P200 family protein IFI204 negatively regulates type I interferon responses by targeting IRF7 in nucleus

Interferon-inducible p200 family protein IFI204 was reported to be involved in DNA sensing, and subsequently induces the production of type I interferons and proinflammatory mediators. However, its function in the regulation of antiviral innate immune signaling pathway remains unclear. Here we reported a novel role of IFI204 that specifically inhibits the IRF7-mediated type I interferons response during viral infection. IFI204 and other p200 family proteins are highly expressed in mouse hepatitis coronavirus-infected bone marrow-derived dendritic cells. The abundant IFI204 could significantly interact with IRF7 in nucleus by its HIN domain and prevent the binding of IRF7 with its corresponding promoter. Moreover, other p200 family proteins that possess HIN domain could also inhibit the IRF7-mediated type I interferons. These results reveal that, besides the positive regulation function in type I interferon response at the early stage of DNA virus infection, the interferon-inducible p200 family proteins such as IFI204 could also negatively regulate the IRF7-mediated type I interferon response after RNA virus infection to avoid unnecessary host damage from hyper-inflammatory responses.

The ALRs belong to the interferon-inducible p200 family proteins (also known as PYHIN or HIN-200 proteins), which contains a pyrin (PYD) domain at the amino-terminus and at least one DNA binding HIN domain at carboxy-terminus [22,23]. Among the p200 family proteins, IFI16, AIM2 and MNDA are from humans, and IFI204, IFI205, IFI209, AIM2, MNDA and MNDAL are from mice. In addition to their functions in DNA sensing pathways [21,[24][25][26][27][28][29][30], the p200 family proteins can also act as modulators of many cellular functions, including cell proliferation, differentiation, apoptosis, senescence and DNA repair [23,31]. In addition, IFI16 has been identified as an ISG that triggers proinflammatory mediators and therefore plays an important role in autoimmune diseases [26,32]. However, Gray and colleagues have reported that ALRs including 13 PYHIN proteins are dispensable for interferon response to viral DNA in mice [33]. Therefore, the functions of p200 family proteins in antiviral responses need to be further clarified.
In this study, we demonstrated that the p200 family proteins, represented by IFI204, which is well known as an ALR and murine ortholog of IFI16 [28,34], were markedly induced in bone marrow-derived dendritic cells (BMDCs) after infection by mouse hepatitis coronavirus (MHV). Interestingly, we found that the IFI204 can specifically interact with IRF7 with its HIN domain and therefore hijack IRF7 in nucleus, consequently blocking the binding of IRF7 with its cognate transcriptional promoters. Moreover, other p200 family proteins that harbor a HIN domain could also bind with IRF7 and specifically down regulate the downstream transcription. Our results reveal a novel mechanism of negative regulation of type I IFNs response by p200 family proteins during RNA virus infection, which may prevent the host innate immune system from over-activation. This indicates that ALRs play two opposite roles during the infection of DNA and RNA viruses.

Higher expression of P200 family protein IFI204 is induced in BMDCs by MHV infection
To identify the function of various cell factors induced by RNA virus infection, we adopted a BMDC infection model by murine hepatitis virus (MHV), which is typical RNA virus with positive single stranded RNA. As shown in Fig 1A, bone marrow cells were harvested from 8-week-old mice (C57BL/6) and induced with GM-CSF for 7-8 days. BMDCs were infected by MHV and collected at 18 hours (hrs) post infection (hpi). The differential expression level of host factors was measured by analyzing the transcriptome and proteome before and after MHV infection. We found that over 1000 genes at mRNA level and 200 at protein level were changed, among which 100 cellular factors were up-regulated or down-regulated both at mRNA and protein levels ( Fig 1A). In this model, the transcription of Ifnb, Ifna1, Ifna4, Ifna6, Tnf and Il6 were significantly activated (S1 Fig), indicating that the IFN response and proinflammatory cytokines were apparently induced. Notably, the p200 family proteins, including IFI204, IFI205, IFI209 and MNDA were expressed at higher level after MHV infection ( Fig  1A). Interestingly, IFI204, which is previously reported to initiate type I IFNs in DNA virusinfected cells, obtained high scores as up-regulated factor both at mRNA and protein levels ( Fig 1B). Quantitative RT-PCR (qRT-PCR) analysis confirmed that the MHV infection induced a 5-fold upregulation of the mRNA of Ifi204 at 18 hpi ( Fig 1B). Moreover, the transcriptional regulation of Ifi204 is consistent with viral growth curve in MHV-infected BMDCs ( Fig 1C). However, Ifi204 was not stimulated in IFN-associated receptor-deficient (Ifnar -/-) BMDCs (Fig 1D), indicating that the upregulation is interferon-dependent. Furthermore, the similar phenomena were observed in wt BMDCs and Ifnar -/-BMDCs infected by a DNA virus (HSV-1) (S2 Fig). These data indicate that the expression of IFI204 is induced by type I IFNs during the infections of DNA or RNA viruses.

