IFNλ is a potent anti-influenza therapeutic without the inflammatory side effects of IFNα treatment

Abstract

Influenza A virus (IAV)-induced severe disease is characterized by infected lung epithelia, robust inflammatory responses and acute lung injury. Since type I interferon (IFNαβ) and type III interferon (IFNλ) are potent antiviral cytokines with immunomodulatory potential, we assessed their efficacy as IAV treatments. IFNλ treatment of IAV-infected Mx1-positive mice lowered viral load and protected from disease. IFNα treatment also restricted IAV replication but exacerbated disease. IFNα treatment increased pulmonary proinflammatory cytokine secretion, innate cell recruitment and epithelial cell death, unlike IFNλ-treatment. IFNλ lacked the direct stimulatory activity of IFNα on immune cells. In epithelia, both IFNs induced antiviral genes but no inflammatory cytokines. Similarly, human airway epithelia responded to both IFNα and IFNλ by induction of antiviral genes but not of cytokines, while hPBMCs responded only to IFNα. The restriction of both IFNλ responsiveness and productive IAV replication to pulmonary epithelia allows IFNλ to limit IAV spread through antiviral gene induction in relevant cells without overstimulating the immune system and driving immunopathology. We propose IFNλ as a non-inflammatory and hence superior treatment option for human IAV infection.

Keywords: immunopathology; infection; influenza; interferon alpha; interferon lambda.

Figure 1. Therapeutic administration of IFNα and IFNλ differentially influences the outcome of IAV‐induced disease

1. A

   Relative antiviral activity of IFNα (circles) or IFNλ (triangles). AEC cultures were stimulated for 4 h with stated IFN at specified concentrations (ng/ml) and induction of indicated ISGs was assessed by qPCR (data shown are representative of four independent experiments, n = 3–4).

2. B, C

   Mice were pretreated with equivalent doses of IFNα (1.45 μg/50 μl) or IFNλ (2.6 μg/50 μl) or Veh Ctrl (squares, 50 μl PBS) 24 h prior to infection with PR8; weight loss and survival were assessed throughout infection (B), and viral load (C) assessed at 4 dpi (data shown are representative of two independent experiments, n = 8–10 (B), n = 3 (C)).

3. D, E

   Mice were infected with PR8 and treated with equivalent doses of IFNα or IFNλ or Veh Ctrl at days 2, 4 and 5 post‐infection; survival and weight loss were monitored (D, data pooled from 4 independent experiments, n = 12–29) and viral load assessed at 4 dpi (E) (data representative of two independent experiments, n = 3–5).

Data information: Significance assessed by log‐rank (Mantel–Cox) test (survival), two‐way ANOVA (weight loss) and unpaired t‐tests (viral load). *indicates IFNα:Veh Ctrl, ⁺indicates IFNλ:Veh Ctrl, and ^°indicates IFNα:IFNλ. *P = 0.0236, ⁺ P = 0.0236 ***P < 0.0001, ⁺⁺⁺ P < 0.001 (B); *P = 0.012, ⁺ P = 0.012 (C); *P = 0.0443, ⁺ P = 0.035, ^°° P = 0.0015 (D); **P = 0.0081, ⁺⁺ P = 0.0066 (E). Symbols on the right of graphs indicate significance of whole curve. Graphs show mean ± SEM.

Figure EV1. Comparison of epithelial responsiveness to IFNα4 and IFNλ2

AEC cultures were stimulated for 4 h with IFNα4 (circles) or IFNλ2 (triangles), and induction of indicated ISGs was assessed by qPCR. Prism 6 software was used to perform a curve fit (Sigmoidal, 4PL) and generate a dose–response curve and a half‐maximal effective concentration (EC₅₀) for each gene assessed for each treatment. A conversion ratio was then generated using the indicated formula for each ISG and a final conversion ratio was taken from the average of all ISGs assessed. Values represent means ± SEM. Data are pooled from two independent experiments where n = 3–7 per experiment.

Figure 2. IFNα treatment correlates with increased inflammation during IAV infection

1. A, B

   Mice were infected with PR8 and treated with IFNα (circles, 1.45 μg/50 μl), IFNλ (triangles, 2.6 μg/50 μl) or Veh Ctrl (squares) as previously stated. Concentrations of stated proinflammatory cytokines in BAL fluid were measured by multiplex cytokine assay (A) and flow cytometric quantification of pDCs and inflammatory monocytes in the lung was performed (B) (data shown are representative of two independent experiments, n = 2–6).

2. C, D

   Lung sections from control and infected mice treated as indicated were stained by TUNEL for apoptotic cells at 6 dpi. Quantification of TUNEL⁺ cells in whole lung slides by Icy‐Spot Detector (ICY‐R3M2Y2) (C) (data shown are pooled from three independent experiments, n = 3–8). Red arrowheads indicate TUNEL signal (D). Scale bar, 200 μM (data shown are representative of two independent experiments, n = 3–4).

Data information: Significance assessed by two‐way ANOVA with Bonferroni post‐tests (where *denotes IFNα:Veh Ctrl, ⁺indicates IFNλ:Veh Ctrl, and ^°indicates IFNα:IFNλ). Symbols on the right of graphs indicate statistical significance of the whole curve. IL‐6 whole curve: **P = 0.0041, ^° P = 0.0144. IL‐6 5 dpi: **P = 0.001884, ^°° P = 0.001645. IP‐10 5 dpi: **P = 0.004897, ^°° P = 0.005354. MCP‐1 5 dpi: **P = 0.007473, ^°° P = 0.002003. Eotaxin whole curve: *P = 0.0235, ^° P = 0.0386. Eotaxin 5 dpi ***P = 0.000149, ^°° P = 0.001975. Mip‐1α 5 dpi: ^°° P = 0.002921 (A). Plasmacytoid dendritic cells: *P = 0.0211, ^°° P = 0.006965. Inflammatory monocytes **P = 0.007842, ^°° P = 0.000895 (B). **P = 0.0011, ⁺⁺ P = 0.0051, ^°°° P = 0.0005 (C). Graphs show mean ± SEM.

