The intracellular DNA sensor IFI16 gene acts as restriction factor for human cytomegalovirus replication

Abstract

Human interferon (IFN)-inducible IFI16 protein, an innate immune sensor of intracellular DNA, modulates various cell functions, however, its role in regulating virus growth remains unresolved. Here, we adopt two approaches to investigate whether IFI16 exerts pro- and/or anti-viral actions. First, the IFI16 gene was silenced using specific small interfering RNAs (siRNA) in human embryo lung fibroblasts (HELF) and replication of DNA and RNA viruses evaluated. IFI16-knockdown resulted in enhanced replication of Herpesviruses, in particular, Human Cytomegalovirus (HCMV). Consistent with this, HELF transduction with a dominant negative form of IFI16 lacking the PYRIN domain (PYD) enhanced the replication of HCMV. Second, HCMV replication was compared between HELFs overexpressing either the IFI16 gene or the LacZ gene. IFI16 overexpression decreased both virus yield and viral DNA copy number. Early and late, but not immediate-early, mRNAs and proteins were strongly down-regulated, thus IFI16 may exert its antiviral effect by impairing viral DNA synthesis. Constructs with the luciferase reporter gene driven by deleted or site-specific mutated forms of the HCMV DNA polymerase (UL54) promoter demonstrated that the inverted repeat element 1 (IR-1), located between -54 and -43 relative to the transcription start site, is the target of IFI16 suppression. Indeed, electrophoretic mobility shift assays and chromatin immunoprecipitation demonstrated that suppression of the UL54 promoter is mediated by IFI16-induced blocking of Sp1-like factors. Consistent with these results, deletion of the putative Sp1 responsive element from the HCMV UL44 promoter also relieved IFI16 suppression. Together, these data implicate IFI16 as a novel restriction factor against HCMV replication and provide new insight into the physiological functions of the IFN-inducible gene IFI16 as a viral restriction factor.

Figure 1. IFI16 protein is a negative regulator of Herpesvirus replication.

A) HELFs were electroporated with a mixture of four different small interfering RNAs (siRNA IFI16), scrambled control siRNA (siRNA ctrl), or left not electroporated (NE), and then infected with the indicated viruses at a multiplicity of 0.05 PFU/cell. Cell-free supernatants were harvested on the indicated hours post infection (hpi) and virus amounts determined by plaque assays. The data shown are the average of three experiments ± SD (*p<0.05, **p<0.01, ***p<0.001, one-way ANOVA followed by Bonferroni's post test). B) HELFs were transfected with siRNA IFI16 or siRNA ctrl or left not electroporated (NE) and IFI16 expression assayed by Western blotting on the indicated days (d) with anti-IFI16 polyclonal Abs. β-actin was included as a loading control. C) HELFs were treated with siRNA as described for panel A and then infected with HCMV at an MOI of 0.05 (left panel) or 1 (right panel) PFU/cell. Cell-free supernatants were harvested on the indicated hours post infection and the virus amounts determined by plaque assays. The data shown are the average of three experiments ± SD.

Figure 2. Effect of dominant negative IFI16 (dnIFI16) overexpression on Herpesvirus growth.

A) Western blot analysis was carried out to detect IFI16 and V5 expression in HELFs stably transduced with the recombinant Lentivirus carrying the full-length IFI16 (wtIFI16), mutated forms of IFI16 (dnIFI16), lacking the PYD domain (ΔPYDIFI16, indicated as ΔDIFI16), the HIN-B domain (ΔHIN-BIFI16, indicated as ΔBIFI16) or expressing the LacZ transgene as negative control (LacZ). β-actin immunodetection was used to control for equal loading. B) HELFs carrying wtIFI16, ΔDIFI16, ΔBIFI16 or the control LacZ gene were infected with HCMV at the indicated MOI. Viral supernatants were collected at 96 hours post infection (hpi) and analyzed by standard plaque assay. The data shown are the average of three experiments ± SD (*p<0.05, one-way ANOVA followed by Bonferroni's post test).

Figure 3. Overexpression of IFI16 reduces HCMV replication by inhibiting viral early and late gene expression.

A) Kinetics of Adenovirus-mediated IFI16 overexpression. HELFs were infected with AdV IFI16, AdV LacZ (MOI of 200 PFU/cell), or mock-infected. At the indicated hours post infection (h), total cell extracts were prepared and subjected to Western blot analysis using anti-IFI16 polyclonal Ab. β-actin served as the internal control. B) HELFs were infected with AdV IFI16, AdV LacZ (MOI of 200 PFU/cell), or left untreated. After 24 hours, cells were infected with HCMV at an MOI of 0.1 (left panel) or 1 (right panel) PFU/cell. Cell-free supernatants were harvested on the indicated days post infection and the amounts of HCMV were determined by standard plaque assay. The data shown are the average of three experiments ± SD.

Figure 4. IFI16 impairs early and late, but not immediate early gene expression.

