The Timing and Magnitude of the Type I Interferon Response Are Correlated with Disease Tolerance in Arbovirus Infection

Abstract

Infected hosts possess two alternative strategies to protect themselves against the negative impact of virus infections: resistance, used to abrogate virus replication, and disease tolerance, used to avoid tissue damage without controlling viral burden. The principles governing pathogen resistance are well understood, while less is known about those involved in disease tolerance. Here, we studied bluetongue virus (BTV), the cause of bluetongue disease of ruminants, as a model system to investigate the mechanisms of virus-host interactions correlating with disease tolerance. BTV induces clinical disease mainly in sheep, while cattle are considered reservoirs of infection, rarely exhibiting clinical symptoms despite sustained viremia. Using primary cells from multiple donors, we show that BTV consistently reaches higher titers in ovine cells than cells from cattle. The variable replication kinetics of BTV in sheep and cow cells were mostly abolished by abrogating the cell type I interferon (IFN) response. We identified restriction factors blocking BTV replication, but both the sheep and cow orthologues of these antiviral genes possess anti-BTV properties. Importantly, we demonstrate that BTV induces a faster host cell protein synthesis shutoff in primary sheep cells than cow cells, which results in an earlier downregulation of antiviral proteins. Moreover, by using RNA sequencing (RNA-seq), we also show a more pronounced expression of interferon-stimulated genes (ISGs) in BTV-infected cow cells than sheep cells. Our data provide a new perspective on how the type I IFN response in reservoir species can have overall positive effects on both virus and host evolution. IMPORTANCE The host immune response usually aims to inhibit virus replication in order to avoid cell damage and disease. In some cases, however, the infected host avoids the deleterious effects of infection despite high levels of viral replication. This strategy is known as disease tolerance, and it is used by animal reservoirs of some zoonotic viruses. Here, using a virus of ruminants (bluetongue virus [BTV]) as an experimental system, we dissected virus-host interactions in cells collected from species that are susceptible (sheep) or tolerant (cow) to disease. We show that (i) virus modulation of the host antiviral type I interferon (IFN) responses, (ii) viral replication kinetics, and (iii) virus-induced cell damage differ in tolerant and susceptible BTV-infected cells. Understanding the complex virus-host interactions in disease tolerance can allow us to disentangle the critical balance between protective and damaging host immune responses.

Keywords: arbovirus; disease tolerance; innate immunity; interferons.

FIG 1

Differences in the replication kinetics of BTV in primary cells are dependent on the type I IFN response. (A) (Top) Virus replication curves of BTV-8 and BTV-2 in ovine (red) and bovine (blue) primary fibroblasts and EC. Data points for independent experiments are shown (for BTV-8, n = 10, and for BTV-2, n = 8, for ovine and bovine fibroblasts; for BTV-8, n = 4, and for BTV-2, n = 3, for ovine and bovine EC). (Bottom) The same replication curves in the presence of 4 μM Rux. Experiments in the presence and absence of Rux were carried out at the same time. (B) Virus replication curves of BTV-8 and BTV-2 in ovine (red) and bovine (blue) EC. Cells were mock treated (solid lines) or pretreated with 500 U of uIFN before infection. Independent experiments were repeated for BTV-8 (n = 3) and BTV-2 (n = 3) for ovine and bovine endothelial cells. For all experiments, cells were infected at an MOI of 0.01, and supernatants were harvested at the indicated time postinfection. Cell-free virus was titrated by endpoint dilution on BSR cells. Values are averages from independent experiments, as indicated, using at least three batches of primary cells independently generated. A one-way ANOVA with the Holm-Šídák post hoc correction test was carried out on the values for area under the curve (AUC) to assess statistical significance between conditions. ns, not significant (P > 0.05); *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 2

A bovine ISG screen reveals gene candidates with anti-BTV properties. (A) Schematic representation of the bovine ISG overexpression screen. 293T cells were transduced with lentiviruses expressing ISGs and RFP. Cells were subsequently infected with BTV-mGFP. At either 16 or 32 hpi, cells were fixed, and the number of infected cells was determined via flow cytometry. Plots representing the relative importance of different cell populations gated on their expression of GFP (virus-infected cells) or RFP (efficiently transduced cells) for an inhibitory or a noninhibitory ISG are schematically represented. (B) Scatterplot representing the percentage of BTV-mGFP replication for each gene normalized to the median. The screen was done at 16 and 32 hpi. Genes restricting BTV by more than 30% are considered inhibitory. The screen at 32 hpi was undertaken in the presence of 0.625 μg/mL puromycin. (C) Validation of antiviral bovine ISGs inhibiting BTV replication at 16 hpi. Data are means and standard errors from three independent experiments. (D) Validation of antiviral bovine ISG inhibiting BTV replication at 32 hpi. Data are means and standard errors from four independent experiments. Some of the schematic images were obtained from Biorender.com under a laboratory license.