IFI204 inhibits RNA virus-triggered type I IFNs responses
To investigate the role of IFI204 in RNA virus-triggered signaling, we designed three siRNAs targeting Ifi204 (S3A and S3B Fig), the one of which being further packaged into lentivirusmediated small hairpin RNA (shRNA) vector to generate an NIH3T3 cell line and BMDCs with stable IFI204 knockdown (S3C and S3D Fig). The qRT-PCR analysis suggested that knockdown of IFI204 stimulated Sendai virus (SeV)-, poly(I:C)-or ssPolyU-induced transcription of Ifnb, Ifna4 and Ifna6 in NIH3T3 cells (Fig 2A), MHV-or SeV-induced transcription of Ifnb, Ifna4 and Ifna6 in BMDCs (Fig 2B), and vesicular stomatitis virus (VSV)-induced transcription of Ifnb, Ifna1 and Ifna4 in RAW264.7 cells (S4 Fig). In addition, the production of IFNα was also significantly enhanced by knockdown of IFI204 in NIH3T3 cells ( Fig 2C) and BMDCs (Fig 2D).
To confirm the similar function of IFI16, which is the human ortholog of IFI204, we generated IFI16 -/-A549 cell line by CRISPR/Cas9-mediated genome editing (S3E and S3F Fig). Consistent with the analysis in murine cells, the knockout of IFI16 stimulated SeV-, poly(I:C)-or ssPolyU-induced transcription of IFNβ and IFNα4 in A549 cells ( Fig 2E). These data indicate that negative regulation of RNA virus-triggered type I IFNs by ALRS is conserved in both mouse and human cells.
Furthermore, previous studies reported that IFI16 acts on STING and cooperates with cGAS-STING pathway to promote the type I IFNs during the DNA virus infection [28,29,35]. To further clarify the function of IFI204 during the infection of DNA virus, we infected NIH3T3 cells with HSV-1. The qRT-PCR analysis revealed that knockdown of IFI204 and double knockdown of IFI204-STING or IFI204-cGAS significantly inhibited HSV-1-induced transcription of Ifnb to different degrees in NIH3T3 cells (S5A Fig). Moreover, we also observed consistent phenomena in HSV-1-infected A549 cells with IFI16 -/-, IFI16 -/-STING knockdown or IFI16 -/-cGAS knockdown (S5B Fig), indicating that IFI204 might facilitate cGAS-STING DNA sensing pathway that leads to IRF3 activation during the infection of HSV-1. These data collectively suggest that IFI204 inhibits RNA virus-triggered type I IFNs, while promotes DNA virus-triggered type I IFNs.