Figure EV2. Immune cell recruitment and lymphocyte activation upon IFNα and IFNλ treatment

Mice were infected with PR8 and treated with IFNα (circles, 1.45 μg/50 μl), IFNλ (triangles, 2.6 μg/50 μl) or Veh Ctrl (squares, 50 μl PBS). Flow cytometric quantification (A) of lymphocyte subsets and neutrophils and (B) of CD69 expression on lymphocytes in the lung was performed (data are representative of two independent experiments, n = 3–5). Significance assessed by two‐way ANOVA where *indicates IFNα:Veh Ctrl and ^°indicates IFNα:IFNλ. IFNλ:Veh Ctrl was not significant. CD69⁺ CD4⁺ T cells: *P = 0.0159, ^° P = 0.0138. CD69⁺ CD8⁺ T cells: **P = 0.002, ^°°° P = 0.0003. CD69⁺ B cells: ****P < 0.0001, ^°° P = 0.0069. CD69⁺ NK cells: **P = 0.0012, ^°°° P = 0.0005 (B). Graphs show mean ± SEM.

Figure 3. IFNα, but not IFNλ, treatment induces pulmonary cytokine secretion through activation of immune cells

1. A, B

   IL‐6, IP‐10 and MCP‐1 concentrations were measured by multiplex cytokine assay in AEC culture supernatants (A) and macrophage, pDC and cDC culture supernatants (B) at 24 h post‐stimulation with IFNα4 (0.725 ng/ml) or IFNλ2 (1.3 ng/ml) or LPS (AEC only) (data shown are representative of two independent experiments, n = 3–6).

2. C

   BAL samples taken from mice treated with IFNα, IFNλ or Veh Ctrl at specified time points (data shown are representative of two independent experiments, n = 5–6).

Data information: Significance assessed by unpaired t‐tests where *denotes IFNα:Veh Ctrl and ^°indicates IFNα:IFNλ. IFNλ:Veh Ctrl was not significant. IL6 pDC: ***P = 0.0004, ^°°° P = 0.0005, IL6 cDC: *P = 0.0102, ^° P = 0.0151. IP‐10 macrophage: **P = 0.0033, ^°° P = 0.0033, pDC: ****P < 0.0001, ^°°°° P < 0.0001, cDC: **P = 0.0013, ^°° P = 0.0013, MCP‐1 macrophage: ****P < 0.0001, ^°°° P = 0.003 (B). IL‐6 10 h: *P = 0.0112, ^° P = 0.0262, 18 h: *P = 0.0314, ^° P = 0.373. IP‐10 10 h: *P = 0.0261, ^° P = 0.0472. MCP‐1 10 h: **P = 0.0081, ^° P = 0.0206, 18 h: **P = 0.0089, ^° P = 0.01 (C). Graphs show mean ± SEM.

Figure 4. Pathogenicity‐related gene clusters are specifically induced by IFNα, not by IFNλ treatment

1. A–D

   Mice were treated with IFNα (1.45 μg/50 μl), IFNλ (2.6 μg/50 μl) or Veh Ctrl (50 μl PBS), and whole lungs were taken at 18 h post‐treatment for global analysis by Illumina.SingleColor.Mouse WG‐6V20R01127 microarrays. Samples (n = 5) were normalized to the median of the vehicle control group and filtered for a fold change of 1.5, yielding 553 genes differently regulated between treatments (one‐way ANOVA, P < 0.01, Benjamini–Hochberg multiple test correction), of which 429 genes are upregulated. K‐means clustering revealed six gene clusters, one of which encompassed genes primarily induced by IFNα4 (A), while the remaining clusters contained genes upregulated by both IFNα4 and IFNλ2 (B). The two clusters of genes were analysed by Ingenuity Pathway Analysis (IPA) (C, D).

Figure 5. IFNα, but not IFNλ, treatment induces cytokine secretion from human immune cells

1. A

   Human AEC cultures were stimulated for 4 h with IFNα (circles) or IFNλ (triangles) at specified concentrations and then assessed for stated ISG induction by qPCR (data are representative of 2 independent experiments, n = 3).

2. B, C

   ISG induction in human PBMCs was assessed at 4 and 24 h post‐IFNα (21 U/ml) or IFNλ (1.2 ng/ml) stimulation (B). PBMC proinflammatory cytokine secretion was measured by multiplex cytokine assay at 4 and 24 h post‐stimulation with IFNα or IFNλ (C) (data shown are pooled from six independent donors).

Data information: Significance tested by two‐way ANOVA where *denotes IFNα:Veh Ctrl and ^°indicates IFNα:IFNλ. IFNλ:Veh Ctrl was not significant. IRF7: **P = 0.0037. Rsad2 4 h: **P = 0.0038, ^°° P = 0.037; 24 h: *P = 0.0234. OAS1 4 h: *P = 0.0328, ^° P = 0.0358; 24 h: ****P < 0.0001 (B). IL‐6: *P = 0.0124, **P = 0.0021. MCP‐1 4 h: ****P < 0.0001; ^°°° P = 0.001; 24 h: **P = 0.0046, ^°° P = 0.005. IP‐10 4 h: **P = 0.0033, ^°° P = 0.0024; 24 h: **P = 0.0011, ^°°° P = 0.001 (C). Graphs show mean ± SEM.