Rotavirus Reprograms Multiple Interferon Receptors and Restricts Their Intestinal Antiviral and Inflammatory Functions

Abstract

Rotaviruses (RV) cause acute severe diarrhea in the absence of substantial intestinal inflammation. They are also highly infectious in their homologous host species. The replication capacity of RV in the small bowel is substantially due to its ability to inhibit different types of interferons (IFNs). Here, we found that during RV infection in vitro, both virus-infected and uninfected bystander cells resist STAT1 phosphorylation and interferon regulatory factor 7 (IRF7) induction in response to exogenous interferon (IFN). Functionally, cellular transcription in response to stimulation with IFN, but not intracellular double-stranded RNA (dsRNA), was inhibited by RV. Further, IFNAR1 stimulation during RV infection significantly repressed a set of virus-induced transcripts. Regulation of IFN signaling in vivo was studied in suckling mice using the highly infectious murine EW RV strain. Kinetic studies indicated that sustained EW RV replication and IFN induction in the small intestine are accompanied by significant decreases in IFN-stimulated transcripts. Lipopolysaccharide (LPS)-mediated intestinal damage, driven by STAT1-induced inflammation, was also prevented in EW RV-infected mice. Remarkably, by ectopically stimulating either IFNAR1 or IFNGR1 in EW RV-infected mice, we could eliminate several intestinal antiviral and inflammatory transcriptional responses to RV. In contrast to infection with homologous RV, infection with a STAT1-sensitive heterologous RV strain induced IFN-stimulated transcripts, inflammatory cytokines, and intestinal expression of STAT1-pY701. Finally, RV strain-specific STAT1 regulation also likely determines the intestinal activation of multiple caspases. The simian RRV strain, but not murine EW RV, uniquely triggers the cleavage of both extrinsic and intrinsic caspases (caspases 8, 9, and 3) in a STAT1-mediated manner. Collectively, our findings reveal efficient reprograming of multiple IFN receptors toward a negative-feedback mode of signaling, accompanied by suppression of IFN-mediated antiviral, apoptotic, and inflammatory functions, during natural RV intestinal infection.IMPORTANCE Rotavirus is a highly infectious pathogen that causes severe diarrhea. Replication of RV in the small intestine is restricted to homologous host species, and host range restriction is substantially determined by the interferon response. In this study, we demonstrate that during infection, RV bystander cells resist exogenous IFN-mediated STAT1 signaling and transcription. In a suckling mouse model, ectopically stimulating different intestinal interferon receptors during RV infection eliminates several innate and inflammatory antiviral responses. Different intestinal inflammatory cytokines were also suppressed by homologous RV, as was intestinal damage in response to endotoxin. The ability of RV to suppress IFN-mediated receptors likely impacts intestinal cell homeostasis, as the cleavage of multiple intestinal caspases during RV infection is mediated by the IFN-STAT1 signaling pathway. Together, our results provide a mechanism underlying both the remarkable interferon resistance of homologous RV and its ability to prevent substantial inflammatory damage to the small bowel.

Keywords: STAT1; caspase cleavage; endotoxin; gut inflammation; interferon receptor; interferons; negative feedback; rotavirus.

FIG 1

Rotavirus inhibits IFN-stimulated cellular responses in vitro. (A and B) HT-29 cells were infected with RRV (or mock infected) at an MOI of 1. At 10 hpi, cells were stimulated with purified IFN-β at the indicated concentrations for 2 h and analyzed by flow cytometry. (C and D) cells were infected with RRV and analyzed by FACS as for panels A and B. (E) Cells were infected in triplicate as for panels C and D and at 12 hpi were stimulated with PBS (black bars) or 500 IU/ml of purified IFN-β (blue bars) for 6 h before analysis by qRT-PCR for IRF7 and hypoxanthine phosphoribosyltransferase (HPRT) transcript levels. *, P < 0.02; ns, not significant. Data shown are representative of two or more independent experiments.

FIG 2

Rotavirus infection inhibits IFN-stimulated feed-forward cellular transcription. (A and B) HT-29 cells grown in triplicate wells were infected with the RRV strain at an MOI of 1 (or mock infected) for 6 h and then stimulated with 2 μg of >200-bp-long dsRNA–liposome complexes or 500 IU/ml of purified IFN-β. At 12 hpi, cells were lysed for RNA purification and then analyzed by qRT-PCR for expression of the indicated host (A) and RV (B) transcripts. ns, not significant; *, P < 0.02; **, P < 0.002; ***, P < 0.0002 (significance is shown relative to results for untreated samples in the mock- and RV-infected groups). (C) Cells were infected as for panel A and B, and the percentage of cells expressing RV VP6 antigen before lysis was determined by FACS. The boxes indicate the gates used to estimate RV-infected cells in the monolayer. Data shown are representative of two experiments.

FIG 3

Effects of homologous RV infection on intestinal IFN amplification responses. (A and B) Suckling mice were infected with murine EW RV or were mock infected (n = 3 to 8 mice per group) for the times indicated, and small intestines were collected to analyze the expression of the transcripts shown. Data shown are fold change in EW RV- versus mock-infected mice. (C) Suckling mice were mock or EW RV infected and 3 days later were parenterally administered purified endotoxin (15 μg/pup) for 6 h before harvesting small intestines for histological examination by hematoxylin-and-eosin staining. Data shown are representative of 2 pups per group.

FIG 4

Rotavirus reprograms intestinal IFNAR1 and IFNGR1 toward a negative-feedback mode of transcription. A schematic of the experimental approach used is shown at the top. Mice infected with EW RV (or mock infected) for 12 h were administered purified universal IFN-A/D (IFN-I), murine IFN-γ (IFN-II), or PBS by the intraperitoneal route. Twelve hours later, mice were sacrificed and small intestinal RNA analyzed by qRT-PCR for different host (A) or RV gene (B) transcripts. *, P < 0.02; **, P < 0.002; ***, P < 0.0002 (significance in each panel is relative to PBS-treated EW RV-infected mice, unless indicated otherwise). Data were measured from 3 to 5 pups per group.

FIG 5

Differences between effects of homologous and heterologous RV strains on IFN-stimulated genes, inflammatory cytokine expression, and STAT1-Y701 activation in vivo. (A) Suckling mice (n = 3 to 5 pups per group) were infected with the homologous murine EW or heterologous simian RRV RV strain (or mock infected). At 12 hpi, mice were sacrificed for subsequent analysis of small intestinal transcripts by qRT-PCR. Data shown are fold relative to mock-infected mice. (B) Mice infected as for panel A were sacrificed, and small intestinal biopsy specimens were analyzed for the expression of the proteins shown by either ELISA (IFN-β) or Luminex-based cytokine assays. (C) Mice infected as for panel A were sacrificed and small intestines analyzed for immunoblotting analysis. Densitometric quantitation of pSTAT1 bands is shown in the bar diagram at the right. *, P < 0.02; **, P < 0.002; ***, P < 0.0002.

FIG 6

Homologous rotavirus inhibits the STAT1-mediated cleavage of multiple intestinal caspases. (A and B) Either wild-type (WT) or STAT1-deficient mice were infected with the EW or RRV RV strain (or mock infected). Twenty-four hours later, mice were sacrificed, and small intestinal lysates were analyzed by immunoblotting (A). Densitometric estimation of bands is shown in panel B. (C) Mice infected as for panels A and B were sacrificed and intestines analyzed for the expression of cytokines by Luminex-based assays. *, P < 0.02; **, P < 0.002; ***, P < 0.0002.