IFI204 inhibits IRF7-mediated activation of type I IFNs responses
As is known, RNA viruses trigger IRF3-and IRF7-mediated activation of type I IFNs, while DNA viruses activate STING-IRF3 signaling pathway [24]. In our MHV-induced BMDCs model, a set of host factors in both IRF3 and IRF7 signaling pathway were stimulated (S6 Fig). Therefore, to find out which signaling pathway is inhibited by IFI204, IRF3 or IRF7 were cotransfected with IFI204 into Irf3 −/− Irf7 −/− mouse embryonic fibroblasts (MEFs), which were infected by HSV-1 or SeV as the model of DNA or RNA virus, respectively. As shown in Fig 3A, in IRF3-transfected Irf3 −/− Irf7 −/− MEFs, the transcription of both Ifnb and Ifna4 could be activated by either HSV-1 or SeV infection. However, IFI204 can only enhance the transcription of Ifnb induced by the infection of HSV-1 in the presence of IRF3 as previously reported. In contrast, in IRF7-transfected Irf3 −/− Irf7 −/− MEFs, the transcription of both Ifnb and Ifna4 could only be activated by SeV infection or by the self-activating IRF7 mutant IRF7-Δ247-467 [36] ( Fig 3A). Intriguingly, IFI204 significantly inhibited the transcription of both Ifnb and Ifna4 in SeV-infected IRF7-rescuing MEFs or IRF7-Δ247-467-rescuing MEFs (Fig 3A). In addition, the ELISA analysis showed that the production of IFNα was also remarkably inhibited by IFI204 in SeV-infected IRF7-transfected Irf3 −/− Irf7 −/− MEFs ( Fig 3B). Consistent with Fig 3A, the infection of HSV-1 could not effectively induce the expression of IFNα in IRF7-rescuing MEFs. However, IFI204 could still inhibit the expression of IFNα induced by the self-activating IRF7-Δ247-467 mutant in the context of HSV-1 infection (Fig 3B). To confirm the above observed results, we further evaluated the transcriptional activation of Ifnb and Ifna4 promoters through more sensitive luciferase assay in Irf3 −/− Irf7 −/− MEFs. As shown in Fig 3C, IFI204 inhibited the transcriptional activation of Ifnb and Ifna4 promoters in SeV-infected IRF7-rescuing and IRF7-Δ247-467-rescuing MEFs but not IRF3-rescuing MEFs. Furthermore, IFI204 could repress the transcriptional activation of Ifna4 promoters induced by overexpression of IRF7 with or without SeV-infection in HEK293T cells (Fig 3D), indicating that the inhibition takes place at the downstream of IRF7 activation. Together, these results indicate that IFI204 can significantly inhibit IRF7-mediated type I IFNs responses, and this inhibition event only occurs in RNA virus-infected cells, which could effectively induce the IRF7-mediated signaling pathway.

HIN domain of IFI204 interacts directly with IRF7
To explore the mechanism, we demonstrated that IFI204 could interact with IRF7 by exogenous and endogenous co-immunoprecipitation and immunoblot analysis (S7 Fig). Domain mapping analysis suggested that the first HIN domain (HINa) of IFI204 (aa 225-428) was responsible for the association (Fig 4A). Truncation mapping analysis showed that the N-terminus of HINa domain (aa 225-241) of IFI204 and the DBD domain (1-134aa) of IRF7 were the key interaction motifs (Fig 4B and S8 Fig). To test whether it is direct association between IFI204 and IRF7, GST-fused IFI204 and MBP-fused IRF7 were expressed and purified in vitro, and the GST pull-down analysis suggested that the association between IFI204 and IRF7 is direct interaction (Fig 4C). To confirm the interaction in vivo, the IFI204 and IRF7 were

IFI204 prevents IRF7 from binding with IRF7-specific promoter
Because the IFI204 and IRF7 were co-localized in the nucleus of RNA virus-infected cells, we hypothesized that IFI204 would hijack the imported IRF7, leading to suppression of the downstream transcription. The nucleo-cytoplasmic separation analysis revealed that knockdown of IFI204 could remarkably reduce the phosphorylation level of IRF7 and decrease the amount of nuclear IRF7 in SeV-, poly(I:C)-or ssPolyU-induced NIH3T3 cells (Fig 5A and 5B). Confocal fluorescence microscopy also confirmed this observation, indicating that IFI204 could interact with IRF7 and increase the amount of nuclear IRF7 ( Fig 5C). These results suggest that negative regulation of IRF7-mediated IFN responses by IFI204 was not due to the inhibition of the phosphorylation and nuclear import of IRF7, but due to blocking a downstream step, such as DNA-binding. Therefore, we adopted the CHIP and EMSA analysis to investigate the binding efficiency of IRF7 and its cognate promoter DNA. As shown in Fig 6, the CHIP analysis demonstrated that IFI204 remarkably inhibits the binding of IRF7 and its promoter DNA. In contrast, the IFI204 mutant (IFI204-Δ225-239) had significantly less inhibition on the binding of IRF7 with its specific promoter DNA ( Fig 6A). The EMSA analysis showed that the IRF7 could bind to PRDI region on the promoter of IFN-β gene as previously reported [36], while IFI204 could not bind with the IRF7-specific promoter ( Fig 6B). This result excludes the possibility that IFI204 directly binds to PRDI promoter to inhibit its transcription. As shown in Fig 6C, the binding of IRF7 and PRDI was remarkably attenuated by the IFI204 in a dose-dependent manner ( Fig 6C). Together, the results indicate that the interaction of IFI204 and IRF7 prevents IRF7 from binding to IRF7-specific promoter DNA.