FIG 3

BTV replication is restricted by both cow antiviral ISGs and their sheep orthologues. (A) (Left) Western blot of KO-IFIT1 cells (clones 3B and 7A) and parental BovFibT (WT) showing expression of IFIT1, phosphorylated STAT1, and GAPDH after IFN treatment. (Right) Quantification of Western blots using Image Studio Lite software (LI-COR Biosciences). Data were obtained from 3 independent experiments. (B) (Left) Western blot of KO-RSAD2 cells (clones 3C and 5B) and parental BovFibT (WT) showing expression of RSAD2, phosphorylated STAT1, and GAPDH after IFN treatment. (Right) Quantification of Western blots using Image Studio Lite software (LI-COR Biosciences). Data were obtained from 3 independent experiments. (C) Graph showing infectivity of BTV-8 mGFP in KO-IFIT1 and parental BovFibT in cells pretreated with IFN (1,000 U), or carrier control, before infection with serial dilutions of BTV-8-mGFP for 6 h. Cells were then fixed, and mGFP expression was determined by flow cytometry. (D) Graph showing infectivity of BTV-8 mGFP in KO-RSAD2 and parental BovFibT in cells pretreated with IFN (1,000 U), or carrier control, before infection with serial dilutions of BTV-8-mGFP for 6 h. Cells were then fixed, and mGFP expression was determined by flow cytometry. (E) Virus replication curves of growth of BTV-8 in immortalized BovFibT, KO-IFIT1 cells, and KO-RSAD2 cells in the presence (dashed lines) or absence (solid lines) of 1,000 U uIFN pretreatment. Cells were infected at an MOI of 0.01, and supernatants were harvested at the indicated time points postinfection. Cell-free virus was titrated by endpoint dilution on BSR cells, and values are means and SEM from at least 3 independent experiments. Statistical significance between −uIFN and +uIFN conditions for each time point was calculated using t tests with Welch’s correction. (F and G) Relative infectivity of BTV-8-mGFP in 293T cells overexpressing bovine and ovine restriction factors at 16 (F) and 32 (G) hpi (n = 8). 293T cells were transduced with lentiviruses expressing the ovine or bovine orthologues for 48 h and infected with BTV-8-mGFP for 16 or 32 h, and mGFP expression was quantified by flow cytometry. BTV infectivity was normalized to the mean obtained from all the negative-control wells. Each dot represents an independent repeat. The mean and standard deviation are presented for each condition. (H and I) Virus titers of BTV-8 in CPT-Tert stably expressing either ovine or bovine IFIT1 (H) and ovine or bovine RSAD2 (I) at the times indicated. Cells stably transfected with an empty lentivirus were used as controls. Cells were infected with BTV-8 (MOI ≈ 0.01), and supernatants were harvested at the indicated time points postinfection. Cell-free virus titers were quantified by endpoint dilution and normalized to titers obtained from the control cell line. Data (n =4) are from 2 independent BTV-8 stocks and 2 independently generated stable cell lines for each gene tested. Multiple t tests were carried out following a Shapiro-Wilk normality test. ns, not significant (P > 0.05); *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

FIG 4

Relative restriction of BTV replication in bovine cells occurs within the first 4 h postinfection. (A) Infectivity of BTV-8-mGFP in primary OvEC and BovEC. Cells were infected with BTV-8-mGFP at an MOI of ~1 (top) or 0.3 (bottom) and fixed at 2, 4, and 6 hpi. mGFP expression was determined by flow cytometry. Data were obtained from 4 independent experiments. (B) Mean fluorescent intensity (MFI) of the infected primary OvEC and BovEC populations at different times postinfection. Data are from 4 independent experiments at two different MOI. (C) Infectivity of BTV-8-mGFP in OvEChTert and BovEChTert cells. Experiments were carried out as for panel A with an MOI of ~1. (D) MFI of the infected OvEChTert and BovEChTert. Data for panels C and D were obtained using the immortalized BovEC from 5 independent experiments using two different clones, and data from immortalized OvEC are from 4 independent experiment using one clone. (E) Binding and entry assays of BTV-8 into OvEChTert and BovEChTert cells assessed by qRT-PCR for BTV segment 10. Cells were synchronously infected with BTV-8 at an MOI of ~10 at 4°C, and samples were taken from the inoculum, unbound virus, or the wash (left) or harvested at 0, 30, and 90 min postinfection (right). Samples were normalized to the inoculum (set at 100%). Data are from 6 independent experiments. ns, not significant (P > 0.05); **, P < 0.01; ****, P < 0.0001.

