Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2012 Jan;10(1):e1001249.
doi: 10.1371/journal.pbio.1001249. Epub 2012 Jan 24.

Stochastic Expression of the Interferon-β Gene

Affiliations
Free PMC article

Stochastic Expression of the Interferon-β Gene

Mingwei Zhao et al. PLoS Biol. .
Free PMC article

Abstract

Virus infection of mammalian cells induces the production of high levels of type I interferons (IFNα and β), cytokines that orchestrate antiviral innate and adaptive immunity. Previous studies have shown that only a fraction of the infected cells produce IFN. However, the mechanisms responsible for this stochastic expression are poorly understood. Here we report an in depth analysis of IFN-expressing and non-expressing mouse cells infected with Sendai virus. Mouse embryonic fibroblasts in which an internal ribosome entry site/yellow fluorescent protein gene was inserted downstream from the endogenous IFNβ gene were used to distinguish between the two cell types, and they were isolated from each other using fluorescence-activated cell sorting methods. Analysis of the separated cells revealed that stochastic IFNβ expression is a consequence of cell-to-cell variability in the levels and/or activities of limiting components at every level of the virus induction process, ranging from viral replication and expression, to the sensing of viral RNA by host factors, to activation of the signaling pathway, to the levels of activated transcription factors. We propose that this highly complex stochastic IFNβ gene expression evolved to optimize both the level and distribution of type I IFNs in response to virus infection.