P200 family proteins possess conserved function in the inhibition of IRF7-mediated type I IFNs responses
The HIN domain is highly conserved among p200 family proteins ( Fig 7A). Therefore, we investigated whether other proteins of the p200 family possess the similar function as that of  IFI204. As shown in Fig 7B and 7C, the representative p200 family proteins, including IFI205, IFI209, IFI202 and MNDAL from mice and MNDA, IFI16, IFI16β and AIM2 from human, all could interact with IRF7 ( Fig 7B) and showed a significant repression of type I IFNs response . Coimmunoprecipitation (IP, with anti-Flag) and immunoblot (IB, with anti-HA) analysis of HEK293T cells that were transfected with plasmids encoding Flag-IRF7 and HA-IFI204 or its mutants. Cell lysate was analyzed by immunoblot with anti-Flag and anti-HA (lower panels in A and B). (C) GST pull-down assay. Purified GST-IFI204 or GST protein was incubated with MBP-IRF7 or MBP as indicated and pulled down with glutathione resin beads. Each sample was detected by Western blotting with the indicated antibodies. GST and MBP proteins were used as a negative control, respectively. (D) Intracellular localization of IFI204 and IRF7 assessed by immunostaining followed by laser scanning confocal microscopy. Irf3 −/− Irf7 −/− MEFs cells were co-transfected with pFlag-IRF7 and pHA-IFI204 stimulated with SeV for 0 h (mock) and 6 hrs (SeV) at 24 hrs post-transfection. Then cells were washed with PBS and harvested for immunofluorescence analysis. Mouse anti-HA antibody was used as primary antibody and CY3-conjugated goat anti-mouse IgG was used to detect IFI204 as shown in red fluorescence. Rabbit anti-flag antibody was used as primary antibody and FITC-conjugated goat anti-rabbit IgG was used to detect IRF7 as shown in green fluorescence. DAPI (4, 6-diamidino-2-phenylindole) is used for nuclei staining as shown in blue fluorescence. Data are representative of three independent experiments. in SeV-induced IRF7-transfected Irf3 −/− Irf7 −/− MEFs (Fig 7C). To confirm this result, IFI205 was overexpressed in BMDCs and the expression level of IFNs were repressed (S10A Fig). In contrast, knockdown of IFI205 stimulated SeV-, or poly(I:C)-induced expressions of IFNs in NIH3T3 cells (S10B Fig). Taken together, p200 family proteins, which contain the conserved HIN domain might play a negative regulation role in IRF7-mediated type I IFN signaling pathway. During RNA virus infection, the activated IRF7 would be bound with HIN domain of Immunoblot analysis of phosphorylation of IRF7 with rabbit mAb against phospho-IRF7-Ser437/438 (CST D6M2I) and β-actin in NIH3T3 cells infected by lentivirus-mediated shIFI204 or shNC as negative control. Cells were uninfected (UI) or infected with SeV or transfected with poly(I:C) or ssPolyU as indicated for 12 hrs. (B) Nucleo-cytoplasmic separation analysis of IRF7 in NIH3T3 cells infected by lentivirusmediated shIFI204 or shNC as negative control. Cells were uninfected (UI) or infected with SeV or transfected with poly(I: C) or ssPolyU as indicated for 12 or 16 hrs. Nuclear lysates were tested by western blotting using anti-IRF7 and anti-lamin b1 antibodies. Levels of IRF7 was quantified by densitometry and normalized to lamin b1 protein levels. (C) Intracellular localization of endogenous IRF7 assessed by immunostaining followed by laser scanning confocal microscopy (left panel). NIH3T3 cells infected by lentivirus-mediated shIFI204 (NIH3T3-shIFI204) or shNC (NIH3T3-WT) were infected with SeV for 12 hrs. Then cells were washed with PBS and harvested for immunofluorescence analysis. Rabbit anti-IRF7 antibody was used as primary antibody and FITC-conjugated goat anti-rabbit IgG was used to detect IRF7 as shown in green fluorescence. DAPI (4, 6-diamidino-2-phenylindole) is used for nuclei staining as shown in blue fluorescence. The white arrow indicates the different protein levels of IRF7 in perinuclear region. Relative level of IRF7 in the nucleus was quantified by densitometry (right panel). � P < 0.05, �� P < 0.01 and ��� P < 0.001. Data are representative of two independent experiments (mean ± SD in A-C).

