Differential induction of interferon stimulated genes between type I and type III interferons is independent of interferon receptor abundance

Abstract

It is currently believed that type I and III interferons (IFNs) have redundant functions. However, the preferential distribution of type III IFN receptor on epithelial cells suggests functional differences at epithelial surfaces. Here, using human intestinal epithelial cells we could show that although both type I and type III IFNs confer an antiviral state to the cells, they do so with distinct kinetics. Type I IFN signaling is characterized by an acute strong induction of interferon stimulated genes (ISGs) and confers fast antiviral protection. On the contrary, the slow acting type III IFN mediated antiviral protection is characterized by a weaker induction of ISGs in a delayed manner compared to type I IFN. Moreover, while transcript profiling revealed that both IFNs induced a similar set of ISGs, their temporal expression strictly depended on the IFNs, thereby leading to unique antiviral environments. Using a combination of data-driven mathematical modeling and experimental validation, we addressed the molecular reason for this differential kinetic of ISG expression. We could demonstrate that these kinetic differences are intrinsic to each signaling pathway and not due to different expression levels of the corresponding IFN receptors. We report that type III IFN is specifically tailored to act in specific cell types not only due to the restriction of its receptor but also by providing target cells with a distinct antiviral environment compared to type I IFN. We propose that this specific environment is key at surfaces that are often challenged with the extracellular environment.

Fig 1. Kinetics of type I and type III IFN-mediated antiviral activities in human mini gut-organoids.

(A-B) Colon organoids were pre-treated with the indicated concentrations of type I IFN (β) or type III IFN (λ1−3) for 2.5 h prior to infection with vesicular stomatitis virus (VSV) expressing Firefly luciferase (VSV-Luc) using a multiplicity of infection (MOI) of 1. Viral replication was assayed by measuring the luciferase activity. (A) The relative antiviral protection is expressed as a percentage of total protection in VSV-infected organoids or (B) as the EC90 corresponding to the concentration of type I IFN (β) or type III IFN (λ1−3) resulting in 90% inhibition (10% infection) of viral replication. (C-D) Colon organoids were treated with type I IFN (β) (2,000 RU/mL equivalent 0.33 nM) or type III IFN (λ1−3) (100ng/mL each or total 300 ng/mL equivalent 13.7 nM) for different times prior to infection with VSV-Luc. Viral replication was assayed by measuring luciferase activity. (C) The relative VSV infection is expressed as the percentage of the luciferase activity present in VSV-infected organoids without IFN treatment (set to 100). (D) Pre-incubation time of type I IFN (β) or type III IFN (λ1−3) required to inhibit VSV infection to 10% (90% inhibition). (E-F) Same as (C-D), except colon organoids were treated at the indicated times post-infection with VSV-Luc. (F) Delayed-time post-infection for type I IFN (β) or type III IFN (λ1−3) to still inhibit VSV infection to 90% (10% inhibition). Data in (A–F) represent the mean values of two independent experiments. Error bars indicate the SD. *

Fig 2. Type III IFNs have a lower transcriptional activity compared to type I IFNs.

(A-B) Human colon organoids were stimulated with indicated concentrations of type I (β) or III IFN (λ1−3) for different times and the transcript levels of the ISGs IFIT1 and Viperin were analyzed by qRT-PCR. Data are normalized to TBP and HPRT1 and are expressed relative to untreated samples at each time point. A representative experiment with technical triplicates, out of three independent experiments is shown. Mean values and SD are shown. (C) Colon organoids were treated with type I IFN (β) (2,000 RU/mL equivalent 0.33 nM) or type III IFN (λ1−3) (300 ng/mL equivalent 13.7 nM) for the indicated times and identification of the IFN-induced ISGs was performed by qRT-PCR. A total of 65 out of 132 ISGs tested were found to be significantly induced more than 2-fold compared with a baseline (mean of untreated controls at the particular time points) for at least one time point by at least one IFN treatment. Data are normalized to TBP and HPRT1 and visualized in a heatmap using R after sorting the fold change of expression in response to type I IFN (β) in decreasing order. (D) Comparison of expression values (log2 (Fold Change)) for all genes induced at the indicated times with type I IFN (β) versus type III IFN (λ1−3). Solid line indicates equivalent expression.

Fig 3. Type III IFNs present delayed transcriptional activity compared to type I IFNs.

(A-D) Human colon organoids were treated with type I IFN (β) (2,000 RU/mL equivalent 0.33 nM) or type III IFN (λ1−3) (300 ng/mL equivalent 13.7 nM) for 3, 6, 12 or 24 hours and the kinetic pattern of expression of the 65 significantly up-regulated ISGs were analyzed by qRT-PCR in triplicates. Data are normalized to TBP and HPRT1 and are expressed relative to untreated cells at each time point. Hierarchical clustering analysis of these genes produced four distinct temporal expression patterns (Groups 1–4) based on the time-point of the maximum induction in response to type I IFN (β) or type III IFN (λ1−3). Color codes have been used to visualize the induction peak per group. (A-B) Gray lines show the normalized kinetic expression of each gene for each group upon treatment with (A) type I IFN (β) or (B) type III IFN (λ1−3). The colored lines are the average of the kinetic profiles for the genes of each group. (C) Gene expression heat map showing the genes clustered in their respective temporal expression patterns groups in response to type I IFN (β) or type III IFN (λ1−3). The genes per group are sorted in decreasing order on the basis of their fold change of expression in response to type I IFN (β) or type III IFN (λ1−3) and only showing the highest expressed values within the temporal groups omitting all other values for visualization. (D) Number of genes belonging to each group.

