Transcriptional profiling of interferon regulatory factor 3 target genes: direct involvement in the regulation of interferon-stimulated genes

Abstract

Ubiquitously expressed interferon regulatory factor 3 (IRF-3) is directly activated after virus infection and functions as a key activator of the immediate-early alpha/beta interferon (IFN) genes, as well as the RANTES chemokine gene. In the present study, a tetracycline-inducible expression system expressing a constitutively active form of IRF-3 (IRF-3 5D) was combined with DNA microarray analysis to identify target genes regulated by IRF-3. Changes in mRNA expression profiles of 8,556 genes were monitored after Tet-inducible expression of IRF-3 5D. Among the genes upregulated by IRF-3 were transcripts for several known IFN-stimulated genes (ISGs). Subsequent analysis revealed that IRF-3 directly induced the expression of ISG56 in an IFN-independent manner through the IFN-stimulated responsive elements (ISREs) of the ISG56 promoter. These results demonstrate that, in addition to its role in the formation of a functional immediate-early IFN-beta enhanceosome, IRF-3 is able to discriminate among ISRE-containing genes involved in the establishment of the antiviral state as a direct response to virus infection.

FIG. 1.

IRF-3 5D-inducible expression of ISG56 gene. rtTA-Neo and rtTA-IRF-3 5D Jurkat cells were induced with DOX for 36 h in the presence of IFN-neutralizing antibodies. Poly(A) mRNA (2 μg/lane) was subjected to Northern blot analysis with ³²P-labeled probes. After hybridization with ISG56 probe, the membrane was stripped and rehybridized with a PLC-γ2 probe, followed by treatment with GAPDH probe for normalization. A second membrane was hybridized with arginase II probe, stripped, and rehybridized with a GAPDH probe.

FIG. 2.

IRF-3 5D-inducible expression of ISG56 gene. rtTA-Neo (lanes 1 to 5) and rtTA-IRF-3 5D (lanes 6 to 10) Jurkat cells were exposed to DOX (1 μg/ml) and anti-IFN antibodies for the indicated times. Whole-cell extracts (50 μg) were subjected to SDS-PAGE and analyzed by immunoblotting with anti-ISG56 antibodies. Membranes were stripped and reprobed with anti-IRF-3 and anti-actin antibodies.

FIG. 3.

IRF-3-induced expression of ISG56 is IFN independent. (A) IFN-unresponsive HEC1B cells were transfected with an empty vector (lane 1) or expression plasmid encoding IRF-3 5D (5 μg) (lane 2) and harvested 36 h posttransfection. (B) HEC1B cells were transfected either with an empty vector (lanes 3 and 4) or with the indicated amount of IRF 3ΔN expression plasmid (lanes 5 and 6). At 36 h posttransfection, cells were infected with Sendai virus (+) and further cultured for 10 h. In panels A and B, whole-cell extracts (50 μg) were analyzed by immunoblotting with anti-ISG56, anti-IRF-3, and anti-actin antibodies. In panel B, the relative levels of ISG56 expression are indicated.

FIG. 4.

Transactivation of the ISG56 promoter by IRF-3. HEC1B cells were cotransfected with the 561-luciferase reporter construct containing the ISG56 promoter and increasing amounts of IRF-3 5D (A), IRF-3 (B), and IRF-3ΔN (C) expression plasmids. At 8 h posttransfection, cells were left untreated or infected with Sendai virus as indicated. The relative luciferase activities were measured at 24 h posttransfection and are expressed as the fold activation relative to the basal level in the presence of the control vector. The results were normalized by using Renilla luciferase. Each value represents the mean ± the standard error of triplicate independent samples. The data are representative of at least two different experiments with similar results.

FIG. 5.

Involvement of ISRE sites in IRF-3 5D-dependent induction of the ISG56 promoter. A schematic representation of the ISG56 promoter mutants generated by overlap PCR is provided. HEC1B cells were cotransfected with the wild-type (WT) or mutated 561-luciferase reporter constructs and 200 ng of IRF-3 5D expression plasmid. The relative luciferase activities were measured at 24 h posttransfection. After normalization for the basal level in the presence of the reporter construct alone and Renilla luciferase, the luciferase activities were expressed as percentages of the activation observed with the wild-type 561-luciferase reporter. Each value represents the mean ± the standard error of triplicate independent samples. The data are representative of at least two different experiments with similar results.