Discussion
IFI204/IFI16, known as ALRs and expressed constitutively in host cells, are involved in the sensing of intracellular DNA derived from pathogens and subsequent induction of the production of type I IFNs as well as proinflammatory mediators via STING-IRF3 pathway [19,24]. However, Gray and colleagues demonstrated that ALRs, including IFI204, are dispensable for the interferon response to intracellular DNA in mice with knockout of all PYHIN proteins [33]. Two later studies reported that IFI16 cooperates with cGAS-STING pathway to promote the type I IFNs during the DNA virus infection [28,29]. In this study, we showed that IFI204 could promote the expression of IFNβ via IRF3-mediated signaling pathway (S5 Fig and Fig 3) but down regulate the IRF7-mediated IFN response. Why inconsistent results were obtained from different studies is still one unanswered question. We propose that PYHIN proteins may not be essential in DNA sensing (as shown by Gray et al.) but involved in the regulation of interferon responses (as shown in this study and by different laboratories). The regulatory function of p200 family members may involve more than one PYHIN protein, and when all PYHIN proteins are knocked out, such regulation could not be observed and, instead, the regulation of interferon response is compensated by other regulation system. Another possible explanation for the discrepancy may be due to the use of different inducers (or viruses) and cell types. More studies are needed to clarify these issues in the future but they are already beyond the range of current work, which focuses to reveal the negative regulatory function of p200 family proteins in IRF7-mediated type I IFN responses (Figs 2, 3 and 7). P200 family proteins are IFN-inducible proteins, belonging to the category of ISGs [37]. Many ISGs were reported to positively regulate the IFN responses. However, only a few ISGs are known to negatively regulate type I IFN responses and proinflammatory mediators. Here, we demonstrated that the IFI204 and several other p200 family proteins were highly induced in both DNA and RNA virus infection (Fig 1 and S2 Fig). Intriguingly, IFI204 specifically inhibits the IRF7-mediated type I IFNs response in RNA virus infection (Figs 2, 3 and 7). The infection of DNA virus HSV-1 could not effectively stimulate the IRF7-mediated signaling pathway as shown in Fig 3 and in previous studies [19,24]. However, in the absence of IRF3, IFI204 could still inhibit type I IFN responses initiated by the self-activating IRF7-Δ247-467 mutant in the context of HSV-1 infection (Fig 3B and 3A). Such results suggest a specific function of IFI204 in suppression of IRF7-mediated type I IFNs responses in whatever way IRF7 is activated.
Negative regulation of IRF7 could take place at the stage of IRF phosphorylation, nuclear translocation and binding with the promoter. Our results showed that IFI204 did not inhibit the phosphorylation and nuclear import of IRF7 but increase the phosphorylation level and nuclear retention of IRF7 to some degree (Fig 5), suggesting that inhibitory function of IFI204

Fig 8. Proposed molecular mechanism for the function of p200 family proteins (IFI204) to inhibit IRF7-mediated activation of type I IFNs responses.
Schematic diagram demonstrates the mechanism of p200 family proteins to suppress type I IFNs responses. Late after RNA virus infection, the highly expressed p200 family proteins, such as IFI204, interact with IRF7 in the nucleus and such interactions lead to the negative regulation of type I IFNs responses. https://doi.org/10.1371/journal.ppat.1008079.g008