Fig 4. Validation of the unique kinetic patterns of ISG expression upon type I versus type III IFN treatment.

(A-D) Human colon organoids were stimulated with increasing concentrations of type I IFN (β) or III IFN (λ1−3) for indicated times and the kinetic pattern of expression of one representative ISG from each temporal expression patterns groups 1–4 was analyzed by qRT-PCR, (left column) type I IFN (β), (right column) type III IFN (λ1−3) treated organoids. Data are normalized to HPRT1 and are expressed relative to untreated cells at each time point. A representative experiment with technical triplicates. Mean values and SD are shown.

Fig 5. Mathematical modeling of type I and type III IFN responses.

(A) Scheme of the mathematical model. IFNs bind to their cognate receptors and activate them; all molecules are also subject to degradation (∅). Active receptors mediate STAT phosphorylation while phosphorylated STAT (p-STAT) drives ISG expression. ISGs may include negative feedback regulator of STAT activation. Dashed lines indicate the potential sources of difference between the two pathways. Red dashed lines show the sources of the difference between the two pathways implemented in the best-fitting model. (B) Model selection. Models fitted to the experimental data were ranked using the Akaike information criterion corrected for small sample size (AICc) and the AICc weight, as a measure of support for the given model by the data. (C-D) The best-fitting model M3 reproduces the Viperin expression dynamics upon treatment with different concentrations of (C) type I IFN and (D) type III IFN (see S3A and S3B Fig for experimental data). In (C) and (D), the solid lines represent the best fits and the shaded areas are 95% confidence intervals. (E) Simulation of the maximum Viperin induction upon treatment with equal concentrations of type I IFN or type III IFN.

Fig 6. Overexpression of type I and type III IFN receptor increases the transcriptional activity of both cytokines.

(A-F) Wild-type T84 cells were transduced with rIFNAR1 or rIFNLR1 to create stable lines overexpressing either IFN receptors. (A-B) T84 wild-type cells (WT) and T84 cells overexpressing rIFNAR1 (WT+rIFNAR1) were treated with type I IFN (β) (2,000 RU/mL equivalent 0.33 nM) or type III IFN (λ1−3) (300 ng/mL equivalent 13.7 nM) for 1h and IFN signaling was measured by immunoblotting for pSTAT1 Y701. Actin was used as a loading control. A representative immunoblot out of three independent experiments is shown. (C-D) T84 wild type cells (WT) and T84 cells overexpressing rIFNAR1 (WT+rIFNAR1) were treated with increasing concentrations of type I IFN (β) for 12 hours or type III IFN (λ1−3) for 24 hours and the transcript levels of the ISGs IFIT1 and Viperin were analyzed by qRT-PCR. Data are normalized to HPRT1 and are expressed relative to untreated cells at each time point. (E-H) Same as (A-D), except T84 cells overexpressing rIFNLR1 (WT+IFNLR1) were used. The mean value obtained from three independent experiments is shown. Error bars indicate the SD. *

Fig 7. Expression kinetics of ISGs are independent of the IFN receptor levels.

(A-D) Wild-type T84 cells were transduced with rIFNAR1 or rIFNLR1 to create stable lines overexpressing either receptors. (blue panels) T84 wild-type cells (WT) and T84 cells overexpressing the IFNAR1 (WT+rIFNAR1) were treated with increasing concentrations of type I IFN (β) for the indicated times and the kinetic pattern of expression of one representative ISG from each temporal expression patterns groups 1–4 was analyzed by qRT-PCR. Data are normalized to HPRT1 and are expressed relative to untreated cells at each time point. (red panels) Same as (blue panels), except T84 cells overexpressing the IFNLR1 (WT+IFNLR1) were used and treated with increasing concentrations of type III IFN (λ1−3). A representative experiment with technical triplicates, out of three independent experiments is shown. Mean values and SD are shown.

Fig 8. Type III IFN mediated expression kinetics of ISGs are independent of differential levels of IFNLR1 receptor.

Wild-type T84 cells were transduced with rIFNLR1-GFP to create a stable line overexpressing IFNLR1 tagged with GFP. (A) WT cells overexpressing IFNLR1-GFP (WT+IFNLR1-GFP) from the same population were separated by cell sorting into three populations: non (neg)-, low- and high-expressing GFP cells. Gates were created based on the auto-fluorescence of WT cells. (B) WT and WT+IFNLR1-GFP cells were treated with type III IFN (λ1−3) (300 ng/mL equivalent 13.7 nM) for 3, 6, 12 and 24 hours prior to sorting in neg-, low- and high-expressing IFNLR1-GFP cells. The kinetic pattern of expression of one representative ISG from each temporal expression patterns groups 1–4 was analyzed by qRT-PCR in each sorted population. Data are normalized to HPRT1 and are expressed relative to untreated cells at each time point. A representative experiment with technical triplicates, out of two independent experiments is shown. Mean values and SD are shown.