FIG 5

Virus-induced global host cell protein shutoff by BTV is faster and more pronounced in ovine than bovine endothelial cells. (A) Immunoblotting of phosphorylated STAT1, total STAT1, Mx1, IFIT1, and BTV NS1 and protein translation rates (puromycin labeling treatment for 1 h) in BovEC and OvEC infected with BTV-8. Cells were infected at 0.04 or 4 MOI equivalents (measured in the indicator cell line CPT-Tert) and harvested at the time points indicated. (B to F) Relative quantification of immunoblots as in panel A obtained from 3 independent experiments. (B) Quantification of relative protein shutoff over the course of BTV infection as determine by puromycin incorporation. All values were normalized to the mock (set to 1) for each time point. (C) Relative expression of BTV NS1 over time normalized to tubulin in OvEC and BovEC infected with BTV-8. (D) Expression of STAT1 (relative to mock infection) over time normalized to tubulin in OvEC and BovEC infected with BTV-8. (E) Relative expression of IFIT1 over time normalized to tubulin in OvEC and BovEC. Cells were infected with BTV-8 using 2 different MOI and harvested at 3 different time points. Data are from 3 independent experiments. (F) Relative expression of Mx1 over time normalized to tubulin in OvEC and BovEC. Cells were infected with BTV-8 using 2 different MOI and harvested at 3 different time points. Data are from 3 independent experiments. (G) Immunoblotting of phosphorylated STAT1, total STAT1, IFIT1, Mx1, and BTV NS1 and protein translation rates (puromycin labeling treatment for 1 h) in OvEC and BovEC pretreated with 4 μM Rux and infected with BTV-8. Cells were infected at an MOI of 0.04 or 4 and harvested at 24 hpi. (H to K) Relative quantification of immunoblots, as shown in panel G, obtained from 3 independent experiments. (H) Relative protein shutoff in OvEC and BovEC infected with BTV as determined by puromycin incorporation. All values were normalized to the mock-infected control. (I) Relative expression of BTV NS1 (relative to tubulin) in OvEC and BovEC infected with BTV-8. (J) Relative expression of IFIT1 (normalized to tubulin) in OvEC and BovEC. (K) Relative expression of Mx1 in OvEC and BovEC. For panels H to K, cells were pretreated with 4 μM Rux for 4 h prior to infection and maintained in the medium after infection. All data are from 3 independent experiments. Statistical significance was calculated using a two-way ANOVA performed using Tukey’s multiple-comparison test. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

FIG 6

Transcriptomic response of BTV-infected OvEC and BovEC. (A) Scatterplot showing baseline regulation of ISGs in mock-infected OvEC and BovEC at 6 and 12 hpi as identified by RNA-seq. Values shown are transcripts per million [log₁₀(TPM + 1)], and those identified in mock-infected OvEC are plotted against the log₁₀(TPM + 1) in mock-infected BovEC. Individual ISGs are plotted in green, and the median values of the ISGs are plotted in purple. (B) Box-whisker plots of IFN-β transcripts in response to BTV-8 and BTV-2 infection in OvEC and BovEC at 6 and 12 hpi. (C) Box-whisker plots of IFN-α transcripts in response to BTV-8 and BTV-2 infection in OvEC and BovEC at 6 and12 hpi. (D) Box-whisker plots of IFNAR transcripts in response to BTV-8 and BTV-2 infection in OvEC and BovEC at 6 and 12 hpi. (E) Scatterplot showing differential expression (log₂FC) of ISGs in OvEC and BovEC at 6 and 12 hpi in response to either BTV-2 or BTV-8 infection. Differentially expressed genes in both BovEC and OvEC are shown in gray, those only in OvEC are in red, and those only in BovEC are in blue. A violin plot of up- and downregulated genes for each species is represented on each axis.