IFI204 inhibits IRF7-mediated Type I IFNs responses
is not due to the suppression of IRF7 activation. By CHIP and EMSA assays, we revealed that IFI204 could specifically interact with the DNA binding domain (DBD) of IRF7 and consequently inhibit its subsequent binding with promoter DNA (Fig 6). IFI204 itself does not have specific binding with IRF7 promoter (Fig 6), indicating that IFI204 does not directly compete with IRF7 in the DNA binding but prevent IRF7 from promoter binding via its interaction with IRF7 DBD.
The type I IFN response is critical for viral clearance. However, the whole signaling pathway should be tightly regulated to avoid overproduction of IFNs and inflammatory factors that can cause tissue damage and even autoimmune diseases [38]. For example, the acute coronaviral infection such as Middle East respiratory syndrome coronavirus (MERS-CoV) and severe acute respiratory syndrome coronavirus (SARS-CoV), and overdosed IFNα in antiviral treatment can cause tissue damage due to the hyper-inflammatory responses [39]. Here we use MHV that belongs to the same genus betacoronavirus as SARS-CoV and MERS-CoV to mimic the acute RNA virus infection. In our MHV-infected BMDC model, both IRF3-and IRF7-mediated pathways of type I IFNs and some subsequent proinflammatory mediators were stimulated (Figs 1 and 3, S1 and S6 Figs). Notably, IRF7-mediated pathway of type I IFN responses is mainly initiated by ssRNA sensor TLR7, and previous studies reported that the over-expression or activation of TLR7 is associated with autoimmune diseases [40][41][42]. The over-expression and activation of TLR7 as well as the stimulation of IFNα were also observed in our MHV-infection BMDCs model (Figs 2 and 3, S6 Fig), which is consistent with clinical symptoms. Our current study revealed a novel negative regulation mechanism in the RNA virusinduced IRF7-mediated IFN responses.
The regulation of type I IFN signaling pathway is dynamic sequential processes [43]. Taken together the results of this study and previous knowledge, we propose a model for the function of IFI204 in negative regulation of IFN responses as shown in Fig 8. Many host negative regulators of IFN responses are constitutively expressed in cells [43]. In contrast, p200 family protein IFI204/IFI16 is up-regulated during virus infection (Fig 1). We may assume that induced expression of IFI204/IFI16 might exert more robust regulation of IFN responses.
In summary, our current work revealed a previously unknown negative regulation mechanism of IRF7-mediated IFN responses by PYHIN proteins including IFI204/IFI16. Such knowledge may help to better understand the accurate regulation system of IFN signaling pathway during viral infection.

Ethics statement
All mice were housed in the specific pathogen-free animal facility at Wuhan University and all animal experiments were in accordance with protocols that were adhered to the Chinese National Laboratory Animal-Guideline for Ethical Review of Animal Welfare and approved by the Animal Care and Use Committee of Wuhan University (NO.15060C). The mice were euthanatized with CO 2 followed by various studies.

Plasmids
The complementary DNA (cDNA) of Ifi204, Ifi205, Ifi209, Mndal and Irf7 were amplified from total RNA obtained from MHV-stimulated BMDCs, respectively. IFI16, MNDA and AIM2 complementary DNA (cDNA) were amplified from cDNA library obtained from Dr. Jia-Huai Han (Xiamen University). Irf7 was cloned into pRK and pMAL-CX2 by recombinase-mediated recombination. To generate the self-activating IRF7 mutant IRF7-Δ247-467 plasmid (pRK-Flag-IRF7Δ247-467), the plasmid pRK-Flag-IRF7 was amplified by PCR to delete the nucleotides 739 to 1401 of IRF7. The PCR product was digested with Dpn1 (Thermo Fisher Scientific) to remove template plasmids and then ligated by recombinase-mediated recombination. The other genes and the mutants of Ifi204 or Irf7 were cloned into pCAGGS or pGEX-4T-1, respectively as mentioned. All the plasmids were confirmed by DNA sequencing. The PCR primers are listed in S1 Table.

Preparation of BMDCs
The BMDCs were prepared as previously described [46]. Bone marrow cells were isolated from mouse tibia and femur. The cells were cultured for 7-9 days in medium containing mouse cytokine GM-CSF (20 ng/ml) to induce BMDCs. The positive rate is 85% by the detection of flow cytometry.

siRNA and gRNA
The sequences of siRNA and gRNA are as following.