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Stochastic IFN and virus-inducible gene expression.
(A) Stochastic IFNβ gene expression detected by ISH using a digoxygenin-labeled IFNβ RNA probe. (B) Percentage of IFNβ-producing cells at different times after SeV infection. (C) Mouse IFNα gene expression in primary MEFs detected by ISH using a digoxygenin-labeled IFNα4 probe. (D) qPCR analysis illustrating the expression levels of different virus-inducible genes in sorted IFNβ/YFP MEFs.
Figure 2
Figure 2. Viral transcription and/or replication are more efficient in IFNβ-producing cells.
(A) qPCR analysis illustrating the relative abundance of viral NP, matrix (M), and L polymerase protein (L) mRNA in sorted IFNβ/YFP MEFs. (B) Western blots showing cytoplasmic distribution of SeV NP protein present in IFNβ-producing and nonproducing cells. (C) qPCR analysis illustrating the relative abundance of SeV DI genome (upper panel), and semi-qRT-PCR analysis illustrating the relative abundance of SeV genomic RNA (lower panel) in sorted IFNβ/YFP MEFs. Reverse transcriptase PCR was carried out to detect viral genomic RNA and host cell β-actin mRNA (control) using gene-specific primers. After 35 cycles (SeV genomic RNA) or 26 cycles (β-actin) of amplification, PCR products were run on a 2% agarose gel. (D) Intracellular staining using SeV antibody and FACS analysis were carried out to determine the correlation between SeV infection and IFNβ expression in IFNβ/YFP homozygous MEFs. IB, immunoblot.
Figure 3
Figure 3. The RIG-I signaling pathway is activated in IFNβ-producing cells.
(A and B) Western blots showing cytoplasmic (C) versus nuclear (N) distribution of different factors present in FACS-sorted cells 8 h.p.i. (A) and 12 h.p.i (B). (C) Western blots showing cytoplasmic distribution of signaling pathway proteins present in FACS-sorted cells 8 h.p.i. Arrows indicate MAVS protein. (D) qPCR analysis illustrating the expression levels of RIG-I, Trim25, and MDA5 genes in sorted IFNβ/YFP homozygous MEF cells 8 h.p.i. IB, immunoblot.
Figure 4
Figure 4. Limiting factors in stochastic IFNβ gene expression.
(A) Different L929 stable transfectants were induced by tetracycline (Tet) for 24 h, followed by SeV infection for 9 h. RNA ISH experiments were carried out to detect the IFNβ mRNA. (B and D) Histograms showing the percentage (mean ± standard deviation) of cells expressing IFNβ from three independent ISH experiments. At least 400 cells were blindly counted and scored for each category. (C) L929 stable transfectant was transiently transfected with expression vectors encoding either GFP (control), RIG-I, or Trim25, then stimulated with tetracycline for 24 h. Cells were then infected with SeV for 6 h, followed by RNA ISH to detect IFNβ mRNA. pt, pt-REX-DEST30.
Figure 5
Figure 5. IRF7 is the significant limiting factor in stochastic type I IFN gene expression.
(A) Different L929 stable transfectants were induced by tetracycline (Tet) for 24 h, followed by SeV infection for 9 h. RNA ISH experiments were carried out to detect IFNβ mRNA. (B and D) Histograms showing the percentage (mean ± standard deviation) of cells expressing IFNβ from three independent ISH experiments. At least 400 cells were blindly counted and scored for each category. (C) RNA ISH experiments were carried out to detect IFNβ mRNA in wild-type (W.T.) or 4E-BP double-knockout (DKO) MEFs infected by SeV for 9 h.
Figure 6
Figure 6. Poly I∶C–induced stochastic IFNβ expression depends on the amounts of poly I∶C and MDA5.
(A) IFNβ/YFP homozygous MEF cells were electroporated with Cy5-labeled poly I∶C, and FACS analysis was carried out 8 h after the electroporation to assay the strength of Cy5 and YFP. The top left panel shows untransfected MEF cells, and the bottom left panel shows the electroporated MEF cells. As indicated by arrows, the two panels to the right represent the “poly I∶C high” and “poly I∶C low” populations, respectively. Data shown are representative of at least three independent experiments. Numbers represent relative percentages. (B) L929-MDA5 or L929-RIG-I stable transfectants were stimulated with tetracycline (Tet) for 24 h followed by transient transfection with poly I∶C. 6 h after transfection, cells were fixed, followed by RNA ISH to detect IFNβ mRNA. (C) Bar plots representing the percentage (mean ± standard deviation) of cells expressing IFNβ from three independent ISH experiments performed as in (B). At least 400 cells were blindly counted and scored for each category. (D) IFNβ/YFP primary MEFs were fixed 8 h after poly I∶C stimulation. Intracellular staining using MDA5 antibody and FACS analysis were carried out to assay the correlation between the expression levels of IFNβ and MDA5. Data shown are representative of at least three independent experiments. Numbers represent relative percentages. Iso-Ctrl, isotype control.
Figure 7
Figure 7. Endogenous variation in the concentrations of components of the RIG-I signaling pathway.
(A and B) Protein distributions in untreated primary MEFs determined by flow cytometry. a.u., arbitrary units. (C) Mouse ISG15 gene expression in MEFs 6 h after IFNβ treatment, detected by ISH using a digoxigenin-labeled ISG15 probe. (D) A model depicting stochastic IFN gene expression. There is a population of ten cells with varying numbers of limiting factors in each cell. Each small, colored circle represents one of the limiting factors, and six limiting factors are shown. Short black lines represent viral inducer. Only two cells in the population have enough of the viral inducer and all six factors to trigger transcription of the IFNβ gene. Iso-Ctrl, isotype control.

Similar articles

See all similar articles

Cited by 49 articles

See all "Cited by" articles

References

    1. Honda K, Takaoka A, Taniguchi T. Type I interferon [corrected] gene induction by the interferon regulatory factor family of transcription factors. Immunity. 2006;25:349–360. - PubMed
    1. Takeuchi O, Akira S. Innate immunity to virus infection. Immunol Rev. 2009;227:75–86. - PMC - PubMed
    1. Kato H, Takeuchi O, Sato S, Yoneyama M, Yamamoto M, et al. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature. 2006;441:101–105. - PubMed
    1. Gack M. U, Shin Y. C, Joo C. H, Urano T, Liang C, et al. TRIM25 RING-finger E3 ubiquitin ligase is essential for RIG-I-mediated antiviral activity. Nature. 2007;446:916–920. - PubMed
    1. Cui S, Eisenacher K, Kirchhofer A, Brzozka K, Lammens A, et al. The C-terminal regulatory domain is the RNA 5′-triphosphate sensor of RIG-I. Mol Cell. 2008;29:169–179. - PubMed

Publication types

MeSH terms

LinkOut - more resources

Feedback