A) HELFs were infected with AdV IFI16, AdV LacZ (MOI of 200 PFU/cell), or left untreated. After 24 hours, cells were infected with HCMV at an MOI of 1 PFU/cell. Total RNA was isolated at 24 hours post infection and assayed by quantitative real-time PCR to determine the relative levels of viral transcripts. Levels of viral mRNA are presented normalized to the levels of cellular β-actin. The data shown are the average of three experiments ± SD (*p<0.05, **p<0.01, one-way ANOVA followed by Bonferroni's post test). B) Cells were infected as described in the legend for panel A. Thirty µg of protein were analyzed by Western blotting for viral protein expression at 48 hours post infection and β-actin was used as internal control. C) Cells were infected as described in the legend for panel A. Viral DNA was isolated at the indicated hours post infection (hpi) and analyzed by quantitative real-time PCR, whereby primers amplified a segment of the E1 gene to determine the number of viral DNA genomes per nanogram of cellular reference DNA (18S rRNA gene). The data shown are the average of three experiments ± SD (*p<0.05, **p<0.01, ***p<0.001, one-way ANOVA followed by Bonferroni's post test).

Figure 5. Effects of IFI16 overexpression on the activity of HCMV UL54, UL44, and MIEP promoters.

A) Luciferase reporter plasmids containing HCMV UL54 promoter segments progressively deleted from the 5′ end and mutated in the IR-1 or DR-ATF elements were transfected into HELFs that were subsequently infected with either AdV LacZ or AdV IFI16 at an MOI of 200. 24 hours later, cells were infected with HCMV (MOI of 0.5) and luciferase activity assessed after a further 24 hpi. B) Luciferase reporter plasmids containing the MIEP (major immediate-early promoter) segment were transfected into HELFs subsequently infected as described for panel A. C) Luciferase reporter plasmids containing HCMV UL44 promoter segments progressively deleted from the 5′ end and the 3′ end were transfected into HELFs subsequently infected as described for panel A. Sp1, putative binding sites for Sp1 transcription factor; oval circles indicates TATA boxes. Experiments were repeated at least three times and one representative result is shown (mean ± SD) (*p<0.05, **p<0.01, ***p<0.001, unpaired t test for comparison of AdV LacZ vs AdV IFI16).

Figure 6. IFI16 impairs IR-1 binding.

A) Nuclear protein extracts from HELFs infected with AdV IFI16 or AdV LacZ at an MOI of 200 for 24 hours and then with HCMV at an MOI of 2 for 24 hours were incubated with a radiolabeled oligonucleotide containing the consensus IR-1 binding site. Competition experiments were performed with 100-fold excess of cold specific oligonucleotide in either the wild type or the mutated form. B) Super-shift experiments were performed by adding polyclonal antibodies against Sp1, control antobody (ctrl). Experiments were repeated at least three times and one representative result is shown.

Figure 7. Interplay between Sp1 and IFI16 down-regulates HCMV replication.

A) Nuclear cell protein extracts from HELFs mock infected (lanes 1, 4, 7), infected with HCMV at an MOI of 2 for 24 hours (lanes 2, 5, 8), or infected with AdV IFI16 or AdV LacZ (data not shown) at an MOI of 200 for 24 hours and then with HCMV (lanes 3, 6, 9) were immunoprecipitated with polyclonal antibodies against IFI16 or control antibody (ctrl). Samples were then immunoblotted with polyclonal antibodies against Sp1 or IFI16. B) Infected cell lysates were immunoprecipitated with anti-IFI16 polyclonal antibodies in the absence (lane 2 or 4) or presence of benzonase (lanes 3 and 5) or with control antibody (ctrl). Proteins from the IP were analyzed by Western blotting using polyclonal antibody recognizing Sp1. C) Nuclear extracts from HELFs mock infected or infected with AdV IFI16 or AdV LacZ at an MOI of 200 for 24 hours and then with HCMV at an MOI of 2 for 24 hours were analyzed by ChIP assay using anti-Sp1 rabbit polyclonal antibody. Sp1-coprecipitating DNA was analyzed by quantitative real-time PCR with IR-1 sequence specific primers. An unrelated rabbit polyclonal anti-serum was used as control (data not shown). Non-immunoprecipitated whole cell extract (input) obtained from AdV LacZ- or AdV IFI16-infected cells was employed to normalize the IR-1 viral DNA subjected to immunoprecipitation. Experiments were repeated at least three times and one representative result is shown (mean ± SD) (***p<0.001, unpaired t test). D) Nuclear cell protein extracts from HELFs stably transduced with the recombinant Lentivirus carrying the full-length IFI16 protein (IFI16wt), or dnIFI16 ORFs (ΔDIFI16 or ΔBIFI16), were infected with HCMV at an MOI of 2 for 24 hours, immunoprecipitated with monoclonal antibodies against flag V5 or control antibodies (ctrl), and immunoblotted with polyclonal antibodies against Sp1 (top panel) or flag V5 (bottom panel).

Figure 8. Suppression of HCMV replication by IFI16 does not require IFN-β antiviral activity.

HELFs were electroporated with a mixture of four different small interfering RNA (siRNA IFN-β) or scrambled control siRNA (siRNA ctrl), or left not electroporated (NE), and then infected with AdV IFI16 or AdV LacZ (MOI of 200 PFU/cell) before subsequent infection with HCMV 24 hours later (MOI of 0.1 PFU/cell). A) Total RNA was isolated 24 hours post HCMV infection and IFN-β mRNA expression was determined by quantitative real-time PCR. Levels of IFN-β mRNA are presented normalized to the levels of cellular β-actin. Experiments were repeated at least three times and one representative result is shown (mean ± SD). B) Cell-free supernatants were harvested at the indicated hours post infection (hpi) and virus amounts determined by plaque assay. The data shown are the average of three experiments ± SD (*p<0.05, **p<0.01, one-way ANOVA followed by Bonferroni's post test).