Lentiviral packaging and infection
The experiments were performed as previously described [47]. HEK293T cells (in 100-mm dish) were co-transfected with 2 μg LentiV2, 1.5 μg psPAX2 and 1.5 μg pMD2.G. Cell supernatants were collected at 48 hrs after transfection and the lentivirus in cell supernatants was conserved at -80˚C after high-speed centrifugation and filtration. Supernatants were used to infect NIH3T3 cells and A549 cells. Cells were infected twice to obtain higher transduction efficiency, and puromycin (NIH3T3 3 μg/ml, A549 1.5 μg/ml) was used to screen the positive cells.

RNA isolation and quantitative RT-PCR
Total RNA was isolated from cells with TRIzol reagent under the instruction of the manufacturer. The mRNAs were reverse transcribed into cDNA by PrimeScript RT reagent Kit (Takara). The cDNA was amplified by a fast two-step amplification program using SYBR Green Fast qPCR MasterMix (Cat No. 11201-11203 � ; Yeasen, China). GAPDH was used to normalize the input samples via the ΔΔCt method. The RT-PCR primers are listed in S2 Table. ELISA Cytokines were detected by using mouse IFNα ELISA kit (R&D) under the instruction of manufacturer.

Purification of recombinant IFI204 and IRF7
GST

Nuclear/cytoplasmic fractionation
Nuclear and cytoplasmic extracts were prepared with nuclear cytoplasmic extraction kit (Applygen Technologies) according to the manufacturer's manual.

EMSA
The purified target proteins were mixed in binding buffer (400 mM Tris-HCl [pH 8.0], 50 mM DTT) and incubated on ice for 30 mins. The CY5-labeled double-stranded oligonucleotides of the IFNβ promoter (PRDI region) (5'-GAGAAGTGAAAGTGGAGAAGTGAAAGTGGAGA AGTGAAAGTGGAGAAGTGAAAGTG-3') and negative control sequence (NC-sequence) (5'-CATGGGTGGCCACAACTTCGCCTTTTCGTCAGTAATACACGATTGGAGAAAGT TAC-3') were add to the reaction mix and incubated at room temperature for 30 mins. The samples were separated in 1% agarose gel at 130V for 1 h and subjected to the analysis with Typhoon FLA9500 (GE).

Chromatin immunoprecipitation (ChIP)
Chromatin immunoprecipitation was carried out using a ChIP Assay kit (Beyotime, catalog number P2078) according to manufacturer's instructions. Briefly, IFI204 or IFI204 mutant were co-transfected with IRF7 into Irf3 −/− Irf7 −/− MEFs, which were stimulated by SeV. 8 hrs post infection, cells were fixed using 1% formaldehyde for 10 min at 37˚C followed by neutralization using 125 mm glycine. The cells were then lysed on ice in SDS lysis buffer (50 mm Tris-HCl, pH8.0, 10 mm EDTA, 150 mm NaCl, 1% SDS) supplemented with 1× proteinase inhibitor mixture (Roche Applied Science). Then the lysate was sonicated on ice to break the genome into 200-1000 bp in size. We used ANTI-FLAG M2 Affinity Gel (Product No. A2220) to allow precipitation of FLAG-IRF7-associated DNA fragments by incubation at 4˚C overnight followed by washing with different washing buffers as follows: low salt immune complex wash buffer, one time; high salt immune complex wash buffer, one time; LiCl immune complex wash buffer, one time; and Tris-EDTA buffer, two times. The DNA components in the precipitation were finally extracted using a standard phenol-chloroform method. PCR amplifications were conducted using the IFNβ promoter-specific primers as belows: IFNβ promoter-forward: 5 0 -AGCCAGGAGCTTGAATAAAA -3 0 IFNβ promoter-reverse: 5 0 -GATCTTCCTTCATGGCCTGT-3 0 GAPDH-forward: 5 0 -CGACTTCAACAGCAACTCCCACTCTTCC -3 0 GAPDH-reverse: 5 0 -TGGGTGGTCCAGGGTTTCTTACTCCTT -3 0
Proteome data processing: The samples were combined and analyzed in triplicates by EASY-nLC 1000 system (Thermo Scientific, Odense, Denmark) and Q Exactive HF mass spectrometer system (Thermo Scientific, San Jose, CA, USA). The preliminary mass spectrum data were analyzed by MaxQuant software suite (v1. 5