A549 cells and 3D HAE cells were each challenged with 1 TCID₅₀ of H7N9 after being treated with rhIFN-α2b (60 IU/ml) or rhIFN-λ1 (250 ng/ml) for 20 h. The viral load was then detected by real-time PCR for 10 consecutive days using the method described above. In the control group without rhIFN treatment, the viral load in A549 cells reached a peak of 3.9 × 10⁹ copies/ml on the sixth day and decreased to 3.5 × 10⁶ copies/ml on the tenth day (Fig. 2a). In control 3D HAE cells, the viral load reached a peak of 1.29 × 10⁹ copies/ml on the third day and was reduced to 1.1 × 10⁶ copies/ml on the tenth day (Fig. 2b). In contrast, the viral load of A549 cells treated with rhIFN peaked on the second day and then gradually decreased (Fig. 2a). For 3D HAE cells, viral replication was inhibited during the first few days after rhIFN treatment; the viral load was reduced to 1.58 × 10⁴ copies/ml (rhIFN-α2b) or 9.55 × 10³ copies/ml (rhIFN-λ1) on the third day. Later, it gradually rose to a peak of 4.26 × 10⁸ copies/ml (rhIFN-α2b) or 6.31 × 10⁸ copies/ml (rhIFN-λ1) on the sixth day and then gradually decreased again (Fig. 2b).

A549 cells, 2D HAE cells and 3D HAE cells were each incubated with serial dilutions of rhIFNs (initial concentration of rhIFN-λ1 and rhIFN-α2b was 4 μg/ml and 1000 IU/ml, respectively) for 20 h, after which these cells were challenged with 100 TCID₅₀ of H7N9. Viral RNA was extracted 72 h later, and quantitative real-time PCR assay targeting the H7N9 HA gene was conducted to quantify the viral load. Among the three cell types, viral proliferation was the fastest in 3D HAE cells, reaching 1 × 10¹⁰ copies/ml, while the HA gene copy number in A549 cells and 2D HAE cells was 2.5 × 10⁷ copies/ml and 8 × 10⁷ copies/ml, respectively. The antiviral activity of rhIFN-α2b was superior to that of rhIFN-λ1 in all three cell types, A549 cells (Fig. 3a), 2D HAE cells (Fig. 3b), and 3D HAE cells (Fig. 3c). As the dose of rhIFN decreased, the antiviral activity also decreased, which demonstrates a dose-response relationship. Compared with the control group, the viral load was decreased to 2.8 × 10⁶ copies/ml and 2.3 × 10⁷ copies/ml, which means 3571-fold and 435 -fold (1 × 10¹⁰ copies/ml) reduction, respectively, when the 3D HAE cells were treated with 4 μg/ml rhIFN-λ1 or 1000 IU/ml rhIFN-α2b. Because of the biggest drop in viral load, the antiviral activity of rhIFN was the strongest when it was applied to the 3D cells (Fig. 3c).

HE staining of ALI-cultured HAE cells revealed that the nuclear position was uneven, there appeared to be multi-layered cells with cilia distributed at the top, and the morphological structure was similar to the pseudostratified ciliated columnar epithelium of human tracheal tissue. In the H7N9 infection group, there were many vacuoles, the number of cilia was reduced, and the cell layer was thinner. However, when 3D HAE cells were incubated with rhIFNs and then challenged with H7N9, the cytopathic effect of H7N9 infection was alleviated (Fig. 4).

The relative expression levels of OAS1, MX1, ISG15, and IFITM3 were detected by real-time PCR using GADPH as the internal reference. Compared with the control group without rhIFNs treatment, cells treated with either rhIFN-α2b or rhIFN-λ1 had significantly upregulated expression levels of all four genes. For A549 cells and 2D HAE cells, the expression levels of these four genes were significantly higher in the rhIFN-λ1 group than in the rhIFN-α2b group. In contrast, the ISG expression levels were significantly higher in the rhIFN-α2b group than in the rhIFN-λ1 group for the 3D HAE cells (Fig. 5). These results demonstrate that the gene expression induced by rhIFNs was different between 3D HAE cells and ordinary 2D-cultured cells.